2013 - Student-Faculty Programs - Caltech

4 downloads 822 Views 2MB Size Report
Regulation of Binding of SynGAP to the Postsynaptic Scaffold PSD-95. Rebecca Hu ...... the site administrator to skirt t
 

2013 Abstract Book

STUDENT-FACULTY PROGRAMS 2013 Abstract Book This document contains the abstracts of the research projects conducted by students in all programs coordinated by Caltech’s Student-Faculty Programs Office for the summer of 2013. Table of Contents Summer Undergraduate Research Fellowships (SURF)

1

MURF Undergraduate Research Fellowships

107

Amgen Scholars Program

115

Laser Interferometer Gravitational-Wave Observatory (LIGO)

125

NASA/JPL Programs  Planetary Geology and Geophysics Undergraduate Research Program (PGGURP)  The National Space Grant College and Fellowship Program (Space Grant)  Undergraduate Student Research Program (USRP)  JPL Student Internship Program (JPLSIP)

137

SURF SUMMER UNDERGRADUATE RESEARCH FELLOWSHIPS

S U R F

  Fabrication of Counter Electrodes for Microprobe Impedance Measurement Iwnetim I. Abate Mentors: Sossina M. Haile and Rob Usiskin A major obstacle to the study of fundamental properties of candidate materials for solid oxide fuel cell (SOFC) cathodes is the morphological complexity of the electrode-electrolyte interface. This complexity prevents a true determination of the catalytic mechanisms. Using well-defined electrode geometries, it is possible to quantify the relative density of two-phase boundary sites to three-phase boundary sites, and so by varying the pattern used to generate the electrode geometry we can determine the primary pathway. Toward this goal, we first made porous composite cathodes of Ba0.5Sr0.5Co0.8Fe0.2O3-δ (BSCF) +Ag or SrCo0.9Nb0.1O3-δ (SCN) +Ag for impedance measurements of well-defined microelectrode SOFCs. Porosity, friability, adhesion to the substrate’s surface, thermal stability and electrochemical properties of the porous films were investigated by optical microscopy and AC impedance spectroscopy (ACIS).Next, we studied whether a chemical reaction occurs between the highperformance cathode material, SCN, and the most conventional electrolyte material, Y0.16Zr0.84O1.92 (YSZ). The conditions that favor the reaction are also determined using X-ray diffraction (XRD). Diagnosis of Cardiovascular Diseases Using Noninvasive Pressure Waveform Amir Abdolrahim Poorheravi Mentors: Morteza Gharib and Neima Pahlevan Study of arterial hemodynamics and waveform dynamics can lead to development of new means which help in diagnosis and prognosis of cardiovascular diseases and complications such as congestive heart failure or arterial stiffness. In the course of a heartbeat, the left ventricle and aorta are initially coupled so the pressure waveform produced in that time interval is affected by characteristics of both of them. After the aortic valve is closed, the aorta becomes decoupled from the left ventricle and the waveform is highly independent of the heart. Study of the waveform in those two intervals yields two Intrinsic Frequencies (ω1, ω2) which are affected by parameters such as arterial compliance and left ventricular health and have been shown to predict some cardiovascular conditions if extracted from aortic pressure measurements. In this study, we investigated how these values change when the distance from the heart increases and for measurements which are done from carotid, radial or femoral locations and formulated these relations. Our results show that ω1 remains constant as the distance changes but ω2 increases as the distance gets larger. We used three sources of data collected from human volunteers, retrospective patient data, and retrospective data from canine subjects. Also, we investigated how ω1, ω2, and ∆ω (=ω1-ω2) are correlated to other established medical indices, extracted the intrinsic frequencies from photoplethysmography measurements (pulse oximeters) and compared them to values extracted from arterial pressure waveforms. To conclude, we came up with a novel medical index for cardiovascular health. High-Throughput Screening of Semiconductors for Solar Fuel Generation by First Principles Calculations Alexander Aduenko Mentors: William A. Goddard, III, and Jose L. Mendoza-Cortes The project is focused in finding semiconductors which have an ideal band gap (Eg) in the visible light region as well as determine their stability in aqueous media for solar fuel generation, i.e. Eg around 1.1 eV for H2 generation and Eg of 1.7 to 2.1 eV for O2 generation in a tandem cell. More than one thousand compounds which can have semiconductor properties were automatically screened. The structures were obtained from the Inorganic Structural database (ICSD). Reduction and oxidation potentials for these compounds were automatically calculated. We created this database of oxidation and reduction reactions using available experimental data for free energies. An approach based on ionic substitution and probability functions for selected semiconductors was used to predict new partially substituted semiconductors for Solar Fuel Generation. We propose a way to determine the properties of the predicted semiconductors which allows us to extrapolate existing data and their correlation. The next goal is to calculate the lifetime of an exciton. Simulation of Thin Shells Aman Agarwal Mentor: Mathieu Desbrun In graphics, the realistic simulation of thin shells such as leaves, fingernails, hats and cans is an important topic. Thin shells are thin flexible structures with a high ratio of width to thickness (>100) that are curved in their relaxed, undeformed state. However, most current models for thin shells assume isotropic behavior. In this project, we implement the Discrete Shells model in SOFA (Simulation Open Framework Architecture). We then further adapt this approach to capture the behavior of orthotropic materials, i.e., materials that have different properties in orthotropic directions, using a recent model called Cubic Shells. To test the effectiveness of both models, we generate demos of thin surfaces involving bending and collisions in SOFA.

1

  Numerical Investigation of Coining Micro-Forming Process via Computational Modeling in Abaqus/Cae Shashank Agarwal Mentors: Guruswami Ravichandran, Aaron Stebner, and Dipankar Ghosh Increasing demands for miniature metallic parts have driven the application of micro forming in various industries. Only a limited amount of research is, however, available on the mechanics of forming miniature features in high strength materials. This study investigates the forming of sinusoidal micro-feature in 304 Stainless Steel, α-Ti and Tantalum by using the coining process. Simulation work has been performed in Abaqus to study the effects of variation in work-piece thickness & material, preform dimension and the interactional properties like friction on the formation of feature size ranging from 2 µm to 50 µm. The characteristic load requirement variation, gradual sample shape variation though the process and the effect of the process on the final coined sample properties have also been studied. Finally, the simulations are validated through comparison with empirical micro-coining data provided by General Atomics, a company that micro-coins these metals for use in laser-fusion energy experiments. Exploring the Viability of Utilizing Synthetic Aperture Radar to Monitor Wildfires Sandia Akhtar Mentors: Mark Simons and Brent Minchew In the common instances when wildfires rage in remote areas such as forests, fire fighting personnel often find it difficult to pinpoint the exact location of the fire. This results in several firefighter deaths per year when they are overtaken by fire. Frequently, optical sensors are not a feasible option in monitoring these wildfires because smoke from the wildfire can obscure the camera. We hypothesize that polarimetric synthetic aperture radar could be successful in mapping wildfires, potentially in real-time. In order to determine the potential of this approach, data for multiple wildfires will be analyzed with a variety of techniques that decompose the measured radar backscatter into the sum of three primary radar scattering sources: ground, vegetation and double-bounce between the ground and tree trunks. The results are being tested to ascertain under which pre-fire vegetation and topography conditions this instrument works. We consider areas with the largest change in the vegetation component of the backscatter to indicate the burned region. This research will not only test whether this concept works, but will also attempt to discern which decomposition method is the most efficient weighing the trade-off between accuracy and computation cost. We have successfully mapped fires in regions with sparse to moderate pre-fire vegetation and high topographic relief, which is very promising for the flexibility of conditions where this instrument can be used. Future research will hopefully have access to data acquired during the occurrence of a fire to assess the degree to which SAR can outline the wildfire boundary during an active fire. Heat Flow Devices on Mars, the Moon, and Venus Narek Akopyan Mentor: Suzanne Smrekar By understanding heat flow on the terrestrial planets, knowledge can be gained concerning the formation of our solar system. In order to calculate heat flow, both the temperature gradient and the thermal conductivity must be known. Measurements on Mars and the moon use a heat flow approach involving reaching into the subsurface using a self-hammering device called the mole attached to a tether, which measures annual temperature variations in the subsurface for one year at a depth below the annual wave, five meters. Tethers, which are responsible for the thermal gradient calculation in the soil, were placed inside the geothermal testbed and temperature was recorded for the RTDs on the three different tether designs. Observing the data, the Short Bridge Tether data seemed flawed and between the other two, the Thermocouple Tether did a better job in calculating the temperature gradient. On the other hand, for Venus, a cylindrical heat flux plate will be deployed for only a couple hours due to Venus’s violent atmosphere, which will reach a steady state temperature difference when the flux across the plate is equal to the planet’s interior flux. The device will be tested in a bell jar reaching very high temperatures, and a DAQ unit must be configured to record the various temperature and voltage sensors continuously. The Neuroscience of Social Influence and Attitude Change Shyam Akula Mentors: Ralph Adolphs and Keise Izuma Cognitive dissonance and cognitive imbalance are two central ideas of social psychology that explain attitude and preference change. The biological underpinnings of these two phenomena have been previously explored using fMRI studies, which implicated the involvement of the posterior medial frontal cortex (pMFC) in producing these effects. This study used a repetitive trans-cranial magnetic stimulation (rTMS) protocol to the pMFC to temporarily inactivate the region to prove the causality of pMFC activity to both cognitive dissonance and cognitive imbalance. Although more data is pending, subjects who received TMS to the pMFC show a trend of less cognitive dissonance compared to controls who either received a) sham TMS to the pMFC or b) real TMS to another region for control (posterior parietal cortex). Thus far, this finding has not been extended to cognitive imbalance, likely due to sample size. As the summer ends, we will continue to collect data with the TMS protocol and solely behavioral data to make robust conclusions with a larger sample size in each of the three TMS conditions.

2

 

3-D Microscopy of Zebrafish Using Super Resolution Imaging Monique I. Alkiewicz Mentors: Long Cai and Antti Lignell The main objective of our experiment is to identify the exact location of individual mRNA by using super resolution imaging on zebrafish. This is achieved by selecting one specific gene of choice, designing oligonucleotide probes for this gene, coupling the probes with a specific dye, hybridizing the fish and imaging them using a SPIM (selective plane illumination microscopy) microscope. We have observed the proper gene expression of STAT3 in the zebrafish and plan on locating the individual mRNA locations. This will be achieved after the auto fluorescence is eliminated from the background of the image. Creating Peptide-Based Capture Agents Through Ruthenium Catalyzed Metathesis Belen Alvarez-Villalonga Mentors: James Heath, Arundhati Nag, and Samir Das Antibodies are currently the most commonly used clinical diagnostics tools. Peptide-based capture agents can be created with affinities and specificities rivaling those of antibodies while being less expensive to produce and more reliable under varying physical and chemical conditions. This project focuses on the development of a cyclic peptide capture agent that can penetrate cells in order to identify and inhibit Akt2 kinase, a protein that is activated, overexpressed, or mutated in various types of cancer. The project involves the use a peptide sequence that has proven in the past to bind selectively to the phospho Ser474 region of Akt2. The sequence is cyclized through Ruthenium catalyzed metathesis; unlike the original peptide which was cyclized through a Copper-catalyzed click reaction. The two peptides were tested in an Enzyme Linked ImmunoSorbent Assay against Akt2 to compare the affinities of the different forms to the target. Testing on the two peptides is ongoing; however, peptides created through metathesis have proven to have better cell penetration, so the project is continuing with the newly synthesized peptide. Inhibitory assays will be performed using the Akt2 protein in addition to a cell culture. Phonon Mean Free Paths of Semiconductors Keshav Amla Mentor: Austin Minnich Phonons are essentially quasiparticles that can be described as wavelength/frequency-associated excitations of condensed matter particles. These phonons are the driving force behind such properties as thermal conductivity, and, at longer wavelengths, sound (from which is given rise the word “phonon” from the Greek word for “sound”). By studying their mean free paths in semiconductors, we hope to better understand the micro-scale effects that cause the thermal conductivity. Using a pump-probe system with various optics, an electro-optic modulator, a lockin amplifier, an initial laser, and a temperature controller, we excite samples of various semiconductors at low temperatures, through which we have determined the thermal conductivity and interface conductance values for Silicon, Gallium Arsenide, Gallium Phosphide, Silicon Carbide, and Sapphire, over the temperature range of 20 K to 300 K, and over the pump spot size range of 20 microns to 60 microns. From this, we hope to better understand the mean free paths of the phonons of these semiconductors. Study of the CMS Potential for Higgs Self-Coupling Measurements for the High Luminosity LHC Da An Mentors: Maria Spiropulu and Si Xie An analysis of the HH → γγ decay channel is presented for the purposes of studying the prospects for measuring the Higgs trilinear self-coupling, λHHH, following the CMS Phase II upgrade for the High Luminosity LHC (HL-LHC). This study uses generator-level Monte Carlo events in 3000 fb-1 of 14 TeV proton-proton collisions. Associated Higgs production and non-resonant background events are considered. In order to approximate the expected upgraded-detector performance for the HL-LHC, efficiency functions are applied to generator-level samples, where photon identification efficiencies and b-tagging efficiencies are parameterized with pT and η dependence. The sensitivity of the HH → γγ channel is determined using maximum-likelihood fits. Development of a Decision Support Tool for Agricultural Economic Analysis of the Production of Bio-Ethanol Robert Anderson Mentors: William Lazarus and Kim Border As carbon-based reserves run dry and energy prices increase, the need for renewable energy sources is immense. Biological based energy sources, such as Short Rotation Woody Crops (SRWCs), represent an area of research that, if proven economical, could greatly advance society in the path towards sustainability in addition to decreasing dependency on foreign fossil fuels. However, in order for these biofuels to be implemented, the product must be economically advantageous. My research focuses on development of Decision Support Tool that utilizes specific user climate and soil input compiled with local economic data to output an enterprise budget for the production of

3

  bio-energy crops. In an aim to broaden the knowledge of agricultural biomass production, this project allows farmers to generate an optimized farm budget that could lead to an increase in the biomass supply for ethanol production based on an economic motivation. Further research needs to be done to advance the production of cellulosic ethanol, which will further increase the profitability and incentive of growing biomass crops. Molecular Dynamics Approach to Calculate Diffusion Coefficient of Gases in Metal Organic Frameworks for Separation and Purification Alexander Anemogiannis Mentors: William Goddard, III, and Jose Mendoza-Cortes Metal Organic Frameworks (MOFs) have received much attention over the last several years for their ability to store gases, but another use emerges when considering the varying rates of adsorption of different gases in a MOF: gas separation. The diffusion coefficients of gases in the MOF determine the differences between the gases’ rates of adsorption and the MOF’s ability to separate particular gases in a mixture. The first step in determining the diffusion coefficient of a gas in a particular MOF is using molecular dynamics simulations to dump the velocities of the gas molecules moving throughout the MOF and then integrating the velocity autocorrelation function calculated from the molecule’s velocities. We specifically looked at the diffusion of H2 and CH4 in MOF-5 and compared our results to experimental values. This molecular dynamic approach can be extended for other systems including other MOFs and membranes for separation and purification. Telescoping Geocell Method: Increasing Ground Acceleration Map Accuracy Marie-Eugenie Arabian Mentor: Robert W. Clayton When an earthquake occurs, a rapid response with key information is crucial to mediate casualties and further damage. One product that can help first-responders is a map that pinpoints peak accelerations within the region. This is the best proxy we have for a damage assessment in the first few hours. However, for various reasons the sensors can produce erratic extreme accelerations, which will distort the accuracy of the ground acceleration map if they are not edited. A method to filter out these anomalies was developed using a telescoping concept. This method utilized the median to filter the peak acceleration of a given rectangular region called a geocell. The geocells telescope down in size and when the base number of devices is reached in a region its median acceleration is updated to its parent’s median acceleration so as to smooth the initial ground acceleration map. The filtered maps result in a greatly improved distribution of peak accelerations as well as a more accurate depiction of high acceleration regions during an event. The technique is applied to the 200-station Community Seismic Network located in Pasadena, CA, and to a 5000-station temporary network that was located in Long Beach, CA. Machine Learning for the LHC Grid Thomas Arnold Mentors: Maria Spiropulu and Yi Chen The data generated by the Compact Muon Solenoid at the Large Hadron Collider requires a significant amount of processing to analyze. The challenge is that such data processing must be spread across multiple sites due to restrictions on the amount of available computer time. Reinforcement learning, a branch of machine learning, provides a potential means of developing an algorithm to determine the optimal way in which such analysis may be distributed amongst multiple sites in order to obtain results as quickly as possible. By writing a scenario describing such a system and integrating it into existing reinforcement learning code known as RLPy, short for Reinforcement Learning in Python, it is possible to determine the effectiveness of reinforcement learning at analyzing such a system. Currently the concept of reinforcement learning shows promise as a means of determining the ideal solution to such a job distribution system. With further research the results generated by the code could have the potential to help streamline the analysis process for data generated at the LHC. Furthermore, additional inference applications can be envisioned. Reconfigurable Optical Physical Unclonable Functions for Cryptographic Key Storage Sidhi Assawaworrarit Mentors: Changhuei Yang and Roarke Horstmeyer In this work, we experimentally demonstrate how random communication keys may be stored error-free in nonelectronic memory over a period of several hours. We use an optical scattering Physical Unclonable Function (PUF) along with a suitably chosen error correction procedure to reproducibly generate gigabits of statistically verified randomness. This stability achievement is facilitated by a uniquely compact optical PUF contruction, consisting primarily of a volumetric scattering medium sandwiched between a spatial light modulator (SLM) and digital detector. The spatial light modulator serves to pattern an incident light beam into a specific PUF challenge, the digital detector serves to measure the intensity of the resulting scattered optical field (i.e., the PUF response), and all optical components are fastened together with an epoxy to ensure the optical system is minimally effected by any mechanical movement. We also demonstrate how the entire random keyspace may be reset through device heating.

4

 

Electrochemical Measurement of Oxygen Nonstoichiometry in Thin Films of Mixed Conducting Oxides Arun S. Asundi Mentors: Sossina M. Haile and Chirranjeevi Balaji Gopal The thermochemical cycling of the nonstoichiometric oxide ceria, CeO2–δ, provides a promising method of splitting water to produce hydrogen fuel. The thermodynamic effectiveness of ceria in thermochemical fuel production is characterized by the nonstoichiometry, δ, which governs the amount of hydrogen produced in one cycle. In particular, a large change in nonstoichiometry, ∆δ, in response to a small change in temperature, ∆T, is desired for maximal hydrogen production. The objective of this project was to determine δ for zirconium-doped ceria (ZDC) as a function of the oxygen partial pressure, pO2, and temperature, T, using alternating current impedance spectroscopy (ACIS). The nonstoichiometry of a mixed conducting oxide can be expressed in terms of a quantity called the chemical capacitance, Cchem, which can be measured using ACIS. The impedance response of ZDC was measured by pulsing alternating currents over a wide range of frequencies across a symmetric cell of ZDC epitaxially deposited onto yttria-stabilized zirconia (YSZ). By modeling the impedance response with an equivalent circuit, the chemical capacitance was extracted from the ACIS measurements, from which the nonstoichiometry was calculated. AND Gated Metabolic Protein Labeling for Selective Analysis of Biofilm Subpopulations Lydia Atangcho Mentors: David A. Tirrell and Brett Babin Psuedomonas aeruginosa is a highly antibiotic-resistant bacteria that forms biofilms, or dense bacterial aggregates, in the lungs of Cystic Fibrosis patients. The most resistant subpopulations of these biofilms are found at the base of the structures where cells are most nutrient-deficient. Using cell-selective BONCAT, a method for cell-selective protein labeling with non-canonical amino acids, it has been shown that proteins can be labeled and isolated from just these regions by expressing a mutant methionyl-tRNA synthetase (MetRS) from a single growth-phase controlled promoter, PrpoS, subsequently allowing incorporation of the non-canonical amino acid, azidonorleucine (Anl). By using a two-promoter controlled expression system, the AND gate, we hope to achieve enhanced labeling selectivity in this complex cellular system. Several P. aeruginosa strains, each with a single endogenous promoter in control of a fluorescent protein were grown and imaged for visualization of targeted regions. Promoters active in the region of interest will be used to construct an AND gate system in which each promoter controls expression of half the mutant MetRS gene. With the AND gate and cell-selective BONCAT, we hope to achieve selective protein labeling in the most resistant regions of P. aeruginosa biofilms. Synthesis and Characterization of Gold Arylisocyanide Complexes for Use as Powerful Photoreductants Jackson Atwater Mentors: Harry B. Gray and Wesley Sattler Gold(I) arylisocyanide coordination complexes have been synthesized and characterized for use as photosensitizers with the ultimate goal of chemically reducing carbon dioxide (CO2). The isocyanide ligands (CNAr) have been prepared by N-formylation of the corresponding aniline (H2NAr), followed by dehydration. Heteroleptic neutral and homoleptic cationic gold complexes, AuCl(CNAr) and [Au(CNAr)2]+, respectively (Ar = 2,6-dimethylphenyl or 2,4,6-tri-tert-butylphenyl), have been synthesized. These complexes have been characterized by various spectroscopic techniques, including 1H and 13C NMR spectroscopies, electronic absorption and emission spectroscopies, as well as single-crystal X-ray diffraction to determine the molecular structures. [Au(CNAr)2]+ complexes are highly luminescent in the solid-state, emitting blue light (max ≈ 430 nm) when irradiated with UVlight (ex ≈ 300 nm). However, in solution, the complexes are only weakly luminescent due to the cleavage of intermolecular Au-Au contacts, commonly referred to as aurophilic interactions. Future work will focus on generating complexes that exhibit luminescence in solution so quenching studies with CO2 can commence. Local Depletion of Macrophages Following Intra-Articular Fracture Karsyn N. Bailey Mentors: Steven Olson and Bridgette Furman Post-traumatic arthritis (PTA) is an accelerated form of arthritis that most commonly develops following fracture of the articular surface of a joint. Following articular fracture, C57BL/6 mice showed significant signs of PTA, whereas a different strain of mice, MRL/MpJ mice did not exhibit these signs, suggesting that inflammation and the infiltration of synovial macrophages plays a role in the pathogenesis of PTA. The Macrophage Fas-Induced Apoptosis (MAFIA) mouse strain expresses the inducible fas-suicide gene, which allows for apoptosis of macrophages following the administration of the dimerizing reagent, AP20187. Macrophages fluoresce via insertion of the Green Fluorescent Protein (GFP) transgene in the MAFIA mice, permitting the visualization and quantification of macrophages. In this study, male MAFIA mice were subjected to a closed intra-articular fracture of the tibial

5

  plateau and treated with AP20187 via intra-articular injection in order to locally deplete macrophages within the joint. Time of injection was evaluated to determine the most effective local depletion of macrophages and reduction of inflammation following joint injury. Depletion of macrophages following joint injury will be used to observe the role of inflammation and macrophages in the development and progression of PTA. The Torque Applied by a Protoplanet to a Surrounding Gas Disk Adam Ball Mentors: Neal Turner and Wladimir Lyra Astrophysics is unique in that it is perhaps the only scientific field in which experiments are never performed. Even observations are often hard to come by. This lack of direct physical evidence necessitates the use of computational models to develop and test astrophysical theories. The Pencil Code is a highly modular high-order finite-difference code intended for use in astrophysical models. The primary goal of my SURF was to implement the effect of a planet’s torque on the gas in a one-dimensional α-disk model within the Pencil Code. The term is based on an analytical formula from a recent paper. However, to make it physically accurate significant modifications had to be made, most notably adding a correction factor to give the overall term a zero mass contribution. Now that this gapcarving term is implemented, the natural next step for further research is to implement gas accretion or planet migration in the one-dimensional α-disk model. Design of Microscale Planar Free Space Focusing Elements Alex Ball Mentor: Andrei Faraon Optical components for focusing light are ubiquitous in modern technology, with applications ranging from imaging, to laser cavities, to CD/DVD drives. However, for applications that require small component sizes, current compatible technologies such as microlenses, Fresnel lenses, and zone plates are difficult to manufacture. A new technology to address this issue is based on aperiodic gratings or post arrays made of high refractive index materials such as silicon. These devices are compatible with standard lithographic techniques and so can easily be used in integrated optical systems. Focusing is achieved by exploiting sub-wavelength feature sizes to induce locally controllable phase shifts to the incoming light. The simulation of such a device is presented: a 200-micronradius lens is used to focus a Gaussian laser beam with a wavelength of 2 microns. The size of the lens makes it computationally difficult to simulate using standard finite-difference time-domain software packages. However, using MIT’s finite-difference time-domain simulation software (MEEP), analytical propagation of optical beams, and several optimization techniques, the lens is successfully simulated and its focusing abilities assessed. The results of the simulation will be used to refine the lens design before fabrication. Additionally, the optimization techniques developed will be used to speed up the simulation of other large planar devices for free space optics. Determining the Optimal Combination of Players on Basketball Teams Susan Ballentine Mentor: Michael Alvarez Increasingly, the world of professional sports is using statistical analysis in building their roster. While this movement has its roots in baseball, there are many opportunities to apply the same types of tools to basketball. This is somewhat harder to accomplish, as in basketball the overall success of a team is less linked to individual player performance metrics and more dependent on team chemistry and play-making ability. The goal of this project was to find a way of determining which set of five players of a complete roster maximizes a team’s performance. Looking at data from the NBA 2012-13 season, we found the effectiveness of each 5-player combination for each team to compare to their playing time and output. This allows us to identify the most efficient combinations that, if played more often, could allow teams to score more points and win more games. Detection and Classification of RR Lyrae Stars With Palomar Transient Factory Sophianna Banholzer Mentors: Judy Cohen and Branimir Sesar RR Lyrae stars act as a useful tool for astronomers and, as such, an accurate method for identifying these stars is needed. A classification system for RR Lyrae stars was designed to classify objects detected with Palomar Transient Factory (PTF) using machine learning techniques. In order to achieve an accurate classification, it proved necessary to divide the object data into two magnitude categories with a division at 19 magnitude. Different light curve parameters were then calculated for the brighter and fainter objects and a random forest classifier was applied. The classifier was further improved by using active learning to create a complete training sample. Through this method, an accurate classification of RR Lyrae stars was achieved. The classifier successfully classified these objects with very little contamination and a high completeness rate for the brighter stars. The completeness decreased for fainter sources but the contamination rate remained low. Applying this classifier to regions covered only by PTF allows for the identification of new RR Lyrae sources.

6

  Study of 2D and 3D Photonic Crystals From Brush Block Copolymers and the Effects of Glass Transition Temperature on Responsiveness to External Stimuli Richard Barz Mentors: Robert H. Grubbs and Benjamín Sveinbjörnsson Brush block copolymers are large, complex molecules that can self-assemble into various morphologies. The selfassembly can generate a photonic crystal with a band gap that can be adjusted by adjusting the properties of the copolymer. Brush block copolymers were prepared using polylactide (PLA) and either polystyrene (PS) or poly(n-butyl acrylate) (PnBA). By varying the internal block size and brush length for PLA-PS copolymers, several copolymers with various properties have been produced. However, it remains challenging to increase the wavelength of the band gap, a goal of current research, for these monomers. Research in PLA-PnBA copolymers holds more promise because of the low glass transition temperature of PnBA. It is expected that this property of PnBA will generate resulting photonic crystals responsive to external stimuli, such as pressure. This interactivity would greatly increase the range of applications for these photonic crystals. A Search for Haze in the Atmosphere of the Cool, Saturn-Mass Exoplanet HAT-P-12b Nathaniel Baskin Mentor: Heather Knutson There are a number of means by which haze can develop in a planetary atmosphere, yet the exact circumstances and mechanisms behind its formation remain poorly understood. Given the variety of pathways by which hazes and clouds can appear, they are likely present in many planetary atmospheres. By using photometric transit observations of various exoplanetary test cases, we can characterize the planets’ transmission spectra. From the spectral data, we can constrain the likelihood of haze formation in an atmosphere, thereby allowing us to test various models for haze formation. In this project, we present the optical spectra of the transiting exoplanet HAT-P-12b taken with the Space Telescope Imaging Spectrograph (STIS) instrument aboard the Hubble Space Telescope (HST). This Saturn-mass planet orbits a 4650K star with a fairly low activity level, allowing us to test for the correlation between haze production, atmospheric temperature, and the level of incident radiation received by a planet’s atmosphere. If the formation of haze were influenced by photochemistry, we would expect to find less haze in the atmospheres of planets like HAT-P-12b. If, in contrast, the haze were composed primarily of silicate condensates, it could still appear in these cooler test cases. Hydrogen and Metal Line Absorption Through Random Sight-Lines in Cosmological Hydrodynamic Simulations Juliette Becker Mentors: Juna Kollmeier and John Johnson We examine the metal distributions in the intergalactic medium as gleaned from random sight-lines in cosmological hydrodynamic simulations centered at z = 0.25. These simulations demonstrate the effect of ion qualities (such as ionization energy) on diffusion scale and plentifulness in the IGM, suggesting what galaxy qualities are implied by the presence of a given ion quality. These results have implications on the galactic outflow process, which is not currently understood. The Role of Cysteine Residues in the Response of p53 to Oxidative DNA Charge Transport Lisa J. Beckmann Mentors: Jackie Barton and Katie Schaefer Tumor suppressor p53 is a cell-cycle regulatory protein that is mutated in over 50% of human cancers. 80% of these mutations are found in the DNA binding domain, highlighting the important interaction between p53 and DNA. Multiple cysteine residues appear to be critical in p53 appropriately coupling to DNA and responding to oxidative genomic stress. The ultimate aim of this project is to determine the roles of redox active cysteine residues within p53 in response to oxidative DNA charge transport (DNA-CT). Mutant p53 with individual residues of cysteine mutated to serine were created by site-directed mutagenesis.. The mutants have been overexpressed in e. coli and purified by FPLC. In vitro, DNA-CT will be induced using consensus sequence oligonucleotides with a covalently tethered anthraquinone photooxidant irradiated while in the presence of bound p53. Differential thiol labeling coupled with protein mass spectroscopy will determine the oxidation state of each cysteine residue after oxidative DNA-CT. We seek to determine the key cysteine residues involved in the charge transport pathway, especially the residues that act as the terminal disulfide acceptor. This work will reveal details about the importance of cysteine residues of DNA binding proteins being responsive toward oxidative DNA-CT.

7

  Interactive Discovery in the Kepler Mission Dataset James Bedell Mentor: Kiri Wagstaff My project focuses on applying automated scientific discovery to Kepler telescope data as an instance of the "big data" analysis problem. The Kepler telescope has observed over 150,000 stars during the past three and a half years in search of transiting exoplanets. The data processing pipeline in place at Ames Research Center preprocesses the optical images, removes instrumental trending in the resulting curves, and then applies a set of filters in an automated process to flag candidates of interest. These candidate stars are then examined through significant statistical testing and manual review to be confirmed or rejected as exoplanet hosts. However, most sources of interest still remain in the "candidate" category, and some exoplanet hosts may not have been flagged at all. I have investigated applying the DEMUD (Discovery through Eigenbasis Modeling of Uninteresting Data) data prioritization algorithm to the preprocessed Kepler light curves to see how we can speed up the supervised classification process, as well as how to expand the DEMUD system to be more versatile and interactive. I will discuss salient features of the stars identified by DEMUD as most interesting. Additionally, I will show how DEMUD has demonstrated success in quickly finding members of rare classes in several data sets. This is beneficial to scientists examining data sets that are not fully explored or understood, and can lead to the discovery of unusual artifacts or important observations otherwise buried in a mass of data. Potential Ground System Risk Analysis Glenn Belen Mentors: Jose Macias and Eleanor Basilio My work consists of identifying and understanding risk items that are encountered when designing a ground system (GS). The method of analysis used to analyze each risk item is Risk Informed Decision Making (RIDM). This method uses human judgment, when there is no concrete testing data to base the risks on. The down side of this method is that it is subjective, which may make the analysis bias. After researching ground systems and risk analysis I built upon an Excel template for risk management. This method analyzes the six potential ground system risk items typically encountered in the Formulation Phase. For each risk item a decision statement and alternatives are generated. The alternatives are analyzed and scored by their importance to the performance objectives. Each performance objective is assigned a weight factor, which is summed to the total score. The scores help determine which alternative best satisfies the decision statement. I created a 5x5 risk matrix on the risks of each alternative, which is based from the risk’s likelihood and consequence. This analysis helps create a base for decision making when designing a ground system. LQG Control on a Channel With a Random Delay Fredrik Bengtsson Mentor: Babak Hassibi Consider a control system in which there is a channel with a random delay with a known probability function between the controller and the process. Methods for optimal LQG control are considered here within. Solutions are derived both for the case where the process applies no control signal if the latest signal from the controller is delayed as well as for the case where the process applies the last control signal it has received from the controller. In the first case we find that the optimal control signal can be derived using a Riccati equation similar to the one used on a channel without a delay. In the second case we find that the optimal control signal not only depends on the states, but also on the previous control signals sent. Moreover instead of a Riccati equation we get a matrix that must be updated with each iteration. The Effect of Aversive Pavlovian Predictors on Highly Incentivized Performance Joseph D. Berleant Mentors: John P. O’Doherty and Vikram Chib It is widely believed that work performance and productivity can be improved through increased monetary incentives. However, a phenomenon known as “choking on the money,” in which performance is worsened rather than improved, can occur when particularly high incentives are offered for completing a difficult task. One explanation for the choking phenomenon is that during performance of a difficult task, an incentive may be reframed as a potential loss, distracting from the task itself. This hypothesis is corroborated by a previous study, which showed that being highly loss-averse was correlated with more severe choking. To test this hypothesis more fully, we present an experimental procedure in which aversive Pavlovian conditioning is combined with monetary incentives during performance of a difficult motor task. It is expected that the presence of an aversive Pavlovian predictor will be viewed as an additional potential loss and should function in a manner similar to an increase in the monetary incentive. Behavioral and fMRI analysis will be used to verify this effect. We have begun preliminary behavioral testing to confirm this hypothesis, and further testing with functional imaging will explore the neural mechanisms that subserve this effect.

8

  Robotic Motion Planning With Complex Tasks James Bern Mentors: Richard Murray and Eric Wolff We implement and test a robust control algorithm for uncertain robotic systems with linear temporal logic task specifications. The system is modeled as an uncertain Markov decision process, a Markov decision process with uncertain transition probabilities. The algorithm generates a control policy that maximizes the probability of satisfying the specification under the worst-case resolution of the uncertainty in transition probabilities. The control algorithm is included in the Temporal Logic Planning toolbox (TuLiP) and interfaced with the Robotic Operating System (ROS) for implementation on physical robots. We used a Turtlebot for experiments, which is a wheeled, open-source robot that runs on ROS. The Turtlebot can navigate autonomously in an uncertain environment by utilizing data captured from an infrared overhead tracking camera. We compare the algorithm's experimental performance to that of pre-existing algorithms, and discuss future work. Connections Between the North Atlantic Jet and the Asian Monsoon Paolo M. Bertolotti Mentor: Tiffany A. Shaw During Northern Hemisphere summer the Monsoons are responsible for significant momentum, moisture and heat transport. Following work by Kushner and Held (1998), the streamfunction was found to be a good metric for measuring the variability of the low-level Monsoon cyclone and its impact on the upper-level jet stream. Significant correlations were found between the low-level streamfunction variance over Asia and the upper-level streamfunction, which is an anticyclone. Significant correlations were also found between the low-level streamfunction variance over Asia with the zonal wind at upper-levels in the Northern Atlantic and Pacific jet regions. The correlations imply that the jet shifts poleward as the Monsoon strengthens. The quantitative impact of Monsoon variability on the position of the Atlantic jet will be examined along with the connection to the momentum and moisture transport. Free-running climate models will be analyzed to understand the connection between changes in the Monsoon and the well-known poleward shift of the Atlantic jet stream in response to increasing carbon dioxide levels.

Multiplexed Mapping of miRNA and mRNA Expression in Whole-Mount Zebrafish Embryos Aditya Bhagavathi Mentors: Niles Pierce and Aneesh Acharya Fluorescence in situ hybridization (FISH) is a prominent technique used to detect and locate DNA/RNA sequences in biological samples. However, conventional FISH is nonquantitative and unable to multiplex short targets such as microRNA (miRNA), thereby limiting our investigative power into the dynamic nature of nucleic acids. To overcome this barrier, we examine the potential of hybridization chain reaction (HCR) in simultaneously detecting short and long targets in biological samples. HCR is a novel programmable amplification mechanism for FISH by which metastable hairpins undergo a triggered polymerization reaction. Its ability to multiplex mRNAs is already documented. After transitioning to DNA-based amplification and optimizing Proteinase K treatment, we have developed a new HCR protocol for multiplexed miRNA detection. We then tested this protocol by multiplexing a combination of miRNAs and mRNAs in whole-mount zebrafish embryos. Despite some initial successful results, inconsistency and a lack of a positive control plagued our efforts to reproduce earlier results.

9

  Fracture Toughness Testing of Nano Trusses Srinivasa Bhattaru Mentors: Julia Greer and Lauren Montemayor In recent years, the field of materials has grown greatly with the advent of architectural design of nano-scale structures. Determining the properties of these structures is incredibly important, largely because they differ so greatly from their macroscopic counterparts. This research is concerned with the fracture toughness of nano-scale structures, in particular of 3D trusses. Projects included determining various features of the architectural design of these trusses such as the relative density, surface area to volume ratio, as well as the predicted Young’s Modulus and yield stress. The trusses were then fabricated for testing using two-photon lithography. The next phase of research involved testing these trusses by indenting them using a nano-indenter to find experimental values for the Young’s Modulus and yield stress. In addition, the project also involved determining the mechanics behind the stresses placed on these lattices, as well as the mechanisms of their collapse. With the experimental values of the parameters and the theoretical work on the stress mechanisms, the research yielded some idea of the toughness of these designed materials. Shrinking and Swelling Dynamics and Equilibrium Configurations of Colloidal Gel Networks Moriah Bischann Mentors: John F. Brady and Charles G. Slominski We derive a model system that describes the shrinking and swelling dynamics and the equilibrium characteristics of colloidal gel networks. Our isothermal-isobaric system has a cubic lattice configuration parameterized by a uniform Hookean bond strength and osmotic pressure. Solvent fluctuations impart thermal energy upon our system’s solute molecules. This energy transfer distorts our system’s network structure and induces a variance in our system’s interparticle bond length and volume with time. Two equations govern our system. The first equation describes the mechanical balance of our system’s ideal gas, Hookean spring, and dynamic osmotic pressure contributions in response to externally applied pressure. The second equation describes the structural shrinking and swelling kinematics of our system’s network configuration in terms of its solvent’s macroscopic velocity and its solute’s interparticle bond length. We manipulate these governing equations to construct a single ordinary differential equation. Then we solve our ordinary differential equation using separation of variables to determine our system’s dynamic motions. Analysis of our inverted ordinary differential equation's vertical and horizontal asymptotes yields insight into our system’s long-time behaviors. Next we utilize the rules of polynomial discriminants, the Fundamental Theorem of Algebra, and Descartes' Rule of Signs to discover critical interparticle bond lengths. Stability analysis of our ordinary differential equation pinpoints the bond strength and pressure parameters that describe the interparticle bond length of our equilibrating system. Finally, we conduct computer simulations of representative colloidal gel systems to confirm our mathematical results. Our insightful model yields a greater understanding of the dynamics and the equilibrium characteristics of complex colloidal gel networks. Climate Analysis Using GPS Radio Occultation Data: A Study of Seasonal and Diurnal Variability of the Planetary Boundary Layer James Blackwood Mentor: Chi Ao GPS Radio Occultation (GPSRO) is a remote sensing technique that allows for precise vertical measurements in the atmosphere and can be performed even in the presence of cloud cover. The Planetary Boundary Layer (PBL) is a crucial component of the atmosphere, with significant impacts on global energy and water cycles. Being the closest part of the atmosphere to the Earth’s surface, the PBL plays a significant role in governing heat exchanges and low-level cloud formations. Using five years (2007-2011) of data from the COSMIC satellite constellation, the seasonal and diurnal variations in PBL height were analyzed for regions in the Northeast and Southeast Pacific Ocean. After developing and modifying algorithms to determine PBL height from GPSRO profiles, the data was then sorted and filtered based on various time-constraints to study seasonal and diurnal cycles. Several signal processing and time-frequency analysis techniques were used and compared to gain insight into the periodic nature of PBL variation. Validation of the GPSRO results was then done through performing similar analyses on radiosonde and weather reanalysis data. Towards Local Topological Quantum Order in 2D: Investigating the PEPS Canonical Form Thom C. Bohdanowicz Mentors: John Preskill and Spyridon Michalakis The so called Matrix Product States (MPS) and Projected Entangled Pair States (PEPS) have recently risen to prominence in describing the ground states of quantum many-body systems in one and two dimensions respectively. The theory of MPS in particular has led to fantastic results in understanding properties of quantum many body ground states in one dimension, ultimately allowing complete classification of gapped one dimensional phases of matter. However, the same level of success has not been enjoyed in two dimensions using PEPS.

10

  Specifically, the absence of a PEPS canonical form, a unique mathematical structure exploitable for all MPS, is the culprit. To make progress on this front, we have studied the detailed mathematical structure of the MPS canonical form in order to better understand how to make progress in formulating a PEPS canonical form. In particular, our goal is to apply such insights to understand how the stability of PEPS under certain natural perturbations relates to the underlying mathematical structure, and to show precisely which mathematical properties (injectivity, gauge symmetries, etc.) are required of a PEPS in order to display local topological quantum order, a property which guarantees the reasonable behaviour of a system described by a PEPS. Identification of Genes That Affect Migration of the Distal Tip Cells in Hermaphrodite C. elegans Larvae Olivia J. Box Power Mentors: Paul Sternberg and Mihoko Kato Cell migration performs an important role in both development and the metastasis of cancer, and genes involved in this process are likely to be conserved between organisms. To find genes involved in the metastasis of cancer in humans, an RNAi screen was performed on C. elegans to find those which affect the distal tip cell migration during development. The migration of the two distal tip cells defines the shape of the hermaphrodite C. elegans gonad; L4 stage worms which had fed on RNAi bacteria were observed under Nomarski and fluorescence microscopy, and their gonad phenotype scored. Common DTC migration defects include abnormal path, timing (failure of the DTCs to turn at the L4 molt) and DTC shape. Approximately 100 genes were selected for the RNAi screen, the homologs of those found in previous studies to be upregulated in metastasis in humans. Genes were considered to have a potential role in cell migration if over 30% of arms had a defect, or if the same type of defect was found consistently. We identified several genes with potential roles in cell migration, such as M116.5, an ortholog of the human gene utrophin which is homologous to the dystrophin gene. The role of these genes will be investigated further by measuring their expression pattern and observing mutants. Analysis of Jupiter’s Thermal Emissions Kimberly Boydstun Mentor: Glenn Orton Throughout the last decade, Jupiter has experienced planet-wide disturbances. Substantial changes were observed in its axisymmetric regions, such as a twice-occurring fade and revival of the South Equatorial Belt, the dark band just south of the equator. Understanding these visual transformations requires understanding the physical processes that underlie them. Temperature, cloud and gaseous composition were determined using images gathered from observations taken at the NASA Infrared Telescope Facility, and should be compared with a timeline of visual anomalies for correlations. Temperature profiles, obtained using the Non-Linear Optimal Estimator for Multivariate Spectral Analysis (NEMESIS), have been used to examine the difference in temperature between Jupiter's northern and southern hemisphere over more than a Jovian year. There exists evidence of subtle change in temperature corresponding to seasonal change. This is bolstered by measurements of brightness temperature, which are obtained from raw data at wavelengths that are sensitive only to temperatures. Substantial work was also done to examine the correlation between calibration approaches and the expectation of temperature changes over a Jovian year as a result of seasonal changes in solar heating. The Effects of Strong Disorder on Topological Anderson Insulators in Two Dimensions Sarah Brandsen Mentor: Xincheng Xie We examine the Anderson transition, where the Topological Anderson insulator (TAI) becomes an Anderson Insulator (AI) under a critical disorder. Electron localization occurs, and the localization length is measured by creating a Matlab program to simulate a two-dimensional TAI under strong disorder. The Transfer-matrix method and tight-binding model are utilized in creating this simulation. Finally, the localization length can be examined as a function of Fermi energy and disorder strength. Discovering New Associations Between Anaerobic Methane-Oxidizing Archaea and Sulfate-Reducing Bacteria Julia Brown Mentor: Victoria Orphan Mutualistic associations between microbes form a major part of nutrient cycling in nature. One of the most important of these associations is the association of anaerobic methanotrophic archaea (ANME) and sulfatereducing bacteria (SRB) to mediate the anaerobic oxidation of methane (AOM). Using fluorescence in situ hybridization (FISH), it is possible to visualize and identify these organisms in the environment and quantify their abundance. Using sequential FISH reactions to target specific groups in environmental samples, we identify the presence of several different bacterial and archaeal groups, including the Gammaproteobacteria and several subgroups of ANME.

11

  Evaluation of Substrate Mode Sensors for Self-Healing Radiators Sidney Buchbinder Mentor: Ali Hajimiri The increased speed and smaller size of integrated circuit radiators has led to greater variation in fabrication between the produced units, degrading system performance. Self-healing radiators monitor the current system performance via on-chip pickup-sensors, and through feedback, utilize on-chip correcting mechanisms and actuation to improve device operation. The pickup-sensors measure substrate modes, electromagnetic oscillations supported as standing waves in the medium. To design useful pickup-sensors, the substrate modes of the radiator must be better understood. Through simulation in HFSS (a high-frequency electromagnetic simulator), supported modes of a proof-of-concept PCB radiator were found for operating frequencies near 5GHz. The same PCB was fabricated on a high-dielectric substrate. Several components of the electromagnetic fields were measured using off-chip antennas, placed near the surface of the substrate to simulate on-chip pickup-sensors. Comparison of the measured and simulated fields suggested which modes were excited in the substrate. While the equivalent simulations and physical measurements did not match exactly, the results indicated that the excited substrate modes depend heavily on the input drive parameters. Furthermore, the differences in the field patterns picked up by various off-chip antennas suggested that the pickup-sensor geometries can be altered to measure different components of the supported substrate modes. Propagator With Positive Cosmological Constant in a 3D Quantum Gravity Toy Model William E. Bunting Mentor: Carlo Rovelli Abstract We study the propagator on a single tetrahedron in a three dimensional toy model of quantum gravity with positive cosmological constant. The cosmological constant and an infrared cutoff are included in the model via q-deformation of the spatial symmetry algebra, that is we use the Tuarev-Viro amplitude. The expected repulsive effect of dark energy is recovered in numerical and analytic calculations of the propagator at large scales comparable to the infrared cutoff. However, due to the simplicity of the model we do not obtain the exact Newton limit of the propagator. This is a first step toward the similar calculation in the full 3+1 dimensional theory with larger numbers of simplicies. The Effect of Baryons on the Distribution of Dark Matter in Galactic Halos Iryna Butsky Mentors: Andrea Maccio and Christian Ott We analyze a series of high resolution hydrodynamical simulations performed within the MaGICC (Making Galaxies In a Cosmological Context) project to study the effect of baryons on DM distribution. MaGICC galaxies are among the most realistic simulated galaxies available today and therefore offer a unique environment to better understand the back reaction of baryons on dark matter. We consider the effect of baryons on several parameters of dark matter halo, including the shape, velocity distribution, velocity anisotropy, and the pseudo phase space density. We find that baryons make more spherical dark matter haloes, more tangential dark matter orbits, a more skewed velocity distribution, and a different power law expression for the pseudo phase space density. However, these effects aren't as strong as predicted by previous studies based on galaxies having a non realistic ratio of stellar to dark matter halo mass. Conformational Preference of Succinic Acid in Protic and Aprotic Solvents Beth C. Campbell Mentors: J. D. Roberts, B. U. Emenike, and W. R. Carroll Succinic acid provides a useful system for studying the conformational behaviour of more complex molecules in solution. Surprisingly the difference in gauche fraction, found by examination of J13/J14 vicinal proton coupling constants, between protic and aprotic solvents is small; 84.4% and 85.6% in water and DMSO respectively. Here we show that, in this instance, the high fraction gauche may be accounted for by a gauche effect rather than hydrogen bonding. This type of gauche interaction may be prevalent in many other systems where a hydrogen bond would traditionally be expected. Planetary Protection Archiving Task of Microorganisms from Prelaunch Spaceflights Anabela Carigo Mentor: Wayne Schubert The archive facility at Jet Propulsion Lab contains over 3500 microbial isolates. These samples were collected during the assembly, testing and launch operations for various spacecraft. This study identifies cultivatable microbes from the archives collected from spacecraft surface assays in order to recognize microorganisms found on earth that can survive the heat shock method outlined by the Planetary Protection policies. Identifying the microbial phylogeny of these microorganisms could potentially indicate if microbes being carried to extraterrestrial environments could endanger future life detection missions. A subset of microbial isolates from the Phoenix Lander

12

  were characterized. DNA extractions were performed using a commercial kit system, followed by Polymerization Chain Reaction to amplify the number of DNA copies. Gel electrophoresis was used to verify the success of the Polymerization Chain Reaction. Standard Polymerization Chain Reaction clean-up was completed and the DNA samples were sequenced by a commercial DNA sequencing service. Glycerol and Cryobead frozen stocks were generated and stored for future reference and study. The Archive database was updated from preceding researchers’ lab work. Results in progress indicate that most isolates are of the genus Bacillus and Paenibacillus. Results are expected to be similar after sequencing and further analysis. The Porous Framework Genome Project Finn Carlsvi Mentors: William A. Goddard, III, and Jose Mendoza-Cortes When predicting gas storage properties of Metallic and Covalent Organic Frameworks, a computational model of the framework structure is needed for verification in simulation. A framework is constructed by combining secondary building units (SBU), which contain multiple connectors each, with linkers that contain exactly two connectors each. Depending on the amount and orientations of the connectors of each SBU, multiple frameworks can be generated when connecting them together through linkers. Generating these structures manually is easy, at best, with a cubic lattice, and has been done many times before. However, it becomes increasingly difficult to create more complex framework arrangements. A software package was created that generates frameworks using every combination of available nets, SBUs, and linkers. Once the frameworks have been created, their energies are minimized to predict the exact atomic structure, and then various computational gas storage simulations are run to predict the framework properties. An online database has also been created, which contains the data for all frameworks. Effects of Decellularized Chondrocyte-Derived Matrix Niche on Transdifferentiation of Fibroblasts Tracey Chan Mentor: Barbara P. Chan Composed primarily of collagens, proteoglycans, and water, and deposited by cartilage-specific cells, called chondrocytes, the extracellular matrix (ECM) in articular cartilage withstands compression and supports normal joint function. As cartilage has little capability for self-healing, despite its strength, flexibility, and complexity, treatments for damaged cartilage are under development to promote tissue regeneration, such as through transplantation of chondrocytes or multipotent cells in engineered constructs that retain chondrogenic phenotype. Because fibroblasts are highly plastic and easily obtained through skin biopsy, fibroblast potential for differentiation toward chondrocytes and application in cartilage tissue engineering was investigated. After culture in collagen type I microspheres for seven days, chondrocytes were removed by detergent sodium deoxycholate; the decellularized microspheres were recolonized with fibroblasts. Microsphere composition before and after decellularization and twenty-one days after fibroblast seeding was analyzed for expression of cartilage proteins collagen type II, aggrecan, Sox-9, and perlecan, using immunostaining and quantitative polymerase chain reaction. Lacking collagen II and glycosaminoglycan expression, chondrocytes cultured in microspheres appeared de-differentiated at 7, 14, and 21 days, perhaps due to the freezing and thawing process as cell stock. Without stimulation by chondrogenic features, seeded fibroblasts, as well as fibroblasts encapsulated in collagen I microspheres, produced Sox-9. These results imply that further study is needed to identify more cartilage-specific proteins in microspheres and to assess fibroblast differentiation in this chondrocytic ECM model. Ultra-Compact Plasmonic Modulator Using Transparent Conducting Oxide Krishnan Chander Mentors: Harry Atwater and Howard Lee Plasmons are waves of electromagnetic fields that occur at the interface between a dielectric and a metal. Such signals, if excited in a waveguide, can combine the compactness of electronic circuits at nanoscale dimensions with the speed of photonic circuits that would not be able to operate at small sizes due to the diffraction limit. The waveguides are constructed with rectangular layers of gold, aluminum oxide and a type of transparent conducting oxide, indium tin oxide (ITO). A path is constructed for the waveguide by making a 300 nm gap in the gold layer that is filled with aluminum oxide and ITO. The gap plasmon is excited by coupling infrared light to a Yagi-style antenna at the end of the waveguide. The charge carrier concentration of the ITO can be varied by sputtering with different oxygen concentrations. A voltage will be applied to the waveguide between the gold and ITO layers to modulate the index of refraction through field-effect dynamics, changing the strength of the propagating signal. The far-field signal intensity has been measured for waveguides with silicon dioxide in place of ITO, and a wavelength around 1540-1550 nm excites the strongest signal. Signals for ITO waveguides will be measured.

13

  The Effect of the Sec YEG Complex on Protein Expression Arun Chandra Mentors: Bil Clemons and Stephen Marshall Membrane proteins are vital to many cellular processes including transport, signaling, and catalysis. However, the structure and functions of most membrane proteins have not yet been clarified. Yields from standard E. coli overexpression trails are often inadequate. Hence, it is vital to address the bottlenecks in overexpression and to find alternative methods for allowing for abundant protein produced. It is likely that targeting and insertion of membrane proteins is in some way inadequate in heterologous expression systems. Previous studies suggest that the translocon pores that facilitate membrane protein insertion are saturated during overexpression, leading to incorrect folding. We propose that by increasing levels of SecYEG in E. coli, we will enhance the amount of correctly folded membrane protein produced. The purpose of my research thus far, is to look at the influence of the expression of the SecYEG gene on the over expression of membrane proteins in bacterial colonies. We are creating artificial SecYEG operons on expression vectors to increase the levels of SecYEG in cells. This is done by annealing four disparate nucleotide sequences together to create an artificial pacyc plasmid with the SecYEG inserted. The effects of the SecYEG complex will be tested by studying expression levels of different membrane proteins that we have pre-determined. Constructing the Catalina Real-Time Transient Survey Quasar Catalog Melissa C. Chang Mentors: George Djorgovski and Matthew Graham Quasars play a central role in modern astrophysics and observational cosmology. Quasars are highly variable and this variability can be used as a technique for selecting quasars. With the new, large data set from the Catalina Real-Time Transient Survey, the possibility of selecting quasars in bulk based on their variability presents itself for the first time. This project aims to develop an accurate method for selecting quasars by variability and to then develop a quasar catalog from data collected by the Catalina Real-Time Transient Survey. To achieve this, four different, already-published methods for quasar selection through variability were analyzed. Some were found to be much less reliable for larger and more variable data sets. However, using a combination of the parameters yielded by these methods, quasars can be selected more accurately. Using the results of the Butler and Bloom (2011) method, the characteristic timescale of the light curve, the slope of the least-squares fit to Slepian variance, and the Teraesvarta nonlinearity p-value, quasar candidates can be identified. Applying this method on data from the Catalina Real-Time Transient Survey would create the largest quasar catalog selected by variability to date. An Analysis of Variabilities in Cloud Cover: Evidence of a Climate Regime Shift Around 1998 Tiffany Chang Mentors: Yuk Yung and Hui Su This study examines how global cloud cover varies over time in the past three decades, with a focus on identifying dominant modes of variabilities and associated spatial patterns. By performing Empirical Orthogonal Function (EOF) analysis on satellite observations of cloud cover, the seasonal cycle and El Niño-Southern Oscillation (ENSO) are found to be the most significant sources of variabilities in cloud cover. In addition, a clear difference in cloud cover is found between the decades prior to and after 1998, suggesting a “climate regime shift” consistent with other recent studies. Climate model simulations of cloud cover are compared with the satellite observations. By improving our understanding of cloud variabilities, we hope to advance climate model simulations and lead to more accurate predictions of future climate. Building Capture Agents for the Detection of Plasmodium Lactate Dehydrogenase (pLDH) as Malaria in vitro Diagnostics Ann Chen Mentors: James Heath and Aiko Umeda Conventional protein-detection methods that identify malaria utilize antibodies for their high affinity and selectivity for their target protein; however, these antibodies are expensive and unstable, especially in the presence of thermal fluctuations. As an alternative to traditional antibody-based capture agents, more cost-effective and stable peptide-based multi-ligand protein-capture agents have been developed through the Huisgen 1,3-dipolar cycloaddition reaction and one-bead-one-compound (OBOC) peptide library screens (“in situ click screen”). In this study, a capture agent against pLDH is being constructed based on the in situ click screen strategy. We chose amino acid residues 285-295 of pLDH as a target epitope for general detection of malarial parasites since this epitope is conserved among the major Plasmodium species. We screened a linear 5-mer OBOC library of D-amino acids against the chemically synthesized epitope to isolate the anchor peptide NH2-hevwh-COOH, which showed moderate binding to pLDH in ELISA. We then screened the second OBOC library of 5-mer cyclic peptides of L-amino

14

  acids using the anchor peptide and full-length pLDH protein. The candidate biligand, hevwh-(YLGHK)cyclic, has been scaled up and is currently being tested for its affinity and selectivity for different strands of pLDH from major Plasmodium species. We expect that peptide-based multi-ligands will serve as drop-in replacements for antibodies as robust capture agents in malaria diagnostics. The Evolution of Sex Determination: Environmental Sex Determination Systems in the Nematode Bursaphelenchus okinawaensis Anthony F. Chen Mentors: Paul Sternberg and Ryoji Shinya The principal objective of this project was to characterize the method of environmental sex determination in the recently discovered nematode Bursaphelenchus okinawaensis. The project primarily consisted of growing the nematodes under various conditions, such as different temperatures, starvation, dauer recovery, length of incubation time, or cross-mating, and observing the resulting ratio of males to hermaphrodites in the nematodes. The other aspect of the project was inducing mutations in these worms by ethyl methanesulfonate mutagenesis, in the hopes of isolating physical mutants or a high incidence of males mutant. This project found that worms grown at 30oC have a significantly higher male to hermaphrodite ratio (when compared to 25oC or 20oC cultures), worms subjected to heat shock in a late L4 stage also have a higher ratio, worms recovered from the dauer stage have a significantly lower ratio, while length of culture time and starvation seem to have no effects. The other experiments are ongoing. There is not too much known about this nematode, but learning more about environmental sex determination in it will establish it as a model organism for similar, parasitic nematodes, which will hopefully address many livestock and environmental concerns in the world. Characterization of 8ANC195 Germ-Line Antibodies Courtney Chen Mentors: Pamela J. Bjӧrkman and Louise Scharf Development of an effective immunogen for a vaccine against HIV-1 has proven difficult for numerous reasons. Through longitudinal studies, it has been shown that HIV-1 viral evolution proceeds faster than the antibodymediated immune response can adapt to the changing HIV-1 antigens. However, viral entry of host cells requires conserved binding sites for host receptors. Through molecular characterizations of broadly neutralizing antibodies (bnAbs), two main regions on the HIV-1 virus have been found to be conserved: the surface envelope glycoprotein gp120 CD4 binding site and surface envelope glycoprotein transmembrane unit gp41 (Burton et al., 2012). These two regions thus define targets of vulnerability for HIV-1. Our collaborators in the Nussenzweig lab (Rockefeller University) have identified a large group of potent bNAbs against the CD4 binding site HIV-1. As a result, recent research in HIV-1 seeks to utilize structural information from these bNAbs to identify correlations amongst the bNAb structures and potent and broad HIV-1 neutralization. A major focus of the Bjorkman group is to elicit potent gp120 specific bnAbs from their putative germ-line precursor B-cells. By understanding how germ-line B-cell receptors recognize the gp120 portion of the HIV-1 viral epitope, critical contacts and differences between germline antibodies and the bnAbs derived from them may be elucidated, guiding immunogen design to elicit potent and broad gp120 binding site antibodies (Scharf et al., 2013). This project thus seeks to characterize, in structure and binding, the germ-line precursor of the bNAb 8ANC195 to provide a basis for developing immunogens that will elicit 8ANC195-like antibodies. Design and Implementation of a Voltage Regulator and PIC Microcontroller Circuit for Direct Digital Synthesizer Based Chirped Pulse Microwave Spectroscopy Oliver W. Chen Mentors: Geoffrey Blake and Ian Finneran Due to recent developments in ultra-high-speed electronics, constructing microwave spectrometers using chirped pulse (CP) technology is now feasible. In CP microwave spectroscopy, a linear frequency sweep polarizes a gaseous molecular sample, and the resultant free induction decay (FID) is recorded. Because the FID is extremely weak, an expensive low-noise microwave frequency power amplifier must be immediately connected to the excitation chamber in order to obtain a reasonable spectrum. We designed and constructed a 15V DC voltage regulator that protects against voltage spikes to shield the amplifier from overvoltage. Furthermore, we integrated a Peripheral Interface Controller (PIC) microcontroller into the spectrometer. When taking data from a CP spectrometer, many samples are averaged to calculate the final value. Without phase coherent pulses, the average will approach zero. Thus, a human controlled digital-to-analog calibrator was used to realign the phase of the pulses every second; however, with the implementation of the PIC microcontroller, simple commands can be sent to the spectrometer via the direct digital synthesizer (DDS) evaluation board. We subsequently constructed a pre-designed microcontroller circuit for automatic recalibration. Finally, we constructed and modified the entire chirped pulse Fourier transform microwave circuit from its discrete components while constructing a compact circuit box to house both the phase-lock loop (PLL) and DDS evaluation boards for ease of access. We built a cost-effective, userfriendly waveguide-based alternative to the traditional, bulky molecular beam-based prototype instrument.

15

  Developing a Dominant Vanillic Acid-Repressible Lethal System for the Suppression of Drosphila suzukii Populations Wen Min Chen Mentor: Bruce Hay Drosophila suzukii is fruit crop pest native to Asia, but is thriving in the United States. Females penetrate the skin of ripe fruits to lay eggs, creating a wound and rendering the fruit unmarketable. As there is no effective method to control this pest, genetically engineering male D. suzukii to produce unviable offspring and releasing them to interbreed with wild females would control the population and reduce crop damage. To achieve this, several constructs were created so that the presence of vanillic acid would suppress toxin production in males, and were injected into Drosophila melanogaster. The survivorship of each transformant line was tracked on both plain and vanillic acid-containing food to ascertain the efficacy of each construct. Currently, the transformation of D. suzukii has started using promising constructs shown by D. melanogaster trials. After these larvae have matured into adulthood, assessment of the constructs would be carried out, and once a suitable line has been identified, field studies may be carried out. Bio-Inspired Adhesive Polymers: Enhancing Tackiness and Wet Adhesion Using Rosin and DOPA Christine Cheng Mentors: Robert H. Grubbs and Hoyong Chung Bio-inspired adhesives have recently seen much advancement, utilizing functional groups to overcome limitations of traditional synthetic adhesives. In this project, novel four-component polymer adhesives are explored. A 3,4-dihydroxy-L-phenylalanine (DOPA) functionality is utilized to improve wet adhesion, and a rosin-containing component increases the polymer’s tackiness. Water swelling is achieved by the incorporation of acrylic acid (AA), which is strongly hydrophilic. Acrylic acid N-hydroxysuccinimide ester (AANHS) was used as a crosslinking segment by rapidly forming covalent bonds with the cross-linker, preserving the catechol moiety. An amine-terminated 3arm poly(ethylene glycol) cross-linking agent was synthesized and tested with various quaterpolymers. Variations in component ratios of the quaterpolymer will be discussed. The prepared polymers were characterized by a variety of techniques, including nuclear magnetic resonance spectroscopy, rheology tests, and tackiness tests. Identification of Gene Targets of MicroRNA125-b David Cheng Mentors: David Baltimore and Alex Steven So Recent studies have shown that overexpression of miRNA-125b induces leukemia in mice. Thus, we are interested in investigating miRNA-125b induces cancerous phenotypes in cells. We made several expression vectors that contain a combination of fluorescent proteins (GFP, TagBFP, Cerulean, E2-Crimson) and genes that suspected to encode for mRNA targets of miRNA125-b. MiR-125b is thought to regulate the expression of IRF4 to control leukemic development. To better understand the role of IRF4 in miR-125b-mediated leukemogenesis, different mutations were made within the IRF4 gene to determine which domains are functionally important for cancer. We then utilized the retroviral vector, pCL-Eco retrovirus packaging vector, to package this expression vector into a retrovirus. Then, bone marrow cells were infected with these viruses, which integrate into the host cell genome and stably express the fluorescent proteins, miRNA 125-b, and various IRF4 mutants within the cell. If the cells that over-express wildtype IRF4 but not particular IRF4 mutants can reverse the hyper-proliferation of HSPCs induced by miRNA-125b over-expression, it indicates that the specific mutated IRF4 domain is important for miR-125bmediated cell growth. Through our work, we have determined likely targets of miR-125b. Cryogenic Nanopillar Compression of Single Crystalline Body-Centered-Cubic Metals Yintong Cheng Mentors: Julia Greer and Seok-Woo Lee At micro- and nanometer scales, materials demonstrate the ability to withstand much higher amounts of stress than bulk samples. In addition, mechanical properties of materials are known to be strongly dependent on temperature. So far, most nanoscale mechanical tests have been restricted to room temperature conditions. However, we have recently developed an in-situ mechanical testing system that works at cryogenic temperatures. In order to understand the effects of temperature on the mechanical properties at the nanoscale, we fabricated BCC niobium and tungsten nanopillars with diameters smaller than 1 micrometer using Focused Ion Beam milling. We then performed uniaxial compression tests on the samples. We discuss our results based on single arm dislocation source model and examine the effect of intrinsic lattice resistance on the strength of niobium and tungsten nanopillars.

16

  Analysis of Binding Affinity of Modified Rhodium Metalloinsertor With DNA Mismatches Evelyn S. Chin Mentors: Jacqueline K. Barton and Alexis Komor DNA mismatch repair (MMR) is a system that is designed to recognize and repair errors within the DNA. However, when a cell is MMR-deficient, the DNA is not checked as efficiently or accurately, leading to the progression of MMR-deficiency and various cancerous diseases. Rhodium metalloinsertors are the key to identifying DNA mismatches within cells that are MMR-deficient, and can ultimately lead to the destruction of these MMR-deficient cells. Their binding-affinity for DNA mismatches can be easily tuned by modifying their ancillary ligands. A particular rhodium metalloinsertor, [Rh(HDPA)2(chrysi)]3+, is very successful in selectively terminating mutated cells over normal cells, but cannot detect the more stable G-containing DNA mismatches. Appending amine groups can potentially provide additional hydrogen bonds with the DNA which could stabilize the interaction between the metalloinsertor and the DNA, and may help in the recognition of G-containing mismatches. A rhodium metalloinsertor that possesses an amine-appended ancillary ligand was synthesized. The complex's mismatch recognition properties will be analyzed by running a photocleavage experiment with 5'-32P-labelled DNA and polyacrylamide gel electrophoresis. Improving Alcohol Tolerance in Saccharomyces cerevisiae for the Production of Biofuels Linda Chio Mentors: Gregory Stephanopoulos, Felix Lam, and Gerald R. Fink A limitation on biofuel, such as ethanol and isobutanol, production in yeast S. cerevisiae is alcohol toxicity. Soonto-be-published data suggest that alcohol tolerance, a multigenic phenotype, can be influenced by the ionic composition of the fermentation medium; furthermore, changes to ionic strength that improve tolerance correlate with the maintenance of neutral intracellular pH (pHi). Therefore, we hypothesize that genetic modifications strengthening pH homeostasis may lead to increased cell viability and alcohol production. To explore this possibility, our research is focused on 1) developing a high-throughput method for measuring pHi in yeast, and 2) investigating the role that membrane ion pumps play in regulating pHi. Through gene cloning and directed homologous recombination, pHluorin, a green fluorescent protein (GFP) derivative whose fluorescence properties change proportionally with respect to pH, has been integrated into yeast strains. Using these strains we can determine how different genes, first by examining membrane ion pump genes, impact pHi and tolerance. If successful, pHluorin can be used as a reporter in larger scale genetic screens to identify new components involved in tolerance. Machine Learning for Earthquake Monitoring Deepti Chopra and Shivam Mani Tripathi Mentors: Julian Bunn and K. Mani Chandy Earthquake monitoring systems consist of sensor networks that detect ground motion and fuse data from multiple sensors to warn about impending shaking and to provide continuing situation awareness as shaking from the earthquake continues. This project explored algorithms based on machine learning for detecting earthquakes by analyzing data streams generated by the inexpensive sensors used in the Community Seismic Network. The project analyzed frequency-domain characteristics of acceleration reported by the sensors and applied Support Vector Machine (SVM) tools to classify acceleration patterns in one of two categories: (1) quiet period, i.e., no shaking due to earthquakes currently, and (2) shaking due to a current earthquake. The project explored the possibility of using neural networks for classifying time-domain patterns into the same categories. The project reports on several experiments for classifying data streams into the two categories and the results are promising. Collecting and Analyzing Pluto's Light Curve for Evidence of Volatile Transport Devin Chu Mentor: Bonnie Buratti The ice distribution on Pluto's surface is naturally uneven. As a result, the amount of sunlight Pluto reflects changes as the dwarf planet rotates, causing Pluto's brightness to fluctuate when seen from Earth. Pluto's brightness can be measured over time, resulting in a rotational light curve. Pluto's rotational light curve is a first-order description of the albedo patterns on the surface. After correcting for viewing geometries, changes in the light curve can provide a direct measurement of volatile transport. This transport is associated with the sublimation of gasses into Pluto's atmosphere, leading to substantial changes in the vapor pressures of methane and nitrogen. Although volatile transport has not been directly observed on Pluto, theory predicts the phenomenon should take place seasonally. Seasonal volatile transport has been observed on Neptune's moon, Triton, and Triton shares many orbital characters with Pluto. Over the course of the summer, Pluto has been observed using the Table Mountain Observatory for multiple nights to compile a light curve. This light curve will then be compared to previous Pluto light curves, and the analysis may reveal evidence for volatile transport. These findings will provide great insight into the understanding of Pluto prior to New Horizons scheduled flyby in 2015.

17

  Progress Toward the Total Synthesis of Cyanocycline A Carolyn Cohen Mentors: Brian Stoltz and Guillaume Lapointe Cyanocycline A is a member of the tetrahydroisoquinoline (THIQ) alkaloids, a class of natural products that display potent biological activity as antitumor antibiotics. Isolated from the fermentation broth of Streptomyces flavogriseus, cyanocycline A has shown activity as both a broad-spectrum antibiotic and an antitumor agent. We propose a total synthesis of cyanocycline A involving aryne annulation and late-stage C-H activation as key steps. The aryne annulation reaction is a method of creating the isoquinoline core that has been used by the Stoltz group in the syntheses of other THIQ alkaloids. The preparation of both substrates for the aryne annulation as well as the annulation reaction itself was addressed. Future work toward the completion of the synthesis includes late-stage C-H activation to create the oxazolidine ring in cyanocycline A. Training Set Construction for Real-Time Vetting of Transients in the Intermediate Palomar Transient Factory Ming Alexandra Cong Mentor: Umaa Rebbapragada The goal of the Intermediate Palomar Transient Factory (iPTF) is real-time exploration and discovery of astronomical transients such as supernovae and variable stars. iPTF succeeds PTF which had an automated classification engine deployed for the purpose of distinguishing true astronomical transients (real) from artifacts of image subtraction (bogus). iPTF features upgrades to its data pipeline that require us to analyze the difference in feature distributions between PTF and iPTF and adjust our training sets accordingly. My report focuses on the work done towards building a balanced, representative training set for the newest classifier on iPTF, including the analysis of feature distributions, the sampling of bogus candidates and methods for mapping real examples from PTF into the iPTF data regime. AirMSPI Polarimetric Aerosol Retrieval Matthew Conroy Mentors: Olga Kalashnikova and Michael Garay The Airborne Multiangle SpectroPolarimetric Imager (AirMSPI) is an instrument developed by the NASA Jet Propulsion Laboratory to image clouds and aerosols in the atmosphere. Information about the size and shape of aerosols is important in order to model their effect on Earth’s climate. Of particular interest is how accurately we can retrieve this information. In order to do this, first we build a look-up table containing the results of radiative transfer simulations for theoretical aerosol particles. Then we find the best match in the table for the real particle data by minimizing a cost function. Once this table is complete, we will be able to perform retrievals using data taken over the San Joaquin Valley in central California. The retrieved parameters will then be compared with data from ground-based aerosol monitoring stations to assess the validity of the results. The end goals for the project will be to not only see how accurately retrievals can be done using this method, but also to theoretically quantify how precise these retrievals can be. Hospital and Laboratory Use of Geobiological Techniques to Investigate Cystic Fibrosis Elise S. Cowley Mentors: Dianne K. Newman and Sebastian Kopf Cystic fibrosis (CF) is a genetic disorder that involves a defect in the transport of sodium and chloride across the epithelium. CF affects the respiratory system greatly, but also impacts every other organ system in the body. CF is characterized by malfunctioning lungs that are the perfect environment for pathogenic bacteria such as Pseudomonas aeruginosa, Staphylococcus aureus, and Stenotrophomonas maltophilia. It is not clear how the microbial communities inside the CF lung survive or how the environment changes over time. The microbial communities have not been studied in situ, but this is critical to understand and treat the disease. Geobiology approaches are used to study microbes in all environments and this makes the techniques uniquely capable of probing the CF microbial environments. Sputum samples were collected from CF patients at Children’s Hospital, LA and a microelectrode was utilized to investigate the oxygen profile of the microbial environment over time. In tandem with hospital studies, development of a model system to establish growth rates with a chemostat was performed in the laboratory. Improved Functionality for the Mobility Mechanics Modeling Toolkit (M3Tk) Jack Craft Mentor: Rudranarayan Mukherjee M3Tk is an open-source mechanics and multi-body dynamics based robot design tool used to develop and validate designs and mechanics when interacting with human-made or natural terrain. Terrain data is collected from a sensor that scans an environment. These data tend to be ‘fuzzy’ and inevitably contain some gaps in information. Functionality was added to M3Tk to automatically smooth these height map data and fill gaps caused by the sensor

18

  missing small areas and recording no value. M3Tk reads in a text file that initializes the dimensions and properties of the multi-body robot or vehicle. This text file was in a format specific to M3Tk and was in need of a more general, accessible, and readable format. The JSON file format was determined to be the best for the purposes of the project, so a program was written within M3Tk to convert the old input files into new JSON formatted files. Functionality was also added into the M3Tk system to read in the JSON files and initialize properties as it would have with the old file format. These new functionalities will streamline the research process and add accessibility in the future so that new models can be read in without specific knowledge of the old input format. MATLAB, M3Tk, and c++ code and compilers were used to complete these tasks. Transformation of Amateur Jupiter Images Tom Craig Mentor: Glenn Orton In 2016 and 2017, the Juno mission will provide new information about the distribution of water and ammonia gas in Jupiter’s atmosphere via the MicroWave Radiometer (MWR) experiment. To better associate this information with what can be gathered at higher altitudes, it will be beneficial to compare it with other experiments, such as Juno’s JIRAM experiment. Because Juno’s CCD camera will not necessarily image the same locations covered by the MWR and JIRAM, the Juno project will be soliciting images from professional and talented amateur astronomers. This project designed the system through which these amateur images will be obtained and then converted to a format comparable to images from professional telescopes. These comparisons yielded data that helped associate Jupiter’s cloud types with colors and reflectivities in the 5-micron range. Probing the Mechanics of Fault Roots: A Search for Tectonic Tremors in Southern California Tudor Cristea-Platon Mentors: Jean-Paul Ampuero and Justin R. Brown Tectonic tremors are long duration, noisy seismic signals induced by natural sources at the deep base of active faults. The phenomenon was discovered in the last decade and has been since observed in a variety of tectonic settings globally. However, spontaneous tremor activity has not yet been reported in southern California. The mechanics of tremors is still not entirely understood and observational seismology is offering an unique window into its origins. Following the deployment of seismic arrays in the San Jacinto area two years ago, it became apparent that a number of events identified as tectonic tremor were in fact ground vibrations induced by freight train traffic in the Coachella Valley. This has raised issues about the possibility of detecting tremors in Southern California. Our goal is to identify significant differences between the spectra of tremor and that of trains, to enable the design of an automatic discriminator and the objective detection of tectonic tremors in this region. Superhydrophobic Nanotubes to Increase the Efficiency of Solar Desalinators by Reducing Salt Buildup Melissa Cronin Mentors: Mory Gharib and Bradley Lyon Solar desalination provides a low cost method of providing clean water without the use of critical resources such as fossil fuels. Due to this, the use of solar desalinators in third world countries with little or no resources is attractive. However, water is required to clean the solar desalinator due to a periodic buildup of salt crystal – reducing the efficiency of the solar desalinator and making its use impractical in low resource situations. In this study we aim to assess the reduction of salt crystal buildup by using superhydrophobic nanotubes. The carbon nanotubes are compared against a range of standard materials to assess the relative reduction of salt buildup by using carbon nanotubes. Carbon nanotubes are a good candidate for use in solar desalinators as they have a superhydrophobic property – they can repel water so strongly that when immersed in water, they form an air film to separate the superhydrophobic carbon nanotubes from the water. It is hypothesized that this air film will significantly cut the rate of salt buildup improving the efficiency of solar desalination operation. Airborne Spectropolarimeter Image Analysis and Data Processing Christopher Culpepper Mentors: Olga Kalashnikova and Michael Garay Aerosols, small particles suspended in the atmosphere, are currently one of the largest sources of uncertainty in global climate models. In order to understand the climate forcing of aerosols, global coverage of aerosol optical properties is needed. The Airborne Multiangle SpectroPolarimetric Imager (AirMSPI) is an instrument capable of delivering this coverage using remote observations of intensity and linear polarization of light at multiple wavelengths. We used data from AERONET sites in the San Joaquin Valley of California to inform the development of a lookup table based retrieval for aerosol optical properties. When complete, we hope to demonstrate the potential for this type of retrieval to be applied on a global scale.

19

  Determination of Microbial Viability Using PMA-qPCR Halley Darrach Mentors: Adrian Ponce and Nicholas Fingland PMA-qPCR is a laboratory technique that can be used to identify viable microbes by employing the use of propidium monoazide (PMA), a DNA-intercalating dye, and quantitative polymerase chain reaction (qPCR). The current model of PMA-qPCR operates under the assumption that PMA is only capable of entering membrane-compromised cells, where it irreversibly cross-links to DNA, precipitates it, and makes it unavailable for amplication via qPCR. However, the exact mechanism behind PMA’s entry into the cell and its interaction with genetic material is not well understood. To better understand PMA’s capabilities, we have tested PMA-qPCR’s effectiveness on bacteria using various inactivation methods to determine what role membrane intactness has on its behaviour and viruses to determine how PMA interacts with RNA. As well, we have examined the effect PMA has on enzyme binding and processivity using endonculeases and exonucleases. Our results suggest that the current model behind PMA-qPCR inhibition is incomplete, in that 1) PMA does not determine viability on the basis of membrane permeability and 2) rather than precipitating the entirety of the DNA, PMA also inhibits enzyme binding in soluble DNA. Shadow Edge Detection Through Illumination Invariance and Adaboost Shival Dasu Mentors: Pietro Perona and Michael Maire It is considered possible to classify ground shadow edges using a manageable set of training examples. However, general shadow edge classification is still an open problem. Our objective was to use the current state of the art— illumination invariant images—along with other feature channels to extend the scope and improve the robustness of shadow edge classification. The Adaboost learning algorithm was trained on the mean-shifted illumination invariant images, hog descriptors, variable scale edge detection, and texton features and achieved an edge type classification accuracy of 90%. However, the illumination invariant method fails on images that have been exposed to the standard post-processing performed in the JPG transformation. Gaussian regression was performed on the Jpg images to restore image quality. Although the entropy curve for the illumination invariant image was smoothed, the algorithm still fails on the majority of dataset Jpg images. Although our algorithm produces useful results for linearized, demosiaced, raw images, compatibility with mainstream image formats is a crucial next step. Using Organic Monolayers to Seed Growth of Stable and Protective Films by Atomic Layer Deposition on Silicon Anne Davis Mentors: Harry Gray and Chris Roske The Holocene period has afforded a stable climate thereby allowing humanity to flourish, but anthropogenic climate change threatens sustained long-term economic development by disrupting the climate’s quasi-equilibrium. In order to combat this looming threat, an alternate source of energy must be developed in order to reduce human CO2 production. Splitting water is one suggested alternative. In this scheme semiconductors absorb sufficiently energetic photons from the sun which are then directed to split water (2H2O  2H2 + O2). This is typically a chemically demanding process, and stable semiconductors under these conditions tend to be large band gap semiconductors. However, small band gap semiconductors such as silicon are necessary to achieve the highest levels of solar efficiency. In this project, we functionalized the surface of the silicon semiconductor using a mixed methyl aldehyde Grignard reagent and then grew protective layers of titanium dioxide in order to eliminate deleterious metal silicides and the undesirable self-passivation of silicon under oxidative conditions. Materials were characterized by infrared spectroscopy, surface recombination velocity, XPS, water contact angles, ellipsometry, and atomic force microscopy. Analysis of New Transition Metal Composites for Cathode Materials for Advanced Batteries Victoria K. Davis Mentor: William West Improved battery systems are needed for future space applications. Materials that can produce batteries with lower mass, greater capability, and longer life are of particular interest to this project. New transition metal composites for cathode materials present opportunity for improved battery lifecycle, specific capacity, and voltage. Following the preparation of cathode precursors Li2MnxM1-xO3 (M= Mn, Co, Cu, Zr, Nb), molten salt synthesis of a 1:1 molar ratio LiMn0.33Ni0.33Co0.33O2 : Li2MnxM1-xO3 (M= Mn, Co, Cu, Zr, Nb), and aspiration of the suspension, a variety of cathode materials were prepared and analyzed. Coin cell studies were performed to determine the specific charge and discharge capacities and cycle life. Three electrode cell studies were carried out to collect cyclic voltammetry data to characterize oxygen evolution during first cycle. Further investigations will continue to determine the effects of transition metal substitution and 2% AlPO4 coating of cathode material systems on battery performance.

20

  Social Judgments of Faces Along Different Dimensions Poonim Daya Mentors: Ralph Adolphs and Shuo Wang In everyday life, people constantly form judgments of others based on purely facial features. The chief objective of this study is to determine how the brain picks up subtle changes in facial structures by investigating perception along different facial dimensions. Established software was used to create sets of faces that vary along one dimension to another: “happy vs. fearful”, “anger vs. disgust”, “anger vs. fearful”, “male vs. female” and more complex social dimensions, like dominance and trustworthiness. We presented subjects with faces and asked them to judge the face as well as rate their confidence while recording their choices, reaction times, and confidence levels. In general, subjects were able to pick up subtle changes in facial structures regarding emotion and gender, based on the relationship between the strength of an emotion and the percentage of times that emotion was chosen. Continuation of this study could involve investigating how people look at faces along different dimensions using eyetracking and the brain structures involved in facial judgments using fMRI. Development of Linear Image Sensor to Probe Magnetized Plasma Jet Formation Maxwell G. De Jong Mentors: Paul Bellan and Vernon Chaplin The Caltech coaxial magnetized plasma jet experiment involves dynamics on sub-microsecond timescales, so fast diagnostics are needed. An expensive fast framing camera is currently used to study jet evolution. To take additional images of jet evolution from different angles, we developed an inexpensive photosensitive detector built around a high performance linear image sensor, the Dynamax ELIS-1024. The linear imager has an array of 1024 pixels with a variable exposure time as small as 10 ns. We designed a circuit to trigger the ELIS-1024 and amplify the output and mounted the circuit on a printed circuit board. The final system will be capable of taking 1D photographs of the plasma or serving as a detector on a spectrometer. Emission spectroscopy identifies the species in the plasma and enables electron temperature estimation through emission line intensity ratios and ion temperature estimation through Doppler broadening. Stark broadening may also be detected. The sensitivity of the detector to these features will be reported. The sensor is also being evaluated for use with a 1D coded aperture imaging system in a separate project. Testing Large Scale Homogeneity Using the SDSS-III Baryon Oscillation Spectroscopic Survey Daniel DeFelippis Mentors: Olivier Doré, Michael Seiffert, and Roland De Putter Given that structure exists in the universe on many scales, including from single galaxies to filaments, it is not immediately obvious that there is a scale at which the density fluctuations become statistically insignificant; that is, a scale at which the universe becomes homogeneous. The existence of such a scale is an important assumption in most current cosmological models, so confirming or refuting it is a fundamental measurement. To investigate the existence of this scale, large redshift surveys are ideal. In the summer of 2012, the Sloan Digital Sky Survey released the first spectroscopic data from its Baryon Oscillation Spectroscopic Survey (BOSS), containing the redshifts of over 300,000 constant-mass galaxies from redshifts of 0.4 to 0.8. Using the JPL supercomputer Zodiac the data from the survey was processed to convert the redshift and weighting data to weighted Cartesian coordinates. Then, using the counts-in-spheres method, the pairwise distances between galaxies were binned in equally spaced distance ranges. By computing where the binned distances begin to increase with the cube of the distance, a value for the scale of homogeneity in the universe was found. Modification by Atomic-Layer Deposition of TiO2, and Spectroscopic and Photoelectrochemical Characterization of n-GaP (111) Surfaces Natalie DeFries Mentors: Nathan S. Lewis and Victoria Tan n-GaP (111) crystals were chemo-mechanically polished, and etched using the following series of solutions: H2SO4, or Br2-methanol and KOH, or Br2-methanol, KOH, and HF. 50, 200, and 500 atomic-layer deposition (ALD) cycles of H2O and TDMAT were performed to deposit varying thicknesses of TiO2 on GaP wafers in order to passivate any surface defects and protect the samples from oxidation. These surfaces were characterized using optical microscopy, ellipsometry, atomic force microscopy, X-ray photoelectron spectroscopy, and electrochemistry. Variously treated GaP samples were made into electrodes tested in an electrochemical cell containing 1 M LiClO4, 10 mM ferrocene, and 0.008 mM ferrocenium in acetonitrile. With 50 ALD cycles, the GaP surface was not completely covered by TiO2. With 500 cycles ALD, the surface was covered, but GaP electrodes exhibited no photovoltage or -current under illumination. This project needs to work on making a thinner, more consistent TiO2 layer that completely covers the GaP surface, but still allows light to reach it.

21

  Involvement of Ikkα Phosphorylation in Rett Syndrome Daniela Díaz-Infante Morales Mentors: Paul H. Patterson and Ali Khoshan Rett syndrome (RTT) is an X-linked neurodevelopmental disorder primarily affecting females. The syndrome is caused by mutations in methyl-CpG-binding protein 2 (MeCP2), which disrupt gene expression and neural development. MeCP2 is phosphorylated by Ikkα at S274, which is in the transcriptional repression domain, and at S326 near the C-terminus. Ikkα also phosphorylates the transcriptional co-repressor SMRT. Using immunoprecipitation, we are investigating the effect of Ikkα on MeCP2/SMRT interaction. Preliminary results suggest that S274A MeCP2, a change that blocks phosphorylation by Ikkα, can bind to SMRT while S274D, a change that mimics phosphorylation by Ikkα, cannot. Thus, Ikkα may regulate MeCP2 activity by influencing its interaction with SMRT. Craniofacial Development in Lampreys Monisha Dilip Mentors: Marianne Bronner and Stephen Green In vertebrates, the pharynx is innervated by the epibranchial ganglia, derived from the epibranchial placodes. Placodes are thickenings of epithelium that are found only in vertebrates. Little is known at this point how epibranchial placodes and epibranchial ganglia, and so by examining them in basal vertebrates like lamprey, we can infer traits about placodal development in primitive vertebrates. In jawed vertebrates, the gene Phox2 is required for epibranchial placode specification, but the regulatory elements controlling expression of this gene are unknown. Multiple upstream elements were identified as potential regulatory regions in chicken but did not drive robust expression in lamprey epibranchial placodes, suggesting that either those regulatory elements are not conserved or that the regulatory DNA was incomplete. A 50 base pair element upstream of Phox2 within the lamprey genome had 93% sequence identity to a noncoding sequence upstream of chicken Phox2b. I cloned this element, termed L-ECR4, out of the lamprey genome and into a GFP reporter construct. I then injected it into the developing lamprey embryo. Preliminary expression of GFP indicates that L-ECR4 might drive native Phox2 expression in the epibranchial ganglia and in axons rising from the dorsal root ganglia. Further injections and better characterization of the neural anatomy of the lamprey are ongoing. DESDynI Solar System Visual Data: Automating Data Into Excel Using Python With MySQL Connections Edward Djrbashian Mentor: Paul Andres The purpose of this project was to develop an algorithm that would automate the process of plotting descriptions, website links, and identification numbers of DESDynI visual data in to an Excel sheet. The crucial essentials of Python programming and its connections with MySQL, various python modules, and regular expressions were used to sort and format the desired excel sheet. The developed algorithm was useful in automating a lengthy manual process of organizing database data. Ideally, this excel sheet could be updated by DESDynI scientists and uploaded back to the MySQL server for global access. The Python code was generalized with methods in a common object class so that future programmers could call them in their programs. Optimization of Galaxy Modeling Software Sean Dolan Mentor: Peter Capak Current methods for determining the structure of galaxies based on the redshifts detected by telescopes for a multitude of photometric bands have been determined to be inadequate for larger sets of data that will be gathered by future telescopes, such as ESA/NASA’s Euclid. Testing has revealed that the bottleneck most frequently lies in the chi-squared fitting loop used to determine the most accurate model as compared to the fluxes measured. We have created a vectorized, parallelized software library to optimize this process, as well as a second library utilizing generalized processing on graphics processing units (GPGPU) code to allow users to make use of all possible hardware configurations. Reduction and Analysis of Jupiter’s Near-Infrared Data Esther Du Mentor: Glenn Orton Near-infrared observations of Jupiter have been made by my mentor for over a decade, primarily in the form of images in the 1-5 micron range using facility instrumentation at NASA’s Infrared Telescope Facility (IRTF), NSFCam and its refurbished version NSFCam2. The data from these instruments enables a continuous monitoring of cloud activity in Jupiter over many years and has proven essential for the analysis of cloud properties. These have included properties of clouds associated with Galileo, New Horizons, tumultuous changes in Jupiter’s atmosphere, and assessment of the debris associated with the asteroidal impact of 2009. Unfortunately, most of these data sets remain unreduced except for the “special events” in the publications noted above. Absolute calibration of the data

22

  has often relied on joint observations of nearby standard stars, but they are not always available on partially cloudy nights. Thus a more general-purpose calibration method was created to calibrate data from both photometric and non-photometric nights. Once calibrated, these data can then be archived with NASA’s Planetary Data System (PDS) and examined for long-term changes in the atmosphere. Investigating the Roles of LPS-Inducible MicroRNAs in Immune Function Xiaomi Du Mentors: David Baltimore and Arnav Mehta MicroRNAs are short noncoding RNA molecules that regulate gene expression by decreasing target protein levels. Every microRNA has several hundred targets, allowing each one to have a plethora of functions. This project focuses on three microRNAs induced by bacterial protein binding to macrophages, a type of immune cell, and how they impact the innate and adaptive immune systems. Two of these microRNAs, miR-146a and miR-155, are essential to the NF- κB pathway, while the third, miR-132, plays a role in B-cell development and immune function and the maintenance of hematopoietic stem cells (HSCs). miR-155 and miR-146a are both induced by NF- κB, but they regulate NF- κB signaling in opposing ways. We hypothesize that their contrasting effects on the NF- κB pathway cause the relative amounts of each of these microRNAs to confer different functional properties to immune cells and have thus developed a technique to quantify microRNA expression in single cells using hybridization chain reaction (HCR) to investigate this phenomenon. In this study, we validate mRNA and microRNA detection in bone marrow and spleen samples from mice. In addition, we explored the effects of miR-132 on B-cells and HSCs by overexpressing miR-132 in mice bone marrow. Dynamic Bayesian Networks for Spatiotemporal Sports Pattern Recognition Sólrún Halla Einarsdóttir Mentor: Tracey Ho Precise tracking data from sports provides a rich and interesting source of data on which to develop spatiotemporal pattern recognition algorithms. Using a basketball data set with hand-labeled occurrences of several types of screens, namely “pick and roll”, “pick and pop” and away-screens, our aim is to develop machine learning algorithms to automatically distinguish among them. Dynamic Bayesian networks such as hidden Markov models have been used successfully for pattern recognition in other types of sequential data. We explore their use in classifying the different types of screens in our data. Investigating the Neural Mechanisms of Anthropomorphizing Animals Emily E. Ellsworth Mentors: Ralph Adolphs and Bob Spunt Anthropomorphism, the attribution of human characteristics to non-human animals and objects, is a very common human behavior. While there has been some behavioral research investigating the anthropomorphism of animals, little is known about the neural mechanisms involved. In this study, we induced both explicit and implicit anthropomorphism in typical adults while they underwent functional magnetic resonance imaging (fMRI) to investigate the brain regions involved in anthropomorphism. Participants passively observed photographs of faces of humans, non-human primates, and dogs; in a subsequent task, they answered questions about the behavioral attributes and mental states of the humans and animals in the same photographs. Preliminary results from a small subject sample suggest that higher order visual processing of animal faces is similar to that of human faces, but that higher cortical processing potentially associated with social reasoning and memory may be different. As the summer concludes, we will continue to gather data from more participants to increase the robustness of this finding. Constraining the High-Energy X-Ray Continuum Cut–Off Energy of MCG-5-23-16 Clarke Esmerian Mentors: Fiona Harrison, Eric Bellm, Felix Fuerst, Mislav Balokovic, and Dominic Walton Active Galactic Nuclei (AGN) are luminous objects characterized by intense emission concentrated in the center of a host galaxy. They are thought to be powered by the accretion of matter onto a supermassive black hole. A power law continuum in the high-energy X-ray band is characteristic of AGN spectra. It is hypothesized that this emission is caused by the Compton scattering of light from an accretion disk off of a plasma corona. The properties of this corona have previously been poorly constrained. With data from the Nuclear Space Telescopic Array (NuSTAR), we have observed a high-energy cut off at ~75 keV in the continuum of active galaxy MCG-5-23-16. This will allow us to constrain the temperature and density of the X-ray emitting corona.

23

  Design of a Thermal Vacuum Test Chamber for the Autonomous Assembly of a Reconfigurable Space Telescope Project and Accompanying Thermal Analysis Erin Evans Mentors: Sergio Pellegrino, Keith Patterson, and Kristina Hogstrom The AAReST Project aims to demonstrate the feasibility of a low-cost telescope design utilizing many small independent spacecraft, each outfitted with its own mirror that can be reconfigured into a single large segmented aperture. To successfully produce a reconfigurable telescope, each mirror will be made deformable with thin glass wafers coated with a reflective metal surface made flexible with multiple piezoelectric membrane and electrode layers. To verify the functionality of these mirrors as well as their actuators and other electronic components in space-like conditions, I designed a thermal vacuum chamber capable of cooling the mirror package in a vacuum to -50 C using multiple Peltier coolers in conjunction with a water-cooled heat sink. A thorough thermal analysis of the design was then performed to ensure the desired temperature could be achieved within one hour. A thermal analysis of the entire telescope structure was also performed in order to verify previously obtained thermal analysis results and resolve conflicts between the expected thermal profile of the deformable mirrors and the prediction from the analysis. Mechanical Properties and Applications of Metallic Glass Microtrusses and Nanopillars Boyu Fan Mentors: Julia R. Greer and David Chen Bulk metallic glass (MG) is a novel engineering material, featuring high strength, large elastic strain, and corrosion resistance. Here, we present two studies of MG structures. First, we simulated the wicking ability of MG microtrusses, for their potential use in field-emission electric propulsion (FEEP) thrusters that are being developed at JPL. These FEEP thrusters require a method to deliver the liquid indium propellant to the thruster emitter. The non-reactivity, high stiffness, and high recoverability of MG make it ideally suited for this task. COMSOL Multiphysics was used to simulate the fluid flow of indium through the truss structure. Various geometries were analyzed in order to determine optimal geometries to maximize capillary flow. The second structures studied were MG nanopillars. A major drawback of MG is its tendency to catastrophically fail when loaded in tension. Here, we fabricated NiP MG nanopillars via electroplating onto an electron-beam lithography patterned polymethylmethacrylate template. Both un-notched and notched pillars will be loaded in tension using SEMentor (an in situ SEM and nanoindenter) to characterize failure methods, study size effects, and determine if any insensitivity to defects exist. Our studies will lay a foundation for further investigation into the applicability of metallic glasses. Sensor Suite for an Underwater Ice Rover Sandra Fang Mentor: John Leichty We are concerned with the study of long term-changes occurring at the ice-water interface in Arctic permafrost lakes, and specifically with the study of methane seeps that occur in these lakes. We have developed a compact sensor suite for an under-ice rover that navigates underwater to sample environmental data. The sensor system is designed to endure long periods of submersion, and is composed of eight different sensors that measure temperature, hydrostatic pressure, pH, electric conductivity, dissolved oxygen and methane content, light intensity and color spectrum. Data is stored in an SD card along with a time stamp, and can be transmitted to a host station via a tether. The system is developed using the Arduino programming language and an Arduino Mega board. To decrease the suite’s size, we have designed and printed a custom circuit board containing all associated sensor circuitry. We have also designed a compact, low power methane sensor that can easily be integrated into our sensor suite, as commercial sensors prove to be too bulky and costly for our applications. The sensor uses a silicone filter to separate the dissolved methane from water, which allows for methane to be detected by an inexpensive semiconductor based gas sensor. It is our hope that this rover and its sensing system will lead to novel mobile sensing applications in other similarly extreme environments on Earth. Riches to Rags: Do NFL Professionals Make Poor Financial Decisions in Their Short High-Paying Careers? Zachary R. Fein and Joshua J. Kim Mentors: Colin Camerer and Kyle Carlson A pillar of neoclassical economics is the life-cycle savings hypothesis (LCSH), which analyzes intertemporal spending patterns and predicts that individuals smooth their consumption profiles over the course of their life. Traditionally, these LCSH models have assumed that individuals are perfectly rational and have abundant willpower. In this project, we update these unrealistic psychological assumptions by analyzing the financial decisions made by NFL athletes. We choose NFL athletes as our sample set due to their highly-variable income

24

  stream and uncertain career length. Unlike previous studies of NFL players, which utilized survey data, we create our own datasets detailing salary histories; injury records; game logs; and personal files such as address history, bankruptcies, crimes, and liens. Using our data, we hope to update current LCSH models with psychological parameters to create a more realistic model of intertemporal decision-making. Using RKSML Telemetry Data to Automate LightWave Animations Aaron Feldman Mentor: Paul Andres The Solar System Visualization team makes animations of the Mars rover Curiosity’s activities. The goal of this project was to develop software to animate the rover’s movements based on telemetry in the RKSML file format. Output animations were encoded in the LightWave Scene (LWS) file format. After studying animated models of the rover, specifications of the actual rover, the RKSML format, and the LWS format, software was written to modify a template LWS file into a nearly complete encoding of the rover’s movements, based on an input RKSML file. Thus, the goal of this project was accomplished. The software developed could be used to produce rover visualizations more easily and accurately, and these visualizations could then be used to increase public knowledge about mars exploration. Direct Digital Synthesis Chirped-Pulse Fourier Transform Microwave Spectroscopy Data Parsing Jerry Feng Mentors: Geoff Blake, Brandon Carroll, Brett McGuire, and Ian Finneran Microwave spectroscopy is an incredibly powerful and essential tool in the field of chemistry. With a molecule’s precise rotational spectra in hand, the chemical formula, the molecular structure, and even the specific isotopologue (that is, HCN versus DCN or H13CN, for example) can be determined. Though a relatively old technique, microwave spectroscopy is still continuing to improve its speed and resolution. A recent development is the integration of the “chirped-pulse” (a microwave pulse lasting less than one microsecond), which results in a spectrometer 4000 times faster than already existing cavity-based FTMW spectrometers. To create an affordable Chirped-Pulse Fourier Transform Microwave (CP-FTMW) the Blake Lab utilizes a Direct Digital Synthesis (DDS) chip as the waveform generator. With the spectrometer complete, we now need to work with the data. Because DDS chips make bandwidth sweeps centered on a Local Oscillating (LO) frequency, the data of the spectra for the molecule are jumbled; we can’t tell which peaks belong to the upper band (UB) and which belong to the lower band (LB). Therefore, scripts were written to categorize peaks as UB or LB, to cut peaks where they do not belong, and stitch together the whole, true spectra. These scripts were placed into an application in order to facilitate ease of use. This new application will allow users to move from data collection and actually start peak assignment quicker. Examining Photometric Orbital Modulations in Kepler Transiting Planet Candidates Tara Fetherolf Mentors: Heather Knutson, John Johnson, and Avi Shporer Photometric light curves of stellar binary systems and star-planet systems reveal critical information pertaining to the systems’ properties. Most notably the radii of the two components are measured using photometry. Traditionally the masses have been measured using radial velocities in the spectra, but telescope time for high resolution spectroscopy is expensive. However, recent studies have shown that masses can be measured from the photometric light curve as well when sinusoidal modulations observed at the orbital period of the system are understood. These periodical modulations are known as beaming, ellipsoidal, and reflection effects. We are using the high precision photometric data taken with the Kepler space telescope and an automated program to extract sinusoidal signals at the known orbital period, specifically for Kepler transiting planet candidates. While several systems have shown to have sinusoidal orbital signals, many are related to stellar activity and are not necessarily induced by the orbital modulations of the two bodies. We turn focus to systems with significant orbital modulations caused by the beaming, ellipsoidal, and reflection effects to further understand their nature. Earth‐Abundant Sulfide Minerals as Catalysts for the Photochemical Reduction of Carbon Dioxide Katherine J. Fisher Mentors: Harry B. Gray and Paul J. Bracher This research investigates the use and efficiency of metal–sulfide catalysts for the photoelectrochemical reduction of carbon dioxide. One of the most fundamental questions of how life began is how inorganic and organic molecules were converted into biologically functional ones. In particular, reduced carbon atoms are essential components of amino acids and nucleotides, the building blocks of life. When irradiated, certain metal sulfides (e.g., ZnS) have been shown to possess sufficient energy to carry out the reduction of carbon dioxide to species such as HCOO− and CH3COO−. Our work focuses on the development of a simple electrochemical method that uses chronoamperometry to screen small samples of metal-sulfides for activity in the photochemical reduction of carbon(IV) species. The full

25

  range of sulfide semiconductors that can mediate these reactions has never been studied—especially binary, ternary and higher-order mixtures of metals (e.g., 1:1:1 ZnS:MnS:FeS). Promising candidates from our screening method were selected for scaled-up reactions that were analyzed by NMR to determine the yield of formate produced. Study of Four-Dimensional Shower Profiles for a Picosecond Time of Flight Detector in CMS at the Large Hadron Collider Cedric W. Flamant Mentors: Maria Spiropulu and Adi Bornheim For the High Luminosity upgrade of the Large Hadron Collider (HL-LHC), the Compact Muon Solenoid Experiment (CMS) is currently studying various upgrade options (CMS Phase II Upgrade). HL-LHC involves raising the rate of collisions (luminosity) by a factor of 10 beyond the LHC's original design value, posing the challenge of an increased rate of collisions per bunch-crossing (pile-up). In high pile-up conditions, it is difficult to assign final state particles to individual collisions. Studies have been conducted on the use of time of flight measurements in the electromagnetic calorimeter (ECAL) of CMS to optimize this assignment. Due to the complex time structure of the hit patterns left by detected particles in the ECAL, defining a viable arrival time requires an understanding of the time and space evolutions of scintillation showers. Through Geant4 full simulation, this project profiles the time distributions and hit patterns of various particle detections with the aim of optimizing the geometry and granularity of a possible time-of-flight detector. In particular, the two distinct approaches of recording the time of the first hit versus recording a time statistic of the shower distribution are compared in regards to their idealized inherent efficiencies and precisions. Synthetic Logic Circuits Using RNA Aptamers Juan M. Flores-Quijano Mentors: Richard M. Murray and Jongmin Kim Synthetic biology involves the design of tools and biological systems that are not found in nature with the purpose of obtaining benefits through an ecological and sustainable way. Plenty of “biocodes” have been generated to program functions in cells, most based on logic gates resulting in delayed computation of a given input. This project aims to demonstrate that synthetic systems can be regulated by employing aptamers, small RNAs with specific sequence and folding that possess the ability to recognize target molecules of interest with high specificity and affinity. In our genetic circuit we implemented an aptamer with the capability of binding to the T7 RNA polymerase (T7 RNAP) and inhibiting its function, intended to allow the transition from one cellular state to the next with smaller delays than bioprogramming only with transcription factor, given that RNA molecules are faster produced and degraded. So far, we successfully emulated in a transcriptional-translational (TX-TL) breadboard with very similar conditions than those of E. coli that the aptamer can be used as a fast-responsive temporal switch to regulate T7 RNAP driven circuits, although in vivo experiments are still required. Additionally, we obtained two variants to the original aptamer that showed increased capability of inhibiting the T7 RNAP through TX experiments, which seems to be very promising for in vivo applications. An Explanation of California Voting: Agricultural Politics and Ballot Initiatives Maxim Foster Mentor: D. Roderick Kiewiet In 2008, California voters were presented with Proposition 2, a bill prohibiting “the confinement of farm animals in a manner that does not allow them to turn around freely, lie down, stand up, and fully extend their limbs.” Proposition 2’s support and opposition were funded almost equally yet the proposition passed by a wide margin. In 2012, California Proposition 37 aimed to change food labeling standards by requiring labeling on raw or processed food made from genetically modified organisms. The opposition spent 81% more than supporters and the measure lost by a small margin. My research explores why similar initiatives with similar actors had contradictory results. This analysis is framed by treating these political actors in light of microeconomic theory as profit-maximizing entities and identifying discrepancies between their optimal, influence-maximizing behavior, and their actual behavior. The differences and similarities between these propositions are striking, and a discussion on how and why characteristics of individual political actors influenced each campaign and result may shed light on how groups maximize influence and vested interests influence policy. Development of a Robotic Laser Tracker Metrology System for the CCAT Telescope Edward Fouad Mentors: Steve Padin and Ross Williamson CCAT is a 25-meter diameter submillimeter telescope that will measure the history of star formation in the universe. The telescope has an active surface of 162 actuated segments, which will constantly adjust the surface profile to correct thermal and gravitational deformations. The initial alignment of the segments will be achieved using a laser tracker, requiring that a retroreflector target be placed at approximately 1000 locations on the surface. We have designed, prototyped, and tested a robot to transport the retroreflector and position it

26

  appropriately. As the robot navigates along the telescope's primary surface, it uses its heading and tilt to determine its position within a few centimeters. At each location, the retroreflector is oriented towards the laser tracker and then set down on the surface. The distance and angle measurements collected by the laser tracker allow the surface profile error to be measured and adjusted to 50 microns. This metrology system results in a surface that is accurate enough for a final wavefront measurement using a celestial source. Using Second Harmonic Generation to Study the Magnetic Ordering of 5d Transition Metal Oxides Neelay Fruitwala Mentors: David Hsieh and Darius Torchinsky In this project, we explore second harmonic generation (SHG) as a technique for studying the general phenomenon of magnetic ordering in crystals. SHG arises because the nonlinear electric or magnetic polarizability in a crystal generates light at double the frequency of any incident light. In our experiments, we shine light onto thin films or bulk single crystals at oblique incidence and measure the intensity of emitted SHG light as the sample is rotated about its normal. The pattern produced by this light can reveal the lattice and electronic (including magnetic) symmetries of the crystal. In order to extract these symmetries based on SHG data, I have developed a program that can simulate the SHG data for any given lattice or magnetic point group. I used this program to help resolve the lattice and magnetic symmetries of various complex oxides including VO2 thin films and Sr2IrO4 bulk crystals which are not amenable to study using the conventional neutron diffraction technique. Modeling the Debris Disk of HD107146 at Sub-Millimeter Wavelengths Constance B. Fu Mentors: John M. Carpenter and Luca Ricci At present, HD107146, a star similar to our Sun, has the brightest debris disk at sub-millimeter wavelengths of any other known G-type star. With the recent building of the Atacama Large Millimeter/sub-millimeter Array (ALMA), the most advanced radio telescope in the world, we are now able to image the debris disk of this young, face-on solar analog at higher sensitivity and angular resolution than in the past. With the ALMA data, we hope to better quantify the star's debris disk, identify any asymmetries in the spatial dust distribution that may indicate the presence of planetary systems, and examine the possibility of a multi-ring debris disk. Using equations and other information from the literature, the star's debris disk was modeled with several free parameters such as inclination, position angle, and inner and outer radius. Subsequently, emcee, a Python-implementation of Goodman & Weare's Affine Invariant Markov chain Monte Carlo (MCMC) Ensemble sampler, was used to sample the parameter space and minimize χ2, thus obtaining the best fit parameters. Inferring Thermal and Mechanical Properties of Celestial Bodies’ Regolith Using (Simple) Low-Tech Tools Joaquin Gabaldon Mentors: José E. Andrade and Alex X. Jerves The rovers that are sent to Mars run into various complications, which include part wear and weathering, but entrenchment in too loose of a soil can halt a mission permanently. This occurs when a rover encounters a surface that is extremely loose and unstable, to the point where the rover sinks into the surface while attempting to traverse it, and cannot free itself. To prevent this, it is necessary for the rover to perform on-site analyses of the properties of the surface it will be traversing, and it will need a specialized tool to perform these tests. This tool must be able to accurately determine various soil properties, such as friction angle and cohesion, and must be durable enough to sustain thousands of uses without loss of accuracy. The goal of this project was the development and testing of such a tool, which in this case is a device designed to mimic the Direct Shear Test, a standard laboratory procedure to test soil properties. Characterization of Cyan Fluorescent Proteins in Mammalian Cell Lines Galen F. Gao Mentors: Michael Elowitz and Fangyuan Ding Fluorescent proteins are commonly used by biologists in a wide array of applications in biological imaging to visualize a cell’s structure, movement, and composition. However, there’s no systematic characterization of all fluorescent proteins in all systems, and, given the many microscopy imaging setups available and the proteins’ performance variability depending on their environment, almost every lab’s requirements for imaging fluorescent proteins is unique. The Elowitz lab, in particular, uses primarily red, yellow, and blue channels to image mammalian cell lines; however, the current blue protein, mCerulean, suffers from low photostability and works poorly. To find an adequate replacement for mCerulean, the coding sequences of different cyan fluorescent proteins were recombined into plasmid backbones, cloned, and transfected into mammalian HEK293 cells. Their fluorescent properties (phototoxicity, brightness, photostability, etc.) will then be characterized by observing cell cycle length and bleaching rate and counting the number of photons detected. The results of these assays will offer

27

  a guideline to choosing the best protein to use under various circumstances. Additionally, besides those of cyan proteins, the DNA sequences of mTagBFP (blue) and mKate2, LSSmKate1, and LSSmKate2 (red) were successfully transfected into HEK cells, and some preliminary data on their fluorescent properties were obtained. Active-Sterile Neutrino Mixing in Big Bang Nucleosynthesis Lauren Gilbert Mentors: George Fuller and Christian Ott In this project, we examine the effect of active-sterile neutrino mixing on standard Big Bang nucleosynthesis (BBN). Sterile (right-handed) neutrinos would decouple at a much higher temperature than left-handed neutrinos. However, given a non-zero active-sterile mixing angle, the thermally-decoupled sterile neutrinos would affect thermally-coupled active neutrinos. Since the neutron-to-proton ratio depends sensitively on the population of electron neutrinos, and BBN on the neutron-to-proton ratio, active-sterile mixing could affect primordial abundances of light elements. Similar conditions exist in the neutrino-driven wind of a core collapse supernova, where the electron neutrino flux is a determining factor in the results of r process. We simulate a simple system in which only electron neutrinos and sterile neutrinos oscillate between flavors. We then run “standard” BBN (Wagoner, Fowler and Hoyle 1967) with a variety of mixing angles and mass differences. This allows us to determine if BBN is consistent with a sterile neutrino species that decoupled much earlier than weak freeze-out. Furthermore, sterile neutrinos are also a possible candidate for hot dark matter in the early universe, transitioning to cold dark matter by the current epoch. By matching observations of primordial elemental abundances and the results of BBN with active-sterile oscillations, we can limit the parameter space for rest mass and mixing angle of sterile neutrinos. NuSTAR Observations of Swift-BAT Selected Active Galaxies Ana Rose W. Glidden Mentors: Fiona Harrison, Eric Bellm, Felix Fuerst, Dom Walton, and Mislav Balokovic Active galactic nuclei (AGNs) are the very luminous centers of some galaxies. This light is caused by the accretion of mass around the central supermassive black hole. Surrounding the accretion disk is a torus of obscuring gas and dust. The X-ray spectrum of an AGN is a power law continuum with photon index around 2, modified by effects due to the geometry of the AGN. These effects can be due to absorption by gas in our galaxy or gas near the AGN and also by reflection, scattering, and fluorescence from the material that the AGN has irradiated. NuSTAR is an orbiting, high energy X-ray telescope recently launched in June 2012 and is the ideal instrument with which to perform a study on AGNs. The space telescope is ~100 times more sensitive at high energies (>10 keV) than previous missions, allowing for much better determination of the broadband X-ray spectrum. Within the NuSTAR band of 3.5-79 keV, the spectrum has a power law continuum with an iron Kα emission line around 6.4 keV from fluorescence and a hump around ~30 keV due to Compton scattering. We analyzed AGNs that have been detected by Swift-BAT, a wide-field survey in the hard X-ray. Improved Thermal Demagnetization Furnace for Paleomagnetism Research Harry Golash Mentor: Joseph L. Kirschvink Successive demagnetization of a rock sample in a thermal demagnetization furnace removes the secondary natural remnant magnetization (NRM) of the sample so its characteristic component of NRM (ChRM) can then be analyzed. The ChRM preserves a record of the Earth's magnetic field and the tectonic movement of the rock over millions of years, and its study is imperative to paleomagnetism research. The old furnace was built with expensive components that are now outdated and inefficient, and its operating software was faulty. It caused overheating and erratic heating and cooling of the samples; temperatures would inconsistently rise and drop. To rectify this, five additional thermocouples were added to the furnace sample tray using multipin connectors secured with a custom aluminum clamp, and a new application to run the furnace was designed in C# to implement PID control of the furnace temperature. A USB device that supports up to eight thermocouple inputs and up to eight TTL output channels is now used to directly control the three heating coils in the furnace via three solid state relays. This allows for a reliable, inexpensive, and easy to maintain setup that ramps up and maintains the sample temperature in the furnace, and provides a thermal gradient, as required for paleomagnetic study. The Great Galactic Oversight Characterizing the Most Numerous Stars in the Galaxy Alexandria S. Gonzales Mentors: John Johnson and Jonathan Swift Stars with masses roughly less than half a solar mass constitute 75% of all stars in the Galaxy. However, due to their enigmatic nature, these small-mass stars remain difficult to characterize. We aim to increase our understanding of the physical properties of these stars through the study of eclipsing binary systems. By utilizing precise photometric data from NASA's Kepler Mission, we characterize all eclipsing binaries in the Kepler dataset with primary stellar mass less than about half a solar mass. After selecting target systems from the Kepler Input Catalog, we classify and verify this sample. Using code developed by John Southworth to model the light curves of

28

  our targets, we vet our sample for prime candidates for radial velocity follow up. Using available time on the Keck I telescope, we obtain radial velocity measurements for our target list from the High Resolution Echelle Spectrometer (HIRES). We are also submitting a proposal for time this fall using the Echellette Spectrograph and Imager (ESI) for follow up double-lined spectroscopic binary systems data. With an absolute mass scale for the eclipsing systems, we then fully characterize the individual stars in terms of their mass, radius, and temperature. From these observables, we can then derive the stars' luminosities. Using these same techniques, we also characterize a dissimilar eclipsing binary system with scientifically interesting properties. Work Towards Yuccaol Natural Products Emmett D. Goodman Mentors: Brian M. Stoltz and Jeffrey C. Holder Total synthesis is an important part of organic chemistry involving the synthesis of naturally occurring molecules. These natural products are usually chemical compounds produced by living organisms that have pharmacological, biological, or other useful chemical properties. Often these molecules are produced in such small quantities that they cannot be extracted in a useful amount for medical and commercial applications. For this reason it is desirable to find a concise synthetic route to construct these valuable chemical compounds. So far, there are no syntheses of any of these five natural products, Yuccaols A-E. This summer was spent pursuing the total synthesis of the Yuccaol class of natural products. The molecules that comprise this class are extracted from the plant Yucca Schidigera, found in the Mojave Desert in areas of California, south Nevada, and western Arizona. These natural products have been found to have antioxidant, radical scavenging, and platelet inhibitory effects as well as antiinflammatory activity. This summer, in particular we focused on synthesizing the core of these molecules. In addition, we made progress towards the actual system present in these products. We were able to synthesize multiple precursors of the core of this class of compounds and hope to cyclize these precursors throughout the school year to furnish this core structure of these natural products. Autonomous Selection of Rooftop Landing Area Eric Gorlin Mentor: Roland Brockers To operate unmanned aerial vehicles (UAVs) safely in complex environments, human operator control can be facilitated significantly if the UAV can execute simple maneuvers autonomously. One example that is particularly important for reconnaissance scenarios is rooftop landing. For autonomous landing on a flat rooftop surface, the UAV needs to initially detect a sufficiently flat roof top landing area and pick a safe landing spot in order to execute a landing maneuver. While an image processing module provided motion stereo depth maps, the goal of this task was to use depth information and image intensities to detect roof edges and segment complete roof surfaces. An edge detection algorithm was run on the rectified input images, and these edges were closed to fill gaps and enclose the contour of the roof. Preprocessed flat surfaces on this roof were used as a starting point to extrapolate the previously sparse depth map into a dense depth map and large flat area from which suitable landing spots can be selected. Sampling Lightning Strike-Induced Magnetic Anomalies on Mars Joseph M. Grappone Mentors: Joseph Kirschvink and Sarah Slotznick Most of the current research on the Martian surface has one common goal: to search for life on Mars, or the presence of water-bearing ancient environments that might have been conducive to it. The next rover is scheduled for 2020 and has this same goal, including the possibility of collecting and holding rocks for eventual return to Earth. One method of searching for previously wet surfaces is to search for air-to-ground lightning strikes, which require non-frozen (wet) ground. A lightning strike is characterized by a strong circular magnetic field around the point of contact with smaller, random lines of magnetization leading out from the center. These fields are strong enough to remagnetize the rocks they contact and can be measured by a fluxgate magnetometer placed at the end of a rover’s instrument arm. We tested this concept by constructing a computer-controlled XY-stage capable of 2D movement of an attached fluxgate. By tilting the stage and manually moving the fluxgate along a track, 3D movement is obtained for a rock sample of any shape. The magnetometer was calibrated in the lab on small dipole magnets and on a manually-remagnetized sample of the Saddleback Mountain Basalt (the flow chosen as the JPL Mars simulation flow). We then modified the unit to run off a series of 12V automobile batteries and took the stage to Flagstaff, Arizona and tested it on lightning-struck basalt basalts on the Black Point Flow, on the Babbitt Spider Web Ranch. We not only found circular fields characteristic of lightning strikes, but also hyperbolic patterns characteristic of the lightning’s current moving randomly through the rock. This demonstrates that precise control of a vector fluxgate over the surface of magnetic anomalies can uniquely identify the fingerprints of ancient lightning strikes and paves the way for the implementation on planetary landers.

29

  A Lower Bound for Torelli-K-Quasiconformal Homogeneity Mark Greenfield Mentor: Vladimir Markovic A closed hyperbolic Riemann surface M is said to be K-quasiconformally homogeneous if there exists a transitive family F of K-quasiconformal homeomorphisms. Further, if all [f] ⊂ F act trivially on H1(M;Z), we
say M is Torelli-Kquasiconformally homogeneous. We prove the existence of a uniform lower bound on K for Torelli-Kquasiconformally homogeneous Riemann surfaces. This is a special case of the open problem of the existence of a lower bound on K for (in general non-Torelli) K-quasiconformally homogeneous Riemann surfaces. Implementation of a Biomolecular Circuit for Tracking Protein Concentration in TX-TL System Aurelija Grigonyte Mentors: Richard Murray and Emmanuel de los Santos The main aim of synthetic biology is to apply engineering principles in biology and develop designs of biological systems with predictable outputs. Scaffold mediated signalling is often used in signal transduction of various organisms. The main aim of the project was to implement scaffold based circuit in vitro TX-TL, cell free environment expression system, as well as to test it in vivo in two different strains of E. coli, BL21 Rosetta and WW62. The circuit of interest contained three major components which were the histidine kinase (Taz), reference protein, GFP, and the response regulator CusR. A simpler model of the circuit of interest was implemented and tested in the TX-TL system. This informed future experimental work on the implementation of the complete circuit in the TX-TL system. Engineering of a More Active 1,8-Cineole Synthase Webster Guan Mentors: Frances Arnold and Thomas Heel Terpene synthases catalyze the synthesis of terpenes, a large class of organic compounds that have a diverse range of applications. 1,8-cineole synthase is a monoterpene synthase that synthesizes 1,8-cineole, a component in eucalyptus leaves that has been shown to have anti-inflammatory effects.1 The pathway to cyclizing these terpenes from the primary substrate geranyl diphosphate (GPP) requires a slow ionization-isomerization step, making them 100 times less active than that of enzymes like PHS1 that can utilize GPP’s cis-isomer, neryl diphosphate (NPP), as a substrate. 2,3 This study aims to take advantage of this fact by engineering 1,8-cineole synthase to become more active on NPP. Using a high-throughput screen for directed evolution developed by the Arnold lab, an error-prone PCR library of 1,8-cineole synthase mutants are screened using a colorimetric assay to determine mutants with improved activity.4 In a first round of directed evolution screening, a mutant of 1,8-cineole synthase referred to as P26H4 has shown 20% increase in activity on NPP over the wild type enzyme. With more rounds of screening of epPCR libraries, we aim to evolve 1,8-cineole synthase to become even more efficient in utilizing NPP. References 1. 2. 3. 4.

Santos, F.A., Rao, V.S.N. 2000. Antiinflammatory and antinociceptive effects of 1,8-cineole, a terpenoid oxide present in many plant essential oils. Phytotherapy Research. Vol 14, 4: 240-244. Davis, E. M., & Croteau, R. C. (2007). Cyclization enzymes in the biosynthesis of monoterpenes, sesquiterpenes, and diterpenes. Institute of Biological Chemistry, Washington State University, Pullman, WA, Retrieved from http://academics.eckerd.edu/instructor/grove/AdvOrg/Terpenes/Review1.pdf Schilmiller, A. L., Schauvinhold, I., Larson, M., Xu, R., Charbonneau, A. L., Schmidt, A., Wilkerson, C., & , R. L. (2009). Monoterpenes in the glandular trichomes of tomato are synthesized from a neryl diphosphate precursor rather than geranyl diphosphate. PNAS, 106(26), 10865-10870. Lauchli, R., Rabe, K., Kalbarczyk, K., Tata, A., Heel, T., Kitto, R., Arnold, F., 2013. High-throughput screening for terpene-synthase-cyclization activity and directed evolution of a terpene synthase. AngewadteChemical International Edition. V52:21:5571-5574.

Assembly of Metal Complexes on Carbon Surfaces for Catalysis of Fuel-Forming Reactions Ayush Gupta Mentors: Harry B. Gray and James D. Blakemore Interfacing well-defined molecular catalysts with electrode surfaces is a key step toward constructing devices for solar-fuel production. Here we describe the synthesis of a Cp*Rh complex (Cp* is pentamethylcyclopentadienyl) bearing a pyrene-functionalized bipyridine ligand (1), which can be noncovalently bound to graphite electrodes via pi-pi stacking between the pyrene moiety and the graphitic surface. The electrodes are prepared on a roughened basal-plane surface of highly oriented pyrolytic graphite that is cleaned and coated with a slurry containing Ketjen black, a high surface area carbon material. After curing at 70ᵒC, the electrode is simply soaked in a solution containing 1 to achieve surface functionalization. Cyclic voltammetry data show the expected redox cycling of 1 on the surface near -1 V vs. Fc/Fc+, a suitable potential for proton-reduction electrocatalysis. X-ray photoelectron

30

  spectroscopy (XPS) studies demonstrate that complex 1 is attached to the graphite electrode both before and after redox cycling. Results on driving proton-reduction electrocatalysis from different acids (proton sources) and product measurement will also be discussed. This research has broad implications for surface catalysis since we describe a method to easily attach any molecular catalyst to an electrode surface. Constructing a Non-Magnetic XY-Stage to Search for Lightning Strikes on Earth and Mars Garima Gupta Mentor: Joseph L. Kirschvink All the missions to Mars thus far have justified themselves as steps in the search for life beyond Earth and most of these missions have focused on searching for the past or present existence of water. One indication of a wet, conductive surface is cloud-to-ground lightning. To prove that there was once lightning, and thus water, on the Martian surface, we must examine Martian rocks for Isothermal Remanent Magnetization (IRM), which is the immortal alignment of the magnetic vectors in a material due to a strong external magnetic pulse. Lightning strikes produce strong, isolated magnetic anomalies, which can be detected by a simple fluxgate magnetometer. On Mars, we could place a fluxgate magnetometer at the end of a rover’s instrument arm and use it to systematically scan over an object or surface. We explored this concept through the construction and use of a computer-controlled, non-magnetic XY-stage that could survey a 75cm by 75cm region. We designed the stage to be used in the field and deployed it first in Flagstaff, AZ for testing on the Black Point Lava Flow basalt, which is similar to the rocks found on Mars. We were able to detect radial and hyperbolic magnetic field patterns, characteristic of lightning strikes. We can conclude from these data that such systematic scanning of magnetic anomalies can uniquely identify the signature of lightning strikes. This concept will be useful not only for testing the hypothesis that Mars once had a wet, conductive surface, but also for avoiding the return of lightning-struck samples to Earth for biogeochemical analysis. Free Will, Decision Making, and the Human Brain: Online Realtime Analysis of Intracortical Signals in Humans Nikhil Gupta Mentors: Ralph Adolphs, Uri Maoz, and Liad Mudrik It has been shown that there is information in the human brain about upcoming actions up to several seconds before the urge to carry out these actions is reported. Recently, Maoz and colleagues were able to develop a realtime prediction system (RTPS) that employs supervised learning to predict which hand subjects would later raise during a matching-pennies game. The system relied on intracranial local-field potential (LFP) signals – mainly from intracortical depth-electrodes – recorded from consenting epilepsy patients, who were implanted with intracranial electrodes for clinical purposes. The system predicted at accuracy above 80% before action onset retrospectively (and close to 70% in real time). However, the system is constrained to make a prediction once, 0.5s before the patient is instructed to raise his/her hand via a go signal that follows a 5s countdown. Now, we aim to develop a continuous real time prediction system (CRTPS) which continuously provides a prediction on the upcoming decision rather than once per trial, along with a confidence on each prediction. This provides subjects with the flexibility to raise their hands whenever they desire, which should enhance our ability to learn about decision formation in the brain and its relation to the reported time of the first urge to move. CRTPS has been implemented as a cascade of the StimOMatic of to an extension of the RTPS, where the former performs the basic single channel data acquisition and processing and the latter performs multi-channel processing. Analysis of Stability and Prediction of Amplitude and Period of Oscillations in Genetic Regulatory Networks With Delays Mattias Gustavsson Fält Mentors: Richard Murray and Marcella Gomez The creation of biological systems with specific behaviors is the goal in the field of Synthetic Biology. For this reason it is important to have a good understanding of the different components in such systems. This project aims to study the properties of some simple genetic regulatory networks with oscillatory behavior, specifically the repressilator, and how they are affected by delays in the system. Although delays are known to be present in the the process of transcription and translation, they are often omitted in the modeling and analysis of these systems. In this project both single deterministic delays, as well as distributed delays are considered to account for the stochastic nature of the systems. The main properties that are studied is the presence of oscillations and their amplitude and period. The analysis is done using tools from the area of control theory such as the Nyquist stability criterion and the describing function method.

31

  Microwave Spectrum of Ethylene Glycol Using Chirped-Pulse Fourier Transform Microwave Spectroscopy Daniel Guth Mentor: Geoffrey Blake Molecular Spectroscopy is considered one of a chemist’s most important tools in studying the structure of molecules. The method we use is known as Fourier Transform Microwave (FTMW) spectroscopy. This approach uses a phase stable chirped pulse to excite the molecular rotation, averages the coherent molecular emission in the time domain, averages the coherent molecular emission in the time domain, and takes a Fourier transform of the molecule emission to yield a frequency domain spectrum. Most FTMW spectrometers require high sample rate oscilloscopes, broadband high power amplifiers, and arbitrary waveform generators (AWGs). Of these, the waveform generators are extremely expensive because they are able to generate complex waveforms within their Nyquist limit. This level of complexity is not required for FTMW spectroscopy, where a linear sweep of frequency versus time is used. Thus, we have tested the use of a direct digital synthesis (DDS) chip which is much cheaper and operates on a narrower band width. This chip was then used to measure the rotational emission of ethylene glycol. Further processing then yielded the microwave spectrum. Future analysis of this spectrum will lead to a greater understanding of the structure and intramolecular interactions of Ethylene Glycol. Determining the Function of GATA3 via Ectopic Expression in B-Cell Precursors James Ha Mentor: Ellen Rothenberg T-cell precursors steadily lose access to all other fates as they progress along the T-cell differentiation pathway. One of the earliest potentials that are silenced is the B-cell fate, which is intriguing given that the B-cell and the T-cell pathways utilize similar sets of transcription factors. It is suspected that the transcription factor GATA3 is responsible for silencing the B-cell fate in T-cell precursors. By forcing B-cell precursors to ectopically express GATA3 using a retroviral vector, and monitoring changes in the expression of key transcription factors, it is possible to determine how GATA3 exerts its effects in early T-cell precursors. Training of a 3rd Generation ENose to Detect Carbon Monoxide Macy Hardley Mentor: Margie Homer The JPL Electronic Nose (ENose) is an event monitor that detects, identifies, and quantifies a set of gaseous analytes that are potentially hazardous to spacecraft crews. It is made of an array of semi-selective, chemi-resistive, polymercarbon black composite sensors that can be trained to detect most compounds. Of specific interest to this project was the ability of a set of ENose sensors to detect and identify carbon monoxide. Previously made sensors known for their relative sensitivity to carbon monoxide were tested for response. An array was selected that responded best to initial tests with carbon monoxide and was then tested for sensitivity. Ideally, the selected array should be able to detect carbon monoxide at or below 20 parts-per-million, the 24-hour Spacecraft Maximum Allowable Concentration (SMAC). Current testing has shown that the selected array cannot yet detect carbon monoxide at 20 parts-per-million. Data on how the sensors respond to carbon monoxide at different levels and humidities will be used to create a fingerprint of response that the ENose computer could utilize to correctly identify future carbon monoxide events. Subsequent work would involve creating an array that could detect carbon monoxide at levels at or below the 24-hour Spacecraft Maximum Allowable Concentration. Simulation of Quantum Photonic Resonators in Silicon Carbide and Other Novel Materials Alexander E. Hartz Mentor: Andrei Faraon Quantum photonic devices promise to enable photonic quantum technologies, such as secure communications and quantum information processing. These devices integrate waveguides which direct the path of light, optical resonators which trap light, and quantum light emitters such as quantum dots or impurities in crystals. By placing a quantum light emitter near the resonance cavity, one can cause the emitter to couple to the resonator. This cavityemitter interaction can be further used to transmit and process quantum information. Devices with high quality factors and small optical mode volumes maximize the interaction between the quantum object and the light by trapping the light within the resonator for many periods and confining the electric field to a small volume. During the last decade, research concerning quantum photonic devices has focused on Nitrogen-Vacancy Centers in diamond and quantum dots. However, novel materials such as Silicon Carbide and complex oxides doped with Lanthanide ions have interesting quantum properties that can be integrated into quantum photonic devices. For example, Silicon Carbide has a range of defects comparable to those in diamond while it is also a material well suited for nano-fabrication of photonic devices. Devices fabricated on top of or in Yttrium Orthosilicate, a complex oxide, can couple to rare-earth dopants, and harness their excellent optical quantum coherence properties. As such, this project aims to design and simulate quantum photonic resonators with high quality factors and small optical mode volumes using Silicon Carbide and other novel materials.

32

  Experimental Investigation of the Relationship Between Backwater Hydrodynamics and Delta Evolution Hima J. Hassenruck-Gudipati Mentors: Michael P. Lamb and Vamsi Ganti Deltas are amongst the most fertile lands with high population density and are extremely vulnerable to anthropogenic and sea-level changes. We lack a mechanistic understanding of deltaic river avulsions, which set the fundamental scale of deltas and can lead to natural catastrophes. Factors of avulsions studied occur in the backwater zone, an area of the river directly affected by sea level. Because of hydrodynamics in the backwater zone, sea-level is thought to affect the morphodynamics of the delta. Using controlled experiments of a coupled river and river plume system, I have been studying the relationship between backwater hydrodynamics and delta evolution. The experimental set up includes an alluvial river section (7 m by 0.07m) that feeds into ocean basin (6 m by 3 m). The set of experiments run this summer held the sea-level, water flux and sediment discharge constant. Documented features are channel bifurcations, self-formed levees, and river avulsions on the delta (set to extend beyond the backwater length of ~2.5m). Future experiments include studying the effects of dynamic sea-level, variable water, and sediment discharges. This study aims to improve modeling of coupled river-plume dynamics and strengthen the predictability of delta evolution, with a focus on understanding avulsions. Stochastic Optimal Control for Brain-Machine Interfaces Using ECoG Signals Bryan D. He Mentors: Lakshminarayan Srinivasan and David Rutledge Brain-machine interfaces are an emerging class of devices that will allow us to directly communicate with computers with neural activity. These devices have the potential to improve the quality of life for paralyzed or disabled patients. To allow brain-machine interfaces to operate in a controllable manner, algorithms that can learn how to decode neural activity into intended motion must be developed. Stochastic optimal control is a method used to describe the interaction between the human brain and the brain-machine interface. For the controller to be trained, predicted movement and a model for the expected neural signals are required. Thus, to improve the performance of the brain-machine interface, we explore various models for intended movement and relationships between movement and neural signals. The algorithms we develop will be tested on real electrocorticography (ECoG) measurements from a human brain. Characteristic Width of Filaments in Molecular Clouds Monica He Mentor: Konstantinos Tassis Recent Herschel observations of molecular clouds have found filaments with an apparent characteristic width, in which most, if not all, protostellar cores are found. However, this characteristic filament width is not present as a signature in the power spectrum of the Polaris Flare cloud (Miville-Deschenes et al. 2010), despite the fact that the ~0.1 pc width is present with central column densities over three orders of magnitude in the same cloud (Arzoumanian et al. 2011). This apparent inconsistency is the focus of the project. We simulated geometrical model clouds with differing radial density profiles to determine whether a characteristic width indeed exists, and whether this width would have a signature in 1D power spectra of the image. Fourier analysis of the apodized model 2D density maps of cylindrical filaments with modified Plummer profiles clearly suggest that the signature is a break in the power law that shifts along the spatial frequency axis. Further research will continue to attempt to lose the power spectra signal through randomization and adjustment of seven input parameters: number of filaments, central radius, filament length, observation angle, central density and differing normalizations, filament misalignment (random walk), and degree of filament overlap. We also tested several methods used in the filament discovery paper to look for any possible artifact in the identification process; none were found. Half-Sandwich Rhodium Hydrides as Hydrogen Transfer Catalysts Emilia Hernandez Mentors: Harry Gray, Wes Sattler, and James Blakemore [Pentamethylcyclopentadienyl rhodium (3,4,7,8 tetramethyl 1,10 phenanthroline)Cl]Cl ([Cp*Rh(Me4Phen)Cl]Cl) and the reduced rhodium (I) compound, Cp*Rh(Me4Phen), were synthesized and characterized. Compounds of this class have been shown to be effective hydrogen transfer catalysts as well as H2-evolution catalysts, so we tried to synthesize and isolate the analogous rhodium (III) hydride, [Cp*Rh(Me4Phen)H]X (X = non-coordinating anion). We were able to generate the rhodium-hydride cation via reaction of Cp*Rh(Me4Phen) with (i) [PhNMe2H][B(C6F6)4] in acetonitrile, and (ii) Ph2SiH2 and CO2 in benzene, each of which was characterized by a doublet in the 1H NMR spectrum, at -9.03 ppm and -10.7ppm respectively. Unfortunately, we have not been able to isolate a rhodiumhydride complex yet. We have also discovered interesting reactivity with Cp*Rh(Me4Phen) with CO2 and various hydrosilanes, and are characterizing these reactions using various NMR spectroscopies and gas chromatographymass spectrometry (GC-MS).

33

  Low Cost Solutions for Optimizing Flow in an Electrochemical Wastewater Treatment Reactor Rachel E. Hess Mentors: Michael R. Hoffmann, Asghar Aryanfar, and Clément Cid In developing countries, diseases caused by ineffective or nonexistent sewage treatment leads to the death of more children yearly than HIV, Malaria, and tuberculosis combined. Access to low cost safe sanitation systems saves lives. As a part of a larger sustainable waste-treating toilet module being developed by the Hoffmann lab for the Gates Foundation’s Reinvent the Toilet Challenge, wastewater is treated electrochemically in a reactor. A major factor that determines the feasibility of the toilet module is the cost of production and maintenance. Designing a cheap, effective, and robust reactor works toward resolving this issue. For the reactor to be effective, the wastewater must not stagnate in the reactor. Pumping patterns and overall geometry of the reactor were studied in SolidWorks Flow Simulation. Simulations of wastewater flow through numerous revisions of reactor designs were conducted and a new design was modified and improved. A prototype of the new reactor design was fabricated. Tests with samples of human waste will point to design features still in need of refinement to achieve the goal of a cheap and effective reactor that reliably sanitizes wastewater. Construction of an Expert Annotation Tool for Static Images Kristján Eldjárn Hjörleifsson Mentors: Pietro Perona and Steven Branson Data annotations are a fundamental factor in the training of many machine learning algorithms used in computational vision. It is therefore vital to have an efficient and consistent way of gathering them. Researchers in computational vision have come to believe that one of the major bottlenecks for problem solving are the datasets used to train algorithms, in particular the relatively small size of the datasets and the limited variety of annotations available. To rectify this, an Expert Annotation Tool for images is in development. It is a sophisticated tool that can be used to gather various types of annotations for static images. Its usage is straightforward for both annotator and researcher, and so it grants less computer-savvy people the means to create complex annotation models without having to know the syntax or semantics of database manipulation. Aside from being a convenient way to gather data according to conventional annotation models, the tool can be used to define new types of annotations. Those can in turn be used to train machine learning algorithms and their output compared to those of existing ones in terms of precision and efficiency. Computational Complexity of Expected Time Nubots Dhiraj Holden Mentors: Erik Winfree and Damien Woods The Nubots model is a cellular automaton-like model that models nanoscale molecular computation. The feature of Nuobts which makes it different than a cellular automaton is the movement rule, where local changes in monomer configuration can cause global changes in the structure of the system. We study the computational power of Nubots both with and without the movement rule, using the notion of agitation, where monomers wiggle around randomly. The notion of agitation in Nubots is analogous to Brownian motion in molecular systems. We prove that if a language can decided by a Turing machine that uses polylogarithmic space, it can be decided using polylog expected time Nubots with movement. Additionally, we show that the movable set of a movement rule, or the set of monomers that move whenever a movement rule is applied, is reducible to planar reachibility, which means that it can be computed in polylogarithmic space. This improves previous results which showed that the movable set was computable in polynomial time. We also show that the Nubots model without movement but with agitation is able to construct shapes of linear size in polylogarithmic time, which implies that random movement works just as well as directed movement in this model. This is an improvement over the previous result, which used movement to show that any language in NC is decidable by Nubots that run in polylogarithmic expected time, and suggests that agitation can do the same things as movement. This suggests that directed movement is not necessary for the computational power of Nubots. Computation of Heegaard Floer Correction Terms for Certain Manifolds Timothy P. Holland Mentor: Yi Ni Based on previous work in the field, we implement algorithms for determining the correction terms of certain classes of manifolds and provide a user friendly environment for working mathematicians to generate the correction terms when needed. The options available are for double branched covers of alternating links and plumbed manifolds with a variety of input options. The program uses a heavy coverage test suite to give a confidence in the correctness of results.

34

  Development of Antennas for Cosmic Microwave Background and Reionization Measurements Samuel Holo Mentors: James Bock, Roger O'Brient, and Zachary Staniszewski We have been developing antenna-coupled bolometers for Cosmic Microwave Background and Epoch of Reionization measurements. The design for these detectors is based off of the 150 GHz detectors used in the Keck Array, SPIDER, and BICEP 2 experiments. We have been adjusting the parameters of these antennas to operate at 250 GHz through HFSS software simulations. In particular, we adjusted the location of the antenna's ports in an attempt to minimize the power reflected over our frequency range. In addition, we have designed and assembled a Fourier Transform Spectrometer with high enough spectral resolution to characterize on-chip spectrometers. We are testing the performance of this spectrometer by testing 90 GHz detectors with both our new spectrometer and Princeton's Fourier Transform Spectrometer. This fall, we plan to use our spectrometer to characterize 220 GHz detectors and on-chip spectrometers. Modeling the Accuracy and Reaction Time in Value-Based Choices Using the Drift Diffusion Model Benjamin Hu Mentor: Antonio Rangel Simple decision-making tasks are a regular part of social science experiments. Recent literature shows that the drift-diffusion model is among the quantitative models that provides a quantitative explanation of various neurophysiological data in simple decision-making tasks. The experimenter wanted to test how well the driftdiffusion model fits data from social science experiments, which consists of data on one- or two-option decisions, including the ratings of the options in all decisions, the actual decisions made and the times it took to make each of those decisions. Specifically, the experimenters wanted to test the hypothesis that a particular parameter (corresponding to each possible choice) in the differential equation describing the drift-diffusion model is directly proportional to the ratings of the choices by looking at the plots of the average drift rates vs. ratings and the graphs of “accuracy” (i.e. how often an object was chosen over an alternative) vs. ratings. Judging from the graphs of the average drift rates v. the ratings and graphs of the averages of all choices v. ratings, it seems that the drift diffusion model does indeed provide a good fit of the data given. It may be fruitful to consider other methods of modeling the distribution of reaction times vs. ratings and the plot of the “accuracy” v. ratings, or to provide more sophisticated data to analyze, such as the history of choices for a given subject. Regulation of Binding of SynGAP to the Postsynaptic Scaffold PSD-95 Rebecca Hu Mentor: Mary Kennedy AMPARs are a type of glutamate receptor that mediates most excitatory synaptic transmission in the brain. They are anchored in the postsynaptic density (PSD) by scaffolding proteins and cytoskeletal elements. The activitydependent variation in synaptic AMPAR number, known as synaptic plasticity, is thought to be the mechanism by which information is stored in neuronal networks. Synaptic retention of AMPARs is believed to occur primarily through binding to PDZ-containing scaffold proteins at the PSD. AMPARs can indirectly bind to scaffold proteins by first binding to TARP proteins, which then bind to scaffolding proteins. Interestingly, synGAP, a Ras-GTPase activating protein that interacts with the PDZ domains of PSD-95, is present in large amounts at the excitatory synapses. Our project tests the hypothesis that binding of Ca2+/CaM (calmodulin) or phosphorylation of synGAP alters its affinity for PSD-95, and might, therefore, free spaces for TARP protein to bind and increase the trapping of AMPA receptors. To test this hypothesis, we used a batch/gravity column combination method to generate HaloTag-PSD-95-PDZ affinity resins for use in synGAP pull down assays with varying parameters (e.g. plus or minus Ca2+/CaM, with or without phosphorylated synGAP). Our results show that adding Ca2+ and CaM (calmodulin) does affect synGAP’s binding to PSD-95. We are currently testing the effect of phosphorylation on synGAP’s affinity to PSD-95. Abnormalities in Courtship Song of Male D. melanogaster Deficient in a Novel Neuropeptide, Natalisin Tianyi Hu Mentors: Youngjoon Kim and Yonghoon Jang Male D. melanogaster flies deficient in a novel neuropeptide, natalisin, have been shown unable to mate with female D. melanogaster flies, despite the lack of physiological difference in natalisin-deficient (nat-) males. However, visual measurements cannot account for the auditory signals, specifically the courtship song, emitted by the males. Combining visual and auditory recordings, the project determines that nat- males have a significantly higher inter-pulse interval (IPI) in their courtship songs. Audio analysis programs coded in Matlab® calculated an average IPI of 44ms for nat- males compared to an average IPI of 35ms. This basic analysis indicates that the courtship song of nat- male fly is quantitatively different and thus contributes to the male’s mating failure with wildtype females. Based on preliminary analysis, it is also interesting to note that all of the songs exhibit an intrinsic sinusoidal variation in the sine to pulse song duration ratio in 30-second window frames. Thus the

35

  courtship song, although dependent on external factors such as the female’s movements, may have an inherent pattern based on the intrinsic neural activities of courtship. A recently developed mutant natalisin strain would allow more controlled background strain crossings, and the analysis would be further enhanced. Development and Design of Situation Awareness Application for Community Sensing and Clouds Edward Huang Mentors: Julian Bunn and K. Mani Chandy Hazard monitoring systems can utilize computing and sensor technology to detect and analyze natural disasters. Creating sensor technology highly accessible to users can allow these systems to maintain consistent data streams in the cloud. The scope of this project provides these sensors with appropriate user interfaces easily obtainable on Android mobile devices. The interfaces will show dynamically updated tile-based situation awareness information and sensor readings in real time. The design requires developing these interfaces in an integrated development environment, utilizing several software development kits to create a compositional framework. The project uses the Java and XML languages to optimize performance and implement attractive features. Furthermore, extensive testing is essential to ensure the robustness of the application. The application’s configurability has largely increased with options easily modified by the user. Stability has similarly increased with the introduction of several bug fixes, reducing the frequency of crashes and issues. Additionally, the project has integrated several social media application programming interfaces to increase its attractiveness to users. Further development of the application requires extensive documentation as well as increased configurability. Modeling Cardiac Flow to Predict Responsive Vibrations From Vortex Formation Indices Vivian Huang Mentor: John O. Dabiri Vortex formation in the heart may provide a robust measure of health. Current methods to observe left ventricular (LV) flow include ultrasound and magnetic resonance imaging, but these are not universally accessible to frontline clinicians. We seek to quantify diagnostic markers based on acoustic vibrations generated by different vortex flows. We model vortex formation in the heart with a calibrated piston/cylinder setup in a water tank. Using digital particle image velocimetry (DPIV), we obtain velocity and vorticity fields correlated with specific piston stroke to diameter (L/D) ratios. Pressure fields are derived from velocity fields using the queen 1.0 Pressure Estimator algorithm developed by Dabiri (2013; arXiv:1303.6966). We implement two approximations in MATLAB, twodimensional and axis-symmetric, to estimate the total force exerted by the flow. Since disease states may manifest in abnormal blood flow, we investigate a range of stroke ratios and LV inflow tract lengths. Eventually, we hope to determine whether markers based on acoustic vibrations may reliably characterize the third heart sound for early stage stethoscopic diagnoses. Characterizing the Kinetic Barrier to Nucleation of DNA Nanotubes Joy Hui Mentors: Erik Winfree and Damien Woods It has been shown that single strands of DNA can be designed to hybridize with others to form a tube-shaped lattice, or nanotube. For an ongoing experiment in the lab, different “species” of nanotubes were designed by varying the number of helices in the circumference in a given nanotube species. One of our goals is to characterize a number of different species that are composed of a single set of shared strands. Formation, also called “spontaneous nucleation,” of nanotubes is sensitive to concentration and temperature. The aim of this project was to characterize the nucleation temperatures of 3 nanotube species at a concentration of 100 nM for a time period between 18 to 24 hours. Here, it is shown that the 3 species of 12, 14, and 16-helix nanotube species have relatively similar nucleation temperatures for a given time period and concentration. This work shows that these 3 kinds of nanotube have the potential to be used together in experiments that require similar nucleation temperatures for time periods of up to one day. A Small-Molecule Screen for Drugs That Affect the Circadian Clock in Larval Zebrafish May Hui Mentors: David Prober and Eric Mosser Sleep is essential for humans to function properly. Yet, despite the approximately 40 million Americans that suffer each year from chronic sleep disorders, the genetic mechanisms that regulate sleep remain largely unknown. The widespread costs of sleep disorders – lost productivity and detrimental health in graveyard-shift workers, for example – give researchers a real incentive to investigate differences in sleep cycles. For this project, a highthroughput circadian rhythm assay using transgenic 7-day-old zebrafish was carried out to determine clusters of drugs that have a similar effect on the circadian cycle. The transgenic zebrafish were produced by using the zebrafish per3 promoter to drive expression of the firefly luciferase gene. A primary screen of ~6000 drugs revealed a group of anti-inflammatory drugs that increase the amplitude and baseline of the per3 bioluminescent

36

  oscillations. qPCR analysis was used to verify that the clock gene was driving the observed changes in oscillations. Repeat qPCR rounds confirmed that the phenotype caused by the anti-inflammmatory drugs is consistent. From these results, treatment of sleep disorders through a concoction of different drugs that counteract disruptions in the circadian rhythm seems promising. Critical Groups of Strongly Regular Graphs Ka Kin Kenneth Hung Mentor: Mohamed Omar Strongly regular graphs (SRGs) are a class of highly symmetric graphs that are not well understood currently. In particular their critical groups, which carry important information about the graphs, are not easily computed. This project develops lemmas and theorems to determine to what extent the parameters (n, k, a, c) of a strongly regular graph dictate the structure of its critical group. The Predicted Ensemble of 3 Dimensional Structures for All Human Olfactory Receptors Sylvia Hürlimann Mentors: William Goddard, III, and Soo-Kyung Kim Humans have ~400 olfactory receptors and each one is activated by a diverse set of odorants. As of yet, there is no experimental data on the structure of any olfactory receptors, but there is a predicted structure for a human olfactory receptor OR1G1. The high sequence and structure homology amongst olfactory receptors means that OR1G1 can be used as a template for other olfactory receptors. Using the BiHelix and SuperBiHelix method, which samples helical rotation, tilt, and sweep angle to generation millions of packings and scores the packings based on energy ranking, we gathered data about the likely structures of the family head of cluster 1 and 2 of the olfactory receptors. Successful structures showed interactions in transmembrane domains 1-2-7 (N1.50, D2.50 and N7.49) and domains 2-4 (N2.45 and W4.50), which have been observed in X-ray and predicted structures for some members of the class A GPCRs, a salt-bridge between the conserved D3.49 and K6.30 in the D(E)RY region, and a hOR specific interaction between the conserved D/E3.39 and H6.40. These structures can be used to calculate structures about the other olfactory receptors in cluster 1 and 2 and, ultimately, to predict the binding of odorants to olfactory receptors. QuakeSim Role-Based Authentication Vy-Luan Huynh Mentor: Andrea Donnellan JPL’s QuakeSim is a computer- and earth-sciences project that models and analyzes earthquake fault cycles and interactions. Currently, users are allowed to manually input data and parameters into QuakeSim models, but at present have no privileges for updating database components. Additionally, users cannot easily share projects with colleagues. It is believed that allowing such open contributions and teamwork will increase science return from QuakeSim . This project aims to improve QuakeSim’s capabilities by allowing better interaction between colleagues and between users and database components. Prior testing indicates that the current weaknesses of QuakeSim are inefficiency in interface and security systems. Through intensive research, we conclude that this issue is best resolved by implementing a new design and infrastructure for the QuakeSim website. The new design was achieved by switching from the original site base to a heavily modified version of Drupal called OpenAtrium. This new site has improved user interfaces and compatibility, allowing people to more easily manage and share their work and manipulate workspaces. The site also now uses LDAP for single sign-on authentication, allowing users to access multiple password-protected tools after logging in just once in order to reduce effort and save time. These changes result in a QuakeSim that is more user-friendly, capable, and efficient, as well as easy to maintain. Development of Hazard Detection and Situation Awareness Computing and Sensor Technology Alina T. Hwang Mentors: Julian Bunn and K. Mani Chandy The Caltech Seismic Network (CSN) uses 1000 community-based accelerometers to create a network of groundshaking data automatically sent to a remote computing system. The development of the CSN is just one of the ways researchers have been working to respond to the demand for technology that allow users to be aware of their surroundings at all times and be warned of imminent environmental hazards. Recently, under Dr. Julian Bunn and Professor K. Mani Chandy, researchers at the California Institute of Technology have developed a small weather station prototype that can gather data on various environmental parameters and send this information to a data storage network located in a cloud computing system at regular intervals. In this work, we continued to research and develop a portable multi-sensor hazard detection device using a Raspberry Pi. We configured the Raspberry Pi to act as a client to the cloud computing system and to communicate with an Arduino, so that it can read data off sensors connected to the Arduino. Further research can be done on modifying the client to create and send picks (unusual data behavior) from the sensors to the cloud and to allow the device to run on a lithium-ion battery.

37

  Regulation of Binding of SynGAP to the Postsynaptic Density Scaffold PSD-95 Ariella M. Iancu Mentor: Mary B. Kennedy Synaptic strength is closely tied to the amount of AMPARs in the post synaptic density (PSD)—the more AMPA receptors there are, the more strongly the neuron depolarizes. AMPARs are retained in the synapse by a diffusional trapping mechanism believed to involve binding sites on scaffolding proteins, such as PSD-95. PSD-95 is a scaffolding protein of the MAGUK family with three PDZ domains, an SH3 domain, and a guanylate kinase domain. AMPARs bind to a phosphorylated stargazin protein, which binds directly to the PDZ domain of PSD-95. PSD-95 also binds to an abundant synapse-specific GTPase-activating protein, synGAP which indirectly regulates AMPARs by regulating GTPase-activating proteins Ras and Rap. We hypothesis that reduction in binding of synGAP to PSD95 upon synaptic activation may increase the trapping of AMPA receptors. Using a pull-down assay, we are testing synGAP’s binding affinity to each PSD-95 domain, and its change in binding affinity after binding of Ca2+/Calmodulin and after post-translational modification by phosphorylation . Thus far, we have found that Ca2+/Calmodulin affects synGAP’s binding to PSD-95. Chemical Reaction Networks: Composition of Deterministic Functions Aakash Indurkhya Mentors: Erik Winfree and Dave Doty A chemical reaction network (CRN) is a formal model of chemical species that react with each other in a well-mixed solution. Recent investigations have characterized the power of CRNs to deterministically compute functions; Chen, Doty, and Soloveichik proved that precisely the *semilinear* functions can be computed by CRNs. We explore the conditions under which CRNs computing functions can be *composed*, i.e., feeding the output of one CRN as the input of another downstream CRN, to compute the composition of the functions they individually compute. If the upstream CRN ever consumes copies of its output, then simply mixing the two CRNs will not necessarily work, since consumption of the output by the downstream CRN could interfere with the correct operation of the upstream CRN. Our goal is a systematic transformation on a CRN to enable its composition with similarly transformed CRNs, with minimal change to the reactions of the original CRN (preserving advantageous aspects of the CRNs such as efficiency in number of species or number of reactions). We present preliminary approaches and identify constrained classes of CRNs on which the approaches work. We discuss limitations to these approaches and the challenges involved in composing arbitrary deterministic CRNs. Signal Processing Optimization for Very Large Radio Arrays Narek Isaghulyan Mentor: Larry D’Addario Very large radio telescopes like the Cosmic Dawn and the Square Kilometer Arrays are heavily dependent on digital signal processing, and the power consumed by these electronics could be impractical. Significant power reductions are possible with specialized integrated circuits (ASICs). The main signal processing components that need to be implemented using ASICs are the Filter Banks, Beamformers and the Correlators. As a SURFer, I have created a realistic Matlab simulation of a filter bank ASIC design including oversampling and fixed point arithmetic. Further research topics I plan to investigate this summer include estimating the area of the chip, the number of copies of the design per chip, the speed, the power consumption and the memory implementation. Furthermore, the power consumption also depends on the array configuration. For instance, according to another Matlab code I have created, the optimum array configuration for the Cosmic Dawn Array, with fixed survey speed N2 Ω = 3.05*107 sr (where N is the number of antennas and Ω is the angular field of view) and one beam per station, is 10 antennas/station and 3070 stations; power consumption is then 15.4 kW. Action Authoring Interface Design for the DARPA Robotics Challenge Gregory Izatt Mentor: Seth Teller The DARPA Robotics Challenge (DRC) is a competition motivating the development of tele-operated, semiautonomous robotic platforms designed for disaster response scenarios. Success in these scenarios requires the planning and execution of complicated interactions with objects in the environment. We extended a prototype interface for such action authoring, emphasizing ease of use and speed of authoring via a series of revisions to both the interface itself, and its motion planning backend. The improved interface allows for a pilot to rapidly generate action plans, guided by nearly real-time feedback from the planner. Generated plans have been successfully verified in simulation.

38

  Implementation of an Online Spike Detection and Sorting System for Electrophysiology Using Tetrode Drives Ruchi Jahagirdar Mentors: Thanos Siapas and Andreas Hoensalaar We aim to develop a software toolchain that provides immediate online feedback from neural data acquired by the recording devices in the lab. The devices are tetrode drives that allow us to position up to 28 tetrodes in brain areas of interest. Immediate feedback using online detection and visualization will allow for adjustment of tetrodes for prime recording quality in a given location. Using the Chaco toolkit in Python, we developed a system that analyzes and visualizes live data in the form of density plots. The system uses approximations to the time-intensive algorithms used for offline spike sorting and detection to keep up with the high bandwidth of the recorded signal. Unsupervised Clustering of Transient Events in the Variable Sky Nishank Jain Mentors: Ashish Mahabal and Ciro Donalek Discovering transients and classifying them is a complex process. The Catalina Real Time Transient Survey detects and publishes all transients that it discovers within a span of few minutes of observation to VOEventNet and SkyAlert. Using the Caltech Time Series Characterization Service, many periodic and non-periodic features are extracted from each lightcurve thereby creating a new dataset to be used for categorization and classification. Lightcurves can show tremendous variation in various characteristics which makes comparisons between them difficult and training classifiers even harder. A way to approach this problem is to characterize a set of lightcurves via a set of common features and then use this representation of lightcurves for analysis and training of classifiers. From the lightcurves of known classes, mathematical representations of the best of these features can be derived and then used as training sets for unsupervised clustering of transients. One such approach using Self-Organizing Maps with the CRTS dataset through application of the SOM Toolbox in Matlab is demonstrated. Predicting Membrane Protein Expression Using Support Vector Machines (SVM) Nauman Javed Mentors: William Clemons, Jr., and Axel Müller Although membrane proteins account for a significant proportion of most organisms’ genomes, their structures and functional mechanisms are notably underrepresented in comparison to soluble proteins due to problems with overexpression. In contrast to soluble proteins, biosynthesis of membrane proteins requires two additional steps: localization of the ribosome-nascent chain complex to a SecY complex in the membrane and co-translational integration into the membrane, both of which are prone to failure in membrane protein overexpression. Expression studies on many different proteins have shown that certain proteins are more expressible in host organisms than others based on characteristics such as the encoding gene’s codon bias and the protein charge distribution. Several python scripts and other protein analyses programs were used to determine 28 nucleotide and amino acid sequence characteristics of thousands of positively and negatively expressed membrane proteins taken from the NYCOMPS(New York Consortium on Membrane Protein Structure). In order to determine whether membrane protein expression could be predicted based solely upon the gene and amino acid sequence, and to determine which features of the protein bore the most weight upon protein expression, SVM binary classifiers were trained on these calculation datasets and used to predict protein expression for several testing datasets. Although the prediction accuracy did not surpass 60% with a Matthews Correlation Coefficient of 0.20, several methods are being implemented in order to potentially improve the classifier, including creating a dual layered classification, adjusting classification cost factors, and implementing new kernel functions. Understanding the factors that govern membrane protein expression through such a predictor would greatly benefit efforts to increase protein overexpression, and thus aid target selection and structure determination. The Stability of Zero Energy Edge Modes in 1D Quantum Chains Adam S. Jermyn Mentors: Jason Alicea and Roger Mong Several one-dimensional quantum systems are known to exhibit localized edge modes. These modes relate degenerate states by transformations which are localized at either end of the system and decay exponentially into the bulk. The existence of such modes is a universal property of these systems, which include the Ising and Potts models, but only occurs in specific regions of their parameter spaces. In addition to being of interest for fault tolerant quantum computing, these modes are also curious because their existence makes a very strong statement about the structure of the energy spectrum of the system beyond the ground state. As a result, understanding them is of significant interest. We have investigated the properties of such modes on the boundaries of regions where they exist, both with analytic tools and with numerical methods. The results, which characterize the stability of these modes, should help inform decisions as to which systems are viable candidates for use in quantum computing systems.

39

  Endomotives of Toric Varieties Zhaorong Jin Mentor: Matilde Marcolli Endomotives are generalizations of motives in algebraic geometry that relate different cohomology theories of algebraic varieties. They are introduced as a way to describe non-commutative spaces that arise in quantum statistical mechanical systems related to number theory. The most basic construction starts with the prototype of Bost-Connes system associated to the one dimensional algebraic torus , and in general the data consist of a projective system of algebraic varieties and an action of a semigoup on the limit lim→ , then the analytic endomotive is given by the C*-algebra ⋊ . These systems have many interesting properties including natural time evolutions and Riemann ς-functions as partition functions, and has applications in the noncommutative geometry approach to the explicit class field theory problem as well as geometry over “the field with one element”. On the other side, toric varieties provide a large collection of “nice” spaces usually considered as a testing ground for theories. We generalize the basic Bost-Connes system to construct endomotives from toric varieties and study their quantum statistical mechanical properties. Bisimulation of Polymer Reaction Networks Robert Johnson Mentor: Erik Winfree The Chemical Reaction Network model is known to be incapable of deterministic Turing-universal computation. When the model is extended to allow formation of arbitrarily long polymers, however, universal computation becomes possible; in particular, a DNA implementation of a stack machine has been constructed. We propose an extended model that is capable of describing that stack machine. We also adapt the concept of weak bisimulation, a notion of equivalence of concurrent systems which has been previously defined for finite CRNs, to the extended model. Bisimulation has been used in the finite CRN model to prove that certain DNA implementations of formal CRNs are valid, and may have a similar use for polymer networks. However, as the extended model is capable of simulating Turing machines, we show that bisimulation is not recursively enumerable, and in fact is ∏^0_2complete. We also prove that the DNA implementation of the stack machine is a valid bisimulation of a formal network which simulates a stack machine. Finally, we define single-locus PRNs, a subclass of PRNs which includes the existing stack machine construction as well as a copy-tolerant Turing machine, and show that any single-locus PRN can be bisimulated by a network using two hypothetical polymer primitives, suggesting a possible future method of DNA implementation. Gender Differences and the Role of IKKB Kinase Complex in Macrophage and Astrocyte Activation, Cytokine Production, and Pathology in Mouse Models of Huntington’s Disease Nichole R. Johnston Mentors: Paul H. Patterson and Ali Khoshnan Neuroinflammation is believed to play an important role in neuronal cell death in Huntington’s Disease (HD). The specifics of such a role, however, are still under investigation. The major regulatory system of inflammation - the IKK/NF-‐κB pathway - is hyperactive in the brains of HD mouse models. The IKK complex directly associates with the mutant protein huntingtin, is found in inclusions of the nucleus in cells with HD, and enhances NF-‐κB activity. Because the IKKβ subunit is the predominant kinase promoting inflammation, we have established a colony of mice with IKKβ genetically disabled to determine its effects on inflammation and pathology. Using immunohistochemistry to compare astrocyte and macrophage levels in wild type versus HD mice with functional IKKβ, we noticed distinct gender and age-related differences in astrocyte activation within the cortex and hypothalamus. Furthermore, there is a striking difference in astrocyte levels of the striatum between the wild type and HD mice. Comparing these groups with the HD IKKβ-deleted mice, initial data suggests that the absence of IKKβ results in an increase of astrocyte activation, but decrease of astrocyte processes. High-Resolution Sea-Level History From Southeast Asia, Over the Past 8000 Years Bhavana Jonnalagadda Mentors: Adam D. Switzer and Aron J. Meltzner Glacial Isostatic Adjustment (GIA) models, which account for the loading and unloading of seawater and ice sheets on continents and continental shelves, are used to forecast future relative sea level (RSL), but they must first be able to accurately show the RSL in the past. Currently, no models match past RSL in Southeast Asia, and accurate sea level data from the region are needed. Previously, RSL data from the Sunda Shelf were scarce and unreliable, despite the region being crucial to GIA modeling with its complicated topography. By analyzing mid-Holocene microatoll coral slabs from Belitung Island, we compiled a comprehensive data set for the RSL on the Sunda shelf. We traced annual growth bands on the x-rays of the slabs, and dated samples using radiocarbon and U-Th techniques. Additionally, we found more realistic uncertainties for the U-Th dates. Reported lab errors on the U-Th dates were too small in comparison with the spread of dates coming from multiple samples from a single coral; in order to make the U-Th dates useable, we statistically analyzed the reported ages, to estimate more realistic errors for each analysis.

40

 

Retinal Circuit Builder Alex B. Jose Mentors: Markus Meister and Hiroki Asari As far as marvelously complex neural assemblies go, the brain typically steals the spotlight. But, as more and more research is revealing, a large amount of non-trivial computation occurs in the retina. In fact, the retina makes for an excellent model system for studying characteristics of neural circuitry – it is far from simple, but still simple enough to avoid some of the confounds that make study of larger-scale neural networks so frustrating. Additionally, unlike many other neural circuits in the brain, we have complete experimental access to the retina, and its inputs and outputs are well-defined. While an excellent model system for study, the retina is still incredibly complex. With its multitude of cell types and internal structures, it can be cumbersome to verify that hypothesized neural circuitry predicts expected retinal output. Thus our objective is to develop a retinal ‘circuit builder’ simulation framework, to expedite and simplify the process of constructing simulations of retinal microcircuits. Ideally, such a tool would augment research into the structure and function of the retina, from which many benefits could arise. Successful retinal prosthetics, for example, entail a thorough understanding of the complexities of retinal circuitry, which this circuit builder could help elucidate. Revealing Hallucinogen-Induced Brain Activity in Mice Using MEMRI Devashish Joshi Mentors: Paul Patterson and Natalia Malkova A key symptom of schizophrenia is hallucinations, occurring in 50-70% of all people diagnosed. Hallucination-like activity can be induced in the mouse brain using the hallucinogenic drug, 2,5-dimethoxy-4-iodoamphetamine (DOI). Our group used Manganese Enhanced Magnetic Resonance Imaging (MEMRI) to show that DOI specifically stimulates accumulation of manganese in the pre-frontal cortex an hour after injection. However, MEMRI of longerterm DOI effects on brain activity have not been studied. Twenty-four hours after injecting manganese, baseline images were collected followed by double injections of DOI/saline in intervals of one hour. The mice were imaged again six hours after the last DOI injection. Statistical analysis was performed to locate new regions of activity induced by DOI and to track manganese transport among neurons over time. These statistical parametric maps are currently being analyzed for differences in brain activity. Once the protocol has been established, the approach can be applied to the maternal immune activation (MIA) mouse model, in which the mice exhibit a schizophrenia-like phenotype, to see if the MIA brain exhibits a different pattern of activity than controls 24 hours after manganese loading. Initial Retention of Water on Mars Daniel Kang Mentor: Andrew J. Friedson Mars' early formation, at its earliest only 1 million years after the formation of the sun, came at a time of intense solar activity with extreme ultraviolet and x-ray radiation greater than ten times their present intensity. The energy of accretion created a magma ocean on the surface, which upon solidification, is thought to release a massive steam atmosphere. In the most recent models, the steam atmosphere is quickly evaporated by the intense solar radiation. Through computational modeling of the early Martian atmosphere, we attempt to verify this assumption by calculating whether Mars, if given time, would have cooled for the steam atmosphere to begin to condense and whether that would have happened before the steam atmosphere could escape, thereby allowing the condensation of a large amount of water on the surface. If so, Mars’ steam atmosphere may have formed a layer of ice that remained on the surface until possible later release. Basemap Creation and Instrument Deployment Study for InSight Landing Sight Downselection Soumya Kannan Mentor: Matthew Golombek In August of 2012, the InSight (Interior Exploration using Seismic Investigations, Geodesy and Heat Transport) mission was selected by NASA as part of the Discovery Program to launch in 2016. The goal of InSight is to study the formation of terrestrial planets present in the inner solar system. The InSight mission will land three instruments on the surface of Mars: the Seismic Experiment for Interior Structure (SEIS), the Heat Flow and Physical Properties Package (HP3) and a precision tracking system (NASA 2013). Selection of the landing site for the InSight mission is currently underway, with the downselection from the original twenty-two ellipses to a final four occurring in July 2013. The original twenty-two sites were identified using available remote sensing data from Thermal Emission Imaging System (THEMIS), IRTM on the Viking mission, and Thermal Emission Spectrometer (TES) to detect thermal inertia and rock abundance, all of which are relatively low resolution sources. In order to perform more detailed analyses of the final four ellipses as the downselection moves forward, a continuous map of higher resolution imagery was created from data obtained from the Context Imager (CTX) instrument aboard the

41

  Mars Reconnaissance Orbiter (MRO). This will function as a basemap for further studies. In addition, analysis was performed of rock fields in the areas of Viking Landers 1 and 2 and the Phoenix lander to model the deployment of the HP3 instrument with respect to cumulative fractional area covered by rocks in the proposed InSight landing terrains. Compact Variable White Dwarf Binaries From PTF and CRTS: Constraining Accretion Activities Through Classical Periodicity Search and Statistical Heuristics Wil Kao Mentors: Shrinivas R. Kulkarni and Sumin Tang We present a catalogue of 174 variable white dwarf (WD) binary candidates and WD-like objects. With a set of hierarchical selection criteria as well as photometric measurements from the Palomar Transient Factory (PTF) and Catalina Real-Time Transient Survey (CRTS), we parsed the light curve, folded light curve and Lomb-Scargle periodogram for each object to isolate the most promising candidates. As a result, 124 periodic candidates (23 strongly periodic), 32 short-term outbursting systems (19 confirmed with PTF images) and 18 objects with long-term variation were identified. As the largest periodicity search for WD binaries to date, we present period estimates for 107 of the 124 periodic candidates, and found that most are likely compact. Our heuristics-based selection methodology, though effective and straightforward, is not without the impact of human factors. Therefore, we relaxed the admission criteria for the final catalogue in the hope that spectroscopic and photometric follow-up will facilitate the precise classification of these objects, and that this catalogue will serve as a valuable source of observational constraints for the development of common envelope theory. Understanding the Faults of the Northern Volcanic Zone of Iceland Using Interferometric Synthetic Aperture Radar (InSAR) Aditya Karan Mentors: Mark Simons and Romain Jolivet The Northern Volcanic Zone in Iceland contains one of the few divergent plates above sea level. Therefore, it offers a rare chance to observe post-rifting displacement. Using interferometric synthetic aperture radar (InSAR) from data gathered by the European Remote-Sensing Satellites (ERS) as well as the Environmental Satellite (ENVISAT), raw images were contrasted using the software packages ROI_PAC and the New Small Baseline Chain (NSBAS). From there, images were correlated together and using these correlated images, interferograms were produced. These interferograms allow one to visualize the net change in the topology of a given region over period of time that the data was acquired. By combining all of these interferograms together, one can produce a continuous timeseries to model the sources of the deformation. Development of a New Flight-Capable K-Ar Geochronology Instrument Jennifer Karolewski Mentors: Paul Asimow, Kenneth Farley, and Julia Cartwright The geologic history of Mars presents many interesting research topics, including the study of climate change and the potential for human colonization. However, while its surface has been extensively mapped and studied by the Mars rovers, the inability of the vehicles to return samples to Earth or perform in-situ age determinations are significant problems for determining absolute ages of the geological features observed. In response to the unique challenges presented by the Martian environment, a novel potassium-argon (K-Ar) dating technique utilizing double isotope dilution has been developed at Caltech. A pilot version of a single instrument to perform K-Ar analysis is currently being designed and built. The instrument includes two ion sources: one for Ar and one for K. With these two sources, a single sample can be dated by K-Ar on a single instrument instead of two separate samples and two separate instruments. In addition, a suite of samples from the Ries Crater, Palisades Sill, and Lonar Crater were selected as suitable Martian analogues and analyzed using electron microprobe for potassium content. Upon completion of the instrument, dating will be performed on the samples to confirm the accuracy and precision of the mass spectrometer. J-tile Entropy Calculations From a Statistical Thermodynamic Model Raj Katti Mentors: Mark Bathe and Erik Winfree The structural integrity and versatility which make DNA an ideal material for encoding the secret of life also make it useful for building nanoscale structures. Applications of DNA nanotechnology range from cell-specific drug delivery to the digital logic circuits underlying molecular computers. Because of its modularity and scalability, DNA tiling shows promise of being the future paradigm of DNA nanotechnology, but the thermodynamic properties of DNA tiles are not yet well understood. Our goal was to derive the thermodynamic parameters of a Holliday Junction tile (J-tile) from first principles.

42

  We developed an analytical model of a Holliday Junction tile and derived the corresponding partition function and entropy. To test the viability of our model, we ran molecular dynamics simulations, and obtained entropy estimates using two different methods, the Schlitter Entropy approximation and the Adaptive Biasing Force (ABF) method. Results of our simulations indicate that our model poorly estimates the entropy of the tile, but further simulations need be run under a variety of conditions to form a definitive conclusion. 2D Combustion Modeling and Analysis: A Computational Study of Soot Formation in Laminar Diffusion Flames Derek Kearney Mentor: Guillaume Blanquart Soot formation is an unwanted product of many combustion reactions. Identifying the source of soot requires both a thorough understanding of flame mechanics and a predictive computer model. The focus of this project revolves around creating a flame model that accurately visualizes an experimental flame with the same setup and configuration. Through the use of multiple programs, the flame model is created in three main stages. The first stage creates over a hundred “Flamelets”, which contain all of the possible flame chemical data under certain initial conditions. The second stage organizes all of these “flamelets” into a “chemtable” that contains an organized data set needed for the flame simulation. The final stage is to run the NGA code, which contains the fluid mechanics and uses the chemtable to simulate the flame model. Once the flame is simulated, it can viewed in Ensight and then all necessary data can be extracted and compared to experiments. Creating an accurate flame model requires one to repeat this process several times, adding tweaks and fixes along the way until it is perfected. Directed Geometric Design of GaAs Nanowire Arrays Christian Kendall Mentors: Harry Atwater and Kate Fountaine GaAs is a remarkable material for solar applications due to its strong absorption in the visible spectrum and superior electronic properties. Arrays of GaAs nanowires (NWs) have been shown to absorb many times more light than planar GaAs per unit volume of material due to strong coupling into resonant waveguide (WG) modes. The NW dimensions determine the peak absorption wavelengths. We sought to design and examine NW geometries to (1) characterize effects of NW dimensions on WG modes and (2) achieve broadband peak absorption across a large part of the visible spectrum. Using Lumerical FDTD, numerous array models were tested with great ease, speed, and low cost compared to experimental fabrication. Simulations and analytical calculations were used to examine and describe the performance of possible NW geometries that can access multiple WG modes and achieve broadband absorption. The useful absorption and generated photocurrent of these models are used as figures of merit. Structures with similar fill fractions to single radius NWs, such as a flat-tipped cone WG, achieved a 192 mA/cm2 increase in generated photocurrent. Simulation experiments and analytical calculations confirmed successful excitation of multiple WG modes. How Does Galaxy Environment Modulate AGN Activity? Emil T. Khabiboulline Mentors: John D. Silverman and Charles L. Steinhardt Black holes growing through accretion, which are the source of active galactic nuclei (AGN), strongly influence their host galaxy. Intergalactic environmental processes are important because they can modulate black hole growth and corresponding AGN activity, but do so in ways that are not well understood. Using the extensive dataset from the Sloan Digital Sky Survey (SDSS), supplemented by improved data on galaxy mergers and clusters, we conduct a study of AGN activity hoping to form a general picture of the role of environment. We apply a continuous measure of AGN activity to all galaxies and then determine how it changes between samples specifically matched to emphasize or marginalize the contribution of key properties. We find consistent merger-induced enhancement and cluster-induced suppression of AGN activity, with clustering dominant over merging. Oxidaton of Endonuclease III by Guanine Radicals Shushan Khorozyan Mentor: Eric Stemp Oxidative damage to DNA is important in the progression of molecular diseases such as cancer and guanine is particularly susceptible due to its low oxidation potential. In vivo, oxidative lesions are often corrected by base excision repair (BER) enzymes such as MutY and it has been proposed that since many DNA glycosylases have FeS clusters that are redox-active once the protein binds to DNA, these proteins might use through-DNA charge transfer (CT) to scan the DNA for defects. Oxidation of guanine leads to MutY oxidation and our goal was to demonstrate that EndonucleaseIII can exhibit similar behavior. Here, we designed oligonucleotides containing 2-aminopurine (a fluorescent base that can photooxidize guanine) and a series of guanines decreasing in ionization potential with increasing distance from the 2-aminopurine, which should helping to minimize back electron transfer.

43

  In a model protein, we observed oxidation of ferrocytochrome c upon irradiation of the 2-aminopurine, consistent with protein-to-DNA electron transfer. With that, our future goal is to monitor reaction between guanine radical and Endo III in real time using nanosecond transient absorption spectroscopy and obtain evidence for permanent oxidation products by gel electrophoresis and uv-visible spectroscopy. Comparing Orbital Chromatic Roots to Classical Chromatic Roots Dae Hyun Kim Mentor: Mohamed Omar Chromatic polynomials are ubiquitous in mathematics, and their symmetrized counterpart, called orbital chromatic polynomials, have been of recent interest as well. Of specific interest is whether the real roots of orbital chromatic polynomials are bounded above by the greatest real root of corresponding chromatic polynomials. Though Peter J. Cameron conjectured this to be true, through constructions, we prove otherwise. However, the root bound still holds for many categories of graphs, including paths, cycles, wheels, and outerplanar graphs with not all cycles even. In addition, we show that some graph operations, including disjoint unions of connected graphs and graph joins with complete graphs, preserve the property. Riches to Rags: Do NFL Professionals Make Poor Financial Decisions in Their Short High-Paying Careers? Joshua J. Kim Mentors: Colin Camerer and Kyle Carlson A pillar of neoclassical economics is the life-cycle savings hypothesis (LCSH), which analyzes intertemporal spending patterns and predicts that individuals smooth their consumption profiles over the course of their life. Traditionally, these LCSH models have assumed that individuals are perfectly rational and have abundant willpower. In this project, we update these unrealistic psychological assumptions by analyzing the financial decisions made by NFL athletes. We choose NFL athletes as our sample set due to their highly-variable income stream and uncertain career length. Unlike previous studies of NFL players, which utilized survey data, we create our own datasets detailing salary histories; injury records; game logs; and personal files such as address history, bankruptcies, crimes, and liens. Using our data, we hope to update current LCSH models with psychological parameters to create a more realistic model of intertemporal decision-making. Time Domain Analysis and Discovery and Characterization of Periodic and Eclipsing Sources With the Palomar Transient Factory and Catalina Real-Time Transient Survey Kyu Bin Kim Mentors: Thomas Prince and Sumin Tang The Palomar Transient Factory (PTF) is a large-area sky survey that looks at 1000 square degrees of the sky daily. PTF searches for variable and transient targets in the sky and from that data new periodic sources such as compact binary stars can be discovered. The objective of this research project was to discover new eclipsing white dwarf (WD) binary star systems using data from PTF because systems such as eclipsing WD binary systems are of scientific interest because close ones are thought to produce gravity waves, provide interesting insights in stellar evolutions, and are test beds for gravity. To search for these stars, a WD catalogue was made by compiling various catalogues from literature and the PTF database was queried according to the compiled catalogue. Using a series of statistical cuts to the data, the approximately 10000 WD’s that PTF had were reduced to 90 eclipsing binary candidates. Conditional entropy, a novel periodicity search algorithm presented in Graham et al. 2013, will then be used to calculate the periods of these eclipsing candidates. So far only one source, PTFS1303n, has been identified as an eclipsing WD binary and follow up observations have been requested, but there are expected to be several more once individual candidate evaluations are completed. Incorporation of 4-Azidoproline Into Fibronectin in Escherichia coli Shi En Kim Mentors: David Tirrell and Seth Lieblich HIV cripples the human immune system by binding to cell receptors and surface proteins to facilitate infection. One of the virus’s envelope glycoprotein, GP120, allows the virus to fuse itself to the cell membrane for viral entry. Previous studies have revealed a peptide which utilizes the non-canonical amino acid 4-azidoproline (4azp) and a ferrocene inhibits GP120 function. This peptide thus acts as a fusion inhibitor. Since proteins make more suitable drugs than peptides as they are more soluble and have a longer blood half-life, our goal is to design a protein HIV fusion inhibitor containing 4azp, utilizing the known GP120 binding peptide in a solvent exposed loop. We will then test whether our engineered protein retains GP120 affinity and entry inhibition. Our methodology includes cloning our protein (here the fibronectin) onto a plasmid, which will be transformed into a proline auxotroph E. coli stain. We will test for 4azp incorporation into the expressed proteins of E. coli. Should the wild type tRNA synthetase be unable to charge 4azp in the place of proline, we will mutate the synthetase such that 4azp can be charged onto tRNAs, allowing for the expression of fibronectin containing 4azp.

44

  Structural and Functional Characterization of Members of the Mammalian Transmembrane Domain Recognition Complex (TRC) Pathway Sumin Kim Mentors: Bil Clemons and Jeeyoung Mock Tail-anchored (TA) proteins undergo post-translational translocation due to their C-terminal membrane anchor and are involved in processes such as protein degradation and translocation. Insertion of the TA proteins to the ER membrane involves several cytosolic factors, including the Bag6-Ubl4A-TRC35 complex and the TRC40-CAML-WRB complex, which are homologous to the Sgt2-Get4-Get5 complex and the Get3-Get2-Get1 complex of the yeast GET pathway. In order to elucidate the role and mechanism of these protein complexes, the Bag6-TRC35 and the CAMLTRC40 proteins were expressed and purified for structural determination, and the interaction between CAML and TRC40 was verified. Wireless Nano-Scale Optoelectronic Neural Interfacing Deepan Kishore Kumar Mentors: Axel Scherer and Akram S. Sadek Interfacing neural signals has been a longstanding trade-off between the number of micro/nano-electrodes and the number of interconnects, due to the inherent wiring problem that arises. We present a novel method of overcoming this problem by using Optical Gradient Force based Nano-Opto-Electro-Mechanical-System (NOEMS) Sensor. We design and simulate a NOEMS sensor that would be able to multiplex various brain signals by tuning the resonant frequency of the NOEMS sensor. The optical gradient force that arises due to the evanescent field coupling between the waveguide and the substrate causes the piezoelectric waveguide to oscillate. This oscillation is fine tuned by superimposing various neural signals on the Peizo waveguide, thus causing a change of resonant frequency due to the change in stiffness of the waveguide. Our results indicate that the Maxwell’s Stress tensor of 0.231E-6 Pa would be driving the resonator. We also study how the stress change affects the stress tensor, and perform a 3D simulation of the entire NOEMS. Synthesis of Substrates for Terpene Synthase Engineering Rebekah Kitto Mentors: Frances Arnold and Ryan Lauchli Terpenes are natural products that have many applications as drugs, essential oils, and biofuels, and are produced in nature by terpene synthase enzymes. We have synthesized modified “surrogate” substrates that react with terpene synthases to produce quantifiable amounts of methanol as a side product. The methanol that is produced is measured using an AOX-Purpald screen and can indicate the amount of primary product present, thereby measuring the overall activity of the enzyme variants. Using an m-CPBA oxidation, a periodic acid cleavage, and a Wittig reaction, we have made surrogate substrates of both neryl pyrophosphate (NPP) and geranyl pyrophosphate (GPP) for testing with monoterpene synthases. We have also developed an improved expdient route to the surrogate form of farnesyl pyrophosphate (FPP), known as surrogate 1. Additional synthetic routes are currently being developed for several novel surrogates. Spectroelectrochemical Assay for Identifying Reactive Oxygen Species Mette Lützen Hoff Kjeldsen Mentors: Harry B. Gray, Michael Hill, and Bryan Hunter More than 500,000 face and neck surgeries are performed each year in the United States. With an average cost of $10,000 dollars per patient, the procedures are invasive and require long recovery times. Along with a surgical team from UC Irvine we are working on a noninvasive technique for reshaping cartilage: electromechanical reshaping of the tissue. We have hypothesized that the oxidation of water and the concomitant acidification of the cartilage is what causes the shape to change permanently. However, significant cell mortality is observed in reshaped tissues. One possible cause is the creation of reactive oxygen species (ROS) at the electrode when water is oxidized. The goal of this project is to develop a new spectroelectrochemical assay to identify different reactive oxygen species formed when electrolyzing water. To this end, we have synthesized photosensitive dyes that “turn on” when coming in to contact with certain ROS. We hope to be able to measure the emission spectra of these dyes in real time, thereby identifying the species that are created upon water electrolysis under different conditions (e.g. electrode materials, electrolyte composition, and applied potential). Simulation of the Light Collection Properties of LYSO Crystals Hannah Klion Mentor: David Hitlin The Mu2e experiment aims to observe and measure the rate of the neutrinoless conversion of a muon to an electron, a process that is not predicted by the standard model. The process causes the emission of an electron with energy approximately equal to the muon rest mass, 105.6 MeV. The energy of the generated electrons will be measured using a combination of a tracker chamber and an electromagnetic calorimeter, which will be composed of

45

  cerium-doped yttrium orthosilicate (LYSO:Ce) crystals. The goal of this project is to simulate the scintillation and light collection properties of LYSO:Ce crystals when interacting with 105.6 MeV electrons and 511 keV gamma rays that will be used for crystal calibration. We use the GEANT4 toolkit, which is designed for the simulation of the interaction of particles with matter. The simulation consists of a crystal and a simplified avalanche photodiode (APD) detector. Hits on the APD are recorded for post-processing. We include all relevant optical and electromagnetic physics processes, such as Compton scattering, scintillation, and optical absorption. The simulation can easily be extended to different crystal and detector properties through a configuration file. We find that the results of the simulation are accurate to first order. BMP Receptor Multiplicity Generates Non-Monotonic, Antagonistic Input-Output Functions Heidi Klumpe Mentors: Michael Elowitz and Yaron Antebi External signals influence cells via different signaling pathways, which generally involve binding to membranebound receptors and activating intracellular transcription factors thus effecting internal functions. Often, multiple signals and receptors activate the same intracellular factor, such as the nearly twenty ligands and eight receptors of the bone morphogenetic protein (BMP) signaling pathway. We hypothesized that these components are not redundant, but make it possible to integrate complex inputs and compute complex outputs that generate unique cell functions. We use flow cytometry to measure a synthetic reporter of BMP pathway activity and find that the transcription factor activation is non-monotonic, and nonintuitively decreases at high BMP ligand concentrations. A mathematical model describing the interactions between pathway components can capture these results and predicts that the decrease is a result of ligands binding to receptors that activate the transcription factor inefficiently. To verify these data and the math model, we tested a counter-intuitive in silico prediction that knocking down the inefficient receptor type would counterintuitively increase the extent of BMP pathway activation. Also, to explore the functional relevancy of Smad activation, the dependence of mesodermal differentiation on Smad activation was measured. Alpha Arylation of Malonate Ester Yoobin Koh Mentors: Brian Stoltz and Seojung Han Oxindole is considered a core scaffold for biologically active natural products and small molecules. In addition, many oxindole derivatives are expected to act as potential anti-oxidant, anti-cancer, anti-HIV, neuroprotective, or antiproliferative drugs. The objective of this experiment is to find the right conditions for arylation at the carbon alpha position of malonate ester of oxindole at C3 position. The expected effect of this project is to further progress the utilization of C3-malonate oxindole, so it can be used to make complex molecule such as communesin. Electromechanical Coupling in a 1-D Crystal Using a Triangular Nanobeam Design Daniel L. Kong Mentors: Oskar Painter and Johannes Fink Here, we present the techniques used to design a wedge-shaped phononic crystal nanobeam optimized for electromechanical coupling. Nanobeams with a triangular cross section can be fabricated from low resistivity bulk dielectrics using angled dry etching, in contrast to traditional designs which rely on multi-layer wafer material. Electromechanical coupling to a microwave LC circuit is achieved using an induced capacitance change due to the displacement of the beam. Of particular interest is the beam’s ‘breathing’ mode, which is a high frequency symmetric standing wave with significant transverse displacement. When the breathing mode is excited, the electrical mode’s frequency shifts by some amount proportional to the strength of the coupling. Applications include efficient coherent conversion between microwave-frequency electrical signals and optical photons in devices using both optomechanical and electromechanical coupling. Modeling the Transient Behavior of Electric Double Layers of Metals in Dilute Saline Solution Nuttawut Kongsuwan Mentors: Axel Scherer and Meisam Honarver Nazari We have studied the formation of electric double layers (EDL) in dilute saline solution of which concentration is comparable to that of hydronium and hydroxide ions in deionized water. Three governing differential equations, Poisson’s equation, Nerst-Planck eqaution and conservation of mass equation, are coupled and solved in one dimension using Comsol multiphysics software package so as to find the time-dependent formation of EDL. Potential differences of range 0.01 - 1 V are applied across the system, and the simulation predicts that there are noticeable differences on the magnitudes and time scales of the EDL formation due to different ions. The results are verified by performing experiment using platinum electrodes of which surfaces are screened by insulated material, SU-8 photoresist, to prevent reactions. We also use Model 1200B Series Electrochemical Analyzer as a power source.

46

  Transition Metal Doped Ceria for Thermochemical Fuel Production Ian Koss Mentors: Sossina Haile and Tim Davenport Ceria (CeO2) is a promising catalyst for thermochemical fuel production, where it can be employed in a two-step process: first, the ceria is heated, inducing the evolution of oxygen; steam is then passed over the ceria as it is cooled, removing oxygen from the steam and generating H2. Previous studies suggested that doping ceria with transition metals such as Ni, Mn, and Fe has the potential to improve the efficiency of the catalyst, but the effectiveness of such dopants was unknown as neither the solubility of transition metals into ceria nor the changes in the catalyst's thermodynamics and kinetic properties had been measured precisely. This study sought to fill in these gaps, synthesizing Ni-, Mn-, and Fe-doped ceria with transition-metal-to-cerium concentrations of up to 1:9 and characterizing the materials with XRD, SEM, EDS, and direct testing of fuel productivity. Free Will, Decision-Making, and the Human Brain: Online Realtime Analysis of Intracranial Signals in Humans Srivathsan P. Koundinyan Mentors: Ralph Adolphs, Uri Maoz, and Liad Mudrik There is a long-standing debate about the role of brain signals that predict action, potentially well before subjects report being aware of having consciously made up their minds. The question we studied in this project was the extent to which unconscious neural precursors of subjects’ decisions can be detected in intracranial local field potentials (LFP) prior to the onset of action, on a single-trial basis, online and in real time. More specifically, we sought to build a continuous real-time prediction system (CRTPS) that would continuously analyze the current neural data and inform the experimenter of its best prediction for an upcoming decision along with a confidence level. We hypothesized that this confidence should gradually increase as the patient deliberates his movement and would reach a maximum around movement onset. To achieve this, we had to come up with classifiers that would train on features from brain activity recorded over previous trials to accurately and instantaneously predict the action for the current trial. Therefore, we spent a considerable amount of time developing and testing multiple automated feature selection algorithms, as hand picking features from LFP data was not practical in real-time analysis. Our ultimate goal here was to isolate combinations of classifiers and features that would provide prediction accuracies of about 80% or more, thereby enabling closed-loop experiments (e.g. changing the stimulus according to some features of neural activity, as it occurs) using the CRTPS. This would open the door to many studies that were so far impossible to carry out, expanding our knowledge about widely debated issues like agency, conscious experience, voluntary action and free will. On the Intersection of Secant Sets in the Hill Cap Nikola Kovachki Mentors: Nets Katz and Karsten Chipeniuk The Hill cap is a unique, up to equivalence, 56-cap K ⊂ PG(5, 3). In the original 1973 paper, outlining the construction this cap, a theorem of Hill implies that for any point P ∈ PG(5, 3)\K there are exactly 10 lines through it which also intersect the Hill cap. We call the set of points in K that are intersected by these 10 lines a secant set of K and denote it A(P). Since we know each line can only intersect K in 2 points, we can easily see that |A(P)| = 20. In a 2000 paper, Hill posed the problem of giving a theoretical proof to the computer result that for any two distinct points P, Q ∈ PG(5, 3)\K, |A(P ) ∩ A(Q)| ≤ 8. Here we give that proof. Constructing a High Bandwidth Microwave Photon to Phonon Coherent Converter Lev Krayzman Mentors: Oskar Painter and Johannes Fink The purpose of this SURF project is to set up measurement hard- and software for experiments on nanoelectromechanical systems in the microwave frequency range. This includes automatization software capable of controlling signal generators, a Vector Network Analyzer, a Spectrum Analyzer and a Field Programmable Gate Array data acquisition board using Matlab. This software allows for the automation of data collection on one instrument while sweeping power or frequency on either the same or a different instrument. Such sweeps are useful in detection of microwave resonance modes as well as mechanical modes in electromagnetically induced resonance measurements. The setup is used to characterize superconducting microwave lumped element resonances, the mechanical resonances of 1D phononic crystal nanobeams as well as their mutual coupling and time response at Millikelvin temperatures. In the future this system could be used to prepare and manipulate phonons on a microchip.

47

  Human Attribute Recognition Model Comparison and Dataset Creation Vansh Kumar Mentors: Pietro Perona and Steve Branson An essential function of computer vision algorithms is to perform object classification within images. A specific problem within the field of object classification is bird species classification, a task which is difficult even for humans. The goal is not to just develop an algorithm in order to solve this specific problem, but rather to extend the techniques used and to apply the knowledge acquired from performing bird species classification to other super categories of problems. An algorithm to solve the problem of bird species classification has already been developed as part of the Visipedia project in Dr. Perona’s lab, but the process to generalize these techniques to other super categories has yet to be determined. To solve the problem of human attribute recognition, first different parts detection models were compared on preexisting benchmark databases. This was done to determine which model would be used to perform parts detection before attribute recognition algorithms could be run on a new dataset. Then, a JavaScript interface was developed to aid in attribute classification for two different datasets – one was a set of fashion-based images and the other was a set of images of pedestrians collected in the Pasadena area. Then, this interface was deployed and attribute annotations were collected for each dataset. The final part of the project was to try training attribute recognition algorithms on the completed Pasadena dataset, but there was not quite enough time to fully test many algorithms, so only popular algorithms such as VLFeat were tested. Behavioral and Electrophysiological Assessment of 3D Object Recognition in Mice Jessica Lam Mentors: Doris Tsao and Tomo Sato The ability to extrapolate a 3D conception from a 2D image relies on border detection cells - certain cells that respond specifically to object edges in order to distinguish between objects in the foreground and those in the background. Such cells have yet to be identified in the mouse visual cortex. This study addresses whether mice have the high level ability to stitch together partially occluded objects and interpret the world as layers of objects that have three dimensional relations. The mice were trained to do a two-alternative forced choice task in which they swiped a conveyor belt to the left or right in response to various stimuli. The mice had to recognize stimuli, which consisted of shapes (e.g. squares and circles) that were partially occluded. Once the behavior tests are conducted, electrophysiological tests will be used to determine the neural mechanisms behind this ability to perceive occluded objects. An Electrostatic Harmonic Potential Ion Trap for Fourier Transform Mass Spectrometry Matt Lappin Mentor: Jack Beauchamp Compact, high resolution, spectrometers that operate in a low mass range that includes hydrogen and helium and consume little power are valuable devices with a variety of applications. Presented is an electrostatic ion trap, incorporating a potential well in which ions oscillate with a frequency proportional to the inverse square root of their mass, which can be constructed to meet these criteria as a Fourier transform mass spectrometer. The trap has been tested using SIMION, an ion trajectory simulation software package, and the ion trap and accompanying ion source that will be used to test the design have been designed and fabricated. To complete the device a vacuum system must be outfitted to test the instrument, and various instrumentation and power electronics still need to be implemented. Following these steps the instrument can be tested with gas mixtures that simulate the atmospheres of Mars and Titan. Applications and Extensions of Structural Topic Models With Covariates Jetson Leder-Luis Mentors: Dustin Tingley and Jean Ensminger Text analysis is a fruitful area of research, as traditional quantitative analytical tools far surpass the ability of social scientists to handle textual data. Topic modeling uses statistical inference to approximate the topical content of a corpus of documents; our Structural Topic Model (STM) performs the same inference but incorporates covariate information about the documents, such as date or experimental treatment. To support our implementation of the STM in both R and C, this research project has focused on the development a set of tools to collect, manage, and parse large political science datasets, as well as display the results of the model via a user-friendly visualization browser. Our results are a set of clean corpora for future analysis, and a set of tools to work from a corpus, through the topic modeling software, to a visualization of the textual content of the documents.

48

  A Novel Real-Time Eigenvector-Based Analysis of Brain Dynamics and its Application to Face Preference Prediction Barclay J. Lee Mentors: Shinsuke Shimojo and Sangwan Lee A computational model was developed and implemented in MATLAB for eigenvector-based analysis of EEG data. The algorithm is able to take in whole-brain dynamics in the form of EEG data and output eigenvectors and eigenvalues characteristic of brain dynamics within a certain time interval. A mathematical method for conversion between row-normalized versus column-normalized eigenvectors was implemented to reduce computation time. This model was applied by prompting subjects with a facial preference task and recording their brain dynamics. In doing so, it is theoretically possible to calculate characteristic eigenvectors for subject preference of either face, and comparing those eigenvectors with EEG data in real time allows prediction of which face the subject is more likely to prefer. Irrelevant activity that might appear during the task, such as blinking or saccadic eye movement, were controlled for prior to the task by establishing blink-characteristic and saccade-characteristic eigenvectors for each subject. These control eigenvectors were able to provide significant positive identification of either blinking or saccadic movement by the subject, which will allow for greater accuracy of facial preference prediction. The QuakeSim Project Chengyi Lee Mentor: Andrea Donnellan While teamwork is vital to everything from research to companies; many projects lack the online infrastructure needed that can simplify note keeping and foster communication. Our focus in particular is QuakeSim, a series of isolated web tools used for earthquake modeling and research. We present and implement a new design for the QuakeSim infrastructure that utilizes a heavily modified OpenAtrium for the intranet and LDAP for implementing single sign-on. This allows a website to have both public and private front-ends and give users freedom to create projects, but still keep website easy to administrate. The new QuakeSim website will give users the freedom to annotate, rate, and share their projects with various users, while the single sign-on system will allow users to seamlessly save the results from the QuakeSim tools directly to their projects, resulting in a streamlined experience. Building the Best Sea Level Curve of the Sunda Shelf Over the Past 8000 Years From Coral Microatolls Grace Lee Mentors: Adam D. Switzer and Aron J. Meltzner We aim to build a high-resolution sea-level curve of the Sunda Shelf over the past 8000 years by analyzing and interpreting coral microatolls from Belitung Island, Indonesia, in an attempt to improve Glacial Isostatic Adjustment (GIA) models. GIA models are geophysical and climate models, which when calibrated would give context to help us better determine the history of isostatic deformation along the Sunda Shelf and anticipate future relative sealevel change. Using graphics software, we mosaicked the coral slab x-rays, traced and counted the annual growth bands, and marked unconformities in the coral slabs. Coral elevations were surveyed relative to one another in the field by Dr. Meltzner. We used each coral’s elevation and growth history to interpret the relative sea-level history during the corals’ lifetime. We are assessing whether any of the corals’ elevations are likely to have been affected by moating, by jointly considering the corals’ spatial distribution, ages, and elevations; this should help us understand how the reef was built. Demonstration of Two-Plasmon Quantum Interference Hyunseok Lee Mentors: Harry A. Atwater and Jim Fakonas Surface plasmons—quanta of light that are sustained by the collective oscillation of a metal’s free electrons—open the possibility for single-photon applications in quantum computing. To explore surface plasmons’ quantum nature, we study two-photon quantum interference (TPQI) in plasmonic waveguides. An essentially quantum phenomenon, TPQI occurs when two photons enter a 50-50 beam splitter simultaneously; they exit the splitter in either output together, but never in separate outputs. We aim to demonstrate TPQI with plasmons instead of photons to test the predicted but not yet observed quantum behavior of plasmons. In preliminary experiments, we observe twoplasmon quantum interference with 64% visibility, showing that plasmons indeed exhibit quantum behavior like TPQI. We expect to gather more extensive and precise data of two-plasmon quantum interference in ongoing experiments.

49

  Py-Im Polyamide-Induced Replication Stress in LNCaP Cells: A Cell Cycle Study Through DNA and Protein Activation Assays Jihoon Lee Mentors: Peter B. Dervan and Thomas F. Martinez Pyrrole-imidazole (Py-Im) polyamides are a class of small, programmable molecules whose structures can recognize and bind to specific sequences in the minor grooves of DNA. They are cell permeable and localize to the nucleus of live cells, where they bind to the chromatin within. As a result, in the cellular environment Py-Im polyamides have the potential to alter gene expression by displacing natural activators and transcription factors; it has also been suggested that the binding may also lead to replication fork stalling. This has important mechanistic implications regarding Py-Im polyamides’ observed antitumor ability. To study the potential effects on replication, androgen-sensitive LNCaP human prostate cancer cells were treated with specific Py-Im polyamides targeting the consensus Androgen Response Element (ARE). The treated cells were then analyzed for the activation of replication fork, DNA repair-related, or apoptosis-signaling proteins using Western blots and luminescent caspase-3/7 detection. Additionally, a flow cytometry assay was also conducted, in which relative cell cycle stage populations were surveyed in order to observe changes in the cycle as a result of possible replication stress. Monte Carlo Studies of Gauge Theories for Topological Phases and Phase Transitions in Two-Loops Model With Statistical Interaction Jong Yeon Lee Mentors: Olexei Motrunich and Scott Geraedts In this work, we study a statistical mechanics model of two species of loops with mutual statistics θ in (2+1) dimensions. Through a reformulation, sign problem was eliminated and we could perform Monte Carlo simulation of this model. We focus on θ=2π/3 and investigated phase transition point between topological and non-topological phase to identify its properties. Using so called dual transformation, which is a change of variables, we can roughly predict phase transition points in terms of gapped dual variables by comparing the dual interaction to well-known 3D XY model. Using this approach, we can find phase diagrams for different θ, which is rational fraction of 2π. We investigate models with different values of θ using Monte Carlo simulation and find phase diagrams. In each case, we could predict different numbers of phases where dual variables are gapped in range of interaction strength, and match these phases with simulation data. Electrocatalytic Properties of Carbon Monoxide Dehydrogenase Catalysed Carbon Dioxide Reduction Katherine Lee Mentor: Fraser Armstrong One key political and scientific challenge is to develop an abundant, clean, and renewable energy source. Currently, energy production is dependent upon fossil fuels, which are a dwindling and environmentally detrimental fuel source. Solar cells can harness sunlight but not store it is as a transportable fuel. One possibility is to use solar energy to drive the reduction of carbon dioxide into hydrocarbons, which could act as a storable fuel. However, there is still no efficient, synthetic catalyst for the reduction of CO2. The bacterial enzyme Carbon Monoxide Dehydrogenases (CODHs) is a reversible catalyst for CO2 reduction which operates at a low overpotential and has a high turnover rate. The objective of this project is to study the potential range of electron transfer in CODH onto an electrode substrate, specifically the n-type semiconductor titanium dioxide. Cyclic voltammetry and electrochemical impedance spectroscopy are used to investigate the electron transfer over a broad range of potentials. Then, these systems are modeled as an equivalent electrical circuit which provides insight into the electron transfer of CO2 reduction in CODH onto titanium dioxide. Dynamics of Surfactant Spreading and Fingering Instability on Thin Films Lisa Lee Mentor: Sandra Troian Abstract: The spreading dynamics of surfactants on a thin liquid film of higher surface tension is of particular interest for many applications. It has been previously shown that surfactants above the critical micelle concentration exhibit a fingering instability, and that the spreading fronts follow a power law growth in time. This project studies the spreading dynamics of the nonionic surfactant oleic acid, as well as the anionic surfactants SDS and AOT, on a thin film of glycerol. Only pure oleic acid was experimented upon. Various concentrations of the anionic surfactants were tested, as well as various concentrations of the solvent, aqueous glycerol. We find that both of these factors play a role in the spreading dynamics. For the most part, decreasing the surface tension of the deposited surfactant (either by raising surfactant concentration or by lowering the amount of glycerol in the solvent) results in more ramified fingers, and then eventually another phenomenon, either square undulations (in AOT), or multiple fronts (in SDS). Additionally, we find that the fractal dimension of the spreading pattern tends to grow, then stay constant, and then drop off. Finally, we verify that the difference between the spreading fronts follows a power law growth in time.

50

  Cell-Type Specific Transcriptome Profiling in C. elegans Michelle Lee Mentors: Paul Sternberg and Meenakshi Doma Caenorhabditis elegans is a nematode used as a model system to study various biological and molecular processes. In C. elegans, current strategies for cell type specific gene expression analysis do not allow study of the transcriptome in specific cells in their native environment. This project explores a new method called thiouracil tagging that allows RNA to be isolated from specific C. elegans cells with spatial and temporal accuracy. Uracil phosphoribosyltransferase (UPRT) from Toxoplasma gondii is an enzyme that catalyzes uracil and incorporates it into RNA. I have created several strains that express UPRT in a cell type specific manner using highly cell-specific promoters. When the worms are fed a modified uracil analog, 4-thiouracil (4TU), UPRT incorporates the 4TU in its newly synthesized RNA. This RNA can then be tagged using biotinylation, purified using streptavidin, and sequenced using RNA-seq. With this technique, cell-type specific transcripts can be identified and rapid changes in gene expression detected by controlling the length of time 4TU labeling is carried out. The main advantage of this approach is that the production of the labeled-RNA occurs in vivo, thereby preserving the gene expression profile in a specific spatial and temporal context. Understanding the Regulation of OGT Sangjun Lee Mentors: Linda Hsieh-Wilson and Andrew Wang O-GlcNAcglycosylation is the attachment of β-D-N-acetylglucosamine to serine or threonine residues of proteins. This unique post-translational modification is found on many classes of cellular proteins including transcription factors, kinases, and cytoskeletal proteins. O-GlcNAc modifications have been shown to play roles in various diseases including cancer, diabetes, and neurodegeneration. The O-GlcNAc modification is regulated by OGT (O-GlcNAc transferase) and O-GlcNAcase. OGT adds β- D-N -acetlyglucosamine to serine or threonine residues of nuclear and cytoplasmic proteins, whereas O-GlcNAcase removes the sugar from the protein. The single OGT enzyme is responsible for the modification, and thus the regulation of OGT is a very important area of research. Our research focuses on understanding the regulation of OGT by protein phosphorylation. The Hsieh-Wilson lab has previously identified phosphorylation sites on OGT, and we would like to understand the function of these modifications. We mutated the phosphylation sites S993, S994 and S481 to alanine to investigate the influence of each site on OGT activity. Affinity purified mutant OGT protein were analyzed in an activity assay and compared with wild type OGT activities to assess whether these sites activate or inhibit OGT. The mutants showed decreased activity, however further optimizing steps of the assay are required. The result will provide further understanding of the regulation of OGT by phosphorylation. Development of High-Throughput Monoclonal Antibody Production Against All Cell-Surface Secreted Proteins Within a Species Yan Ping Lee Mentors: Kai Zinn and Michael Anaya Antibodies serve not only as essential reagents for protein analyses, but also hold applications in targeting critical pathways for disease intervention. The creation of therapeutic proteins requires cell‐based immunogens, successful production of which involves optimized antigen presentation to the immune host. Presently, monoclonal antibody (mAb) production against novel targets remains largely restricted by low‐throughput screening methods. This project attempts to develop stably transfected 3T3 murine cell lines that express an array of extracellular (XC) domains to ultimately stimulate production of a pool of hybridomas capable of secreting mAbs against the various targets. Thus, it allows for recognition of many different native antigens whilst eliminating the need to separately purify antigens prior to injection into murine models. The investigation utilized molecular cloning techniques such as PCR, vector construction, immunoassays and flow cytometry to generate murine cells containing a range of human protein tyrosine phosphatase receptor (RPTP) constructs. The approach has yielded stably transfected cell lines of a portion of the desired cell surface secreted XC domains and further investigation may elucidate the role of RPTPs in developmental cell fate decisions upon their stimulation of mAbs. Testing the Hypothesis of Late Cretaceous True Polar Wobble Grace Leishman Mentors: Joseph Kirschvink and Sarah Slotznick True polar wander is the motion of solid earth with respect to a planet’s spin axis; it occurs when mass anomalies from subduction zones or mantle plumes are moved around. Earth’s magnetic dipoles randomly wander around the rotational axis on geologically short time periods, and can hence be used to detect true polar wander. Thus far, it’s been argued that true polar wander occurs much slower than the rate of overturn of the earth’s tectonic plates. However, samples from the Scaglia Rossa in the Italian Apennines provide evidence of oscillations in magnetic field over very short time scales--~4°/million years—suggesting that a different type of true polar wander, noted true

51

  polar wobble, can occur. By testing the magnetostratigraphy of South Dakota’s Pierre Shale—a layer which formed ~80 Ma during the Cretaceous period, the same time the Italian Scaglia Rossa formed—we can confirm the occurrence of a true polar wobble. Recognition of this rapid polar wander would in turn imply similarly rapid sealevel cycles, and call for the reconstruction of a more precise paleogeographic record. Investigation of Ice Crystal Growth Rates Between -10C and -20C for a Quantitative Morphology Diagram and to Test Predictions of Structure Dependent Attachment Kinetics Cameron Lemon Mentor: Kenneth Libbrecht The many-body dynamics controlling the growth of a crystal from the vapour phase, are particularly complex, even for water. The interplay between diffusion processes and anisotropic attachment kinetics result in a wide range of morphological structures. A recent theory, Structure Dependent Attachment Kinetics (SDAK), has been able to explain these structures at a qualitative level, but there is still much left unexplained. To obtain a quantitative morphology diagram we use two adjacent diffusion chambers. In the first we grow electrically-enhanced ice needles and transfer these to the second chamber for better control of both the temperature and water vapour supersaturation. The growth on ice needles avoids systematic errors from substrate interactions. Growth rates are extracted from time-stamped photographs, allowing for a quantitative morphology diagram. These results are hoped to be reproduced when advances in cellular automata computer modelling have been achieved. Furthermore, we investigate predictions from the SDAK model by first growing needles at -5C such that any chemical contamination is removed by a quasi-liquid surface layer, which is not present at lower temperatures. The temperature is then reduced and the growth of the clean needle viewed and analysed. Developing a Capture Agent Against P. falciparum Histidine Rich Protein II by the Process of Epitope Targeting Bianca Lepe Mentors: Jim Heath, Samir Das, and Arundhati Nag Protein capture agents are vital diagnostic tools that combine the high-affinity and high-specificity characteristics of antibodies with better accuracy and stability. A capture agent is being developed against Plasmodium falciparum Histidine Rich Protein II (PfHRPII), a specific biomarker protein for the deadly strain of P. falciparum that causes malaria. To create the capture agent, a cyclic peptide library was pre-screened with two fragments from the N-terminal region of PfHRPII and then screened with the essential 21mer fragment from the C-terminal region of the same protein. Further testing will be done to the hit peptide sequences obtained from the screen to identify the best candidate peptide to be used in the final stages of binding against the full PfHRPII protein. Photo-Crosslinking Viscoelastic Protein Hydrogels Using 4-Azido-L-Phenylalanine Fanfei Faustine Li Mentors: David Tirrell and Larry Dooling Proteins are one of the most important structures in biology and valuable from a material engineering standpoint. Some proteins with a coiled-coil motif can exhibit self-assembly into hydrogels. These gels could be used in the fields of tissue engineering, wound repair, or biomedical research. This study seeks to optimize the mechanical properties of protein hydrogels by using aryl azide photo-crosslinking as a method to introduce more rigidity. More rigidity in the gel is hypothesized to promote cellular proliferation and migration within the gel. The aryl azides are introduced into the protein by the incorporation of a non-canonical Phenylalanine (Phe) analog, Azido-LPhenylalanine (Azf). Preliminary studies have found that Azf can be successfully incorporated into coiled-coil proteins. More investigation using rheometry will be done to investigate the solidity of photo-crosslinked gels. Progress has been made toward creating a modified protein sequence with additional Phe residues. It is hypothesized that additional Phe residues increase crosslinking density. In this way, the mechanical properties of the hydrogel can be optimized for biological roles. Nanoelectromechanical Membrane Mass Sensing Using Matrix-Assisted Laser Desorption/Ionization Jarvis Li Mentors: Michael L. Roukes and Peter Hung Nanoelectromechanical systems (NEMS) represent the next wave in miniaturizing various electrical and mechanical devices used in a variety of fields, such as physics, biology, and engineering. In particular, NEMS devices have high surface area to volume ratios, low power consumption, low mass, and extremely small footprints. These properties allow NEMS to explore more fundamental regimes of matter. Current NEMS mass sensing advancements have only utilized doubly-clamped beams and cantilevers. However to push the mass sensitivity limits and increase the potential dynamic mass range of devices, we utilize a membrane geometry because they may have higher quality factors and lower Allan deviations compared to doubly-clamped beams and cantilevers of similar masses. For this experiment, we utilize circular membranes of stacked aluminum nitride and molybdenum. For mass deposition, we utilize a technique known as matrix-assisted laser desorption/ionization, which focuses a pulsed UV laser onto the

52

  desired sample embedded in a corresponding matrix. The energy causes a plume of particles to desorb off towards the device. As a particle lands on the device, we are able to deduce its mass from the shift in the resonant frequency. Here we show the viability of our detection setup, mass deposition setup, and our mass deposition results. Modeling the Halo Occupation of Star-Formation Through Abundance Matching Jiaxin Li Mentors: Jamie Bock and Marco Viero Studies have shown that there is strong correlation between star formation rate and stellar mass of the majority of star-forming galaxies. Using large-scale cosmological simulation of dark matter, a cosmological model can account simultaneously for the galaxy luminosity, mass, and velocity distributions. In this project, we use the technique of abundance matching to establish a connection between halo mass and stellar mass. This is accomplished by connecting galaxies across all different epochs to halos in dark matter simulation, which requires populating simulated dark matter distributions with galaxies using the assumption of halo mass and stellar mass functions. The generated model successfully infers the infrared luminosity distribution on the sky because galaxy luminosity depends on where the galaxies reside which is related to their stellar mass. Using the power spectrum, we model the luminosity function and its evolution with redshift, and therefore predicts the (cross-)power spectra of the cosmic infrared background and the correlation functions of the resolved infrared galaxies. Studying the correlation between dark matter halos and star-forming galaxies allows a clear illustration of the abundance and density matching of different galaxies and thus predicts which and how galaxies generate stars in the process of star formation. Curiosity Animation Automation Joanne Li Mentor: Paul Andres One goal of the Solar System Visualization team at JPL is to produce some of the graphics and animations that are released by NASA, including animations of the MSL Curiosity Rover exploring Mars. These animations are created and matched up to rover telemetry data manually in a rather time-consuming process. The animation format, known as the Lightwave scene format, was studied and deconstructed so that a Python script could be written to automatically generate scenes accurate to Curiosity’s movements on Mars. Telemetry data is stored in a database called PLACES, which is accessed by the script before it generates a scene file. The script was written to be easily extensible to any Lightwave model of the Curiosity rover, and is intended to be compatible with Lightwave models of other missions. Optimal Distributed LQG State Feedback With Varying Communication Delay: Two Plant, N-Delay Case Kevin K. Li Mentor: Tracey Ho In this paper, we extend previous work on the distributed two-plant LQQ feedback problem. In particular, we find a controller that accommodates for arbitrarily large propagation delay N between the two plants, and the communication delay varying from 1 to N. Our scheme assumes that the channel capacities of the links are large enough to recover from holes in information transmitted. Investigating Physical Methods of Protecting Linear DNA Fragments in Cell-Free Expression Systems Monica Li Mentors: Richard Murray and Anu Thubagere An in vitro system for translation and transcription (TX-TL) has been developed such that genes can be expressed without the complications of live cells. Rather than inserting the gene onto a plasmid, transfecting cells with the plasmid, then waiting for the cells to grow, this 3-4 day process can be shortened into 1 day with the TX-TL system. This system would be more useful if genes amplified via PCR could be tested directly, but the exonucleases in the system degrade linear DNA starting at the ends. The goals of this project are to find a physical means to protect this linear DNA from degradation. We first replicated the results where circularized DNA by creating complementary overhangs with the employment of uracil-containing primers. The efficiency of this method is rather low, and other processes are being investigated. We evaluate several other techniques, including rolling circle amplification, (a process viruses naturally use,) and addition of the lac operon, biotin and neutravidin to DNA ends.

53

  Understanding the Molecular Mechanisms Regulating Nicotine Response in C. elegans Dennis N. Liao Mentor: Paul Sternberg Nicotine is a highly addictive psychoactive drug. We employed the model organism Caenorhabditis elegans to study which genes specific to the AVA neuron, which is responsible for backwards movement, affect response to nicotine. We selected genes of interest by finding enriched genes from RNA-seq data from AVA neurons and conducted behavioral assays on mutants lacking these genes. Behavior of the mutants on and off nicotine was compared to see which worms responded differently to nicotine based on how often and for how long they moved backwards. In addition to behavior, we selected genes that were enriched only in AVA to identify markers specific for AVA neuron. We created a fusion between these specific gene promoter and GFP and injected these transgenes into the worms to visualize their fluorescent expression patterns. Our research suggests that the genes investigated play a role in response to nicotine. The Carbery Conjecture and Related Problems David Lichko Mentors: Nets Katz and Karsten Chipeniuk There is an emerging field of study in combinatorics concerning substructures of point sets in the plane. This paper focuses on the Carbery conjecture, which states that if a subset of the unit square does not contain the vertices of a rectangle with area larger than , the area of the subset is . This is known under certain conditions, which is an improvement over the general bound of

log

. For this conjecture and various related ones, the best

known bound comes from a Cauchy-Schwarz argument which is resistant to improvement. The Cauchy-Schwarz argument suggests that the worst cases consist of certain sparse structures at many scales. We created programs that attempt to fill in these cases stochastically. The results indicate that if the Carbery conjecture is false, the counterexamples are highly complex. Geometric Representations of Last Passage Percolation Models Bryant Lin Mentor: Tom Alberts Last passage percolation is a model in statistical mechanics that is simple and elegant yet centrally important to many other fields of study. Non-negative i.i.d. random variables are attached to the vertices of , and the length of any nearest-neighbor lattice path is defined as the sum of the variables encountered along the path. The passage time is defined as the maximal length of all directed paths with given starting and ending points. There are two main questions of interest: what are the statistics of the passage time as the ending point moves off to infinity in a specified direction, and what is the typical location of the maximal path? For this SURF we study both questions for the two dimensional lattice using the embedding of weights as elements in . In particular, we study properties of the Voronoi cell , where the path is maximal, as well as computational results from for small lattices. simulations and the geometric description of Electrochemistry on Iron Sulfides at Hydrothermal Vents Chung-Kuang Lin Mentors: Michael Russell and Laurie Barge Hydrothermal vent is a likely geological feature for the origin of life, for the interface between the sulfidecontaining alkaline hydrothermal fluid and the iron-containing acidic ocean of the prebiotic earth could result in an iron sulfide porous membrane, which may catalyze a series of reactions with molecules such as carbon dioxide and bicarbonate that eventually produce oxygen. Some icy bodies of the solar system, such as Europa, have been observed to have similar hydrothermal vent features. Iron sulfide membrane was made in the laboratory with ironcontaining acidic ocean solution interfacing sulfide-containing alkaline hydrothermal fluid through a porous membrane (e.g. dialysis paper). The formed iron sulfide membrane was characterized through cyclic voltammetry to analyze its conductivity in ocean-simulating and hydrothermal-vent-simulating solutions. The results showed that the iron sulfide on the working electrode was the catalyst inducing the current in cyclic voltammetry experiments. Certain substances, such as Nickel (II), were observed to enhance the catalysis. For the continuation of the project, various cyclic voltammetry solutions should be tested, iron sulfide membranes containing other catalytic substances should be made, and other methods of making the iron sulfide membrane (e.g. on carbon cloth) need to be developed to enable different means of analyses.

54

  Investigating Effects of Synaptic Learning Rules on Neuronal Network Topology and Investigating Added Spatiotemporal Resolution of Microwire Electrocorticography Electrodes for Visual Responses Randall Lin Mentors: Gabriel Kreiman and Michael Cross The layers of the visual cortex possess functionally and developmentally important directionality in connectivity. We manipulate spike timing dependent plasticity rules in various configurations to generate useful directionality in connectivity. With minimal assumptions in assigning learning rules to connections between layers, we derive certain canonical subcircuits. Separately, microwire electrodes are physically much smaller and finer than standard electrocorticograph electodes. Using similar data analysis methods, we show microwire electrodes offer higher spatial resolution for visual responses. Comparing the Reliability of Far-Side Active Region Predictions From HMI and GONG Using STEREO Far-Side Images Xintong Lin Mentors: Paulett C. Liewer and Eric M. De Jong Helioseismology, the study of wave oscillations through the Sun to reveal the Sun’s internal structure, has applications that have been used to determine strong magnetic regions on the far side of the Sun. These determinations lead to the production of far-side active region prediction maps from analysis of helioseismology data collected by the National Solar Observatory Global Oscillation Network Group (GONG). Recently, data from the Helioseismic and Magnetic Imager (HMI) aboard the Solar Dynamics Observatory has also been analyzed to produce far-side active region prediction maps. Since active regions are the sources of solar flares and coronal mass ejections, we want to predict these regions on the far side to mitigate space weather’s effects on Earth. The reliability of the GONG maps has been investigated previously (Liewer et al., 2012), and we now extend the same analysis to HMI maps. We also investigate the success of GONG and HMI maps for predicting large far-side active on the East limb before they can be viewed from Earth. Due to evolving HMI data processing methods, we analyzed the reliability of reprocessed HMI predictions over the reliability of the initial HMI predictions as well as comparing reprocessed HMI and GONG predictions. 3D Printing of Tubular Polymer Scaffolds for Engineering Biologically Powered Impedance Pumps Anna Liu Mentors: Morteza Gharib and Hesham Azizgolshani Tissue engineering has recently become a field of interest in biomedical engineering. Systems of synthetically engineered tissue have promising prospects in both research and clinical applications. Current methods in 3D tissue engineering are dependent on the development of an appropriate scaffold to provide structure for tissue growth. In the current work, we investigate computer-aided 3D printing as a method of on-demand 3D scaffold. Development of a suitable scaffold requires a material that supports cell attachment and a deposition method that can attain our desired tubular form. Towards this end, the 3D printer is calibrated to print our desired structure with multiple materials. Alginate hydrogel was used as a scaffold material because it is physiologically similar to the extracellular matrix and gels rapidly in the presence of an ionic cross-linking agent, CaCl2. In order to facilitate cell attachment to the scaffold, porcine skin gelatin was mixed with the alginate. Due to its ability to gel at room temperature, gelatin also lent structure to the scaffold during the alginate gelation process. Cells seeded on top of a gel did not exhibit a significant ability to penetrate the gel's surface. Therefore, cells were suspended in the gels to simulate tissue. The results of this project indicate that 3D fabrication of cells suspended in alginate mixed with gelatin may be an invaluable method for the production of artificial tissues. Retrieval of Atmospheric Concentrations Using AIRS Measurements Jiabin Liu Mentor: Yuk L. Yung Remote sensing of Earth’s atmosphere has been an important method to monitor chemical concentrations as well as to characterize climate changes. The Atmospheric Infrared Sounder (AIRS) is an infrared remote sensing instrument to capture the atmospheric temperature and moisture and to trace the greenhouse gases. In our study, Line-by-line Radiative Transfer Model (LBLRTM) and a two-stream analytical model are combined to retrieve the water vapor, temperature, and CO2 in the boundary layer, based on the data from the AIRS mission. Degrees of freedom (DOF) and information content (IC) of the retrieval are also calculated to show the efficiency of vertical profile retrieval for each gaseous species. Based on the efficiency information, we suggest the ideal retrieval windows and resolution for a Moderate-resolution Infrared Imaging Sounder (MIRIS), a conceptual instrument for future at a comparatively lower cost with equally good or better performance with AIRS.

55

  Quantum Limited Photo-Detection Using Homodyne Techniques Jonathan Liu Mentors: Nergis Mavalvala and Alan Weinstein Due to the extreme sensitivity of gravitational wave detectors, highly precise techniques must be used to measure signals from the lasers involved in the detectors. The method of balanced homodyne detection is one way to measure quantum-limited noise in laser light – that is, the theoretical minimum amount of noise in a laser system due to the Heisenberg Uncertainty Principle. In essence, the technique combines a signal laser with a powerful local laser at the same frequency, splits the beam into two, and subtracts the two signals to remove the classical noise, leaving behind the quantum-limited noise. Although the technique has existed for years, this project implements a new circuit design to improve upon existing homodyne detectors for use in the Laser Interferometer Gravitational Wave Observatory (LIGO) collaboration. The finished detector is expected to directly measure the shot noise of light, and as a result be useful in measuring “squeezed” noise in LIGO’s squeezed-light experiments, which create sources of light that manipulate the Heisenberg Uncertainty Principle to improve sensitivity in the gravitational wave detector. Nickel-Catalyzed Asymmetric Alkylation of -Halo Pinacolboronic Esters Tianxiang Albert Liu Mentors: Gregory C. Fu and Junwon Choi Enatiomerically enriched secondary alkyl pinacolboronic esters are important synthetic intermediates because they can be further derivatized into chiral alcohols, carboxylic acids, and many other functional groups. A nickelcatalyzed process to asymmetrically alkylate α-halo pinacolboronic esters is under development. The condition which currently offers the best yield (85%) as well as enatiomeric excess (87%) is shown in the scheme below. This condition tolerates ester groups and TBS protected alcohols on the nucleophile.

Development of a Cryogenic Probe for Rapid Testing of Single-Photon Detectors on Optomechanical Cavities William Livingston Mentors: Oskar Painter and Justin Cohen Nano-scale optomechanical cavities are studied through the optical transduction of mechanical motion. The mechanical oscillators in these systems have recently exhibited quantum phenomena. Further study of these systems require superconducting nanowire single photon detectors (SNSPD), which operate at very low temperatures. I design and build a cryogenic probe that uses a bath of liquid helium to rapidly cool the system and characterize the optical properties of these detectors. I also test electrical equipment used to read signals from the SNSPDs using model optical setups. Roots of Unity and the Field With One Element Catharine Wing Kwan Lo Mentor: Matilde Marcolli Much research has been conducted on the hypothetical field with one element. In this project, the hypothetical field

F

is investigated. The first attempt is to find a polynomial, which when evaluated at

However, very few

n

, where

which is necessary since #X(

n

F

n

F , #X(  ) is obtained.

is the nth root of unity, have been found to give non-negative integer values,

) is the class of

F

. Hence, the restriction is loosened. The extension

F n 

of

F

is

Fn F considered, using the polynomial interpolation applied to 1 of 1 . Electron Response of STEREO High Energy Telescope Through GEANT4 Modeling Daniel Y. Lo Mentors: Edward C. Stone and Mark E. Wiedenbeck Accurate measurements of solar electron energies and fluxes are crucial in understanding the processes that accelerate solar energetic particles and drive space weather. Using GEANT4 libraries, we developed an instrument model for the High Energy Telescope (HET) on the Solar TErrestrial RElations Observatory (STEREO) to study its

56

  electron response. Such a theoretical calibration is required to translate observations into a quantitative description of the distribution of solar electron energies. Using the model, we also studied the effects of the instrument housing on the electron response, and these findings help inform better instrument designs for future sensors to optimize the quality of the electron measurements. Graphene Optics: 2D Superabsorbers and 3D Photonic Crystals Josue Lopez Mentor: Harry Atwater The control of infrared light (IR) with nanoscale structures will lead to improvements in present day military, industrial, and scientific applications. Recent theory has predicted that arrays of graphene nanostructures on top of a dielectric layer with a metallic backreflector can absorb 100% of incident light in the infrared (IR) light regime. This high absorption can dramatically increase the efficiency of IR detectors and could allow for useful light management in telecommunication and energy applications. Herein, we demonstrate the fabrication and investigation of electrostatically tunable graphene superabsorbers that can absorb ≥ 25% of incident IR light. Our demonstration of tunable optical responses in graphene superabsorbers opens up the possibility for novel optical properties via more complex structures in three-dimensions (3D). We demonstrate this potential through our efforts for the fabrication of the first experimentally demonstrated 3D photonic graphene crystal. The fabrication of 2D and 3D graphene structures demonstrates the tremendous potential of graphene for the control of light. Using Flow Simulation to Optimize Reactor Design Mark Lorden Mentors: Michael Hoffmann and Asghar Aryanfar The best design of an electrochemical reactor to achieve the optimal flow patterns has been investigated. The goal of the reactor is to purify the waste water from the toilet, and the more contact time that the waste has with the electrochemical plates, the more thoroughly it will be purified. Therefore the design which causes the flow to cover the most distance over the plates is optimal. It was determined that a design with a “cover” over the electrode plates to force the flow to circulate within the compartment created. Incorporating Nanotruss-Structured Si Anodes Into Lithium Ion Batteries Alexander X. Lozano Mentors: Julia Greer and Lucas Meza Li-ion batteries are an indispensable part of managing energy with uses ranging from grid level storage to portable electronic devices. Si based anodes have been explored extensively in the literature as they have a capacity of 4200 mAh g-1, almost ten times the storage capacity of the currently used graphite anode. That being said, Si expands up to 400% upon lithiation causing cracking and a loss of electrical contact ultimately leading to a low cycle life. To solve this problem, new nanotruss-structured Si Anodes have been proposed that may be able to avoid strain by expanding locally at designed structural points rather than globally. The goal of this investigation has been to incorporate these structures into a working lithium-ion battery and study their capabilities as anodes. This has been accomplished by designing and testing the function of a compatible thin film cell that can easily integrate Si nanotruss anodes in the future. Preliminary tests have been carried out demonstrating certain key functions of these batteries and plans for integrating the new anode structures have been completed. Much progress has been made, however further exploration into the practical fabrication of such a cell is required so that it may have a gravimetric charge capacity competitive with those of commercial batteries. Multiple Scattering Optimization of MCViNE With Mystic and Cluster Analysis Genesis Lung Mentors: Brent Fultz and Jiao Lin In spallation research, the excitation of crystalline solids known as phonons is a subject of great interest, as fundamental properties can be derived from acoustic and optic modes. As there is no currently satisfactory program for optimizing the scattering and material model that takes into account both the behavior of excited phonons and the multiple scattering, and can generate data that matches experimental results, our goal was to create such a simulator/optimizer combination. By marrying optimization routines from the Mystic python module created by Mike McKerns and cluster programming with the MCViNE suite created by the DANSE project, we will be able to generate an extension of the software that should produce results better suited to what we would expect from experiments. It is hoped that these capabilities will eventually be integrated into the Virtual Neutron Facility project being worked on by the Fultz lab which allows users to perform a large range of complex simulations. Given coefficients for a data point, our program will find the corresponding location of the curve point, and upon fitting to the specified curve, will generate the coefficients for all other points on the curve.

57

  Simplified Model Spectra and the Search for SUSY Particles at the LHC Anthony Lutz Mentor: Maria Spiropulu With the recent confirmation of the appearance of the Higgs Boson at the Compact Muon Solenoid (CMS), much of the research at CERN has been focused on the search for particles that appear beyond the Standard Model. Popular candidates for such new particles are the Supersymmetric (SUSY) partners, which theoretically connect the two classes of elementary particles, bosons and fermions. To describe possible scenarios in which SUSY particles appear, Simplified Model Spectra (SMS) are used, which reduce a particle decay resulting from a collision to only the components which may directly be involved with SUSY. The research for this internship is concerned with the analysis of three distinct SMS. For each of these models, computational analysis was used to decay compressed data and determine resultant details of importance. Qualities such as the mass and identity of decay daughter particles, as well as the ratio of mass imparted from the parent particle, were recorded and converted to graphical representations so that correlations would become more apparent. The finished data has now been sent to the CERN database for use as a guideline for future pertinent experiments, and may possibly lead to the discovery of new physics over the next few years. Preparing the Chemical Laptop for Field Analysis Eugene Lynch Mentors: Peter Willis and Amanda Stockton The Chemical Laptop, co-engineered with Los Gatos Research, is a portable, miniaturized instrument designed to detect molecular signatures of life in liquid samples. Operated by a tablet computer, it uses microcapillary electrophoresis (µCE) coupled with laser-induced fluorescence detection (LIFD) to analyze labeled samples pneumatically prepared and transported on a multilayer microfluidic chip. Several modifications were made to the instrument’s software and hardware to prepare for a completely automated end-to-end field analysis. Existing software was revised to reduce the complexity of spectral data and optimize automated sample processing cycles. New software was written, allowing for automated, on-chip colorimetric determination and adjustment of sample pH. A variety of fabrication methods, materials, and post-fabrication procedures were explored to develop a lightweight, air-tight, pneumatic manifold resistant to deformation under high pressure. Different micropumps were tested to supply sufficent pressure to this manifold. New hardware was developed to make the instrument fielddeployable. A simple mounting system was installed to position a USB microscope above the microfluidic chip, allowing for colorimetric pH control and remote observation. Miniature, modular, refillable cartridges were created to dispense all necessary reagents required for the analysis of numerous compound classes by a simple clamping and switching mechanism. Identifying the Interaction Domain Between DNA2 Nuclease and FANCD2 Harinee J. Maiyuran Mentors: Judith L. Campbell and Kenneth Karanja The nuclease DNA2 has been shown to interact with FANCD2 in the replication process of DNA. Identifying the interaction domain between both DNA2 and FANCD2 is the primary goal of my work this summer. We confirmed that we had the correct DNA2 constructs using restriction fragment analysis. Next, we ensured, by western blot, that endogenous DNA2 can be depleted using shRNA-containing virus. This will serve as a control showing that without endogenous DNA2, FANCD2 cannot be immunoprecipitated. We then tested the effectiveness of DNase I in digesting DNA using purified DNA plasmid and in whole cell extracts in the presence or absence of potassium, magnesium, and calcium divalent ions. We found that DNAse I digested plasmid DNA efficiently after the addition of potassium, magnesium, and calcium in the buffer. Next, we evaluated whether DNAse I can digest genomic DNA in whole cell extracts instead of purified plasmids. We optimized these conditions to determine whether DNA2 and FANCD2 interact directly by digesting genomic DNA in our cell extracts. Finally, we will express truncated DNA2 proteins in human cells and perform immunoprecipitation experiments to determine which domain interacts with DNA2. Studying the Metabolic Activity of Proteins in the Heme-Copper Oxidoreductase Superfamily Allison Maker Mentors: Woodward Fischer and James Hemp The rise oxygen in the atmosphere 2.35 billion years ago was one the most important events in the history of life on Earth. It profoundly changed the geology of Earths surface and lead to the origin of aerobic respiration, the most exergonic metabolism known. The rise of oxygen is attributed to Cyanobacteria performing oxygenic photosynthesis, however a new mechanism of producing O2 has recently been discovered: nitric oxide dismutation. Our goal was to isolate and characterize the protein that catalyzes nitric oxide dismutation, as well as other proteins in the heme-copper oxidoreductase superfamily. During the research period, the genes for nitric oxide dismutase were cloned into several different types of expression vectors, which were then transformed into either

58

  the host bacteria or E. coli for protein production. The metabolic activity of the proteins will then be analyzed to determine if they perform nitric oxide dismutation as predicted, and phylogenetic and structural analysis of the proteins will help to determine their evolutionary history. This will allow us to determine whether nitric oxide dismutation evolved before or after oxygenic photosynthesis. Using CRISPR-Mediated Genome Editing to Create a Selfish Genetic Element Chaitanya (Chat) Malladi Mentors: Bruce Hay and Omar Akbari The goal of this research project is to develop a transgenic line of Drosophila melanogaster expressing a selfish genetic element based on CRISPR genome-editing technology. From the progeny of an injected line of flies, one female fly was identified expressing CRISPR knockdown of the white marker gene, thereby confirming that such a transgenic line can be procured; this result in itself is significant since only transient expression of the CRISPR system has been reported thus far. The plasmid for creating the selfish genetic element is still being cloned. This DNA molecule will contain CRISPR elements that can cut themselves out and insert themselves at a targeted point in the genome via the guide RNAs. Once the plasmid is completed and injected into flies, its ability to spread throughout a population will be assayed. If successful, this application of CRISPR would provide a method to spread a particular gene throughout the population by making it a selfish genetic element; such technology could be used for conferring disease refractoriness among a population of insects that act as vectors. Purification of E. coli uvrC, a DNA Repair Protein With a Predicted [4Fe-4S] Cluster Janani Mandayam Comar Mentors: Jackie Barton and Michael Grodick Through a process termed DNA-mediated charge transport (CT), DNA can conduct charge through its overlapping π-orbitals of the stacked base pairs. DNA CT can occur over extensive molecular distances (~34 nm), but its ability to do so is very sensitive to the uniform and continuous stacking of the base pairs. Small perturbations, such as mismatched base pairs, attenuate CT. An enzyme of interest that we studied here is UvrC, a protein in the nucleotide-excision repair (NER) pathway. A sequence alignment of UvrC from organisms across the phylogeny, including gram-negative and positive bacteria and archael organisms, reveals the presence of four highly conserved cysteine residues, opening up the possibility that UvrC contains a redox active [4Fe-4S] cluster that would facilitate CT. In this project, we expressed and purified UvrC from E. coli to see if it contains a [4Fe-4S] cluster. In addition, we created a UvrC knockout in the Inverted A strain to investigate if UvrC can signal with other DNA repair proteins containing [4Fe-4S] clusters. Investigating the Use of Harmonic Lasing to Produce Short-Wavelength X-Ray Radiation in a Free Electron Laser Ajay Mandlekar Mentors: Tor Raubenheimer and Ryan Patterson The LCLS (Linear Coherent Light Source) is the world’s most powerful X-ray laser, and it has witnessed unprecedented success since its construction. Now, with the prospect of building the LCLS-II, a more powerful successor to the LCLS, it may be possible to make the transition towards shorter wavelength and higher energy radiation. Harmonic lasing might be one such way to realize these goals. Here, we investigate whether harmonic lasing will be a useful tool in the LCLS-II by considering the optimal way of generating 18keV radiation. We compare producing 18keV radiation as the 3rd harmonic of 6keV radiation to producing 18keV radiation in the fundamental. We conducted this study using Genesis simulation code and Matlab as a scripting tool to optimize various parameters and run several simulations. In optimizing the harmonic lasing case, we had to understand the radiation generation mechanism and the transverse radiation profile and optimize the transverse electron beam profile and the magnetic lattice. Future work will focus on verifying our findings and implementing them into the LCLS-II design. Optoelectronic Properties of Zn(Sn,Ge)N2 Semiconductors Aashrita Mangu Mentors: Harry A. Atwater, Prineha Narang, and Nathan S. Lewis The II-IV-N2 compounds are closely related to the wurtzite-structure III-N semiconductors with a mixed A-site composition where the choice of different group II and group IV elements provides chemical diversity to tune the structural and electronic properties. Specifically, ZnSnxGe1-xN2 alloys with optical band gaps ranging from 2-3.1eV can be tuned to span a large portion of the solar spectrum, and could therefore be a viable earth-abundant light absorber and replacement for InGaN in nitride optoelectronic devices. They exhibit local order as demonstrated via X-ray absorption fine structure spectroscopy (EXAFS) and a linear relationship between the (002) peak position and composition in X-ray diffraction studies, indicating continuous access to the entire range of band gap values without phase separation or the need for more complicated growth strategies. The bowing parameter is 0.29 eV for the measured band gaps and 0.67 eV for the calculated band gaps, both significantly smaller than that of In1-xGaxN,

59

  indicating that the ZnSn1-xGexN2 alloy band gaps can be tuned almost monotonically by controlling the Sn/Ge composition. In this presentation we will describe theoretical studies of the optoelectronic behavior, carrier dynamics of ZnSnxGe1-xN2 alloys and defect physics of the ZnSnN2 series as elucidated from photoluminescence and pump-probe spectroscopy. Application of GPS Measurements to the Characterization of Groundwater Variations From Atmospheric River Events in California Jennifer Marion Mentors: Duane Waliser, Yuk Yung, Frank Webb, and Angelyn Moore GPS observations offer a new means to indicate groundwater variation through small deformations of the ground surface. GPS detects small positive displacements due to recoverable land subsidence indicating groundwater recharge in addition to negative displacements due to both recoverable and permanent land subsidence indicating lowered groundwater levels. A high resolution of GPS data allows for a time series analysis of the signals from groundwater recharge seen in conjunction with Atmospheric Rivers. An analysis using daily vertical GPS data and Atmospheric River dates characterizes the response of aquifers throughout California to isolated Atmospheric River events for WY1998 to April 30, 2011. The analysis uses Atmospheric River landfall location and hydrological data to match Atmospheric River events to relevant GPS stations. The aquifer responses for geographic regions of interest in California are compared. Figures for temporally and spatially averaged time series for periods surrounding Atmospheric River events are discussed. These figures show an increase in vertical land height which is correlated with Atmospheric Rivers given sufficient data and specificity of landfall location. This analysis provides a means to understand the connection between atmospheric science and groundwater hydrology. In conclusion, a network of GPS stations provides a means to monitor underground water supply. Structure Prediction and Ligand Binding for the Human MT1A and MT1B GPCRs Lasya Marla Mentor: Ravinder Abrol GPCRs have been implicated in a myriad of signal transduction pathways in the human body. A firmer understanding of GPCR structure will undoubtedly facilitate disease prevention and treatment. Activated MT1A and MT1B GPCRs initiate signal transduction pathways, which may be instrumental in the search for a cure for pancreatitis. An accurate structural prediction for the MT1A and MT1B GPCRs will facilitate a better understanding of these signal transduction pathways. Unfortunately, GPCRs have a high degree of conformational flexibility making structure prediction nearly impossible because they are constantly switching states spontaneously. So, using knowledge of chemistry and general properties of GPCRs a mathematical methodology was developed and implemented to predict likely structure of these GPCRs. The resulting structural predictions were combined with visualization programs to develop potential ligands to induce the active state and consequently the signal transduction pathways of interest. Mitigating Image Persistence in Hubble Space Telescope’s Wide Field Camera 3 Melissa Marshall Mentors: Jason Rhodes, Roger Smith, and Susana Deustua HgCdTe infrared detectors, such as the one aboard the Hubble Space Telescope’s Wide Field Camera 3, suffer from image persistence – a remnant signal seen most often after bright exposures. When light hits a detector, new regions of each pixel are exposed to charge. Within these regions are charge traps that retain charge after a pixel is reset to a standard voltage. Over time, the traps release their charges, which quickly rejoin areas of similar polarity within the pixel. The motion of these charges induces a small voltage change in the pixel which mimics a signal. This spurious signal confuses measurements of the brightness and shape of imaged objects. In particular, weak gravitational lensing studies are dependent upon very precise photometry of background galaxies. Understanding and mitigating image persistence is essential to improving the quality of this research. Although image persistence is understood theoretically, it is not characterized well enough to mitigate its effects. We present initial findings about the behavior of image persistence, and potential avenues of further study. Supervised Learning With Generative Stochastic Networks Eric Martin Mentor: Yoshua Bengio Recent work introduced the generative stochastic network (GSN) into the ranks of generative deep models such as the Deep Boltzmann Machine. Generative models like the GSN can be used for both unsupervised and supervised learning, but previous work had only investigated the GSN for unsupervised problems such as sampling. This work extends the GSN model by placing a supervised layer such that the hidden layers lie between it and the input layer. Various training schemes for the supervised GSN were evaluated on the MNIST handwritten digit dataset. Although room remains for the model to be improved, initial results indicate that jointly trained supervised GSNs perform comparably on black box classification to models such as the denoising autoencoder, with the advantage that the GSN does not require a multi-step training process that incorporates steps like supervised fine-tuning.

60

 

Testing the Efficacy of Spin-on Glass as an Alternative Method for Booting Silicon Microwires Jacqueline Maslyn Mentors: Nate Lewis and Heather Audesirk Silicon microwires are being investigated as the photocathode material for a photoelectrochemical cell. Booting with silicon oxide is necessary to protect the substrate and bases of silicon microwires and prevent short-circuiting, or shunting, of the sample in photoelectrochemical cell conditions. The existing oxide booting process is time and resource inefficient. Spin-on glasses are a class of silicon- and oxygen-based compounds which are suspended in liquid, but can undergo condensation polymerization to form inorganic glasses. Spin-on glass booting is accomplished using a spin coater to layer spin-on glass at the base of a wire field. Microwires with silicon oxide or various spin-on glass boots are tested for shunting and electrical performance using cyclic voltammetry in methyl viologen. Measurements and Modeling of the Dynamics of Epigenetic Silencing Kayla McCue Mentor: Michael Elowitz and Lacramioara Bintu The study of epigenetic marks—changes to the chromatin that affect the expression levels of genes in the cells—is important in understanding cell processes such as differentiation. However, such studies have been limited in the past by techniques that are merely correlational. By recruiting different epigenetic modifiers to the promoter region of a fluorescent reporter gene, a clearer cause-and-effect relationship can be established. Using such a system and flow cytometry, we characterized both the silencing of the gene and its heritability after the stimulus was removed. We found that for different types of epigenetic silencers, the establishment of different marks leads to different dynamics of both silencing and recovery of gene expression after silencing. Particularly, silencing by histone deacetylation is lost within 1-2 cell generations, silencing by DNA methylation is maintained over many weeks, and silencing by histone trimethylation lies somewhere in between. With these data in mind, we developed a simple model of silencing where marks on a certain region of chromatin can be gained, lost or spread to neighboring regions with rates that might vary among chromatin marks. While this model still needs refinement, it has thus far been successful in reproducing some of the experimental data. High-Performance Clock Distribution and Transceivers for Chip-to-Chip Optical Interconnects Omar Mezenner Mentor: Azita Emami-Neyestanak As the speed and complexity of integrated circuits grow, the amount of information relayed between them also necessarily grows. However, the rate of scaling for chip-to-chip interconnect bandwidth has lagged behind the growth of chip performance and speed. Traditional approaches to minimize this disparity have either increased the output data rate per-pin or have increased the number of pins used for I/O. However, problems arise when the electrical channels used to transmit and receive data have high frequency-dependent loss, and thus decrease the amount of usable bandwidth. The solution that is being explored is the use of optical, rather than electrical, interconnects. Optical channels have negligible frequency loss, and therefore are able to support higher bandwidths than their electrical counterparts. Additionally, optics allow for dense information transfer due to the ability to focus very short wavelengths into small areas, eliminating the observance of crosstalk between data lines that may be present in high-speed electrical circuitry. Investigating the Origin of the C-2 Hopanoid Methylase Alice J. Michel Mentors: Dianne Newman and Jessica Ricci Earth history has been dominated by microbial life, for which fossil evidence is often lacking. Hopanoids comprise a class of bacterial lipids that are well preserved as biomarkers, though their physiological functions and taxonomic distribution are not clear. One member of this class, 2-methylhopanes, fossilized derivatives of 2-methylhopanoids, have a rich and diverse distribution in the rock record, indicating that understanding their evolutionary history may reveal information about ancient life on Earth. With the recent discovery of the gene, hpnP, that encodes the C-2 hopanoid methylase, HpnP, the evolutionary origin of 2-methylhopanoids can be reconstructed. Although attempts have been made to root the phylogenetic tree, previous research has not indicated a definitive origin in any one phylum. Using newly available genome sequences, bacteria cultures, and inverse PCR to obtain additional hpnP sequences, we have constructed an improved phylogenetic tree for hpnP.

61

  Synthesis of Correct-by-Construction Control Protocols for Switched Systems Using Partial State Information Oscar Mickelin Mentors: Richard Murray and Necmiye Ozay Hybrid systems, which contain both discrete and continuous variables, have become increasingly common at the interface of electronics and engineering, creating a need to synthesize controllers which force hybrid systems to satisfy certain high level specifications, e.g., on reliability and performance. This project establishes frameworks to allow for the synthesis of control protocols for hybrid systems with dynamics determined both by uncontrollable switches in the environment and controllable switches initiated by the supervisory control. Further, we provide theory necessary to expand the techniques to when only partial state information is available and apply this to an industrial problem in the form of an air management system of an aircraft. Lastly, we formalize the problem of transforming a continuous system into a discrete representation needed for the synthesis procedure and present an optimal algorithm which retains the maximum amount of transitions within the system. Performance of Janus Particles in the Self-Assembly Fabrication of Nanoparticle-Decorated Honeycomb-Structured Polymeric Films via the Breath Figures Patterning Technique Sophie Miller Mentors: Julia Kornfield and Wei Sun The breath figures (BFs) patterning technique represents a simple, robust method for producing microstructured honeycomb-patterned porous polymer films. Nanoparticles can self-assemble by spontaneous adsorption onto the interfaces between the templating water droplets and the polymer solvent in order to achieve controlled micrometer and nanometer scale hierarchical structures. Amphiphilic Janus nanoparticles are expected to optimize the films with regards to nanoparticle decorating density within the pores as well as ordering degree of the honeycomb structures given their dual role as Pickering emulsion stabilizing agents and as highly effective solid surfactants, resulting in a significant improvement over what has been achieved using isotropic nanoparticles. Exposed surfaces of silica particles embedded in spherical wax structures (colloidosomes) are chemically modified to render half of the particle hydrophobic. A reliable method for producing optimal colloidosome coverage has been established. Once successful synthesis of amphiphilic Janus particles is confirmed, they will be employed in the BFs templating method for the first time. Characterization of the resulting films’ morphologies will be reported and will help draw conclusions on the interfacial behavior and self-assembly performance of the Janus particles, as well as the general mechanism of structure formation in the BFs patterning technique. HeartWatch Ritvik Mishra Mentors: Shwetak Patel and Sidhant Gupta The goal of this project was to build a compact device that can fit into the form of a watch, and can accurately track the user’s heart rate over the course of a day with minimal interference with everyday life. Most heart rate monitors are either obtrusive and uncomfortable such as the devices that must be worn around the user’s chest, or lack the accuracy and battery life to accurately track the user’s heart rate for the course of a full day. Doctors have suggested that if they could have access to a user’s 24-hour heart rate, the diagnosis of a number of ailments would become much easier as they have access to much more knowledge of the patient’s condition. We have created a working prototype of this device, where raw data is written to a microSD card, which can later be removed and plugged into a computer for post-processing to extract the actual heart rate data. A Hard X-Ray Study of Ultraluminous Source NGC 5204 X-1 With NuSTAR Eric S. Mukherjee Mentor: Fiona Harrison Ultraluminous X-ray sources (ULXs) are among the most mysterious sources in X-ray astrophysics. Their observed super-Eddington luminosities indicate either a new intermediate mass class of black hole or a different accretion mechanism than accepted models. We use combined observations of ULX NGC 5204 X-1 from XMM-Newton and NuSTAR to produce a spectrum between 0.2 and 20 keV. This spectra can be described by a number of continuum models. We find the best fit model to be being a cool ~0.25 keV multicolor blackbody disk with power law component with an exponential cutoff at 4.9 keV. We also measure flux and fit spectra from potential contaminating sources approximately 15” and 25” from the source center detected by Chandra that are unresolved by NuSTAR and XMM-Newton. We will use these spectra to further constrain ULX accretion models.

62

  Bounding the Roots of the Orbital Chromatic Polynomial Alexander Mun Mentor: Mohamed Omar Given a graph  and some G a subgroup of Aut(G), we can define the orbital chromatic polynomial OPG (x) by interpolating over the number of colorings of  up to permutation by some element of G. It was conjectured by Cameron and Kayibi that the real roots of OPG (x) are bounded above by the roots of the chromatic polynomial of  for all G. This is proven for common graphs including cycles, complete graphs, and wheels, as well as for outerplanar graphs which contain an odd cycle. The conjecture, however, is not true for general outerplanar graphs, and counterexamples are presented. We additionally establish several lemmas about the chromatic polynomials and orbital chromatic polynomials. Using Representations of SL2(q) to Construct Low Coherence Frames Vidya K. Muthukumar Mentors: Babak Hassibi and Kishore Jaganathan A problem that repeatedly comes up in coding theory, compressed sensing and quantum measurements is the designing of a matrix of given dimensions with the lowest possible coherence. The coherence of a matrix is defined as the maximum value of the inner product between two columns. For a general matrix, this coherence is always greater than or equal to the Welch bound, which depends on and . We want to design carefully constructed matrices whose coherence is as close to the Welch bound as possible using the properties of a group of order . Previously, an elegant construction of these matrices was done using abelian groups. The objective of this project was to obtain results for non-abelian groups. Using representation theory, we analyzed the representations of SL2(q), and used cleverly chosen induced representations to construct our matrix. We showed that our results were asymptotically similar to the abelian case. Currently, we are exploring how to construct the more complex cuspidal representations. In future, we will be extending our analysis to GL2(q) and studying the applications of these matrices more comprehensively. Anyons, Topological Quantum Computing, and Noncommutative Geometry John Napp Mentor: Matilde Marcolli Anyons have found important applications in topological quantum computing. The simplest universal model of topological quantum computing utilizes Fibonacci anyons. Possible fusion paths in a Fibonacci anyons system are described by a Bratteli diagram, which uniquely determines an approximately finite dimensional (AF) C*-algebra related to the noncommutative torus of modulus the golden ratio, which admits real multiplication. The generators of the K0 group of the AF algebra have a clear interpretation in the physical system, as does the generator of selfMorita equivalences. Through such connections, we reformulate properties of the Fibonacci anyons in terms of modules over the noncommutative torus of modulus т and quantum theta functions. This dictionary between the physical system and the noncommutative torus allows us to exhibit novel properties of the system. Hydraulics for Micro-Robotic Actuation Chiraag M. Nataraj Mentors: Jaret C. Riddick and Joel W. Burdick The motive of the proposed student research is to investigate hydraulic actuation for small and micro robotic systems, and develop control algorithms for their autonomous operation. The objective is to implement this idea in a test system to prove the viability of the concept. In order to ensure that the pump is as efficient as possible, equations and parameters are used in order to optimize the efficiency and force/pressure output of the pump. Furthermore, parametrized CAD drawings of the different parts of the pump are utilized in a programmatic manner to facilitate changing parameters. Once the pump has been optimized and fully developed, it will be used to power an actuator which can trigger an action such as e.g. an arm bringing down a hammer. For the case of an arm and hammer, pistons can be placed both above and below the joint to allow the arm to be raised and lowered. Time-of-Flight (ToF) Sensors and Their Application to Animal Social Behavior Recognition Santiago Navonne Mentors: Pietro Perona and Xavier Burgos-Artizzu In an attempt to understand the mechanisms involved in animal social behavior, biology research groups are often interested in monitoring the interactions of small animals over long periods of time. This is normally achieved by recording their activity through standard cameras and then manually scoring their behavior, a time-consuming task. Recently, attempts at replacing this tedious task with an automatic visual recognition system have been made. However, these methods struggle due to the standard cameras’ inability to provide scene information. In this work, we study the application of Time-of-Flight (ToF) sensors such as the Microsoft Kinect and the Creative

63

  Interactive Gesture Camera for this purpose. We first analyze both sensors’ performance on newly gathered videos of interacting mice in traditional experimental setups. Then, we propose a novel approach to tracking the animals and estimating their poses based solely on z-axis data. Our approach is able to successfully estimate the orientation and identity of two interacting mice over long periods of time, thus proving the usefulness of ToF sensors. Current methods being researched in the computer vision laboratory will benefit greatly from these results and from the ad-hoc software developed throughout the project. Understanding and Quantifying the Effect of Compositional Context on Biocircuit Performance Andrew Ng Mentors: Richard Murray and Enoch Yeung Compositional context, coupled with host context and environmental context, make up the major causes of failure in synthetic biological systems. Understanding the effects of context will greatly improve the ability of synthetic biologists to implement more complex biocircuits within biological systems. This project explores the effect of compositional context in the form of orientation and positioning of adjacent promoters in simple biocircuits. Several simple circuits were assembled consisting of a repressor protein and a fluorescent reporter protein driven by an inducible promoter, arranged in tandem, divergent, and convergent orientation while also varying the relative 5’ to 3’ orientation of the two genes. In this work we characterize the variability in gene expression as a result of differing circuit layouts. Identification of Transcription Factors and Their Regulatory State Domains During Gut Formation in the Sea Urchin Larva Gracia Ng Mentors: Eric Davidson, Isabelle Peter, and Jonathan Valencia A gene regulatory network (GRN), which documents regulatory gene interactions as they direct cell specification, has previously been constructed for Strongylocentrotus purpuratus during early development prior to gastrulation. However, the transcription factors and their interactions that contribute to the formation of the pluteus gut during larval development remain unknown. In an endeavor to assemble a late endodermal GRN during gut organogenesis, candidate regulatory factors that may participate in gut formation have been under analysis. 20 of the remaining genes left to be examined were assayed using whole mount in-situ hybridization experiments to determine their spatial expression. A time-course was generated using embryos from 24 to 72 hours postfertilization at 6-hour intervals to observe the dynamics of gene expression throughout various stages. All of these regulatory genes except z166 were expressed at some point during this time frame, therefore, suggesting that they play a role in endodermal specification during gut formation. These genes were placed in their respective regulatory state domains, bringing this crucial component of the late endodermal GRN of the sea urchin embryo near completion. Real-Time PCR Thermal Cycler: Miniaturized Heating Integrated Chip Design and Fabrication Loc T. Nguyen Mentors: Axel Scherer and Imran Malik Real-time Polymerized Chain Reaction (PCR) technique has been used vastly in molecular diagnosis industry for several years. However, in order to make the machine portable and more accessible to many remote regions of the world, the Nanofabrication Group at California Institute of Technology has been developing a completely new, fast, and compact diagnosis device based on real-time PCR technology. All components must be miniaturized and optimized to meet requirements of portability, robustness, and accessibility. A heater, thermal cycling chip, one of the main components, demands a completely new design for integrity and speed. In this project, we have designed and fabricated the next generation Polymerized Chain Reaction (PCR) thermal cycling core chip on a single flexible integrated printed circuit board for a portable real-time molecular diagnosis machine. With integrated Resistance Temperature Detector (RTD) on a single chip, the heater can achieve ultra-fast temperature ramping rates of heating and cooling within 10 seconds per cycle at circuit board thickness of 50 micrometers for an active area of 1.25 cm by 2.25 cm for one reaction cartridge with highly precise temperature control. We believe that by reducing the size without scarifying reliability, we could reach the limit of the chemistry of PCR reaction to create the next generation molecular diagnosis device that is more portable, more robust, and more accessible for everyone. Finite Sets With No Singleton Intersection Tian Nie Mentor: Richard M. Wilson Let A 0,1, . . . , q 1 be an alphabet of q elements. Let C ⊆ A be a q-ary code of length k. The Hamming distance , , … , , and , , … , , is the number of coordinates i, 1 i k, with d , between two code words . Let D be the set of all Hamming distances d , between any two code words a, b in C. Delscarte proved . However the bound is not tight. In this project, we focus on the case that if |D| s, then |C| ∑ 1 |D| k 1, i.e. just forbidding one distance. We determine the maximum attainable size of C for different values of restricted distance for small values of k and subsequently extend the results to general cases.

64

 

Kinetics Improvement for Water-Splitting Perovskite Materials Patrick Nikong Mentors: Sossina Haile and Chih-Kai Yang Using a thermochemical cycle, strontium-doped manganese oxide (LSM), a perovskite material, can split water to produce hydrogen gas, which can serve to store energy. This project seeks to investigate methods for improving the kinetics of this reaction. Three methods were tested. The first method was increasing the concentration of lanthanum and strontium in the material. The second was depositing ceria and samarium-doped ceria (SDC) nanoparticles onto the surface of the material using cerium nitrate and samarium nitrate solutions. A third, similar, method involved the deposition of active metal catalysts such as rhodium and platinum onto the surface of the LSM. X-ray diffraction (XRD) and scanning electron microscope analysis indicated that we were successful in depositing ceria, SDC, and other metal catalysts on the surface of LSM pellets. XRD analysis also suggests that the increase in lanthanum concentration created secondary phases within the LSM material. Thermochemical tests so far indicate a slight improvement in the kinetics of the reaction with increased lanthanum and strontium concentration. Further testing with different compositions and heat treatments will confirm the efficacy of these methods. Volatile Transport on Pluto: A Photometric Investigation Ariel O'Neill Mentor: Bonnie Buratti As Pluto moves with respect to the sun, the changing temperature and atmospheric pressure may cause methane and nitrogen ice to sublimate and redeposit seasonally. It has been determined that this volatile transport occurs on Neptune's moon Triton. Triton and Pluto have similar orbital characteristics, atmospheres, and chemical composition, but we do not yet know whether volatile transport occurs on Pluto. Through observations at Table Mountain Observatory's 0.6 meter telescope, plots of Pluto's surface albedo with respect to phase angle were obtained in the B, V, and R filters. These light curves were compared to historical data. If Pluto's surface is static, then light curves from different years should be identical after accounting for the relative positions and orientations of Pluto, Earth, and the sun. If volatile transport occurs on Pluto, then the surface albedo will change and the measured light curve amplitude will deviate. Measuring the visual light curve amplitude provides insight into that nature of Pluto's surface before NASA's New Horizons spacecraft encounters Pluto in 2015. Razor Variable Search for a High-Mass Higgs Decaying to WW in the Leptonic Final State With an Extension to a Boosted 2H  4W Analysis Georges Obied Mentor: Maria Spiropulu We will report the results of a search for a high-mass Higgs at the LHC using data collected with the CMS detector corresponding to an integrated luminosity of 19.6 fb-1. This will be done by investigating a high-mass Higgs that decays to WW which in turn undergo leptonic decays giving rise to opposite flavor (OF) or same flavor (SF) leptons and missing energy. A razor variable approach is particularly suitable for this decay process due to the presence of missing energy in the final state. Appropriately defined razor variables would give sensitivity to the presence and the mass scale of a higher mass Higgs. This study will then be extended to a boosted 2H  4W decay in which the pair of W bosons from the first Higgs decays leptonically and the other pair decays hadronically producing merged jets. Proximity Ligation Assay to Measure α/β T Cell Receptor Pairing Meghana Pagadala Mentors: David Baltimore and Michael Bethune The heterodimeric α/β T cell receptor (TCR) is the sole determinant of T cell specificity. In TCR gene therapy, patient T cells are transduced with tumor-specific α and β TCR genes to induce anti-tumor immunity. However, introduced TCR α and β chains can mispair with the transduced T cell’s endogenous TCR β and α chains, respectively, reducing the number of tumor-specific TCRs on the surface and potentially generating autoreactive TCRs. We are developing a quantitative assay to measure the extent to which mispairing occurs for TCRs of clinical interest. Proximity ligation assays (PLA) use oligonucleotide-linked antibodies to convert the detection of a protein interaction to a nucleic acid readout. Antibody-linked oligonucleotides are brought into proximity only if the antibodies’ targets interact. Enzymatic ligation of proximal oligonucleotides creates a novel oligonucleotide that can be detected via quantitative PCR. We identified two antibodies that bind the constant portions of the endogenous TCR α and β chains. These epitope sites were mutated in the transduced TCR chains and synthetic epitope tags were added to the N-terminus of these chains to enable their orthogonal recognition by a second set of two antibodies. Thus, with four antibodies, all possible α/β pairings can be distinguished.

65

  Synthesis of Ruthenium 3-Way Junction DNA Intercalator Lisa Pangilinan Mentors: Jacqueline K. Barton, Ariel L. Furst, and Michael G. Hill DNA-based electronics have taken on a larger role in the development of new circuitry platforms. The selfselectivity, fast electron transfer rates, and specificity of DNA make it an ideal molecule for molecular electronics. However, this sensitivity makes it difficult to direct charge around a corner in DNA. This project focuses on the synthesis of a metallointercalator complex that will connect two arms of a 3-way junction and enable charge transport around a corner. Progress was made towards the synthesis of the metal complex and successful formation of a 3-way junction. Although the final biruthenium complex is still being constructed, future work includes determining the electrochemistry of the metal complex alone and testing the complex both for binding affinity and electrochemistry in a 3-way junction. Random Walk Simulation for Particles and Cilia Grace E. Park Mentors: Mathieu Desbrun and Eva Kanso The random walk can describe the motion and diffusion of small particles and molecules within organisms. The concept of diffusion has several applications in physics, chemistry, biology, sociology, economics, and other various fields. In this work, we are particularly interested in the random walk of an ensemble of particles in a heterogeneous environment. Our main motivation is to try to understand the effect of oscillating cilia on the diffusion of molecules. Cilia are hair-like microstructures observed on the surface of many biological systems such as the human lungs, and mainly serve as the sensory organelles. Here, we model the cilia as oscillating circular obstacles and we develop a visual program to study the collective behavior of diffusing particles in such an environment. In the absence of any obstacles, an ensemble of particles undergoing a random walk would generate a normal Gaussian distribution centered at the origin. However, the obstacles that absorb, hold, or reflect the particles change the outcome and reveal interesting distributions, which clearly define shadowed area where particles are unlikely to be able to reach. Synthesis and Characterization of Vertically Aligned Carbon Nanotubes Nicholas Parker Mentors: Chiara Daraio and Ramathasan Thevamaran We grow of vertically aligned forests composed of single and multi walled carbon nanotubes (CNTs) using a thermal chemical vapor deposition process. We control the CNTs’ properties by tuning the partial pressure of the carbon source, acetylene, from 0.3 mbar to 16.5 mbar and by controlling the chamber temperature from 720°C to 775°C. To control the partial pressure, we regulate the overall pressure of the reaction chamber (80 mbar to 720 mbar) and the composition of the feedstock (nitrogen, hydrogen, acetylene, argon). We observe that an increase in the partial pressure of acetylene results in an increase of the CNTs’ terminal growth length. We attribute this effect to an increase in the initial CNT growth rate. Beyond 3 mbar of acetylene partial pressure we did not observe an increase in terminal growth length. In addition, higher hydrogen content in the feedstock reduces the CNTs’ bulk density. We use an Instron E3000 compression testing machine for quasi-static characterization, and an impact testing system based on moiré interferometry for dynamic characterization of the resulting single and multilayer structures. A high-speed camera coupled to a microscopic lens is used to visually observe the deformations. Implementation of the Sipper Probe for the Mass Spectrometric Analysis of Submerged Interfaces Aleena Patel Mentors: J.L. Beauchamp and Daniel Thomas A novel nano-electrospray source (nanospray), the Sipper Probe is proposed to selectively sample from submerged interfaces for mass spectrometric analysis. While current methods of nanospray rely on external pumping to introduce sample through a nanospray emitter, the Sipper Probe relies on capillary action and electro-osmotic pull to drive the continuous flow of sample through a tapered silica capillary tube. An electric field generated by high voltage applied to a nanospray emitter tip initiates soft electrospray ionization of the sample at the tapered tip. The nanospray emitter remains orthogonal to the liquid sample contained in a glass vial. This design permits the construction of a submerged interface that forms between two immiscible liquids in the sampling vial. Changes in volume of the liquid in the bottom phase cause the vertical motion of this submerged interface. As the interface is moved vertically past the nanospray emitter, we continuously obtain sample through the interface from one phase to the next.

66

  Fabrication of Silver Photonic Nanocrystals Shane Patel Mentors: Julia Greer and Victoria Chernow Photonic Crystals utilize the periodicity of dielectric constants to exert control on the propagation of photons. Many different techniques were tested with the goal of fabricating hollow, silver nanotrusses deposited on a transparent substrate. These nanotrusses provide an ideal medium for photonic crystals because of their repeated physical features, which are on the same size order as the wavelengths of the visible light spectrum. Polymer nanotrusses were created using a nanofabrication technique called two-photon lithography. These trusses were then coated in silver, using thin film methods such as sputtering and chemical vapor deposition. We created nanotrusses of varying unit cell sizes, between two µm and ten µm, and heights, between 2.5 and 10.5 unit cells. Reflectance and transmission measurements are being gathered in order to determine the presence of a photonic bandgap – a forbidden range of photonic wavelengths. Future thrusts of this research will include fine tuning the bandgap by varying truss shape and size. Quantum Fourier Transforms on Continuous Groups Evan J. Patterson Mentors: John Preskill and Gorjan Alagic The quantum Fourier transform computes the discrete Fourier transform (DFT) as the coefficients of a quantum superposition. In the general setting of harmonic analysis on groups, the DFT may be viewed as the Fourier transform on the (finite) cyclic group. Motivated by the success of this algorithm, we develop quantum algorithms for Fourier transforms on continuous groups. We present an exact quantum algorithm for the Fourier transform of a band-limited function on the circle group. We then consider several extensions and an application to trigonometric interpolation. We also investigate quantum Fourier transforms on nonabelian continuous groups, such as the special unitary group and the rotation group. Axial Patterning and Blastomere Fates of the Early Petromyzon marinus Embryo Sarah Pearce Mentors: Marianne Bronner and Stephen Green Cleavage stage developmental fatemaps identify which tissues arise from individual blastomeres of the early embryo. They are a crucial tool in developmental biology that allow for the accurate description and interpretation of experiments, tissue-specific studies in vivo, and the discovery of lineage determination mechanisms in development. Surprisingly little is known about the early development of Petromyzon marinus. Also known as the sea lamprey, this species is located at a key phylogenetic position, poised to reveal mechanisms of vertebrate evolution and lend valuable information to biomedical studies. A fatemap for the species should prove an invaluable tool. Thus, through experiments consisting primarily of fluorescent lineage tracer injections at early cleavage stages and analyses of staining in later stage embryos, we assigned prospective fates to blastomeres at two-, four-, and eight-cell stages. These data suggest that events in body plan development, including the early establishment of the dorsal-ventral, anterior-posterior, and left-right axes, occur by the first and second divisions post-fertilization. Furthermore, we find that these axes appear to be oriented along characteristic early division planes. We plan to further analyze experimental embryos in order to continue assembling a prospective cleavage stage fatemap which will greatly aid crucial future studies in lamprey. Stacking Galaxy Spectra: Developing an Averaging Procedure to Detect Faint Emission Lines and Create Combined Spectra John Pharo Mentor: Chuck Steidel An important step in understanding Big Bang Cosmology is observing the behavior of baryonic matter both within the dark matter haloes of galaxies and in the intergalactic medium. Cold accretion of gas onto galaxies is widely used in theoretical models of the development of structure in galaxies. In order to actually observe such structural development, one relies on the relative strengths of lines in emission spectra. This can be difficult, because many observed lines have low signal-to-noise ratios. This can be alleviated through “stacking” the spectra, or combining them in ways to increase this ratio. Developing a procedure to perform this on weak-line spectra can allow for actual measurement of the weaker lines, which can lead to actual inferences about the baryonic flow and structural development of the objects. This process can further be used to make combined spectra and aid in other data analysis.

67

  Characterizing the Spatiotemporal Expression and Function of HDAC4 During Neural Crest Development Jacquelyn Phillips Mentors: Marianne Bronner and Crystal Rogers Histone deacetylases are a class of epigenetic modifiers that strengthen the bonds between the DNA backbone and histones, thereby reducing the expression of the genes encoded by the DNA wrapped around those histones. Here, we isolate and identify the spatiotemporal expression pattern and function of histone deacetlyase-4 (HDAC4) during avian neural crest development. HDAC4 competes with the activity of histone acetyl-transferases and can be recruited to specific genomic regions by transcription factors such as MEF2. MEF2 has been widely studied and is expressed in neural crest cells, suggesting that HDAC4 may regulate neural crest-specific genes through interactions with MEF2. Mutations in human HDAC4 are associated with neural crest-related disorders such as, cleft palate and brachydactyly mental retardation syndrome, and studies have shown that it may be expressed in cranial neural crest cells in zebrafish. There is very little information defining the expression or function of HDAC4 during embryonic development, but we hypothesize that it may be expressed at the right stages and in the right tissues to be involved in neural crest development in chick. Our preliminary studies suggest that HDAC4 transcripts are expressed ubiquitously with regions of higher expression in the developing nervous system and neural crest cells. Sensitivity of the Spin Measurement of the Higgs-Like Boson in the LHC Run 2 Dataset at CMS Jeff Picard Mentors: Maria Spiropulu and Alex Mott Results are reported on the expected sensitivity in the diphoton decay channel for measurement of the spin-parity of the recently discovered Higgs-like resonance in an sqrt(s) = 14 TeV dataset from the CMS experiment at the LHC. The expected improvements in sensitivity for spin-parity measurements from proposed upgrades to the CMS Electronic Calorimeter are also presented. Monte Carlo simulations inform the generation of toy data from which we plot the angular distribution of the photons in the resonance rest frame and observe the corresponding uncertainty and test statistic. The analysis compares the Standard Model Higgs with a spin-2 graviton-like model with minimal couplings. The diphoton invariant mass is used to separate signal and background via the sPlot method. Bismuth Vanadate–Silicon Tandem Photoelectrodes for Solar Water Splitting Alex Pien Mentors: Nathan Lewis and Matthew Shaner Tandem junction devices have the highest theoretical efficiency for unassisted solar water splitting. Stable, transparent, conductive materials are needed to provide electrical connection between the two semiconductors. Additionally, high aspect ratio structures, such as silicon microwires, are preferable for water splitting as they minimize ion-transport and therefore conformal coatings of these contact materials are necessary. Full tandem junction devices are synthesized by sequential spray deposition of fluorine-doped tin oxide as a stable, conductive contact followed by tungsten-doped bismuth vanadate as the second semiconductor. Finally, a cobalt oxide catalyst is deposited through photo-assisted electrodeposited onto the photoelectrodes. A tandem junction photoelectrode without a catalyst exhibits a short circuit current density of 1.4552 mA-cm-2 at the oxygen evolution reaction potential (OER) and open circuit potential of -.3796 V with respect to the reversible hydrogen electrode (RHE). Additionally, cyclic voltammetry indicates the tandem junction photoelectrode operates with a current density of .3784 mA-cm-2 at the hydrogen evolution reaction potential. Identifying Regions of Lhcb5 Involved in Binding in the Lhcb5-cpSRP43 Complex Samantha Piszkiewicz Mentor: Shu-ou Shan Protein homeostasis is essential for all cells and requires the proper control of the folding, localization, and interactions of all proteins. The misfolding and aggregation of proteins are detrimental to cells and have been found to be the root cause of numerous age-related diseases. The cpSRP pathway can be used as a model system to study the general mechanisms by which a chaperone protein (cpSRP43) can solubilize a typical hydrophobic membrane protein (Lhcb5). We have probed for binding interactions between Lhcb5 and cpSRP43 at various residues in Lhcb5 using N-ethyl-maleimide labeling of single cysteine mutants. LC-MS allows us to observe the accessibility of each residue to the solvent in the form of the fraction of reacted protein, from which we can infer which regions of Lhcb5 are involved in binding. Characterizing the Hot Kepler Objects of Interest Ellen Price Mentors: John Johnson and Leslie Rogers Planets around retired A stars (hot stars that have evolved off the main sequence) tend to be further away from their stars than planets around cooler, Sun-like stars. This could be due to the fact that the star is evolved; as stars evolve and expand, they may engulf closer-in planets, or the planets’ orbits may tidally decay until they are destroyed. Alternatively, differences in planet formation processes around these more massive stars may account

68

  for the observed result. To distinguish between these two possibilities, we characterize main sequence A stars with transiting planets detected by Kepler. We identified likely A stars in the Kepler Input Catalog by their estimated stellar effective temperatures. To verify the classification of these stars, we measured their spectra using Palomar Observatory’s Dual Beam Spectrograph and Polarimeter (DBSP) and collected high-resolution images with Keck NIRC2. We determined the physical parameters of the transiting planets’ orbits by fitting the Kepler transit lightcurves with Markov Chain Monte Carlo. By constraining the semi-major axis and eccentricity distributions of planets orbiting A stars, we gain insights into the role of stellar evolution in hot star planetary systems. Analysis of the Cellular Origin and Role of Epigenetic Regulators in Plant Regeneration in Arabidopsis R. Edward Pursifull Mentors: Elliot Meyerowitz and Kaoru Sugimoto Arabidopsis regeneration occurs from growing cell mass called callus, which is induced from various types of tissues by hormonal treatment. Xylem pole pericycle (XPP) cells are present in the plant’s root vasculature and known to be an origin of root-derived callus. However, it is not yet described in detail whether other tissues outside of roots contain this type of cell, and what molecular mechanisms provide the cells with the competency to form callus. In order to determine these, reporters are being constructed to map expression for promoters unique to the XPP cells. This will allow further characterization of the cells, such as cell type-specific genome analysis. As epigenetic state modifications are implicated in reprogramming process in various systems, I am also investigating the role of epigenetic regulators in Arabidopsis callus formation. Observation of defective mutants for epigenetic controllers identified the line with SETDOMAIN GROUP 15 mutation as exhibiting severe decrease in Arabidopsis ability to form callus tissue from the roots, indicating that it is important to processes that occur in callus tissue formation. These results can be compared with callus formation phenotypes of aerial tissues to determine similarities between the processes of callus formation from different tissues. The Effects of Transcriptional Driver, TCF1, on the Regulation and Maintenance of T-Cell Specification and Differentiation Shuyang Qin Mentors: Ellen V. Rothenberg and Sagar Damle One transcription factor, T-cell factor 1 (TCF1), encoded by Tcf7, is thought to be critical in the regulation and maintenance of the T-cell commitment program. The progression of T-cell differentiation is described through changes in surface markers cKit and CD25, in which DN1 cells are phenotypically characterized by cKit+/CD25-, DN2 by cKit+/CD25+, DN3 by cKit-/CD25+, and DN4 by cKit-/CD25-. Tcf7 has been found to be heavily upregulated during the DN1-DN2 transition and is present in large amounts during later stages of development. Here, we show that different levels of Tcf1 expression not only change the transcription of important T-cell specific genes, but may also affect the pace of commitment. Specifically, we observe that downregulating Tcf7 expression in mice fetal liver hematopoietic progenitors decreases CD25 expression, which may signify either a retardation of the T-cell program or a failure to expand the DN2 population. We aim to discover, through quantitative RT-QPCR, both direct Tcf7 target genes as well as indirect target genes that may function downstream in the early T-cell commitment process in order to further understand and confirm the regulatory roles of Tcf7. Study of Infrared (IR) LEDs for Skull Tissue Penetration Brynan Qiu Mentors: Azita Emami-Neyestanak, Krishna Settaluri, Mayank Raj, and Manuel Monge In cases of brain trauma, research has shown that side effects can be minimized if the affected area of the brain can be treated with drugs within one hour of the event. Previous simulations and preliminary data indicate that infrared light is capable of penetrating skull/brain tissue, which has the potential to enable targeted drug-release in the brain to treat cases of brain trauma quickly and reduce undesirable side effects. The objective of this study is to verify previous results and examine the light scattering characteristics of skull/brain tissue. For this, a test board has been designed and built that can be used to test selected high power infrared LEDs at varying intensities. Sensor Platforms for Deployment in Rural Areas Nishant Rao Mentors: Julian Bunn and K. Mani Chandy This project explores the improvements in technology and designs for small sensors to be placed in remote rural areas. The present generation sensors being powered through external wired plug-ins and mostly communicating via 3G network with the cloud, it becomes highly inconvenient to deploy them in rural areas of India where about 70% of the population resides. This project dealt with developing and improvising the initially modeled stand-alone earthquake sensor box, to make it withstand the technological and geographical conditions in India. The algorithm to monitor data communication considering the power available in the sensor was developed. Besides, the containers provide power to the sensors through batteries coupled to solar cells and communicate with cloudcomputing services through 2G cellular networks. Moreover, the functionality of the container and its sensors must degrade gracefully when the battery power gets low and solar energy is unavailable. A key constraint was the cost-

69

  effectiveness of the container. The containers and batteries are designed for deployment in areas such as Kutch and Saurashtra in Gujarat State in India; these regions have access to a 2G cellular network and have adequate sunshine during most months of the year. A Secure Digital High Capacity (SDHC) memory card was introduced for storing critical data in case power was insufficient for the sensor to communicate. Understanding Fluid Transport Through Carbon Nanotubes for Optimisation of Microneedle Design and Performance Neeru Ravi Mentors: Morteza Gharib and Bradley Lyon Microneedles offer significant advantages over conventional hypodermic syringes. They are of minute dimensions and allow almost instantaneous drug delivery due to their powerful delivery capabilities. This system also improves patient compliance and safety, and most importantly is virtually painless. The drug delivery system comprises a polymer skin patch and an array of microneedles made of a carbon nanotube, polyimide composite. The focus of this project is to develop a model for the fluid wicking behaviour through bundles of vertically aligned carbon nanotubes, and compare it to theoretical predictions of capillary action. Initially tests are carried out on the unpatterned nanotube material using glycerol, and theory later verified using the actual microneedles and polyimide. The fluid flow behaviour during spin coating, in which both capillary action and centripetal forces act is also an area of investigation. Understanding the fluid transport phenomenon will enable different polymers (other than polyimide) to be used to construct the microneedle, and possibly aid the development of dissolvable microneedles for prolonged, local drug delivery. Thus the result of this study will be to optimise the design and performance of the microneedles. Simulation of Comets Using Cellular Concrete Caelan Reed Mentor: Greg Peters Comets are made of bits of ices and minerals left over from the original formation of our solar system and thus examination of these primordial materials would be very instructive in the understanding of the origins of our solar system. To this end engineers and scientists at JPL are developing sample acquisition systems capable of comet sample return missions which necessitate an adequate comet simulant with mechanical properties similar to real comets at non-cryogenic temperatures for the testing of previously mentioned systems. It is generally accepted that comets have a bulk density ranging from 0.2 to 0.5 g/cc and a shear stress ranging from 1 to 125 kPa, so the comet simulant should span this range of properties. The simulant must also behave as a comet would when struck, that is it will hold itself together as a comet would due to its various internal forces but would be granular in nature when struck. To achieve this, a Portland cement mixture with Vermiculite as the aggregate and a foaming agent called Varimax is used to create a light cellular concrete material. Using this cellular concrete mixture the range of shear stresses can be achieved with a density of 0.5 g/cc. Structural Study of TRPA1 Ankyrin Repeat Domain Modification by Covalent Linked Noxious Electrophilic Compounds Michelle Soto Reid Mentors: Douglas C. Rees and Aron Kamajaya The structure of the Transient Receptor Potential channel (TRP) is interesting due to its ability to integrate signals from a variety of sensory stimuli such as light, sound, chemicals, temperature, and touch. The structural region by which these signals are integrated is not well understood. In mammals there are six related families of TRP genes: canonical (C), vanilliod binding (V), melastatin related (M), ankyrin repeat (A), polycystin (P), and mucolipin (ML). The cytosolic N and C terminal domain of the transmembrane segments participate in channel gating and regulation and sense information about cellular state. TRPA1 (TRP Ankyrin 1 known for its extensive ankyrin repeat domain) is known to respond to a variety of noxious stimuli including electrophilic compounds (found in wasabi, horseradish, mustard oils, garlic, cinnamon oil, etc). It was shown that these electrophilic compounds activate TRPA1 by the covalent modification of five cysteines residues on the ankyrin repeat domain (ARD) of the ion channel. The actual conformational change of the TRPA1 ARD that is covalently linked to electrophilic compounds is unknown. In this study, the overexpression and purification of the full-length human TRPA1 ARD (66kDa) and truncated human TRPA1 ARD (44.8kDa and 29kDa) were designed containing all five of the highly reactive cysteine residues in E. Coli. Analyzing the crystal structure of these constructs while bound to pungent electrophilic compounds has the potential of mapping the structural modification on the TRPA1 ARD. An IR FMCW Coherent Camera Angad S. Rekhi Mentor: Ali Hajimiri An IR (1550 nm) frequency-modulated continuous-wave (FMCW) coherent imaging system is presented. The crux of the system is a Mach-Zehnder interferometer (MZI) whose reference arm lies in fiber and whose imaging arm lies in free space, where the object being imaged is suspended. A tunable laser amplitude-modulated by a linearly

70

  chirped electrical signal is fed into the input of the MZI. A 4x4 grating coupler array integrated in silicon is used to receive light scattered from the object being imaged. For each grating coupler, the light from both arms is recombined and photodetected on-chip; the resulting electrical signals are filtered and amplified off-chip, captured using a digital oscilloscope, and sent to a computer, where the data is used to reconstruct the object being imaged. Derivations and demonstrations of 3-D transmissive, 3-D reflective, and 2-D phase contrast imaging are shown. Magnitude and Timing of the Mid-Holocene Sea-Level Highstand at Belitung Island, Indonesia Andrea J. Ritch Mentors: Adam D. Switzer and Aron J. Meltzner Following the termination of the last glacial period, changes in relative sea level (RSL) have varied based on location with respect to the glacial centers. Ice-induced crustal deformation governs the RSL signal in areas nearest the glacial centers; in far-field regions, such as the Sunda Shelf in Southeast Asia, the eustatic meltwater signal dominates. Models of this glacial isostatic adjustment (GIA) have implied a sea-level maximum, or highstand, in far-field locations as the production of meltwater decreased during the mid-Holocene. However, there are no reliable RSL records from Southeast Asia, leaving GIA models uncalibrated and the magnitude and timing of the sea-level highstand unresolved. Analysis of the annual growth banding of coral microatolls, whose vertical growth is limited by sea level, allows us to quantify mid-Holocene sea-level change around Belitung Island on the Sunda Shelf of Southeast Asia. Our chronology is based on both 14C and U-series dating of the corals, but there are complications associated with each method. Inconsistency in the U-series dating leads us to prefer the 14C results, though we will need to determine a more accurate marine radiocarbon reservoir age correction in order to constrain the timing of the sea-level highstand more precisely. Working Towards a Solution to the Ân Standard Form Problem Thomas Robinson Mentor: Michael Aschbacher In the attempt to classify 2-fusion systems of component type, it is necessary to prove an analogue of the component theorem for finite groups. We consider some problems on alternating and symmetric groups needed to solve the standard form problems for 2-fusion systems with components which are 2-fusion systems of the covering group of an alternating group. In particular, we fill in details of a proof by Solomon and relate his theorem to fusion systems. Thermoelectric Properties of Sr5In2Sb6 Andrew Romine Mentors: Jeff Snyder and Alexandra Zevalkink Research into thermoelectric materials exists in order to find materials with specialized properties and high zT values, a quantitative measurement that describes the material’s ability to convert a heat gradient into electricity or the ability to turn an electrical flow into a solid state cooling devise. Research on Sr5In2Sb6 began because of its similarity in lattice structure and chemical formula as Ca5In2Sb6, which has been shown by Alex Williams in the Snyder lab as a viable thermoelectric with a zT value up to 0.9 when doped with Zn in the In location. Research into Sr5In2Sb6 has proved that it could be a useful thermoelectric. After creation of the material, density tests, XRD tests, and SEM tests were performed to assure Sr5In2Sb6 was created. These tests had positive results and further testing into the speed of sound across the material, the high temperature Seebeck value of the material, the Hall value of the material, and the thermal conductivity of the material were performed. While Zn doped samples of Sr5In2Sb6 still need to be tested to find a high zT value the tests performed on the non-doped samples show that Sr5In2Sb6 shares many characteristics of Ca5In2Sb6 and should be a useful thermoelectric. Structural and Functional Characterization of Bruton’s Tyrosine Kinase Connor E. Rosen Mentors: John Kuriyan and Qi Wang Bruton’s Tyrosine Kinase (Btk) is a key signaling molecule in B-cell activation, and misregulation of Btk signaling has been linked with both immunodeficiency syndromes (X-linked agammaglobulinemia) as well as cancers. Using X-ray crystallography, we have determined the structure of the Btk kinase domain bound to a small molecule inhibitor. Using this structure as a guide, we have investigated the binding of small molecules to Btk that specifically activate or inhibit the kinase, by structural modeling as well as mutational analysis in in vitro enzymatic assays with small molecules. Future investigations will focus on determining the mechanism of specificity for the small molecule-mediated regulation of Btk activity. Our work enhances the understanding of Btk regulation and how modulation of Btk activity may be achieved in a scientifically and clinically relevant context.

71

  Plasmonic Behavior of Gold Shape-Memory Alloys Tatiana A. Roy Mentors: Axel Scherer and Dvin Adalian We are interested in the plasmonic behavior of Spangold, a shape-memory alloy. This alloy exhibits a crystal phase transition at a semiconductor device-compatible temperature. We are investigating the change in plasmonic behavior between these two crystal phases in order to learn more about the principles of surface plasmons. Gold has excellent visible wavelength surface plasmon behavior as evidenced by its yellow color, which stems from its orbital structure. To create these alloys, we built a rotating and heated stage to sputter while crystallizing the metals onto a substrate. To observe the SMA transition behavior, we will measure the resistivity change of the material, as well as changes in the absorption spectrum using a far-field transmission/reflectance visible wavelength spectrometer setup. We are searching for changes in the reflectivity and transmissivity of the SMA to study the plasmonic resonance behavior. Development of Selective Ethylene Oligomerization Catalysts Iva Rreza Mentors: John Bercaw and Aaron Sattler The development of catalysts that selectively oligomerize olefins for uses in polymers and fuels remains of interest to the petrochemical industry. A new series of transition metal compounds implementing the Fujita ligand (FI) (FI = phenoxy-imine) framework has been synthesized and characterized with NMR spectroscopy and X-ray crystallography. The air and moisture sensitivity of these complexes stipulates the use of Schlenk line, highvacuum line and glovebox techniques. Interestingly, a tantalum(V) complex, namely (FI)TaMe2Cl2, dimerizes ethylene at room temperature to produce 1-butene when activated with Et2Zn. Further studies will aim to determine the active species in this system, in addition to exploring other variants of the catalyst and the oligomerization of other α-olefins. Analysis of the Enhancer Region Responsible for c-Myc Expression in the Premigratory Neural Crest Cells Elizabeth Ryan Mentors: Marianne Bronner and Laura Kerosuo The neural crest is a pluripotent population of stem cells that form in the dorsal neural folds during early embryo development. After specification, the neural crest cells go through epithelial to mesenchymal transition, detach from the neural tube and migrate to different locations in the embryo giving rise to various cell types including melanocytes, the peripheral nervous system and bone and cartilage of the face. The gene regulatory network (GRN) underlying the correct identity of the developing neural crest cells has been intensively studied in the Bronner laboratory. Here, we have dissected the 200kb enhancer region surrounding the gene coding for c-Myc into 22 pieces and cloned them into a fluorescence reporter construct with a minimal promoter (ptkEGFP). By using in vivo electroporation of gastrulating avian embryos, we found two conserved regions that drive the expression of the c-Myc gene in the neural crest. Comparison of the expression of the enhancer to the WT expression of c-Myc reveals that the enhancer driven domain is broader than the expression of the actual c-Myc RNA. In ongoing experiments, we are further dissecting the active enhancer regions into smaller, 500bp pieces in order to find the functional domain. By using the JASPAR software we will identify the putative binding sites of transcription factors and by using site directed mutagenesis we will then identify the active sites. Unpublished data in the lab suggest that c-Myc is regulating self-renewal of the premigratory neural crest. This study will provide us with valuable data of the upstream regulators controlling the self-renewal process. The Viability of Lasting Formation of Terrestrial Binary Planet Systems Keegan Ryan Mentor: David Stevenson The formation and evolution of binary star systems are both common in the universe and well researched. However, little research has gone into the possibility of binary planet systems containing two bodies of similar mass. No such systems have yet been observed. However, in a grazing collision between two earth-like bodies, conservation of angular momentum dictates that completely merging the two bodies would be unstable, resulting in multiple bodies. We began by examining the initial conditions which would encourage formation of a binary planet system, and then we simulated those scenarios using smoothed particle hydrodynamics. In the cases where the masses did not escape, one large body of approximately 90% of the total mass would form, with most of the remainder of the mass orbiting around it. Thus if terrestrial binary planet system formation is possible, the initial conditions cannot be in the space spanned by the conditions we tested. Additional simulations would be useful here, to find if there are outcomes other than partial accretion or escape.

72

  Fold-Change Detection Circuitry: Probing the Regulatory Mechanism of Temporal Ratio Computation in Wnt Signaling via Protein-DNA Interactions Bryan Ryba Mentors: Lea Goentoro and Jae Hyoung Cho The Wnt pathway plays a significant role in determining the phenotypic development of Xenopus embryos. Past studies have shown that the siamois target gene of Wnt signaling is governed by fold-change detection of betacatenin. An 11-base pair (bp) promoter region found within 1.3 kb of siamois is known to be responsible for foldchange detection. The purpose of this project was to characterize the protein-DNA binding interactions occurring at this 11-bp promoter. Electrophoretic mobility shift assays (EMSAs) were performed with an infrared-labeled 11-bp promoter, embryo protein extracts, and various unlabeled competitor DNAs in order to characterize binding specificity and competition. Mutations as specific as single-nucleotide substitutions were shown to drastically decrease target protein binding to this region. Subcloning and bacterial expression of several TCF-related proteins were used to study binding of TCF-related factors to the 11-bp promoter. Wnt stimulation via lithium chloride treatment and subsequent luciferase assays were then used to study the effect of embryo-injected mRNAs on foldchange detection (results not available at time of publication). Continuation of the aforementioned experiments will aid in characterization of the mechanisms governing fold-change detection, while additional assays will be used to help identify the target protein(s) binding the 11-bp promoter. Phase Retrieval Algorithm for Improved Resolution in Digital In-Line Holography on a Chip Donghun Ryu Mentors: Changhuei Yang and Guoan Zheng We present a unique phase retrieval algorithm to improve the resolution performance of a lensless digital in-line holography setup. Digital in-line holography is a technique that enables a recovery of the complete description of a sample after it interacts with light illumination. Digital in-line holography has advantages on its compactness and cheap cost since it does not require any lenses, bulky optical/mechanical components or coherent sources such as lasers, making it a convenient technique to achieve chip-scale microscopy. Using a partially coherent 632nm LED, we first obtain holograms of red blood cells. Second, in digital post-processing, we reconstruct the complex field map of a collection of red blood cells from the recorded hologram intensity image. Our algorithm is based on the well-known phase-retrieval process. However, unlike traditional phase retrieval, we introduce a procedure to divide each pixel in hologram image into four pixels computationally. This modification in algorithm allows us to obtain improved resolution of reconstructed image even if we have information in original one pixel only. This modified algorithm from traditional one will be useful in that we can achieve improved resolution without any hardware changes. Evaluating the Role of Secondary Metabolites on Microbial Gene Expression Cristian F. Salgado Mentors: Dianne K. Newman, Nicholas Shikuma, and Megan Bergkessel For decades, secondary metabolites have taken the backseat in bacterial metabolic studies. Although it is difficult to understand the importance of chemicals that have limited influence on microbial growth, growing evidence suggests that these molecules are more relevant than previously thought. Historically, interest in secondary metabolism has been fueled by the assumption that many of these compounds possess antibiotic capabilities. Although this can be true, natural concentrations of these metabolites are usually present in sub-lethal levels for other bacteria; suggesting additional physiological functions. Among these metabolites are violacein and hopanoids. Previous research on these chemicals has focused on their antibiotic capability and geological prevalence, respectively. However, their significance towards microbes that produce them remains unresolved. In this study, the transcriptional response to supplementation of either metabolite was evaluated. Using Pseudoaltermonas luteoviolacea and Rhodopseudomonas palustris as model organisms, the genome-wide effects of violacein and hopanoid treatments were tracked. Ultimately, the pejorative title of secondary metabolites is challenged, and their importance for cell survival is emphasized. Flow Through Channels With Deformable Obstructions John G.W. Samuelsson Mentors: Beverley McKeon and Mitul Luhar Many internal flows of biological (e.g. cardiac and tracheal flows) and engineering (e.g. design of passive flow regulators) interests are characterized by the presence of deformable obstructions. In this project, the static and dynamic deformation of such obstructions was investigated via laboratory experiments. Experiments were conducted in the Graduate Aerospace Laboratories of the California Institute of Technology with flexible polyethylene walls of varying thickness placed normal to the water flow in the Cann Water Tunnel. The deformation of the flexible walls was captured using digital videography across a range of flow speeds. The flow speed was estimated using Digital Particle Image Velocimetry, DPIV. Results will be presented.

73

  Analysis of Flow and Vortex Formation Around an Airfoil With Rapidly Changed Asymmetric Angles of Attack Alejandro Sanchez Mentors: Mory Gharib and Emilio Graff Conventionally it is accepted based on steady flow dynamics, when an airfoil rapidly changes its angle of attack twice with opposite magnitudes two vortices will form corresponding to each motion. However, new mathematical models which consider unsteady flow dynamic effects propose, given the proper conditions, not two but three vortices will form. This phenomenon occurs due to the rapid nature of the motion from a high angle of attack, 5 degrees, to a lower angle of attack, 2 degrees. The first vortex will form as predicted at the trailing edge of the first motion. Due to the rapid change in angle, this first vortex will be cut off before compensating for the change in angle of attack and a second vortex will form possessing the opposite direction of the initial vortex. However, since the first vortex has yet to leave the immediate vicinity, a third vortex with the same polarity as the first will form and will reduce the magnitude of the second vortex predicted by steady flow dynamics. In the past the existence of this third vortex has been inconclusive, however, with modifications to the old set-up the results are clearer. The new set-up emphasizes on consistent repeatability between trials giving clearer data. With the newest results, the existence of the third vortex could neither be confirmed nor denied; however it is clear that should the third vortex exist it is being lost to noise canceling. Microorganisms Residing on Mars-Based Spacecraft: Characterization and Understanding the Potential for Survival Under Mars-Simulated Conditions Nicholas A. Sanchez Mentor: James N. Benardini Spacecraft destined for Mars are required to adhere to international planetary protection policies aimed at reducing the potential for forward and reverse biological contamination.This policy allows for a minimal bioburden to be present on a spacecraft surface upon launch, which is comprised, of organic matter and bacterial spores. During the assembly, testing, and launch operation phase of the Phoenix Lander and Mars Exploration Rovers, bioassays were conducted to verify the missions compliance with the bioburden requirements. Endospore-forming organisms from the genera Paenibacillus, Filibacter, Sporosarcina and Bacillus have frequently been isolated from sanitized spacecraft surfaces due to their high viability in extreme conditions. In this study, isolates collected from the Phoenix Lander and Mars Exploration Rovers, bioassays were characterized via16S gene sequencing. Bacillus, Paenibacillus, Brevibacillus, and Sporosarcina have been the prominent genera of the isolates tested. In addition, glycerol and cryobead stocks were generated to expand the current archive of Phoenix and MER isolates into six copies so that they can be appropriately stored for future researchers. Characterization of isolates collected from sanitized spacecraft is the an important step to improve contemporary spacecraft sanitation procedures as well as understand extremophile organisms on earth that now may be present on Mars. Brain Microstructural Correlation Between Cognitive Decline and Parkinson’s Disease at 7.0-Tesla Magnetic Resonance Imaging Laura Santoso Mentors: Geoffrey Kerchner, Kathleen Poston, and Paul Patterson Parkinson’s disease (PD) is a major neurodegenerative disorder typically characterized by abnormal motor symptoms, but patients often experience accompanying cognitive impairment. This dementia may be attributed to other neurodegenerative diseases, such as Dementia from Lewy Bodies (DLB) or comorbid Alzheimer’s Disease (AD). However, presently available clinical tests and imaging often do not provide sufficient information to make clear diagnoses, particularly when patients exhibit a mixture of neurodegenerative symptoms. Understanding the precise pathology of cognitive decline is critical for targeting potential therapies for dementia. This project will attempt to identify biomarkers of cognitive decline in 7.0 Tesla magnetic resonance images, as this new imaging approach provides more detail in regions of interest that have been suggested to be related to PD. Sixteen participants with varying levels of PD and some controls were selected for this study and were all imaged at 7T. A methodology was developed to blindly and manually trace the substantia nigra (SN), which are not clearly delineated in 3T images. Statistical analyses on the volume and symmetry of the SN were then performed. These features of the substantia nigra were then correlated with clinical measures of PD – the Unified Parkinson’s Disease Rating Scale (UPDRS), Mini-Mental Status Exam (MMSE) and the Montreal Cognitive Assessment (MOCA). Understanding correlations between imaging biomarkers within the substantia nigra and clinical measures of PD will help future physicians make stronger patient diagnoses that can help target therapies. Configurational Factors Contributing to Sum-Frequency Generation on Liquid Water Samuel Savitz Mentors: Rudolph Marcus and Kai Niu Snaphsots from molecular dynamics simulations of water were decomposed into layers using a novel definition which takes surface fluctuations into account. Structural elements relevant to sum-frequency generation (SFG) spectroscopy were then analyzed for each of these layers. Preliminary results suggest that the new definition

74

  reveals more of the structure leading to the SFG spectrum and restricts most of the SFG to a thinner band of surface molecules than definitions naïve of surface fluctuations. Furthermore, the contribution of each of the layers to the SFG spectrum can be understood in terms of configurational factors, so the SFG spectrum can be almost entirely explained by these statistical results. Elucidating the structure of the water surface has not only fundamental theoretical importance, but also has applications to understanding on-water catalysis, which is a promising technique in green chemistry and is believed to influence ozone-depleting reactions in the atmosphere. Automated Classification of Flare Stars Kartik Saxena Mentor: Ashish Mahabal Transients are astronomical objects that exhibit a change in their brightness over short period. Flare stars are variable stars that can undergo unpredictable dramatic increases in brightness for a few minutes. The Catalina Real-time Transient Survey (CRTS) can detect flare stars easily owing to its high cadence. This project aims to develop a system that can automatically classify flare stars observed by CRTS into M dwarfs and their their sub-types. A basis for classification is developed by visualizing and analyzing the properties of objects, such as their brightness, over different spectral regions, obtained using different sky surveys, from a set of pre-classified objects, and finding trends that can help distinguish one type from another. Another aspect of this project is to identify objects that do not follow these trends and study them in detail as they may correspond to rarer and more interesting objects. Applications of Secure Multiparty Computation in Secret Sharing for Tactical Environments Mahrud Sayrafi Mentors: Simon S. Woo and Edward T. Chow In cryptology, secret sharing refers to the methods that enable us to split a secret between few participants. The shareholders can then reconstruct the secret if a sufficient number of shares are combined, but no information about the secret can be gained otherwise. For most intents and purposes secret sharing is a solved problem, but is there a way to keep the shares private while using them to reconstruct the secret? Secure multiparty computation is a another field of cryptology with many open problems that intend to answer the same question, but since it is a relatively new field, not many multiparty functions exist yet. In this research we analyzed different methods of secret sharing, specifically Adi Shamir's method and KarninGreene-Hellman method, to see which ones are efficiently compatible with currently discovered multiparty functions in order to deploy them in a technology designed by researchers at JPL called Autonomous Information Unit (AIU). This technology, then, can be used in tactical environments with unreliable resources to provide nearly perfect forward secrecy for confidential data. Nanofabrication of Multi-Electrode Arrays for Neural Interfacing Elizabeth C. Schroder Mentors: Axel Scherer and Akram S. Sadek Much of the work to be done in neuroscience over the next decade, most importantly the Brain Activity Map project, will require high density, high resolution recording of neurons in every part of the brain. Current dual-sided neural probes may have up to 32 electrodes but probes featuring hundreds or even thousands of electrodes on a single shaft would vastly increase resolution of neural signal recording or stimulation and would thus be indispensable in such applications as the Brain Activity Map. As a preliminary investigation into the fabrication of such probes we attempt the fabrication of 16-electrode neural probes and experiment with the process. The probes are 9 mm long and taper from 5.5 mm to 55 µm wide with 16 100 µm2 electrodes and are fabricated using standard nanofabrication techniques such as photolithography, chemical vapor deposition, and reactive ion etching for which we vary the parameters and observe the results. Multi-Domain Overset Mesh Spectral Implicit-Explicit Solvers for Maxwell's Equations Simon Schulz Mentors: Oscar P. Bruno and Aditya Viswanathan The project revolved around the numerical analysis of Maxwell's Equations for Classical Electrodynamics and the types of Partial Differential Equation (PDE) solvers used in one-dimensional and two-dimensional problems. Initially, we investigated the underlying theory and checked existence, uniqueness and regularity of solutions to the problems under consideration. We then moved on to developing explicit and implicit numerical algorithms for solving electrodynamics problems where we implemented Perfect Electric Conductor (PEC) boundary conditions and Perfectly Matched Layer (PML) boundary conditions in multiple dimensions. The final three weeks of the project were spent developing code to solve a multi-domain Maxwell system with overlapping meshes. Here, we used an

75

  explicit method in one domain (specifically a fourth-order Adams-Bashforth time-stepping method with Fourier Continuation spatial differentiation) and an implicit method in the other domain (second-order Backward Differentiation Formula time-stepping with Chebyshev Collocation spatial differentiation). Applications for such solvers are found in antenna design, sonar, and many more devices. Supercontinuum Generation in Ultra-Low-Loss On-Chip Waveguides David A. Sell Mentors: Kerry J. Vahala and Dongyoon Oh Supercontinuum generation is a result of extreme spectral broadening as a consequence of nonlinear interactions within a medium. The most successful observations of this phenomenon have been observed in fiber media. Fabrication of on-chip waveguides with 0.08 dB/m loss rates and compact spiral structures of path lengths of about 5 meters provides a potential source of integrated supercontinuum generation. This possibility is tested by free space coupling light from an optical parametric oscillator consisting of 130fs pulses with an 80MHz repetition rate and up to 350mW average power. Simulation data suggests that 100mW of such light propagating through a 5 meter spiral waveguide can produce nearly 800nm of spectral broadening around the zero dispersion wavelength of 1250nm. Efforts to achieve the required coupling efficiency for such an observation are currently underway. In-Vivo and Integrated CMOS Sensors for Electrochemical Glucose Detection Mehmet Sencan Mentors: Axel Scherer and Muhammad Mujeeb-U-Rahman Diabetes is a significant health problem. Continuous blood glucose monitoring guided treatment has shown to be extremely effective in mitigating the symptoms. Current systems for continuous blood glucose monitoring involve centimeter-scale implants which tether through the skin to an outside transmitter. It is proposed that a micronscale, untethered implant would be more comfortable and effective. To this end, micron-scale electrodes fabricated on silicon and CMOS substrate were assembled, functionalized, and packaged for glucose sensing in laboratory rats. The sensors will be tested in rats for specificity as well as biological compatibility. Protective Effects of Nitrite Treatment Against Necrotizing Enterocolitis: An Investigation of Expression Levels of the NEC-Related Genes iNOS and iFABP Neera M. Shah Mentors: Arlin Blood and Henry Lester Necrotizing enterocolitis (NEC) is a disease that affects about 20% of premature infants and has a 20-30% death rate. This disease results in the necrosis of the intestinal tract and is most commonly associated with low birth weight, artificial breast milk nutrition, and ischemic insult; however, the etiology is unknown. Dietary nitrite has been shown to be protective in adults by being converted to NO, which increases blood flow and mucus production in the gastrointestinal tract. Thus, it is hypothesized that nitrite treatment may prevent the progression of NEC in infants by increasing blood flow in the intestine. Using intestinal tissues previously collected from a rat pup model of NEC, the gene expression patterns of inducible nitric oxide synthase (iNOS) and intestinal fatty acid binding protein (iFABP), will be used as an index of disease severity. iNOS and iFABP are two proteins known to be highly expressed in the intestines of infants with NEC. Expression will be quantitatively measured via real-time polymerase chain reaction (qPCR) and qualitatively shown via immunohistochemistry. Success of the nitrite treatment will be based on relative comparisons of tissues from treated vs. healthy and diseased pups. Effect of Chronic Hypoxia on Expression of Neuroglobin and Cytoglobin in Brains of Fetal and Adult Sheep Nehaly Shah Mentor: Arlin Blood Neuroglobin and cytoglobin, two heme-containing proteins, have been shown to increase cell survival in conditions of hypoxic and oxidative stress. Like other heme-containing proteins, neuroglobin (which is mainly expressed in neurons) is thought to play a role in controlling cellular oxygen levels. Studying expression of neuroglobin and cytoglobin in hypoxic and ischemic tissues has medical relevance, as the fetus experiences significantly lower oxygen tensions relative to the adult. We hypothesize that chronic hypoxia results in increased transcription of these globins, and that the fetal brain will have higher expression than the adult. Tissue samples were obtained from ewes and fetal sheep near the end (139±1 days) of gestation (147 days) at high (3800 m) or low (700 m) altitude. Samples were collected from the cerebral cortex, medial cerebellum, midbrain brainstem, thalamus, and neck muscle (non-nervous control). Gene expression was measured using real-time reverse-transcription PCR, and results were normalized against the expression of reference genes appropriate for the conditions studied. Observing increased expression of these globins in fetal versus adult sheep and in hypoxia versus normoxia would be consistent with the hypothesis that neuroglobin and cytoglobin play a role in regulating cellular responses to hypoxia.

76

 

Reconfiguring a Flight Cryostat for the Ground-Based TIME-Pilot Elly Shao Mentors: James Bock and Zak Staniszewski TIME-Pilot is a proposed imaging spectrometer that will observe the very first galaxies during the epoch of reionization. For my SURF, I worked on an existing flight cryostat that will be used to cool the TIME-Pilot detector array. This system, developed as the testbed for the space-based BLISS mission, must be modified to accommodate TIME-Pilot’s detectors. This summer, I tested the cryostat’s cooling performance and made some modifications allowing it to be used more easily. Synthetic Biological Circuit Design Implementing Protein Degradation in vitro Rohit Sharma Mentors: Richard Murray and Zachary Sun In-vitro systems shorten engineering time by eliminating the need to work with intact viable cells and facilitate rapid testing of genetic circuit modules e.g. toggle switches, oscillators. These synthetic modules are regulated by instantaneous concentration of the proteins that act as circuit activators and repressors. Although the switches may change input states, it takes a lot of time for the already formed activator/repressor and for the reporters to “degrade” via dilution due to cell division, and hence the circuits to actually switch their state. In cell free expression systems such dilutions can be mimicked by degrading the protein by employing unfoldases. We proposed the in vitro implementation of classical Stricker-Cookson-Bennett oscillator in which activators, repressors and reporters have an ssrA degradation tag and addition of purified ATPase Clpx enables rapid protein degradation emulating cell dilution. Firstly, we verified the oscillatory circuit in vivo and then switched to TX-TL platform. The optimal performance of the circuit was obtained by adjusting temperature and inducer levels. Introduction of Clpx brought down the concentration of the regulator proteins but periodic oscillations could not be observed. Further work in this area is required to obtain definitive oscillations and to establish rapid switching in genetic circuits. Characterization of Quantum Cascade Laser Wafers Aaron L. Sharpe Mentor: Ryan M. Briggs Quantum cascade lasers have emerged the most versatile and practical technology for generating coherent radiation at mid- to long-wave infrared wavelengths. One of the most important applications of quantum cascade lasers is laser absorption spectroscopy for monitoring of gas concentrations using molecular absorption lines throughout the infrared. The primary objective of this project is extensive characterization of quantum cascade laser structures grown epitaxially on wafer substrates, which are used by the semiconductor laser group at JPL to fabricate infrared lasers. The temperature dependence of laser emission was used to study the efficiency and performance of the wafers. Additionally, structures from different epitaxial growth runs were analyzed to determine process repeatability and the effectiveness of design modifications. The ultimate goal of this effort is to develop low power consumption quantum cascade lasers for robust environmental monitoring absorption spectrometers. A Search for Phototrophic Bacteria Capable of Using 1300 nm Light Amanda Shelton Mentor: Jared Leadbetter Anoxygenic phototrophs are a diverse group of bacteria that use light in their energy metabolism. Most have distinct light niches in the visible and near-IR. The lowest energy niche known is centered at ca. 1025 nm, targeted by bacteria containing Bacteriochlorophyll B (BchlB). But is that the far limit for utilizable light in biology? There yet may be potential niches at longer wavelengths, and novel photosystems to exploit them, as such photons in principle have sufficient energy to be harnessed by cells. To begin to examine if such microorganisms might exist, mineral salts media were inoculated with samples from terrestrial, aquatic, and marine environments, and incubated under illumination with narrow bandwidth 1300 nm LEDs. No phototrophic growth has yet been observed under illumination with such light. In positive control experiments, purple non-sulfur phototrophs were enriched from terrestrial, aquatic, and marine samples at 850 nm, which confirms that the medium is in principle sufficient to enrich phototrophs. But a lack of growth in positive control incubations at 1050 nm indicates that additional sampling locations should explored for reliably recovering known organisms with BchlB, e.g., before ruling out the existence of bacteria with novel light-harvesting features. Protocol Reliability Testing for Delay/Disruption-Tolerant Networking Emily Shih Mentor: Jordan L. Torgerson Reliable data transmission is necessary for efficient communication between earth and space. Current internet protocols are insufficient for deep space networking due to the inconsistency and time lag of earth-space communication links. Delay/Disruption-Tolerant Networking (DTN) is a new architecture that provides dependable

77

  delivery of data and has been selected for future human space flight missions. For now, it is still in testing. In this project, we observe the performance of DTN for varying models of data transmission. A network of five DTN nodes was implemented with the Interplanetary Overlay Network (ION) software and used to simulate communication between earth and space. Sample files were sent between nodes, and a network emulator was used to simulate the loss and delay that occurs in deep space links. Varying factors included the type and size of the files, the amounts of loss and delay, and the rate of transmission. The efficiency of transmission was measured by amount of data transmitted over time. The results suggest how DTN might behave when applied in real-life scenarios analogous to the simulations. The Change in Activity of GATA3 Caused by PU.1 Natalie Shih Mentors: Ellen Rothenberg and Marissa Morales Del Real The mammalian T lymphocytes precursors have the potential to either become T cells, which are involved in the adaptive immune system, or divert into myeloid cells, which are a part of the innate immune system. T-cell precursors are not committed into the T-cell fate until after the hematopoietic progenitors have migrated to the thymus. The overexpression of PU. 1 causes the T-cell precursors to divert into macrophage. However in the presence of Notch signaling the cells do not divert. The levels of the GATA3 protein are downregulated by the presence of PU.1 and loss of Notch signaling. Thus, we want to study how PU.1 changes the level of GATA3 and the localization of GATA3. By using Western blot and probing for GATA3, we found that GATA3 was being degraded in the cytoplasm. This shows that not only a change in localization was occurring but that protein levels decreased due to degradation. Characterization of SuperSpec: A Kinetic Inductance Detector for the Millimeter/Sub-Millimeter Band Corwin Shiu Mentors: Matt Bradford, Steve Hailey-Dunsheath, and Erik Shirokoff The millimeter and sub-millimeter wavelengths contain a wealth of astronomical information from high red-shift galaxies to the Cosmic Background Radiation. SuperSpec is a compact, on-chip spectrometer consisting of superconducting micro-resonators. Its small size, ease of fabrication using common lithography techniques, and scalable frequency multiplexing make it a promising technology to study very high red-shifted galaxies. Here, we report on a prototype device with 77 channels operating in the 200-300GHz band. Incident radiation on a metal feed horn propagates along a niobium microstrip. The radiation then passes through a filter bank made from halfwave niobium resonators. Each filter-bank channel is then coupled to a power detector consisting of a titanium nitride lumped element kinetic inductance detector (KID). We discuss on-going testing of these devices using a newly commission multi-channel readout system, including measurements of their spectral response and noise equivalent power (NEP). Computation of Heegaard Floer Correction Terms Laura Shou Mentor: Yi Ni The Heegaard Floer correction terms is an invariant for rational homology spheres, introduced in 2003 by Ozváth and Szabó as part of Heegaard Floer homology. We develop a computer program to compute the Heegaard Floer correction terms for two types of 3-manifolds: the double branched cover of an alternating link L in S^3 and a Seifert fibered rational homology sphere specified by data {e; (p_1,q_1),...,(p_r,q_r)}. We generate a quadratic form, which is then used to compute the correction terms as described by Ozváth and Szabó. Progress has also been made towards computing the correction terms of a manifold S^3_{p/q}(K) obtained by Dehn surgery with slope p/q on a knot K in S^3. The method would use grid diagrams and the combinatorial description of knot Floer ∞ homology in a paper by Manolescu, Ozsváth, and Sarkar to compute the homology of the chain complex CFK (S, K), and then a paper by Ni and Wu to compute the correction terms. Empirical Study of the Multi-Axial, Thermo-Mechanical Behavior of Nickel-Titanium (NiTi) and NickelTitanium-Hafnium (NiTiHf) Shape Memory Alloys (SMA) Dhwanil Shukla Mentors: Guruswami Ravichandran, Aaron Stebner, and Dipankar Ghosh Elements like Hf have been substituted for some Ti in NiTi alloy (most widely used SMA for low temperature applications) to increase the transformation temperatures for high-temperature applications forming NiTiHf Alloy. Limited studies have been performed to understand the microstructure and compressive stress-strain behaviour of NiTiHf alloys. However, their mechanics remain largely uncharacterized. Highly non-linear and anisotropic responses of SMA due to presence of phase transformation along with elastic and plastic deformation make it impossible to infer three-dimensional behaviour of the new alloys from uniaxial tests. Thus the project’s primary aim was to characterize the multi-axial response of these new NiTiHf alloy by experimentation. The experimentation included loading thin walled tube NiTiHf alloy sample in tension, torsion, compression, and combined tension-torsion and compression-torsion at different temperatures and measuring strains on sample 78

  surface using stereo digital image correlation (DIC) cameras thus creating data set that can be used for further study on the alloy. Further, the project will be extended to doing numerical study of the structural response of the new NiTiHf alloys applying the SMA constitutive model, which has been formulated to simulate all shape memory behaviours, which will be calibrated using literature and the experimental results. The Longitudinal Dependence of Solar Energetic Particle Release Times and Its Relationship to Coronal Mass Ejection Properties Emma R. Shupper Mentor: Christina Cohen Radiation from solar energetic particle events can cause significant damage to spacecraft subsystems and Earthbased technology. Implementation of protective measures requires formulation of reliable space weather forecast methodology. Thus, the acceleration and transport of particles must be understood. The relationship between solar particle release time and longitudinal separation of the spacecraft and source region is currently of interest. The main goal of this research is to determine any organizing factor between release time and properties associated with coronal mass ejections (CMEs) and flares. A compilation of events observed by the Advanced Composition Explorer (ACE) and the Solar Terrestrial Relations Observatories (STEREO) were selected from 2011 to 2013. Particle release times and path lengths were extrapolated using intensity onsets at multiple energies. Differences between release times and various solar time indicators, such as CME liftoff and flare start/maximum, were analyzed as functions of spacecraft longitudinal position relative to the solar sources. Our results demonstrate that time differences are organized by magnetic connection of the observing spacecraft with the event source. That is, larger time differences are observed by spacecraft west of the source at large longitudinal separations, while smaller time differences are observed by spacecraft at small longitudinal separations. Detection and Differentiation Among Four Different Bacterial Genera Using a Deep UV Raman Fluorescence Spectrometer Shakher Sijapati Mentors: Luther Beegle and Rohit Bhartia Differentiating among various genera of bacteria is a necessary process that may allow for a better means of detection and distinction of species. Currently, approaches to distinguish between different types of bacteria are limited to methods such as qPCR and microarray based identification. However, spectroscopic analysis of bacteria allows for a non-contact detection system that may allow for nondestructive spectral analysis of various bacteria. This study uses the Deep UV Raman Fluorescence Spectrometer (MOBIUS) to collect data on four different types of bacteria: Pseudomonas aeruginosa, Bacillus subtilis, Shewanella oneidensis, and Escherichia coli. The process to acquire the needed spectrum includes inoculating the cells in LB media followed by a re-inoculation into M9 media—a defined media. Once grown, each cell genera is taken and cleaned three times using phosphate buffer solution (PBS) and re-suspended in PBS. A portion of the cells are taken and streaked onto cleaned stainless steel coupons and dried for various amounts of time. After drying, each cell streak is looked at using on the MOBIUS instrument to acquire UV RAMAN and UV Fluorescence spectra. With this data, we hope to be able to distinguish between different genera based on their specific spectra. This new method will hopefully aid in efficient and effective uses for current identification methods. Photoelectrochemical Oxygen Evolution With Catalytic Co3O4 Nanoparticles on Earth-Abundant Transition Metal Oxide Photoanodes for Solar Water Splitting Timothy S. Sinclair Mentors: Harry B. Gray and Astrid M. Müller Harvesting energy directly from sunlight and converting it into storable fuels, such as making oxygen and hydrogen from water and sunlight, offers a clean and sustainable way to meet energy demand in the future. Water splitting devices that will provide a solution for the world’s energy crisis have to be low cost and can only be realized with highly active hydrogen and oxygen evolution catalysts, and efficient photocathodes and photoanodes, using earthabundant materials. Highly active nanoparticulate Co3O4 water oxidation catalysts, which were recently developed by Müller et al., were mechanically deposited to the surfaces of earth-abundant semiconducting transition metal oxide photoanodes, such as TiO2 and BiVO4, as the next step in creating a functional water splitting device. Multiple methods, such as spin-coating with varying conditions, spray pyrolysis, and electrodeposition, were used to prepare BiVO4 electrodes, which were characterized by optical spectroscopy, x-ray photoelectron spectroscopy, powder x-ray diffraction, atomic force microscopy, and photoelectrochemical cyclic voltammetry. Photocurrent generation was also assessed with three different oxygen evolution catalyst mass loadings. Light-driven water oxidation performance of the assembled electrodes was evaluated as a function of Bi:V ratio at the surface, catalyst mass loading, film thickness and morphology, and compared across the different preparations.

79

  Expanding the Scope of the Aldehyde Selective Wacker to Oxygen Functionalized Species Kacper Skakuj Mentors: Robert Grubbs and Zachary Wickens This project expanded the substrate scope of a novel aldehyde-selective Wacker oxidation. The aldehyde-selective Wacker system allows for the catalytic transformation of terminal alkenes to aldehydes (anti-Markovnikov addition) over the more common methyl ketones (Markovnikov addition) produced by the Tsuji-Wacker conditions. This study focused on probing the effects of steric profiles and protecting groups on compounds containing alcohols. Substrates with steric bulk, cyclic structures, tertiary and phenyl substituted alcohols were tested. Most substrates gave aldehyde yields comparable to the Tsuji-Wacker with reversed selectivity. Some mechanistic aspects of this oxidation were explored using two sets of diastereomers, giving support for oxygen chelation to the palladium metal and a chair transition structure. A synthesis of the drug Atomoxetine probed the retention of enantiomeric excess as well as the practicality of the method. Voting Rights in Peril: A Database of Racial Discrimination in Voting Laws (1957-2012) Matthew D. Smalley Mentor: J. Morgan Kousser In 1870, the ratification of the Fifteenth Amendment to the United States Constitution guaranteed, to all citizens, that no American would be denied the right to vote on the basis of race, color, or previous condition of servitude. Nearly 100 years later, the passage of the Voting Rights Act finally made this promise real. Unlike all previous Civil Rights Legislation, the Voting Rights Act required jurisdictions with a history of racial discrimination to pre-clear any and all election law changes with the Department of Justice. This past June, the Supreme Court of the United States declared unconstitutional the formula used to determine subjected jurisdictions, citing that the data used was too old to be accurate. In this decision, the Court urged Congress to pass a new formula based on more modern data. Admittedly, racially discriminatory voting laws have dramatically decreased as a result of the Voting Rights Act. By documenting all voting rights cases, thwarted attempts to pass discriminatory laws, and the circumstances surrounding such events, our database is able to reveal modern patterns of racial discrimination in voting laws. This database can be used to produce maps to show existence of discrimination, and potential formulas to prevent its return. Quality and Electrical Transport Properties of Room-Temperature Grown Graphene on Different Substrates Jun Ho Son Mentors: Nai-Chang Yeh and Chen-Chih Hsu Graphene, a honeycomb lattice of monolayer carbon atoms, has been an important research topic in the field of condensed matter physics for the last decade due to its unique physical properties and wide range of possible applications. However, there was no method to grow high-quality graphene on large area. Recently, Professor Yeh’s group developed a new method to grow large-area graphene at room temperature and with minimized defects. My goal of research is to assess the quality of room-temperature grown graphene by Raman spectroscopy and electrical transport measurements. By analyzing the D, G, and 2D peaks in Raman spectroscopy, the number of layers and relative amount of defects can be determined. Electron mobility and charge carrier density of graphene are measured in transport measurement. Through performing these analyses on multiple samples, the quality of graphene grown by the room-temperature method will be compared to that of graphene grown by the traditional method at 1000 C, and factors that affect the quality of graphene produced by the new method will be determined. Toward a Novel Integrative Algorithm for the Detection of Cancer Driving Genes Elijah Sorey Mentor: Hyunju Lee During the progression of cancer, certain genetic changes can lead to abnormal expression of the associated genes. Thus, today’s research aims to detect these affected genes as potential targets for therapy and further study. A popular approach to identifying such genes involves integrating the gene copy number and the gene expression data sets to determine which genes have their expression levels tied to the genetic changes observed in cancer. Unfortunately, past attempts to develop such an algorithm have had mixed results, often missing previously-known cancer genes or reporting many false positives. Here, we endeavor to produce a more effective algorithm by modifying the relatively successful breast cancer algorithm known as S2N. In the process, we extend the applicability of the algorithm to include data from glioblastoma multiforme, the most common type of brain cancer, and verify results from an algorithm based solely on the copy number data set. In addition, this study leads naturally to potential directions for future research.

80

  Inverse Opal Hydrogel Sensors for the Detection of Endospore Viability Tara Sowrirajan Mentors: Adrian Ponce and Heather McCaig Endospores are considered one of the toughest life forms, and their hardiness makes them ideal as biological indicators of sterility for applications such as healthcare, food safety, and spacecraft assembly. Uniquely, endospores contain high concentrations of dipicolinic acid (DPA), which is released upon endospore germination. This research develops a new method to detect endospore viability through the release of DPA. Created from a microstructured hydrogel, the sensor expresses a visible color change upon exposure to DPA, resulting in rapid, low-cost detection of live endospores independent of external equipment. Inverse opal hydrogel photonic crystal sensors were synthesized by creating a polymer colloid crystal with polystyrene beads (220 nm). The interstitial space was infiltrated with a pre-polymerization solution, including DPA, to allow for molecular imprinting, followed by photopolymerization to form the hydrogel. A solution of pyridine was used for template removal and DPA dissolution. The remaining inverse opal hydrogel sensor was exposed to solutions of varying pH and aqueous solutions of DPA. The color change will be characterized through UV-Vis spectroscopy to determine concentration detection limits of DPA and sensor response time. The minimum detectable number of endospores as measured by color change of the sensor will be determined. Analysis of Interseismic Deformation Using Geodetic Methods in the Ventura Basin Ollie Stephenson Mentors: Mark Simons and Romain Jolivet The Ventura basin, lying 60 km to the northwest of Los Angeles, is a densely populated sedimentary basin in the western Transverse Ranges province of Southern California. The basin is bounded by the Oak Ridge fault to the south and the San Cayetano fault to the north. Previous studies suggest that convergence across the basin is in the range 7-10mm/yr, highlighting the potential for a major earthquake and hence the need for a reassessment of the seismic hazard in the region. Here, we use a spherical wavelet based multiscale approach to estimate a continuous velocity field from GPS data, suggesting that convergence is accommodated by the Oak Ridge and San Cayetano faults. Velocity profiles across the basin indicate convergence rates of 6-10mm/yr, consistent with previous estimates. We now plan to use spatially continuous Interferometric Synthetic Aperture Radar images to derive a time series of the ground deformation, extract a high resolution velocity field and explore how each of the bounding faults accommodates the convergence across the Ventura basin. Solar Activity at Mars and Its Influence on Martian Atmospheric Temperature Jamesa L. Stokes Mentor: Armin Kleinbӧhl This project establishes a time series of particle activity at Mars from the last solar minimum into the present, 2006 to 2012 (corresponding to three Mars years), and catalogs solar particle events using the High Energy Neutron Detector (HEND) instrument aboard the Mars Odyssey orbiter. The neutron flux detected by HEND is largely caused by galactic cosmic rays within our solar system, a decrease of which is expected due to the increase in solar activity. Overall, the neutron flux in HEND decreased from a maximum of 0.25 cm-2sr-1s-1 to 0.10 cm-2sr-1s-1. Five solar particle events with fluxes above 1.0 cm-2sr-1s-1 are observed within this time period, with the largest event reaching a maximum of 135 cm-2sr-1s-1 . The particle activity at Mars is correlated with solar activity measured on Earth. The solar radio flux has increased from 80 to about 120 ·10-22 W m-2 Hz-1 from 2006 to 2012. In addition, the 10 MeV background proton flux measured by the Geostationary Operational Environmental Satellite (GOES) decreased from 0.0005 to 0.0001·108 cm-2sr-1day-1 , and it correlates well with the decrease in the background neutron flux measured by HEND. Profiles of Martian atmospheric temperature are obtained from the Mars Climate Sounder (MCS) instrument aboard the Mars Reconnaissance orbiter in order to find influences of solar activity on the temperature of the upper atmosphere of Mars. The inter-annual variability at 80-90 km in altitude in the fairly repeatable seasons of Martian northern spring and summer is investigated. In addition, the influence of individual solar particle events on the temperature within this altitude range is evaluated. During northern spring, there was a change of 0.961 K and 1.816 K, at day and night, respectively, between Mars year (MY) 29 and MY 30. During northern summer, there was a change of 1.556 K and 4.378 K, at day and night, respectively, between MY 28 and MY 29. During a solar particle event in January 2012, there was a change of 2.226 K and 1.438 K, at day and night, respectively. However, these temperature changes cannot be clearly associated with changes in particle activity. Expressing and Characterizing Novel Heme-Copper Oxidoreductase Proteins Ye Juliet Su Mentors: Woodward Fischer and James Hemp The sudden rise of oxygen 2.35 billion years ago was the most important environmental change that occurred in Earth’s history. Understanding the mechanism of this transformation is vital to determining how life on Earth evolved. The rise of oxygen has been attributed to the evolution of oxygenic photosynthesis in Cyanobacteria, however it has recently been discovered that molecular oxygen can also be produced by nitric oxide dismutation. It

81

  was predicted that the enzymes responsible for nitric oxide dismutation belong to the heme-copper oxidoreductase superfamily, a group of enzymes that also play central roles in aerobic respiration and denitrification. To characterize these previously unstudied membrane proteins, the genes for six nitric oxide dismutases were cloned into inducible plasmid vectors and transformed into E. coli C43 strains, which are optimized for the production of membrane proteins. These proteins will then be isolated and purified using a nickel affinity column that bind to the histidine tags located on the enzymes. Enzyme activity can then be assessed, and future experiments will investigate the protein structure and ultimately determine the role that nitric oxide dismutation played during the course of evolution. Transferable Force Fields for van der Waals Forces Yubo Su Mentors: William A. Goddard, III, and Jose L. Mendoza-Cortes Quantum mechanical calculations offers an accurate description of atom and molecules, however this approach cannot deal with large amount of atoms (>3,000 light atoms). Thus on this regime computational chemistry often rely on force fields which is a Newtonian approximation to the quantum mechanical results. In generating force field parameters there is always a tradeoff between number of parameters and accuracy, and these are parameterized for certain phenomenon or environments. More specifically, current general force fields have been tested in predicting adsorption in various porous frameworks such as Metal-Organic Frameworks and Covalent Organic Frameworks. In the present paper, we develop force field parameters for adsorption interactions by fitting parameters to energies generated via quantum mechanical calculations. We demonstrate that the force field parameters predict well both structure and energy of systems generated by quantum mechanical calculations. Furthermore, the data are validated against experimental adsorption data via Grand Canonical Monte Carlo simulations. The techniques set forth here can be used to generate and validate additional parameters to cover the entire periodic table and better predict adsorption in a wide variety of frameworks. Rapidly Initiating Ruthenium Olefin Metathesis Catalysts Alexandra H. Sullivan Mentors: Robert Grubbs and Vanessa Marx The synthesis of rapidly initiating olefin metathesis catalysts to be observed via variable temperature (VT) NMR spectroscopy for the confirmation of Z-selective ruthenium olefin metathesis catalysts’ intermediates structure is ongoing. Towards this end, 1 was found to be C-H activated in the presence of sodium pivalate. While metallacycles could not be formed of 3, the C-H activation of the mesityl group on 1 indicates that C-H activation on the adamantyl group in future rapidly initiating Z-selective catalysts should be facile. Efforts to prepare analogous complexes to 3, that are foreseen to initiate rapidly, are ongoing.

N

Cl Cl

1

N

Mes

THF/MeOH (1:1)

N

RT, 1 hr

Ru PCy3 BF4

PivO

N

Mes

N

60 C, 2 days

Ru PCy3 BF4

PivO

THF

Cl

N

Mes

Ru PCy3 BF4

Cl

2

3

Investigations Into the Activity of Ruthenium Olefin Metathesis Catalysts Coordinated With Alternative Six Membered Chelating NHC Ligands Benjamin A. Suslick Mentors: Robert H. Grubbs and Myles B. Herbert Olefin cross metathesis, a transition metal catalyzed process, involves the reaction of two terminal olefins to produce a new internal olefin and ethylene. This reaction, however, produces two different isomers of the final product: E and Z. Since selectivity is essential for any industrially used process, investigation into preparing stereoselective catalysts was performed. It has been recently shown that ruthenium metathesis catalysts containing fivemembered chelating N-heterocyclic carbene (NHC) ligands were able to catalyze highly Z-selective cross metathesis reactions. Current five membered chelating catalysts, however, are limited in their ability to form trisubstituted olefins and bulky olefins with high Z-selectivity. In an attempt to address these limitations, the syntheses of catalysts containing a six membered chelating NHC architecture were carried out. The steric properties of the NHC ligands were changed to favor a geometric conformation in the intermediate that correlates to the Z-isomer, with more open space around the catalyst’s active site to accommodate increased steric bulk. NHC ligands were synthesized using several different multistep routes either previously reported or derived for this study.

82

 

Supercritical CO2 Organic Extraction and Efficiency Analysis Stephanie M. Tan Mentors: Fang Zhong and Ying Lin Martian atmosphere is composed of 95% CO2 so it is convenient for the operation of a wet chemistry lab utilizing supercritical CO2 fluid extraction and analysis technology. It has been known that supercritical CO2 is capable of recovering organic materials through extraction. The parameters to achieve highest extraction efficiency, however, are not thoroughly identified and studied. In order to understand the extraction and analysis instrument, these parameters, such as pressure, temperature, flow rate and duration of extraction, need to be pinned down. After ensuring the operation of a commercial supercritical fluid extraction (SFE) system and renovating a homemade SFE system, we identified 500 um glass beads as our medium for organic materials. We coated the glass beads with bio-interest molecules, such as coronene and cholesterol and inserted them glass beads into the pressure vessel for superciritical CO2 extraction. The extraction was collected by dissolving in a particular solvent such as DMSO and acetone. We expect to analyze the extractant solution through capillary electrophoresis and mass spectrometry in order to find out the extraction efficiency. We anticipate that with varying temperature, pressure and flow rate, the extraction efficiency will be different and therefore we can determine optimal range for each parameter. Imaging the Akt Pathway via Synthesis of Akt Capture Agents With Fluorescent Dye and Cell Penetrating Peptide Grace Y. Tang Mentors: James R. Heath and Joseph Varghese Akt is a serine/threonine-specific protein kinase involved in important processes such as glucose metabolism, apoptosis, and cell proliferation. Recent studies have shown that the Akt pathway is targeted by amplification, mutation, and translocation far more than any other pathway in cancer patients, making it a key pathway to study for cancer diagnostics and therapeutics. Through the use of in situ click chemistry and epitope targeting, the Heath Group has successfully developed two Akt-specific capture agents which are capable of activating and inhibiting Akt. Via solid phase peptide synthesis, these capture agents were resynthesized with an additional fluorescent dye molecule and TAT cell penetrating peptide to be used for Akt imaging in living cells. A HIF-1α degradation signal was also added, catalyzing the ubiquitin-dependent proteolysis of the protein. ELISAs conducted on OVCAR3 cells treated with the capture agents demonstrate that we are able to activate, inhibit, and destroy Akt. Finding of a Synthetic Capture Agent for KRAS Through In-Situ Click Chemistry Kevin Tang Mentors: James R. Heath and Ryan K. Henning KRAS is RAS proteins that is involved in important tissue signaling in humans. As such, mutations of KRAS proteins are behind a large portion of cancers. In particular, 80% of pancreatic cancer involves a mutation in the KRAS protein. Traditional antibody methods of creating capture agents have shown to have low robustness, as well as difficulty of consistently creating the same antibody. This project utilizes a cyclic 5 amino acids long OBOC (one bead one peptide) library to screen against mutant and wildtype KRAS epitopes. In particular, we addressed the G12D mutation, the most prevalent of the oncogenic mutations. As such, amino acid chains of epitope of wildtype and mutant KRAS, each containing the G12D mutation point were synthesized. Analysis from MALDI-TOF mass spectrometry indicated that the epitopes were synthesized correctly. The synthesized library and the KRAS epitopes are now in the process of screening for targeted hits. This process will first involve an antiscreen, where beads binding to wildtype KRAS will be thrown out, followed by a full screen, where the final candidates for the capture agent will be selected. These final candidates will be tested for binding affinity, specificity, as well as stability. Using Observational Datasets to Test Theories of Tropical Circulation Nicholas Tang Mentors: Andy Ingersoll and Da Yang The Madden-Julian Oscillation (MJO) is a large-scale, eastward propagating intraseasonal variability in the tropics. Nearly 50 years after its discovery, a successful MJO theory has not been found. Yang and Ingersoll (2013) use a shallow water model to simulate the MJO. They propose that the MJO is the result of an interference pattern between eastward (EIG) and westward (WIG) propagating inertia gravity waves. To test this theory, I have been comparing the model results to results derived from observational datasets. These datasets include NOAA oncedaily mean outgoing longwave radiation (OLR) and 4x-daily zonal wind. Results from the OLR data are used as a reference to ensure that the entire method is sound, while results from the zonal wind data are compared quantitatively to the model output. Although testing with the OLR data is nearing completion, further research is needed to determine whether or not the observational datasets produce a strong positive correlation between MJO and IG wave signals.

83

 

A Look at the Hard X-Ray Spectra of LMC X-4 by NuSTAR Rebecca Tang Mentors: Fiona Harrison, Felix Fuerst, and Eric Bellm The X-ray emitting pulsar, LMC X-4, is a High Mass X-ray Binary (HMXB) consisting of a massive star orbiting an accreting neutron star. During accretion, matter falls from the optical companion to the neutron star, producing X-rays of intensity proportional to the amount of accreted matter. By analyzing the emitted spectrum, one can study the accretion process. Unlike many HMXB neutron stars, cyclotron absorption lines have not been found for LMC X-4 by previous observations, an interesting point of study. NuSTAR is sensitive from 3-79keV, making it ideal for the search for cyclotron lines. Spectra and light curves were generated from NuSTAR data and modeled using standard astrophysical software. Timing analysis confirms a pulse period of approximately 13.4992s. The phaseaveraged spectrum can be described by a power law with a high-energy cutoff at ~20.4keV, photoelectric absorption, and two iron emission lines. This represents a model that describes the data, but does not conclusively provide evidence for a physical cyclotron line. We will fit phase-resolved spectra to search for evidence of such absorption lines. Verification and Functional Studies of Genes Expressed in the AWC Neurons in Caenorhabditis elegans Zuofu Tang Mentors: Paul Sternberg and Yen-Ping Hsueh Caenorhabditis elegans senses different attractive odours with a single pair of olfactory neurons, AWC. There are thousands of genes expressed in these AWC neurons, which are required for chemotaxis to volatile odours. Here I try to identify genes that are required for fungal sensing in the AWC neurons. Fungi produce volatile organic compounds that are attractive to C.elegans. By running chemotaxis assays, I show that only certain compounds are attractive to C.elegans. The attraction is displayed by different numbers of nematodes concentrated on one side of chemotaxis plates where one microliter of a specific compound is placed on. The more nematodes on that side of the plate, the more attraction the compound shows. I am also testing mutants of various G-protein-coupled receptors to see whether they have different reactions comparing to the wild type. Information I gather here expresses their behavioral responses to different chemicals and through which I can decide which genes in the AWC neuron are responsible for these behaviors. Von Neumann Entropy and Thermodynamic Semirings in Renormalization Algebras Nicolas Tedeschi Mentor: Matilde Marcolli The Connes-Consani approach to geometry over the field with one element led to the development of thermodynamic semirings, which are MaxPlus or MinPlus semirings deformed according to an entropy functional such as the Shannon entropy. It’s natural to consider whether the formalism of thermodynamic semirings can be extended to quantum information measures, and what properties would be inherited from the classical case given the involvement of quantum phenomena such as entanglement. The von Neumann entropy is the quantum analogue of the Shannon entropy from classical information theory, and so we use it as the natural candidate for exploring these properties. MaxPlus semirings have also been suggested as a potential object of consideration in the study of renormalization and computation. Hence we also examine the ways thermodynamic semirings can be used to construct an algebraic structure replacing the Rota-Baxter algebra in the Connes-Kreimer formulation of renormalization, “renormalizing” computation time as opposed to Feynman amplitudes. A Computational MD Study of an Engineered Sensor Gated siRNAs Switched on by Disease Specific mRNA Michael S. Teng Mentors: William A. Goddard and Si-Ping Han As DNA and RNA nanotechnology progresses, computational studies are being used to complement the experimental results seen in trials. As part of an ongoing effort to examine the effectiveness of engineered signal activated siRNAs (saRNA) in treating diseases, a molecular dynamics (MD) simulation of various artificial saRNAs is carried out to determine structural and conformational changes of the constructs in solution. With the main question of stability in mind, MD is done on variations of a 25 base pair siRNA domain (as two 12/13 base pair duplexes) linked by either a sensor duplex or a single stranded RNA connection. Data to be collected. Quantitative Investigation of Ice Crystal Growth Rates and Morphology Between -10C and -20C Sarah Thomas Mentor: Kenneth Libbrecht Ice crystals provide an excellent case study for crystal growth dynamics from the vapor phase. A dual diffusion chamber is used to grow and observe ice crystal growth over a wide range of temperatures and supersaturations. In the first chamber, a strong electric field is applied to a metal wire to grow ice needles which form due to an

84

  electrical growth instability. These electrically enhanced needles are transferred to a second chamber where the temperature and supersaturation can be set to the desired conditions, and crystal growth occurs at the tip of the ice needle. A series of high magnification images is later analysed to measure the growth rate of the crystal and determine its morphology. Detailed measurements over the temperature range -10C to -20C and a wide supersaturation range provide a more quantitative morphology diagram, with growth rates as a function of these two parameters as well as morphological transition points. These data give insight into the instabilities governing ice crystal growth and appear to support the theory of Structure Dependent Attachment Kinetics. It is hoped that parameters extracted from these data will provide physical inputs and understanding to aid progress towards a computer model describing 3D crystal growth. Adaptive Wavelets Methods for Predicting the Mechanical Behavior of Polycrystals Adam Thor Thorgeirsson Mentors: Kaushik Bhattacharya and Gal Shmuel In recent years, new development in materials characterization techniques led to a huge amount of data on the microstructure of polycrystals. Simultaneously, improvement in the computational capabilities has enabled accurate full-field simulations for the micro mechanical fields developing in the polycrystalline aggregate. These show that in elastic phenomena including plasticity and phase transformation, localized bands of deformation percolate in complex way across various grains. This work is motivated by a desire to find an efficient representation for such fields. In particular, we focus on adaptive wavelets-based analysis, motivated by the similarity between the localized nature of the mechanical fields and the basis functions in wavelets space. We will present the main characteristics of these functions, and demonstrate their efficiency in representing exemplary data. Study of the Impact of Stellar Multiplicity on Planet Occurrence and Properties Rachel Thorp Mentors: Jean-Michel Desert and John Johnson Stellar multiplicity in a exoplanet-host star system is likely to affect planetary formation and evolution. Direct imaging of a host star environment using Adaptive-Optics (AO) techniques is a focus of this project to assess stellar multiplicity. The new approach of this program is to use a robotic instrument which allows observations of a larger number of objects—in the case of this project, 200 targets. In order to better understand these processes, the data collection of 200 exoplanet host stars is being analyzed for possible stellar companions. For each object, a search for the presence of stellar companions is performed using several techniques. The detection limits of each technique are then transformed into physical parameters, for which, any possible stellar companion meeting those standards will be taken to be undetectable. Following this, the questions of density number and spectral types of stellar companions are addressed. To properly derive statistics from our sample, we also compare our findings to a reference sample that contains no known exoplanet host stars. Development and Testing of Drive-Tube Penetrometer for Reduced Gravity Sediment Experiments Leslie Timms Mentor: Brian Trease A well-developed understanding of sediment-mechanism interaction is crucial to any sample return mission. This project focused on prototyping a drive-tube testing apparatus for sediment-mechanism interaction studies aboard Reduced Gravity Aircraft to study this interaction in low- and microgravity environments. The prototype consists of an actuated drive-tube in line with a digital force gauge. In addition to developing an automated prototype, this project took data at 1g using various sands and simulants. This 1g testing serves not only to establish a baseline of data on this testing apparatus, but also to verify that the automated data collection system collects reliable and repeatable data. Data was taken with numerous independent variables including sediment type, testing speed, and drive-tube dimensions. Various tests were compared for reliability and test circumstances were compared to understand how changing different variables changed the recorded data. Generation and Reaction of ortho-Quinone Methides Under Mild Basic Conditions Christopher Tonge Mentor: Brian Stoltz The focus of my proposed research through the SURF program this summer is the generation of ortho-quinone methides under mild basic conditions. Ortho-quinone methides (o-QMs) are a class of reactive molecules that are prone to dimerization and are often unisolable. They are therefore generated and reacted—typically by conjugate addition or Diels-Alder reactions—in a single operation, and have been applied in a variety of total syntheses. Various carbonate substrates are used to demonstrate the scope of the reactivity of mild bases (eg. alcohols, imines, mild carbon nucleophiles, and amines).

85

  Machine Learning for Earthquake Monitoring  Shivam Mani Tripathi and Deepti Chopra   Mentors: Julian Bunn and K. Mani Chandy  Earthquake monitoring systems consist of sensor networks that detect ground motion and fuse data from multiple sensors to warn about impending shaking and to provide continuing situation awareness as shaking from the earthquake continues. This project explored algorithms based on machine learning for detecting earthquakes by analyzing data streams generated by the inexpensive sensors used in the Community Seismic Network. The project analyzed frequency-domain characteristics of acceleration reported by the sensors and applied Support Vector Machine (SVM) tools to classify acceleration patterns in one of two categories: (1) quiet period, i.e., no shaking due to earthquakes currently, and (2) shaking due to a current earthquake. The project explored the possibility of using neural networks for classifying time-domain patterns into the same categories. The project reports on several experiments for classifying data streams into the two categories and the results are promising.  Air-Righting Behavior in Mice Huey-Ru (Debra) Tsai Mentors: Markus Meister and Melis Yilmaz There remain many questions surrounding the mechanisms that link neural processing of visual information in the retina to behavior. The first step to better understanding these mechanisms is to expand the currently limited literature on visually guided behavior. Because of its genetic accessibility, the mouse is a popular experimental animal in visual neuroscience. Here we explore the role of vision in air-righting behavior during a fall. We show evidence that air-righting might involve using visual cues. However, this may be dependent on the age of the animal. Frictional Response of Molecularly Thin Liquid Films Charles Tschirhart Mentor: Sandra Troian The behavior of a fluid is strongly dependent on the surface-to-volume ratio of the fluid sample. This gives rise to many interesting phenomena in fluid mechanics; the properties of a fluid can vary wildly as the size scale over which the fluid is investigated is varied. Classical fluid mechanics has been used to successfully describe macroscopic samples of fluid for centuries, but it was not until recently that scientists gained access to experimental equipment capable of investigating extremely small samples of fluid. Over the past few decades, research into the exciting and bizarre phenomena exhibited by extremely thin films of fluid has blossomed. It was discovered relatively recently that fluid viscosity, previously thought to be a constant property of a fluid for a given temperature and pressure, changes unpredictably at such small size regimes. This effect is thought to be a result of the influence of surface forces on the behavior of the molecules within the fluid, and it is in some sense a product of the breakdown of the continuum hypothesis at such small scales. This phenomenon is extremely difficult to measure using conventional experimental methods in hydrodynamics; methods exist for determining fluid viscosities, but most are either unreliable or unsuitable for very small quantities of fluid. In our research, we apply shear stresses to extremely thin films of fluid and examine the resulting changes in the film profiles in order to determine the effective viscosities of the fluids at molecular scales. We are interested in particular in the effects the air-liquid and liquid-surface boundary layers have on the effective viscosity of the fluid. We hope to explain these effects based on the chemical structure of the molecules composing the fluid. The experimental setup we have constructed for this work includes a dip-coater for laying down micro- and nanoscale fluid films on silicon substrates, a Michelson-interferometry scanning system for investigating the thickness profiles of our microscale films, and an ellipsometer for determining thicknesses of nanoscale films. Previous research has found that thinfilm viscosities differ strongly from bulk viscosities of the same fluid, and as a result we expect to encounter interesting behaviors in the dynamics of thin films at such small scales. Measurements of nonuniformities in the measured viscosities of small quantities of fluids would have major effects on many fields, including (but not limited to) microfluidics, tribology, lubrication theory, and cellular transport in biological systems. Depth Perception in Mice Christine Tseng Mentor: Doris Tsao The goal of this project is to determine whether mice have the ability to perceive depth, and if they can, to use mice as a model to solve the binding problem. Five male mice are fixed with a headplate on a treadmill and trained to recognize a target shape among other shapes by operant conditioning with sugar water. The mice are then presented with the same shapes partially occluded by blobs. The mice have not been tested with the occluded shapes at this time; however, if they do recognize the target shape underneath the blobs, then it suggests that mice can perceive depth and are appropriate models for solving the binding problem.

86

  Transmitting Information and Receiving Automated Instructions Over SMS Jean-Alexandre Turban Mentors: Michael Hoffmann and Clément Cid Caltech’s sun powered toilet prototype is in its testing phase before launching in India this December. The transmission of data and commands via SMS is a desirable service in the science field as well as to us due to the versatility and control it provides. The proposed system lets us receive automated hourly information updates from our prototype container. This process involves receiving SMS (Short Message Service) via a GSM (Global System for Mobile) modem connected to a laptop which then handles certain actions such as opening valves, or shutting down. The incoming message is processed via scripts and triggers certain actions in the container. Furthermore, the modem serves as a means to send automated system updates involving power and maintenance to a central computer back in Caltech's lab. One of the goals being versatility, the system is carefully designed for use across the world – the modem can access all four telecommunication frequencies and the system has been developed with open source software and common scripting languages. The Generation of Maps for the Situation Awareness Framework Nitesh Udhani Mentors: Julian Bunn and K. Mani Chandy This project developed mapping tools for displaying information about geospatial events such as earthquakes and hurricanes. The maps are dynamic showing how the intensity of geospatial events, such as the magnitude of the acceleration in an earthquake, varies over the region and over time. The project also developed an intuitive way of displaying acceleration magnitudes across a region by using an acceleration heat map. The project used several tools including AngularJS and JavaScript for web application user interfaces, and GIT for version control. Using Razor Analysis of Top Quark Pair Production in SUSY Search Gautam Upadhya Mentor: Maria Spiropulu We present a study of two simplified Supersymmetric (SUSY) models, corresponding to pair production of stopsquarks, and show how they can be distinguished from a Standard Model background process using the razor kinematic variables. With data collected from the Large Hadron Collider (LHC) at 8 TeV using the Compact Muon Solenoid (CMS) detector in 2012, we expect to be sensitive to events containing a stop-squark, the superpartner of the top quark. Before looking at LHC data in this search, we first conduct the same analysis on MadGraph generated Monte Carlo samples. Stop-pair production is difficult to distinguish from top quark pair production, since both signal and background processes result in two b-quarks and two leptons as the only visible final state particles. The missing transverse energy (MET) is carried off by squarks and neutrinos for stop-pair production, and by only neutrinos for top-pair production. We find that the razor variables are optimal for discriminating background from signal, as well as approximating the mass difference between the stop-squark and the LSP in signal events. We explore other potentially discriminating variables, such as lepton angles and the correlation between reconstructed and observed MET, and characterize the distribution shape for some razor variables. Sex Myoblast Migration in C. elegans Anand Upadhyaya Mentors: Paul Sternberg and Mihoko Kato Cell migration is an important mechanism in both development and diseases such as metastatic cancer. I have curated a gene list based on genes that are upregulated both in human metastasis as well as the migration of the male linker cell in C. elegans. I have been scoring the migration of another cell, the C. elegans sex myoblast which migrates to the center of the hermaphrodite gonad, in order to identify genes core to cell migration in general. Using an RNAi screen on the curated gene list, I have been scoring the sex myoblast’s migration by its position relative to the anchor cell in the gonad. Through 49 genes, however, I have not observed any defect in migration; all sex myoblasts migrate to the correct position on top of the anchor cell. I hypothesize that the migration of the sex myoblast is controlled by different genes than that of the linker cell. As such I have compiled an an alternate list of muscle-related genes, which I will screen. Inferring Thermal and Mechanical Properties of Celestial Bodies Regolith Using (Simple) Low-Tech Tools Mariia Usoltceva Mentors: José Andrade and Alex X. Jerves The project aims to develop and test new tools and algorithms for in-situ characterizations of thermal and mechanical properties of regolith on celestial bodies. The new tools and algorithms will be used onboard a landing vehicle to assess properties of regolith including friction angle, cohesion, dilatancy, thermal conductivity, and specific heat. The primary objective will be to obtain as many engineering and scientific properties with simple lowtech tools. The apparatus built on basis of these tools and related post-processing algorithms will provide clues

87

  about the nature of fundamental morphological processes on celestial bodies by providing crucial mechanical properties. The goal of this project has also a significant impact on mission cost and success, where minimizing energy, eliminating high-tech equipment failure, and optimizing functionality and data extraction are the key issues to tackle and overcome. Ice Target Thermal Test Bed for the Compositional Infrared Imaging Spectrometer (CIRIS) Preethi Vaidyanathan Mentor: Daniel Berisford CIRIS is a spectrometer designed for use on a future Europa orbiter, with possible applications for missions to other worlds. It has a compact and rugged design compared to traditional interferometers. As part of the technology development and TRL progression effort, this device must undergo thermal-vacuum environmental testing. This involves analyzing icy targets under cryogenic vacuum conditions with the spectrometer. The test bed will be used to validate the performance of the CIRIS under Europa-like and Mars-like conditions. A dewar will be filled with liquid nitrogen using a fluid feed through, and placed inside the vacuum chamber. A cold plate mounted to the dewar will freeze/ contain sample targets. An external blackbody light source will illuminate the targets via a mirror/ window assembly, and a second mirror will direct the reflected light into the spectrometer. The test bed is compatible with both flat and concave mirrors in order to vary the viewing area of the sample seen by the spectrometer, and can be moved in six degrees of freedom for optical alignment. The versatility of the test bed allows testing of many different optical configurations and target materials, and will be used to advance the development of the CIRIS instrument. Analysis and Validation of MISR Cloud Height Data Products Alexandre Van Anderlecht Mentor: Michael Garay The Multi-angle Imaging SpectroRadiometer (MISR) is an instrument that has been flying aboard NASA's Terra Satellite for over 13 years. With nine cameras, MISR retrieves height and motion vectors of clouds using stereoscopic methods. To validate MISR retrievals for the recently delivered TC_CLOUD product, cloud top heights generated from MISR have been compared to ground-based radar and lidar measurements. MISR has an estimated accuracy of 560 m based on a single pixel mismatch and the geometry of the cameras used in retrievals. The MISR TC_CLOUD product is expected to have improved vertical precision due to the inclusion of sub-pixel matching in the retrieval algorithm. Cloud top heights were compared to MISR from the Atmospheric Radiation Measurement (ARM) sites of the Southern Great Plains (SGP), Tropical Western Pacific (TWP), and North Slope of Alaska (NSA). The TWP consists of three separate facilities on Los Negros Island in Manus, on Nauru Island, and in Darwin, Australia. These sites provide years of retrievals where MISR data is also available that can be used for comparing heights. The Top Height of Hydrometeor Layer from Composite (MMCR/Clothiaux et al. MPL) Algorithms from the ARSCLBND1CLOTH products available on the ARM Climate Research facility website were used for the intercomparison. The ARM Active Remote Sensing of Clouds (ARSCL) value-added product used in this study consists of data merged from millimeter-wave cloud radar (MMCR), ceilometer, and micropulse lidar (MPL), providing data with a 45 m vertical height interval grid. MISR and ARM heights were collocated temporally and spatially, using the latitude and longitude of ARM sites and a time window of 15 minutes for the MISR crossover of ARM sites. Gamma Ray Burst Progenitors: A Study of Host Metallicities and Magnetar-Driven Light Curves David Vartanyan Mentor: Palli Jakobsson Long gamma ray bursts (LGRBs) are highly energetic explosions associated with Type Ibc core-collapse supernovae. LGRB light curves (LCs) consist of prompt gamma ray emission followed by X-ray afterglow that may display flares. During prompt emission, gamma rays are generated as a driven relativistic jet undergoes internal shocks or magnetic dissipation. Subsequent external shocks with the interstellar medium produce an X-ray and optical afterglow. The currently accepted theories to explain LGRB central engine activity are the collapsar model and the magnetar model. The former relies on a star collapsing into black hole with rapid accretion driving a relativistic bipolar jet. In the magnetar model, the bipolar jet is formed by a neutrino wind driven by the rapidly rotating magnetar's dipole radiation. Measurements of progenitor metallicity and mass help distinguish between these two models. Typically, lower metallicity corresponds to both increased LGRB occurrence and outburst energy. By analyzing LGRB X-ray LCs from Swift, we consider whether this claim implies a correlation between metallicity and the average photon hardness and the flare count of the outburst. Furthermore, we examine noncanonical light curves consisting of an extended internal plateau followed by steep decay as possible indicators of magnetar central engines. These studies allow us to discriminate between the two proposed central engine models and further constrain the highly-contended neutron star equation of state.

88

  Archival Search for Fast Radio Bursts Adrian Vatchinsky Mentors: Sarah Burke-Spolaor and Joseph Lazio We present a search for fast radio bursts (FRBs) in legacy VLA Archival data from June 2003 at 330MHz, motivated by the recently published 2013 Thornton et al. paper on the discovery of four millisecond long FRBs in archival Parkes Observatory data. Our two candidates do not exhibit any periodic behavior and are uncorrelated with known nearby radio emitting sources. The candidates are also located within a few arcsecond of cataloged galaxies. In both cases, we have split the data in frequency and confirmed that the candidates are visible in both frequency sub-bands. Furthermore, we have used a threshold of 6.5 standard deviations indicating these candidates are unlikely to be background noise or instrumental error. If these two candidates prove to be FRBs they can serves as evidence that FRBs are not localized or somehow related to the Parkes observatory from which all FRB have been reported so far and should motivate further studies of archival data to better understand the distribution of these events. This research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration. Investigation of the Viscoelastic Properties of Microbubbles Using a Pressure Chamber Malvika Verma Mentors: Eleanor Stride and Paul Rademeyer Surfactant stabilized microbubbles are well established as highly effective contrast agents for ultrasound imaging. The properties of the surface coating affect their behavior, but these are poorly understood, and this presents a barrier to fully realizing the potential of microbubbles in applications such as targeted drug delivery and gene therapy. The purpose of this project is to observe the behavior of bubbles with different coatings and the conditions under which coating material is lost due to compression. The microbubbles were placed in a pressure chamber and imaged with a combination of brightfield and fluorescence microscopy in order to characterize the mechanical properties of the coating. Interesting phenomena, such as the buckling and wrinkling of the bubble surface, were observed. Additionally, the behavior of the bubble varied with different compositions of the coating as expected. Large-Scale Informatics Approach to Understanding Decision-Making Computations in the Human Brain Ratnalekha V. Viswanadham Mentors: Antonio Rangel and John Clithero A major challenge of cognitive neuroscience research is to synthesize the results of many studies in order to understand the underlying structure that relates brain function to mental processes. Meta-analysis affords the opportunity to aggregate studies of similar nature in order to test common hypotheses that require sets of data larger than those of individual studies. Our goal is to apply meta-analytic methods to large collections of functional magnetic resonance imaging (fMRI) studies to identify common and distinct mechanisms for perceptual and economic decision-making in the human brain. We first categorized the types of decision-making studies into value-based decision-making studies and perceptual-based decision-making studies. We applied the Activation Likelihood Estimate statistic to the voxels collected from the studies to identify areas of the brain in these decisionmaking processes. We know that areas of the brain such as the ventromedial prefrontal cortex (vmPFC) are involved with value-based decision-making. Further work would include constant addition of data from published studies and implementation of data visualization techniques for understanding the relationship of fMRI decisionmaking studies to mechanisms in the brain involved in decision-making. Photophysical Characterization of Rose Bengal Adsorbates Matthew Voss Mentors: Enrique San Roman and Mitchio Okumura Upon excitation by light, common organic dyes lead to a singlet excited state. From this state, intersystem crossing may produce a long-lived triplet state. This long-lived triplet excited state, in the presence of molecular oxygen, can lead to a reactive species named singlet molecular oxygen by a process called photosensitization. Singlet molecular oxygen has both pharmaceutical and environmental relevance because it constitutes a central reactant in photodynamic cancer therapy and in the inactivation of microorganisms. Singlet molecular oxygen is highly reactive, and can significantly alter cellular biochemistry, making it an effective tool for incapacitating both cancerous and microbial cells. To further the development of molecular oxygen solid state photosensitizers, Rose Bengal dye was adsorption-deposited from solvent onto a solid particulate supporting material. Photophysical characterization was performed by reflectance and fluorescence spectroscopy, light induced optoacoustic spectroscopy (LIOAS), and diffuse reflectance laser flash photolysis (DRFLP). The effect of molecular aggregates, which build up at high concentrations of the dye, on the formation of the triplet state is analyzed.

89

  Combating Tropical Disease Through Chromosomal Rearrangement in Insect Vectors James M. Wagstaff Mentors: Bruce Hay and Anna Basalova Buchman Insect-borne diseases are a major health concern in many parts of the world. Classical control strategies such as releases of sterile males, distribution of nets and medication, and insecticides, demand large and continuous capital investments. A superior solution would be a one-time release of insects carrying transgenes that inhibit transmission of disease in such a way that transgenic insects replace the wild population. One mechanism by which disease resistance genes can be spread is engineered underdominance, whereby heterozygotes have lower fitness than either homozygote; the only stable situation is then fixation of one or other allele; an extreme example of this is a translocation where a ‘heterozygote’ will almost certainly have zero fitness. Insects carrying engineered translocations will replace localised wild populations if released at population frequency greater than 50% (assuming no fitness cost and low migration), and likewise can easily be removed from small populations by release of wild-type insects at frequencies greater than 50% (a key advantage over other schemes). This project seeks to generate specific translocations with closely linked genes of interest in D. Melanogaster and demonstrate threshold dependent population replacement in a laboratory setting. Testing Calibration for the Compositional Infrared Imaging Spectrometer (CIRIS) Alex Wallach Mentor: Daniel Berisford The CIRIS is currently under development to study the surface composition of Europa and other worlds within the solar system. As part of a thermal-vacuum testing effort for the device, an optical calibration device was designed for integration with the test setup. The calibration source is designed to emit blackbody radiation at temperatures of 150-350K via interchangeable apertures in order to provide spectral and radiometric calibration capability. The device is able to calibrate the spectrometer before measurement and then move away during measurement of other targets. To maintain an emissivity as close as possible to that of an ideal blackbody, the calibration source was designed to be a conical cavity coated with highly emissive material within a solid cylinder. 80-20 material provides rails for vertical and horizontal adjustment, which positions the calibration source in and out of the field of view of the spectrometer. The cavity is made from aluminum with heaters attached to ensure isothermal conditions. Ceramic standoffs connecting the calibration source provide a long thermal path allowing it to be at high temperatures while the spectrometer and other hardware are cryogenic. This calibration device is to be constructed and implemented for future testing of the CIRIS. Razor Variables and Applications on Searches for New Physics at the LHC: SUSY Scenarios Ann Wang Mentors: Maria Spiropulu and Javier Duarte The high energy collisions at the Large Hadron Collider (LHC) allow us to test supersymmetric (SUSY) extensions in the Standard Model (SM), which are important to explaining phenomena such as dark matter. Supersymmetric searches can be performed using razor kinematic variables, which help distinguish SUSY signal events from SM background. Specifically, we will examine stop pair production in T2bw and T2tt Monte Carlo generated signal, which are topologically interesting due to their similarity to the ttbar background. We will develop a razor variable analysis as well as various fit models for distinguishing variables to help set new limits for T2bw and T2tt events. Discovering the Enzyme(s) That Catalyze the Reduction of Phenazine-1-Carboxylic Acid (PCA) in Pseudomonas aeruginosa Ben Wang Mentors: Dianne Newman and Nathaniel Glasser Pseudomonas aeruginosa is a Gram-negative opportunistic pathogen and is most well-known for causing lethal lung infections in persons with cystic fibrosis (CF). These bacteria synthesize redox-active compounds called phenazines that not only serve as virulent factors but are also directly enable the survival of the producing organism. Phenazines are often found in a reduced state; however the enzyme that catalyzes the reduction of phenazines is currently unknown. This project will identify the enzyme(s) that catalyze the reduction of Phenazine-1-carboxylic acid (PCA). So far, experiments have shown that P. aeruginosa lysate reduces PCA in the presence of NADH, α-ketoglutarate, and pyruvate. Dihydrolipoamide dehydrogenase (DLDH) is found in both the α-ketoglutarate dehydrogenase and pyruvate dehydrogenase complexes and can also receive electrons from NAD/NADH, which it can then hypothetically donate to PCA. Thus, the current hypothesis is that DLDH is the enzyme that catalyzes the reduction of PCA. The hypothesis will be further tested by purifying DLDH and generating mutants deficient in its production. As phenazines are important virulence factors for P. aeruginosa, understanding the reduction pathway of phenazines can provide a novel avenue for treatments against the pathogen in chronic infections.

90

  Using Multispectral Information for Source Detection and Measurement Daniel Wang Mentor: Peter Capak Current object detection algorithms for astronomical imaging data examine only single wavelength images even though multi-spectral images are usually available. This single waveband method does not optimally detect astronomical sources because objects with very different spectral properties can be overlapping or not present in the detection waveband. We present a software tool that improves upon this process by using the multispectral information for detection, photometric measurements, and deblending of overlapping sources. Given a multispectral data set, this program outputs flux measurements and object detections using information from all provided wavelengths to improve accuracy. We base our algorithm on model fitting, using Gaussian profiles to represent the galaxies in each image. Then, using this model, we derive a measurement of flux, as well as color information for each object. We present test cases where this algorithm performs well, as well as cases where more improvement is necessary. Finally, we propose future development and improvements to the software, such as replacing the Gaussian profile with a more advanced model. Demand Response Capacity of Thermally Controlled Loads Esha Wang Mentors: Steven Low, Cole Hershkowitz, and Ka Suen The overarching objective of this research is to explore the potential of demand response capacity and determine whether or not this capacity will be enough to regulate supply and demand on the energy grid. Studying the flexibility of thermally controlled loads (TCL’s) such as refrigerators and AC’s proves useful because TCL’s have the capability of storing a variable amount of thermal energy, and thus can influence the energy grid while still maintaining basic operability. Furthermore, California ISO and PJM publishes data on pricing signals that are sent to generators and other industrial loads that help monitor the balance between supply and demand of energy. By comparing the energy usage of TCL’s with the frequency regulation data provided by these two companies, we invented methods to more effectively control supply and demand on the overall energy grid. Using demand response techniques, our research has confirmed that it is possible for generators to save a significant amount of money every year, and at the same time to advocate green energy technology for society. Development and In Vitro Characterization of an Angiogenesis-Promoting System for Subcutaneously Implanted Glucose Sensor Leads Kening Connie Wang Mentors: William M. Reichert and Varadraj N. Vernekar Diabetes mellitus (DM) is a disease that affects over 25 million Americans today, about 8 percent of the population. An indwelling, closed-loop insulin delivery device that monitors blood glucose levels and delivers insulin accordingly would represent a significant improvement in quality of life for these individuals; however, reliable long-term glucose sensing has proven to be a major roadblock. Currently available glucose sensors survive less than a week before poor vascularization around the sensor lead results in their failure. The Reichert group has previously developed a porous, dexamethasone-eluting polyurethane coating for the sensor lead that aims to inhibit scar tissue formation around the lead, but no work has thus far been done to promote new blood vessel formation. In this project, we developed a method to incorporate the elution of ginsenoside Rg1 (an angiogenic drug) and immobilized RGD peptide (an adhesion peptide) into the existing sleeve design with the goal of promoting blood vessel formation and retention. The design modifications were characterized using scanning electron microscopy (SEM), confocal microscopy and HPLC. Additionally, the angiogenic ability of the sleeve design was assessed using a cell-based tubulogenesis assay. Identification of the Genes Involved in the Maintenance of Stem Cell Populations in Arabidopsis thaliana Shoot Apical Meristem Lawrence Wang Mentors: Elliot Meyerowitz and Yun Zhou Continuous organogenesis in plants is dependent on the maintenance of stem cell populations. In Arabidopsis thaliana, stem cell populations in the shoot apical meristem (SAM) are maintained by a negative feedback loop involving homeodomain protein WUSCHEL (WUS) and CLAVATA (CLV) signaling pathway. Studies show that WUS promotes stem cell identity, whereas CLV restricts stem cell identity. However, the molecular mechanisms behind this negative feedback loop are poorly understood. The focus of this study is to shed light on the molecular mechanisms by identifying genes that interact with WUS and its promoter, which has been divided into ten regions. In particular, we perform yeast-one and yeast-two hybrid screens to identify the genes that interact with WUS promoter regions 1 and 2 (WP1 and WP2) and WUS, respectively. Multiple novel genes have been shown to interact with WP1, WP2, and WUS.

91

  Role of Mcl-1 Anti-Apoptotic Protein in Pancreatic Cancer Siyuan Stella Wang Mentor: Ravinder Abrol The Bcl-2 family of apoptotic proteins induce or inhibit apoptosis, regulating cell survival and death. Anti-apoptotic proteins in the Bcl-2 family inhibit apoptosis by disabling the release of apoptogenic factors. Previous analysis on the proteins in this subfamily (Bcl-xL, for example) has shed light on their interactions with other Bcl-2 proteins and has lead to the creation of apoptotic mutants with binding capabilities that correlate to their abilities to induce apoptosis in cancer cells. The Mcl-1 anti-apoptotic protein is a pro-survival protein in the Bcl-2 with little confirmed regarding its specific roles in causing cell survival. Using molecular dynamics software, complexes of Mcl-1 and proapoptotic proteins in the family were computationally analyzed and their binding characteristics determined. From this, Mcl-1's effect on apoptosis was better understood and mutants of the protein were designed to be produced and experimentally tested for their effectiveness in inducing apoptosis in pancreatic cancer cells. Coding For Two-Player LQG Plants With Missing Information Yizhen Wang and Kevin Li Mentor: Tracey Ho This paper attempts to solve a hard variation of the two-player LQG plants model in which each plant has incomplete knowledge about the other plant. Two major issues are discussed: the most appropriate metrics for coding schemes for this purpose is found; a new coding scheme is developed to avoid undesirable erasure pattern so as to maximize the control performance. Localization of Complement Component C3 in Autistic Cerebellum Yuchen Carrie Wang Mentor: John Allman Autism spectrum disorders (ASD) describe a wide range of developmental disabilities associated with behavioral deficits including sociobehavioral, cognitive, linguistic, and perceptuomoter abnormalities. Abnormal cerebellar structure and function in autism are well documented. Purkinje cells are the sole output of the cerebellar cortex. The molecular layer contains the dendrites of the Purkinje cells and the axons of granule cells that form trillions of synapses with them. Autistic individuals have shown a significant decrease in the number of Purkinje cells, which suggests that the synaptic density in the molecular layer will also reduce in autistic cases. The mechanisms through which Purkinje cells and synapses are pruned are still unclear. Gene expression studies showed elevated expression of C3, which is involved in the complement cascade, crucial for complement effector functions such as elimination of pathogens, debris, and cellular structures. The elimination of target structures is mediated by triggering phagocytosis through C3 receptors on microglial neurons, which are crucial to immunity within the central nervous system. Staining for C3 proteins on microglial neurons and for synapses in the granule cell layer in the cerebellar tissues showed that inappropriate complement activation during synapse development could alter neural connectivity by excessively targeting synapses for elimination and hence, destabilizing neural networks. Indeed, the impaired synaptic pruning is observed in autism. Modeling Sea-Level Variability From Hours to Years: A Test Case in Indonesia David A. Warrick Mentors: Adam D. Switzer and Aron J. Meltzner In order to validate geophysical and climate models, accurate estimates of past sea-level change are required. Numerous researchers have used coral microatolls, disc-shaped inter-tidal corals with living sides and flat, dead tops that grow near the surface at low tide. When exposed to air, the exposed portion of the coral dies; this is useful, as past relative sea level can be tracked by investigating the timing and magnitude of coral diedowns. However, this methodology has not been adequately validated. Using data from satellites and tide gauges, sea level over the past 20–30 years can be compared to modern corals in order to verify the correlation between sea level and a coral’s highest level of survival (HLS). We are building a model to investigate the extent to which factors beyond low water level affect a coral’s HLS; such factors may include the time of day at which low water level occurred, cloud cover, or wind speed at the time of low water level. This will allow us to better understand how coral HLS relates to sea level, and it will help us test the idea that corals are a precise method of determining past sea level. Lunar Cold Spots: Characterization and Possible Formation Mechanisms  Clifford Watkins Mentor: Paul Hayne   The Diviner radiometer on the Lunar Reconnaissance Orbiter (LRO) identified a previously unrecognized class of small craters surrounded by unusually cold ejecta patterns, termed “cold spots”. The cold spots are distinguished by near-crater deposits with evidence for lateral, ground-hugging transport. More distal, highly insulating surfaces surround these craters and do not show evidence of either significant deposition of new material or erosion of the

92

  substrate. The near-crater deposits can be explained by a laterally propagating granular flow created by impact in the lunar vacuum environment. Further from the source crater, at distances of ~10-100 crater radii, regolith surfaces appear to have been decompressed without the accumulation of significant ejecta material. We developed a power law fit to the density profile of the cold spot region as a function of crater radii, to model the distribution of the excess material compared to the expected ejecta blanket. From these data, several models for cold spot formation have been proposed, including degassing of volatiles in the regolith by ballistically-emplaced ejecta particles. In order to test this hypothesis, we developed a soft-sphere discrete element model (DEM) to analyze the potential effects of volatile release from impacted grains and connect the large scale LRO maps with the surface material physics. Modeling of Plasmon-Mediated Inter-Emitter Interactions Carla Watson Mentors: Harry Atwater and Ruzan Sokhoyan Surface plasmons supported by metallic structures have been known to alter the quantum optical properties of quantum emitters, such as spontaneous emission or energy transfer rates. The plasmonic properties of such metallic waveguides may promote an epsilon near zero (ENZ) environment that can mediate long-range interactions. I investigate and model the inter-emitter interactions involved with an elliptical waveguide of dielectric core and silver cladding. By adopting a dyadic Green's function approach involving field quantization and FDTD analysis, I numerically developed the spontaneous decay enhancement and local density of states for a quantum dipole emitter coupled with an elliptical waveguide. Fabrication and Characterization of Macroporous Durable Separator to Improve Safety of Lithium Batteries Rochelle Weber Mentors: Michael Hoffmann and Farshid Roumi The growth of lithium dendrites during the repeated charging and discharging of lithium batteries poses a safety concern because the dendrites can puncture the battery separator, create an internal short, and lead to the rapid discharge of the battery and potential ignition of the flammable organic electrolyte. This paper proposes the use of durable materials like Kapton and mylar as a separator to eliminate the risk of dendrite puncture. The separator utilizes laser cut holes in Kapton or mylar (to allow ionic conductivity) and layers of nonwoven polyester (to retain electrolyte). The wettability, porosity, and thermal conductivity of this two-layer separator is comparable to commercial Celgard 2325 separator. In addition, coin cells and rolled 18650 size batteries built with this separator exhibit similar capacity fade and longer cell life compared to similar batteries built with Celgard 2325 separator. Predicting Off-Target Binding of Nucleic Acid Probes Eli N. Weinstein Mentors: Niles Pierce, Brian Wolfe, and Maayan Schwarzkopf The use of short, complementary nucleic acid strands as probes to detect specific mRNA is a core technology in the study of gene expression. However, as the sensitivity of detection strategies increases, researchers are confronted more and more with the problem of non-specific binding. Here a strategy is developed for predicting these offtarget effects. A software program models, in moderate detail, the energetics of probe-target interactions. Its predictions on the relative off-target binding rates are then used to design new probes. These predictions are tested using the Northern Blot technique. As an interesting side product, by comparing the predictive power of various energetics calculations we find evidence for the hypothesis that potential rearrangements are a critical factor in determining the stability of non-complementary interactions. Shifting Implicit Social Biases Nancy Wen Mentors: Ralph Adolphs and Damian Stanley We rely heavily on previously stored internal representations of social groups to make rapid social judgments (i.e. whether a stranger is trustworthy). These internal representations are formed through experience and are malleable through associations of certain social groups with positive or negative outcomes. One measure of internal attitudes is the Implicit Association Test (IAT) which measures the strength of association between two social categories (e.g. black and white) and two concepts (e.g. positive and negative). We chose to use the race IAT as a measure of people’s implicit attitudes towards Black and White people. Our goal is to develop and test a paradigm to shift implicit race biases by controlling the exposure of subjects to emotional faces of specific racial groups. I created a Rapid Serial Visual Presentation (RSVP) paradigm in which 500 Black- and White American faces were presented to participants for 500msec each. We manipulated the emotional content of the faces by changing the proportion of angry to neutral faces for each race. The overall experimental paradigm included a race IAT, RSVP (counterbalanced over subjects so half received angry Black and half received angry White), and then a second race IAT. The data from our first pilot (10 subjects) indicated that exposure to angry faces of a specific race shifted

93

  preference toward that race. It is possible that exposure to a race-specific angry faces would make the neutral faces of that race used in the IAT seem more pleasant. However, this effect was not shown in the second group of participants (20 subjects). We are in the process of collecting more data to determine the validity of the effect observed in the first pilot. Higgs Physics Through Two Decay Channels, Higgs to Diphoton and WW Final States Ian Wessen Mentors: Maria Spiropulu, Emanuele Di Marco, Alex Mott, and Cristian Peña Unquestionably, one of the most important break throughs in experimental physics of this generation has been the discovery of what could be the Higgs boson. While there is strong evidence that the candidate boson discovered is the Higgs, there are many other properties which need to be understood to confirm it's behavior and see exactly how it fits into the Standard Model. We examine two decay channels of the Higgs, diphoton and WW. Superior resolution of ECAL and the clean kinematics of photon final states enable kinematic reconstruction of parent particles with great accuracy. In diphoton, I want to parametrize a fitting function against Monte Carlo Standard Model background processes and then compare this to data. The pulls on the fit parameters suggest how well the Standard Model and the data from CMS agree. The decay to two W’s contains a lot of kinematic information, but there are many challenges in distinguishing our signal from background. Background sources of missing energy and dilepton final states require a series of discriminating cuts. My analysis focuses on vector boson fusion Higgs production with a characteristic 2-jet final state. By carefully choosing events with these signatures, we can then apply kinematic analysis à la razor techniques to learn about the Higgs. Accuracy Improvement of AFM Measurement Using an Additional Sensor Theodore J. Wilkening Mentors: Kyihwan Park and Jiseong Jeong The Atomic Force Microscope (AFM) utilizes piezo-electric (PZT) actuators to scan samples at the micro- and nanometer level. However, PZT actuators have an inherent non-linearity that is severely apparent when measuring samples with large height (on the order of several micrometers). To negate the error caused by this non-linearity, an additional sensor must be applied to the system in order to measure the absolute displacement of the cantilever in addition to the conventional optical sensor that is used for a feedback loop. Analysis of sensor properties and the characteristics and properties of the PZT generated error are given. Experimentation with the application of an additional sensor to the AFM system has also been done. Protein Catalyzed Click Capture Agents for Altering Protein Function: Characterization and Augmentation of Protein Capture Stephanie Wong Mentors: James Heath, Blake Farrow, and Kaycie Deyle Protein catalyzed click (PCC) capture agents have the ability to alter protein function when bound to specific domains. Two examples of proteins these capture agents can target are the protective antigen (PA) exotoxin secreted by anthrax and the E17K mutant pleckstrin homology domain (PHD) of the Akt1 kinase that is sufficient to cause cancer. PA is involved in the anthrax infection cycle and the translocation of lethal factors through cell membranes. A PCC agent targeting PA has been created that has inhibitory effects on lethal factor translocation. One aim of this project is to determine the specific binding site of the PCC agent using enzymatic digestion and mass spectrometry of target proteins covalently labeled by the PCC agents, to better understand the mechanism of inhibition. This method was also applied to verify the binding site of a PHD PCC agent to the epitope containing the oncogenic E17K mutation. The PA PCC agent does not completely inhibit the translocation of the lethal factor; thus, the other aim of this project is to synthesize multiple antigenic peptides (MAPs) that would bind to the target much more robustly and then test for stronger inhibition. Improving the Non-Common Path Calibration Method for Robo-AO to Enhance the Adaptive Optics Correction Chatarin Wong-u-railertkun Mentors: Christoph Baranec and Reed Riddle As pre-observing procedures, calibration of the instrument has to be done using internal telescope and source simulator; however, even with the smallest single-mode optical fiber available, the emitted light beam is so large that it decreases the sensitivity to the static wavefront errors in the system. This non-common path errors are the reason why Airy rings, on the science camera, are not circularly symmetric. During the installation process of a pinhole at the tip of fiber to reduce the aperture, the pinhole is damaged and becomes a single slit instead. To delicately observe the aberrations introduced by the internal optical system, wavefront sensor is installed. By slowly adjusting the deformable mirror, a new flat map is developed, reducing the wavefront residual by half. By running the adaptive optics system on sky with wavefront sensor attached, a new slope offset, created to nullify the time-averaged wavefront residual, measured in terms of several Zernike modes, yields a reconstructed point spread function with greater Strehl ratio and more circularly symmetric.

94

 

A 3d High-Speed Probe for Measuring the Magnetic Components of a Whistler Wave Pakorn Wongwaitayakornkul Mentors: Paul M. Bellan and Xiang Zhai In the Caltech astrophysical jet experiment, observations show that a burst of wave activity in the whistler frequency regime (10-30 MHz) occurs at the time of a fast magnetic reconnection. The whistler wave magnetic component is expected to be circularly polarized even for oblique propagation and also contains most of the energy*. An inductive whistler wave detector has been designed using B-dot probes to measure the 3D high frequency magnetic field fluctuation. Each probe component consists of two miniature commercial oppositely oriented inductor coils connected to a miniature transformer. The transformer subtracts the signals of the two coils to cancel the unwanted capacitive component and retain the inductive component. The three coil pairs are arranged orthogonally and are adjacent to each other. The probe has excellent rejection of capacitive coupling and should resolve the whistler wave polarization. The measurements will be compared with other diagnostics, namely a capacitively coupled probe, an ultra-high-speed camera and a EUV detector. * Bellan, P. M. (2013) Circular polarization of obliquely propagating whistler wave magnetic field. (submitted for publication) STS Studies of Magnetically Doped Topological Insulator Heterostructures Nicholas Woodward Mentors: Nai-Chang Yeh and Chien-Chang Chen Kane and Mele predicted 3D topological insulators (TI), characterized by four Z2 topological invariants. Bismuth alloys were theoretically predicted to be TI, due to their large spin-orbit coupling and inverted band-gap. ARPES experiments later confirmed the odd number of Dirac cones in the band structure which is characteristic to strong TI in Bi2Se3, Bi2Te3, and BixSb1-x alloys. For non-magnetic impurities, unitary density of states (DOS) resonance peaks around impurities have been observed. It has been predicted that in the presence of magnetic impurities, the time-reversal symmetry breaking will cause a spin-dependent energy splitting near the Dirac point and create dual spin-polarized DOS resonances around the Dirac point, provided that the interaction among magnetic impurities is negligible. Additionally, in the limit of high-density magnetic impurities, it has been theoretically predicted that the gapless Dirac cone associated with the surface states of the three-dimensional TI will become gapped due to global time-reversal symmetry breaking by the magnetic impurities. We report our scanning tunneling spectroscopy studies on Bi2Se3/Cr-Bi2Se3 heterostructure MBE-grown epitaxial films grown on GaAs. Both dual and single impurity resonances in the low doping density limit were observed. We also found a surface Curie temperature TC ≈260° K indicated by a surface gap of up to ∆max≈ 0.8eV, which is much larger than the observed bulk Curie temperature of ≈ 7°K. This could be due to the formation of Cr-Se ferromagnetic domains phaseseparated from pure Bi2Se3, similar to the Fe-Se ferromagnetic compounds observed in Fe- Bi2Se3 crystals. These observations are in accordance with theory. Additionally, I spent time rebuilding a superconducting-magnet transport measurement system, which is a practical setup to characterize the electrical transport properties (such as the carrier density and the electrical mobility) of graphene and topological insulator samples for the laboratory through longitudinal and Hall resistivity measurements. The Effect of the MJO on Global Ocean Wave Heights Benjamin Wu Mentors: Duane Waliser and Yuk Yung The MJO, or Madden-Julian Oscillation, is an organized, low frequency oscillatory storm system that propagates eastward across the tropics in the Indian and Pacific Oceans. It is often linked to anomalous precipitation events, wind patterns, and surface warming. It is also predictable with lead times up to 2-4 weeks due to well-documented observations since 1974 in concert with advanced weather/climate forecast models capable of simulating many aspects of the MJO. In this paper, we strive to find a link between the MJO and significant wave heights over the global oceans. To do this, we use daily MJO indices, which contain the date, phase, and amplitude of MJO events, and ocean wave data from 2005 to the present, which contains the daily estimates of significant wave height at all global ocean points. We composite the daily ocean wave maps according to the phase of the MJO (1-8), including a composite control group of weak/no MJO, and do this compositing for both boreal winter and summer. The findings illustrate a definite link between MJO events and global wave heights and the next step is to determine the cause of the link by comparing wave height and various aspects of the MJO, particularly the atmospheric circulation and near-surface wind characteristics.

95

  Finite Element Modeling of Thin Film Flow in Grooved Microchannels for E-Jet Printing Applications Nancy Wu Mentor: Sandra M. Troian Electrohydrodynamic (E-jet) printing uses inks such as metals, polymers, DNA, and carbon nanotubes for fabricating 3D structures without supporting material, DNA spotting for biological assays, and printing metal interconnects for microelectronics. To optimize the hydrodynamics for nanoprinting applications, we investigate the flow of thin films of perfectly conducting liquids in grooved microchannels of constant cross section. These conduits meter the flux necessary for ion or droplet mode emission by Taylor cone destabilization at the groove terminus. In this part of the study, we focus exclusively on the conduit flow and use finite element simulations to solve a coupled system of nonlinear partial differential equations describing the influence of Maxwell and capillary forces on the shape of the flowing liquid film. We study the influence of the electric field distribution and groove opening angle, with particular focus on the most suitable initial and boundary conditions for providing the necessary flux to the Taylor cone region. Our studies indicate that while different initial conditions all lead to the same final flow configuration, the particular choice of boundary conditions can significantly alter the film shape and volumetric flux, thereby enhancing or repressing the influence of Maxwell stresses near the Taylor destabilization point. Characterization of Splicing Mechanisms of NF-kB Regulated Genes in B Cells Conway Xu Mentors: David Baltimore and Devdoot Majumdar RNA splicing plays an essential role in the function of most eukaryotes by being one of the major processes that refines immature RNA transcribed from DNA, so it can eventually be translated into proteins. Previously, it has been shown that introns separating constitutive exons are usually excised in a 5’ to 3’ fashion while the pre-mRNA is still be synthesized. While the overall process of RNA splicing has been elucidated over the years, the state of RNA splicing pertaining to genes in the immune system are still unclear. The goal of this project is to investigate splicing mechanisms of NF-kB regulated genes in B cells. By conducting time course stimulation experiments on B cells, purifying RNA, and running quantitative PCR on the corresponding RT-PCR products, we aim to measure the relative expression of introns in comparison to mature RNA. As a result, we have come across several genes including CD86 where the supposed 5’ to 3’ excision pattern of introns is not observed. If we can characterize the splicing mechanism shared among NF-kB regulated genes in B cells, we will be able to better understand a key regulatory mechanism within the immune system. Functional Knockdown of the Novel Protein Olfactin Jing Xu Mentors: Marianne Bronner and Ankur Saxena Olfactin is a novel protein that is expressed in the olfactory region, mid/hindbrain boundary, and ear in developing zebrafish embryos. Through confocal microscopy with live embryos, we can image olfactin by using a novel protein trap line. In this line, olfactin is fused with either citrine (green) or mCherry (red) fluorescent proteins. In the olfactin-mCherry line, olfactin is truncated and missing the last exon, possibly creating a partial loss-of-function protein. We are characterizing this line for any possible phenotypes. Furthermore, we are expanding our experiments with a splice-blocking morpholino that prevents the correct splicing of the second exon. By comparing homozygotes to heterozygotes in semi loss-of-function experiments of olfactin, we have seen a preliminary result that may suggest aberrant development of the olfactory region. Further work will involve a more complete loss-offunction using a translation-blocking morpholino targeted against olfactin. The External Control of Free-Swimming in Scyphomedusae Nicole W. Xu Mentors: John O. Dabiri and Janna C. Nawroth Scyphomedusae, invertebrates comprising a bell and subumbrellar muscles, are model organisms for studying biological propulsion and can be used to engineer biorobotic vehicles and model cardiac flow. Previous studies have demonstrated the ability to control free-swimming in tissue-engineered juvenile jellyfish by stimulating pacemakers to regulate pulsation frequency; however, this is limited to one stereotypic mode of execution. More studies need to be conducted to demonstrate control of free-swimming, including turning, in larger organisms. To do so, stimulation tests were conducted on Aurelia aurita using a wire electrode connected to a DAQpack system to send 6.6-V stimuli to the exumbrella. In-dish and free-swimming tests using paired wires and/or newly designed electrodes will determine the optimal stimulation location and signal amplitude and frequency to initiate contractions at the nearest pacemakers. Wire designs have been successfully attached to free-swimming jellyfish at the bell center and margin using gauze, Opuntia spines, and mussel-inspired adhesive. Video analysis of freeswimming tests will compare the kinematics of natural and apparatus-controlled swimming motions.

96

  Assembly of a Combined Scanning Tunneling Microscope and Scanning Electron Microscope System for Studying High-Temperature Superconductors Cassidy Yang Mentors: Nai-Chang Yeh and Marcus Teague Scanning tunneling microscopy (STM) characterizes the electronic and topographical properties of sample surfaces with atomic resolution at cryogenic temperatures via quantum tunneling. However, its limited field of view and slow scanning speed make the study of small features of interest in a macroscopic sample very difficult. We seek to combine STM with a scanning electron microscope (SEM), which has a larger field of view and much faster scan speed. In the combined STM/SEM system, the SEM will be able to locate particular regions of interest, so that the STM can be quickly aligned with the region for further investigation. This project was started last summer; the vacuum chambers, pumps, gauges, cryo-cooler, and STM head have been assembled. However, the spectroscopic and topographic capabilities must be tested and optimized at liquid nitrogen and liquid helium temperatures to ensure that all STM components are operational before the SEM is attached. We must also minimize electrical and mechanical noises and implement temperature control. The completed system will be used to study high temperature superconductors and other interesting condensed matter materials. Remote Analysis of Mineral Composition Using LIBS Spectra on Mars: Improving Data Analysis of LIBS Spectra Taken by the ChemCam Instrument Suite Under Martian Conditions Jade Yang Mentor: Bethany Ehlmann The rover Curiosity carries the instrument suite ChemCam. ChemCam uses Laser-Induced Breakdown Spectroscopy (LIBS) to collect data on the Martian mineralogy. It fires a laser at a specimen and records the spectra emitted by the resulting plasma. In order to improve the quantitative composition predictions estimated from emission lines extracted from spectra, a procedure for radiation continuum removal was developed. The procedure utilizes the Savitzky-Golay filter applied the local minima of the spectrum to remove background Bremsstrahlung radiation. Wavelengths useful for element detection and concentration prediction were also identified using a series of preexisting measurements of mixtures with mixtures with predetermined compositions. Linear fit statistics were generated for each wavelength using this database. Significant wavelengths were chosen based on sensitivity, low error, and correspondence with wavelengths previously identified as significant. The existing ChemCam team partial least squares (PLS) procedure used for generating the predicted element concentrations was also improved. The database on which the PLS analysis was run was expanded to include more samples. More improvements on the PLS algorithm are still being worked on. These improvements will help enhance the accuracy of Curiosity’s understanding of the elemental composition of the Martian surface. Investigating Temporal Localization of Choice-Induced Preference Change Kevin H. Yang Mentors: Ralph Adolphs and Keise Izuma Studies in social psychology have repeatedly demonstrated that when individuals make a choice between two equally likeable alternatives, they tend to express higher preference levels for the alternative they chose, and lower preference levels for the alternative they rejected. It is believed that having to reject something they like or having to choose something they don’t like causes an inconsistency of choice and preference, which the subject resolves by lowering their preference for the rejected item. This phenomenon, known as choice-induced preference change, is a centerpiece of cognitive dissonance theory, but has recently been challenged as a confounding effect due to error in the experimental paradigm. In this study, by using a new experimental paradigm, in which preference and choice were decorrelated, it was observed that subjects truly do raise their preferences for selected computer wallpapers because of their choice. Temporal localization though electroencephalography (EEG) will also be used. By measuring activation levels of the posterior medial frontal cortex (pMFC), it can be established that the pMFC plays a key role in feelings of cognitive dissonance, and therefore plays a key role in choice-induced preference change. In-Flight Calibration of Level 1 Data From JAXA’s GOSAT Sensor Yun Ju (Sharon) Yang Mentors: Florian Schwandner and Carol Bruegge The detector sensitivities on Japan Aerospace Exploration Agency’s (JAXA) Greenhouse gas Observing SATellite (GOSAT) are subject to degradation. To correct for this degradation, the radiances measured by GOSAT for atmospheric greenhouse gas detection require episodic calibration. Annual in-flight calibration is carried out at a ground reference site, to derive top-of-atmosphere (TOA) radiances. As part of this calibration process, the JPL team measures bidirectional reflectance factors (BRFs) on the ground at specific wavelengths, using the Portable Apparatus for Rapid Acquisition of Bidirectional Observations of the Land and Atmosphere (PARABOLA) instrument. Raw BRF data collected from the instrument are extracted and compiled by IDL programs to output specific BRF products. The final BRFs are plotted on polar graphs and stored in reduced data tables for JAXA’s calibration team.

97

  JAXA’s team uses these BRFs to derive TOA radiances and correct for degradation of GOSAT. Existing code was debugged and improved, and new codes documented. The overall process was significantly streamlined and better documented, to increase automation and integration, and thus enhance operational readiness for the same procedures for NASA’s OCO-2 satellite (launch 2014). The NMR Investigation of Electrostatic Gauche Effect in Solution Zelal Yavuz Mentors: John D. Roberts, Bright Emenike, and William Carroll The conformational preferences of X-CH2-CH2-Y molecules where X = F, CI, Br and Y = NH2, OH were analyzed and the halide electrostatic gauche effects were investigated in their protonated analogues. In the investigation of electrostatic interaction between X and protonated Y the experimental vicinal proton-proton and proton-fluorine coupling constants were obtained from gNMR fittings. The required 3JHH and 3JHF values for the calculation of the gauche fraction were derived from the Haasnoot-Altona equation for the H-H couplings and the generalized Karplus-type equation for the F-H couplings, respectively. The experimental results indicate that the effect of electrostatic attraction between the partially negative-charged halide and the positive-charged NH3 or OH2 dominates the effect of intramolecular hydrogen bonding and that electrostatic interactions are responsible for the high gauche preferences. However, the fraction of gauche differs corresponding to the solvent polarity and to the solute substituent electronegativities. Building a Virtual Caltech & Data Visualization in OpenSim Stacy Yeh Mentors: George Djorgovski and Ciro Donalek This project builds on previous work done in the Djorgovski lab. Using OpenSim virtual reality development software, we continue construction on a virtual Caltech campus as well as exploring existing and new methods of data visualization in a virtual world. In this project, we spent time constructing replicas of current Caltech buildings. These buildings will eventually become part of a virtual Caltech campus that can be used as a forum for educational and scientific discussion and interaction. Additionally, we examined and built on existing scripts for the visualization of “big data,” which contain extremely large numbers of data points. These scripts allow us to plot these “big data” sets in innovative ways that allow for easier interpretation and visualization. In the future, using these scripts to plot data in a 3D virtual world will allow users to interact with the data and each other, as well as provide a new way of organizing and presenting large amounts of data. Studying the Effects of Chromogranin A and Catestatin Knockout on Cardiac Structure and Ischemic Preconditioning in Mus musculus Kevin Yei Mentors: Sushil K. Mahata and Ellen Rothenberg Cardiac ischemia/reperfusion (I/R), the sudden restoration of blood flow to the heart following constriction or obstruction of coronary arteries, is associated with injury leading to myocardial infarction, cardiac arrhythmia, and contractile dysfunction. Ischemic preconditioning (IPC), brief cycles of sublethal I/R prior to prolonged ischemia, protects the heart from I/R injury. However, the mechanisms underlying this innate protective mechanism are incompletely understood. Since ischemic heart disease, the leading cause of death in the world, is associated with cardiovascular risk factors including hypertension, diabetes and insulin resistance in humans, we examined the relation of IPC to a secretory prohormone overexpressed in hypertensive individuals called chromogranin A (human CHGA, mouse Chga) and a 21-amino acid peptide fragment of CHGA called catestatin (CST: human CHGA352–372) which exerts antihypertensive and cardioprotective actions in humans and rodents. Using hypertensive and hyperadrenergic Chga knockout (Chga-KO) and CST knockout (CST-KO) mice to gain better insights into the mechanisms regulating IPC, we discovered that whereas insulin-sensitive Chga-KO mice provided IPC-induced cardioprotection comparable to that of WT mice, insulin-resistant CST-KO mice were unable to provide IPC-induced cardioprotection against I/R injury. We then examined the cardiac structural and functional differences between Chga-KO and CST-KO mice using a Visual Sonics Vevo 2100 echocardiogram in order to better understand their differential capacity for IPC. We found that in 3-month-old mice with regular cardiac function, Chga-KO mice were hypertrophic with significant increases in left ventricular mass and wall thickness, while CST-KO mice displayed cardiac structure and function comparable to that of WT mice. We hypothesize that insulin sensitivity makes ChgaKO hearts more resilient to I/R injury despite increased cardiac muscle mass while insulin resistance makes CSTKO hearts more susceptible to I/R injury despite regular cardiac structure, and we are currently using Western blot analysis to characterize the signaling changes involved in this process.

98

  Development of Rhodium Platinum Conjugates as Tri-Functional Therapeutics to Target DNA Mismatches Jessica S. Yeung Mentors: Jacqueline K. Barton and Alyson Weidmann The accuracy of DNA replication is vital to cellular development and survival. Failure of the cell to repair DNA bases mismatches arising from polymerase errors or DNA damage can lead to diseases such as cancer. The coordination compound cis-[PtII(NH3)2Cl2], or cisplatin, is one of the most effective commonly prescribed anticancer drugs with an impressive efficacy shown by a 90% cure rate for testicular cancer. Nonetheless, toxicity, adverse side effects, and acquired drug resistance limit the potential of many platinum-based anticancer drugs in cancer treatments. Within the Barton group, several rhodium complexes have been developed that recognize and bind specifically to DNA mismatches with high selectivity in vitro as well in cellulo. Thus, the synthesis of a rhodium platinum conjugate to increase the potency of Rh metalloinsertors, improve Pt effectiveness in mismatch repair (MMR) deficient cells, and maintain the high selectivity of Rh metalloinsertors would be a valuable tri-functional therapeutic tool to target DNA mismatches and ultimately cancer. By covalently attaching known DNA platinating agents to a mismatch selective rhodium complexes and appending a cell penetrating peptides (CPPs) to direct the constructs into cellular nuclei, we anticipate improved cellular uptake and an increase in selectivity of platinum based chemotherapeutics to target MMR-deficient cancers. Heterologous Expression of G-Protein-Coupled Receptors in Yeast Cells: A Study of the Interaction Between Fungus and Nematode Pheromones Dorothy Yim Mentors: Paul Sternberg and Yen-Ping Hsueh Some fungal species trap nematodes as prey. The method through which these nematophagous fungi sense the presence of its prey is by detecting ascarosides, small molecules that nematodes use to communicate and regulate behavior. The detection of these molecules then induces the fungi to produce traps that catch nematodes for consumption. Therefore, these ascarosides must play a crucial evolutionary role for nematode fitness, especially if it compromises their survival to increased fungal predation. The recognition of these signals is critical to understanding predator-prey relationships in ecology, ultimately shedding light on them mechanisms of coevolution. By cloning different fungal G-protein-coupled receptors (GPCRs) that respond to nematode ascaroside and inserting them into yeast, the receptors can be tested for their response to ascaroside outside of the fungal system. Through different assays, the receptors are treated with ascaroside and measured for their interaction, either through a beta-gal enzymatic assay or a growth assay. Combining these two surveys allows for an analysis on which fungal receptor responds to the nematode ascaroside and under what conditions the worm trapping mechanism is induced. Design and Integration of an Intelligent Controller for a Fourier Transform Spectrometer Patrick Yiu Mentors: Didier Keymeulen and Dan Berisford This paper presents the design and integration of an intelligent controller for CIRIS (Compositional InfraRed Interferometric Spectrometer) on a stand-alone field programmable gate array (FPGA) architecture. CIRIS is a novel take on traditional Fourier Transform Spectrometers (FTS) and replaces linearly moving mirrors (characteristic of Michelson interferometers) with a constant-velocity rotating refractor to variably phase shift and alter the path length of incoming light. This design eliminates the need for periodically accelerating/decelerating mirrors inherent to canonical Michelson designs and allows for a compact and robust device that is intrinsically radiation-hard, making it ideal for spaceborne measurements in the near-IR to thermal-IR band (2-12 m) on planetary exploration missions. The instrument’s embedded microcontroller is implemented on a flight-qualified VIRTEX-5 FPGA with the aim of sampling the instrument’s detector and optical rotary encoder in order to construct an interferogram. Subsequent signal processing, including a Fast Fourier Transform (FFT), noise reduction/averaging, and spectral calibration techniques are applied in real-time to compose the sample spectrum. The instrument is characterized in simulated space-flight conditions and we are able to demonstrate that this technology is capable of meeting the strict volume, sensitivity, and power consumption requirements for implementation in scientific space systems. Recognizing 3-Colorability Infeasibility With Degree Two Nullstellensatz Certificates Michael M. Young Mentor: Mohamed Omar A graph is called 3-colorable if each vertex can be assigned one of three colors such that no two adjacent vertices are the same color. 3-colorability turns out to be an NP-complete property. In “Recognizing Graph Theoretic Properties with Polynomial Ideals” by J.A. de Loera, et al, the authors examine methods to detect non-3-colorability in graphs through the use of degree-one Nullstellensatz certificates. We continue this exploration by characterizing the non-3-colorable graphs that have degree-two Nullstellensatz certificates. Therefore, we can detect such graphs in polynomial time.

99

 

Export From the Weddell: Monitoring the Surface Circulation Using Drifters Madeleine Youngs Mentor: Andrew Thompson The complex export pathways between the Weddell Sea and the Antarctic Circumpolar Current (ACC) influence water mass transformation, nutrient fluxes and ecosystem dynamics. In January 2012, 40 surface drifters, equipped with temperature sensors, were released into the Antarctic Slope Front in the northwestern Weddell Sea to study these exchange processes. Comparison of this data set with a similar deployment in January 2007 provides insight into the interannual variability of the surface circulation in this region. Despite a greater northward extent of sea ice at the time of deployment in 2012, the percentage of drifters, released into the ASF, that travelled from the Weddell Sea to the ACC was greater in 2012 than in 2007. This change primarily resulted from a relaxation of the Coastal Current in 2012, such that no drifters entered Bransfield Strait. Variations in surface wind patterns and interactions with the cryosphere are identified as potential drivers of the change in the surface circulation. This shift is consistent with modeled changes to the surface circulation in response to variations in the Southern Annular Mode. The drifter trajectories in 2012 were also associated with warmer surface temperatures and lower satellite-derived chlorophyll levels, suggesting a link between local biological productivity and the surface circulation. 2V287L Mutation Increases Sensitivity of 42 and 542 nAChRs Through Altered Assembly and Trafficking Caroline Yizhu Yu Mentors: Henry Lester and Weston Nichols Autosomal dominant nocturnal frontal lobe epilepsy (ADNFLE) is linked to mutations in genes that code for the 4 and 2 nicotinic acetylcholinergic receptor subunits, but the pathophysiology is unknown. 42 exist in two stoichiometries, which have different sensitivities to acetylcholine. To study the effect of stoichiometry, we generated fluorescently tagged 4, 5, 2V287L and 2 subunits. The dose response curves obtained from Flex station experiments confirmed an increased sensitivity to acetylcholine in N2A cells for both 4 2VL and 54 2VL. Oocytes were transfected in such a way to bias assembly towards either the low or high sensitivity receptor stoichiometry. The resulting dose response curves revealed a decreased response in the low sensitivity receptors only. TIRF imaging of 542VL and 42VL tagged with pH sensitive fluorophore was used to quantitatively compare plasma membrane expression levels of the various subunits, showing a twofold increase in 5 membrane expression, twofold decrease in b2 expression and a threefold decrease in 4. Moreover, ER expression levels remained constant, suggesting a preferential assembly or trafficking of the high sensitivity stoichiometry. Overall, our experiments show that 2V287L increases receptor sensitivity due to a shift in stoichiometry due to altered assembly. Drugs that bias assembly towards the low sensitivity stoichiometry therefore are novel and exciting potential therapies for ADNFLE patients. Deciphering the Roles of Covalent Protein Modifications in Neuron and Muscular Disease Pathogenesis Shengyang Kevin Yu Mentors: Paul W. Sternberg and Amir Sapir The increase in lifespan in the modern world is balanced by age-induced neurodegenerative diseases. Understanding such diseases—Alzheimer’s, Huntington’s, Parkinson’s, and ALS—has emerged as one of the fundamental medical problems of our time. These diseases are associated with the accumulation of disease-specific misfolded proteins and ultimately interfere with cellular processes. The mechanisms for the formation of such aggregates and their role in causing disease is still poorly understood. However, for each disease, there are existing Caenorhabiditis elegans models which express these such aggregation proteins along with fluorescent markers. We are interested in examining the role of small ubiquitin-like modifiers (SUMO) and other regulatory proteins on disease pathogenesis and the strain aggregations. Screening by RNAi gene knockout of smo-1, a C. elegans ortholog of SUMO and other genes will allow us to pinpoint various connections between the roles of the smo-1 protein and disease pathway. Isolating various connections will prompt further investigation of protein interactions through IP experiments using tagged smo-1 to decipher the cellular landscape of sumoylation and other protein modification mechanisms in aggregate forming strains. This could ultimately lead to a better understanding of pathway mechanisms in the neurodegenerative disease as well as the regulatory proteins. Seeking Symmetry: An Investigation of Healing in Aurelia aurita William Yuan Mentor: Lea Goentoro The moon jellyfish, Aurelia aurita, has a curious ability to rearrange its body to reform radial symmetry after sustaining trauma, such as the loss of a leg, rather than simply regrowing the leg, as seen in other invertebrates. The study of this phenomena was divided into two main parts: the use of chemical inhibitors to determine the role of Wnt signaling in the symmetrization mechanism, and the formation of a mathematical model to determine the morphological changes that occur in recovering jellyfish. In the former section, jellyfish specimens had legs

100

  removed, and were incubated in 6-Bromoindirubin-3'-oxime and lithium chloride solutions, both Wnt pathway activators. The specimens that were exposed to them did not exhibit symmetrization. For the mathematical analysis, the precisely measured rotating jellyfish were tracked, and the data entered into a geometric model to determine the amount of volume lost during symmetrization. The volume loss may be attributed to necrosis or programmed cell death. The model also provides insight into how individual legs move with respect to the rest of the body through the process of symmetrization. Overall, insight into the unique healing mechanisms exhibited by A. aurita can provide a platform for its potential use as a biomaterial. Single-Cell Dynamics of Bcl11b Expression and T-Cell Lineage Fate Kevin A. Yuh Mentors: Ellen V. Rothenberg and Hao Yuan Kueh The lymphatic immune system relies on the development pathways of hematopoietic stem cells to T-cell commitment. However, the mechanism by which DN2a cells commit to T-cell fate is still unknown. The expression patterns of Bcl11b, a crucial component of the gene regulatory network for commitment, are an indicator of commitment in DN2a cells. We analyzed the single-cell dynamics of pertinent cell lineages in a previously-collected 86.91-hour timeframe of cell development, where yellow fluorescent protein (YFP) gene was inserted into the Bcl11b locus. Using custom scripts for image analysis, we calculated YFP expression, removed noise using curvesmoothing and signal-processing methods, and analyzed the difference in expression between cells, particularly among sister cells. Preliminary results suggest that daughter cells in a lineage will share Bcl11b expression trends, suggesting that T-cell fate is determined long before the expression of other key commitment genes. In addition, trends in the rate of change of YFP expression suggest that Bcl11b can be expressed at a wide variety of rates within a given lineage. Protocols and scripts here establish a framework for analyzing single-cell dynamics in future experiments. Neurobiological Metrics for Simple Choice Span Perceptual and Values-Based Decisions Mitchell Zak Mentor: Antonio Rangel Time and time again, the Drift-Diffusion Model (DDM) has been shown to fit two-alternative forced-choice task data quite well. Its predictive power spans a wide array of tasks from perceptual decision-making, in which there exists a correct and an incorrect answer, to values-based decision-making, in which personal values determine the correct choice. What remains to be examined is the relationship between parameters of the model across (rather than within) types of decisions. Preliminary results indicate strong correlations between parameters of the DDM for subjects making choices about food preferences and distinguishing between patterns. Having redesigned the experiment to represent a larger proportion of the types of decisions that exist, it’s likely that these parameters will correlate even more highly. Subjects will make choices about food and gambles both together and separately. If the experiments are carried out as planned, this would be further evidence for the model’s robustness and its applicability in the extrapolation about a subject’s neurological behaviors from simple response data. Testing the Reliability of Reported Results Using the P-Curve Method Klavdia Zemlianova Mentors: Colin Camerer and Gideon Nave Research publications often report results that support their hypothesis and omit the data that does not. Several studies about scientific reproducibility have asserted that a combination of investigator practices and publication biases in a variety of sciences could lead to publication of false positives that do not replicate. The p-curve, a distribution of statistically significant p-values for a set of findings, was created to help determine whether or not a given data set does actually allow one to reach the conclusion that the study claims. The aim of this project is to survey current research in economics under this lens. Database Development for Microorganisms Residing on Mars-Based Spacecraft Farah Zerehi Mentors: James N. Benardini and Wayne W. Schubert Populating a database of the isolates found in cleanrooms was a majority of the work done this summer and it includes information such as physical location, source of origin, and biological characteristics, this database will allow for easy reference of isolates kept. This project also consists of keeping a physical archive of samples found on spacecraft and various associated hardware at the many cleanrooms around the country. A set of assigned microbial isolates were processed this summer consisted of 36 Mars Science Laboratory and about 22 Phoenix Faring isolates. DNA was isolated using an DNA purification system. The 16S rRNA gene was amplified by polymerase chain reaction and gel electrophoresis was used to confirm the DNA. After DNA cleanup and purification, samples were sent for sequencing. For safekeeping, each isolate also had five glycerol stocks and one cryobead working stock made and stored in a -80 ˚C freezer. Once the sequencing is completed, a phylogenetic tree will be made, which compares new samples to other identified microorganisms. The information will be used to inform Planetary Protection and JPL regarding the nature of the microorganisms found in cleanroom environments.

101

 

Building a Virtual Caltech Jerry Zhang Mentors: George Djorgovski and Ciro Donalek With further development, Virtual Worlds can be excellent platforms for multi-dimensional data visualization, telecommunication, scientific collaboration, and education. Virtual worlds are effective in regard to visualizing large and complex data sets because they allow the data to be displayed intuitively, understandably, and collaboratively. Our research aims to create more flexible and intelligent data visualization software, as well as virtual environments in which Caltech students and instructors can interact in a scholarly setting. To do this, we used the OpenSimulator application to develop a Virtual Caltech containing replica buildings and landmarks that create a comfortable environment for virtual interaction. In addition, we used the in-world scripting language to create scripts that can generate visual representations of several variables, and apply them to actual data sets. We successfully replicated buildings such as Beckman Auditorium and Page House, and after several trials and tests, we added new features and capabilities to the data visualization script that allow it to be more powerful and versatile. Investigation Into the Conformational Preferences of 1, 4-Butanedioic Acid Dianion in Different Solutions Jingxian Zhang Mentor: John D. Roberts The conformational preferences of the 1, 4-butanedioic acid (succinic acid) dianion in different protic and aprotic solvents and the role of solvent effect in its unexpectedly high fraction gauche were investigated. 25 mg samples of succinic acid were prepared at a 1:2 molar ratio with tetrabutylammonium base in methanol and then dried extensively; after dissolving the sample in a trial methanol solution and taking the proton NMR on the Fid600 machine, integration of the sample peaks to counterion peaks at a ratio of 1:8 in MestreNova ensured that dianion was the sole product prepared. The dianion was then dissolved in a variety of solvents in NMR sample tubes and proton NMR spectra were obtained with the Fid600 machine. The observed spectra were then fitted with stimulation spectra in gNMR to determine the J13 and J14 couplings, which served as input for the Haasnoot-Altona equation (3JHH = 14.63 cos2 (θ) – 0.78 cos (θ) + 0.60 + Σ λi x {0.34-2.31 cos2[si (θ) + 18.4 |λi|]} ) to calculate the fraction gauche of the succinate dianion in a particular solution. Ten fraction gauche values from different protic and aprotic solutions along with their solvent parameters were regressed with the Kamlet-Taft equation (K=A0+Aα+Bβ+P(π *)+D∆) to attempt pinpointing trends in parameter coefficients that explain how the gauche preference comes about despite significant steric hindrance and charge repulsions between the two negative carboxylate groups. The trendline obtained from these ten solutions was Ln(Fg/(Ft/2)) = -0.52 + 0.56α + -2.2β + 0.47π * + 1.6∆ with an adjusted R2 value of 0.96, and the parameters β and ∆ (hydrogen bond accepting basicity and polarization correction term) being the most significant contributors. After taking into account the pKa of the solvent and its ability to sustain the dianion in solution, the only significant outlier to the trendline from the dataset of solvents is THF, a phenomenon that warrants further investigation. Stochastic Version of a Classical Spin Model of the Electron and Its Comparison With Quantum Mechanics Paul Zhang Mentor: James Beck This SURF studies the properties of a non-relativistic electron governed by classical spin equations in a force field. The central force and linear force fields were explored for stationary orbits with energies corresponding to those given in quantum mechanics. A repulsive linear force scenario was explored to replicate tunneling effects through the potential wall. The full connection between this classical spin model and quantum mechanics does not seem achievable without expanding to relativistic particles. Investigating Polarity Effects for Residue-Specific Incorporation of Escherichia coli L-Asparaginase II Evan Zhao Mentors: David A. Tirrell and Katharine Y. Fang Incorporation of a noncanonical amino acid through replacement of a natural residue using endogenous cell machinery can lead to destabilization of the target protein or loss of biological activity. The analogue residue is usually nonpolar, but researchers in the Tirrell lab have shown that a new noncanonical amino acid, glutamic acid hydrazide, can be used to replace glutamine, a polar residue. It would be advantageous to use polar residues because they usually reside on the surface on a protein. Thus, replacement of a polar residue could have substantially different effects on protein activity and stability than replacement of a nonpolar residue. To investigate the effects of the polarity of the replaced residue on protein stability and activity, we have chosen a model protein, L-asparaginase II, a cancer drug produced naturally by Escherichia coli in its periplasm. We will use azidohomoalanine, a methione, a nonpolar residue, analogue, and glutamic acid hydrazide to demonstrate this effect. Biological activity will be measured through a Nessler’s reagent based asparaginase activity assay.

102

 

Mapping Geological Structures in LA Basin With InSAR Deformation Time Series Analysis Taokun Zheng Mentors: Mark Simons and Romain Jolivet The focus of this project is to understand how aquifers influence ground deformations in the Los Angeles Basin using time series analysis of Synthetic Aperture Radar interferometry. The InSAR measurement of ground deformation include two major components: a secular subsidence resulting from the persistent, inelastic deformation of aquifers, and a seasonal oscillation resulting from the periodic pumping and recharging of underground water every year. We analyzed three InSAR products, derived from the time series analysis: 1) the secular velocity field; 2) the spatial distribution of the phase and amplitude of the seasonal oscillations. We observe that the confined aquifers is bounded by fault systems. Furthermore, the area undergoing the largest subsidence is bounded to the north by the so-called forebay line, corresponding to the boundary between confined and unconfined aquifers. Further research should be conducted to understand and model the influence of confined and unconfined aquifer on the ground deformation of the LA Basin area, including the role of tectonic faults. Experiments With Dynamic Obstacles and Correct-by-Construction Controllers Yuqian C. Zheng Mentors: Richard M. Murray and Scott C. Livingston A major challenge in designing autonomous robots is automatically synthesizing a high-level controller for selecting actions to allow a robot navigate to a destination through a field of dynamic obstacles. Specified as temporal logic formulas, algorithms have been proposed to generate controllers for high-level motion planning while satisfying desired properties. As these models assume perfect conditions, the controllers may fail during implementation on a robot, as uncertainties arise internal and external sources. Due to the inherent probabilistic nature of the inputs and outputs, we test the controllers in situations with varying degrees of uncertainty. Simulations are run in Stage, and for physical testing, a Turtlebot is used as the agent, which is controlled using the Robotic Operating System (ROS) Groovy. The specific conditions we deteriorate are freedom of movement by introducing denser dynamic obstacle fields and sensor fidelity by increasing noise and limiting data intake. To quantify correctness, we look at metrics such as collision rate and time to complete an obstacle course. We thus investigate the probabilistic guarantees that can be made about the system and characterize under what conditions the original guarantees of correctness fail. A Tale of Two Toggles: The Costs and Benefits of Various Designs of Biochemical “Decision Engines” Tiffany Zhou Mentors: Richard Murray and Dan Siegal-Gaskins Bistable gene-regulatory circuits are involved in a variety of cell processes, such as cell cycle regulation, signal transduction, and apoptosis. Given the large diversity of bistable circuit topologies, understanding these differences will assist in identifying motives behind specific circuit designs and in the engineering of novel genetic circuits. In my project, I compared the costs and benefits of two circuits previously predicted to exhibit bistability, named the monomeric toggle and the dimeric toggle. Various circuit properties, including energetic costs, response times, and size of the bistable region of parameter space, were characterized using a combination of experimental and computational tools. Experimentally, a new monomeric repressor protein required for one of the circuits and intended to function like a dimer was assembled and characterized. Induction tests using open-loop versions of the circuits were conducted to determine differences in repression, and the complete circuits were then tested in vitro using a cell-free environment called TX-TL. Computationally, models and simulations were developed using ordinary differential equations and the TX-TL modeling toolbox to explore relevant parameter ranges and assist in the understanding of experimental observations. The characterization methods used in my project lay a foundation for the future analysis of more complex bistable circuits. The Characterization of Binding Kinetics of Short DNA Fragments Through the Use of Rapid Single-Molecule Force Spectroscopy Julia Ziac Mentor: Ozgur Sahin Atomic Force Microscopy (AFM) is a form of Scanning Probe Microscopy (SPM) used to explore the properties of surfaces with a proximal probe. The unique abilities of AFM to image and manipulate nanoscale objects makes it ideal for investigating biological samples. By functionalizing the AFM probe and scanning surface with short fragments of single-stranded DNA (ssDNA), we have explored the sequence-dependent binding kinetics of DNA, lending itself to applications in biomolecule sequencing and biochemical imaging.

103

MURF MURF UNDERGRADUATE RESEARCH FELLOWSHIPS

M U R F

Arithmetic of the Quaternions Over Finite Rings Ayah K. Almousa Mentors: Dinakar Ramakrishnan and Mohammad Hadi Hedayatzadeh Most scientists are familiar with the complex numbers, which are of the form a+bi where a and b are real numbers and i ^2 = -1. William Hamilton extended this idea to four dimensions in 1843 when he described what are now called Hamilton’s quaternions: numbers of the form a+bi+cj+dk where a, b, c, and d are real numbers and i^2 = j^2 = k^2 = -1, where multiplication is not commutative. The idea of quaternions can be extended so that i^2 = u, j^2 = v, and k = ij = -ji, and where u and v are fixed and a, b, c, d are all elements of a general number system R, called a ring (a set with multiplication and addition defined for its elements). The study of quaternion algebras has been well-documented when R is a field, which is a commutative ring allowing division, but not necessarily in the case where R is an arbitrary ring that may have strange properties like having non-zero elements that equal zero when taken to some power. We analyze the structure of the quaternions over such rings and seek cases where the quaternions can be represented as a matrix algebra, a useful property. Given that quaternions over fields arise naturally in theoretical physics, we wonder if the analogues over arbitrary rings will also have a physical relevance. Cdc42 Effector Protein 3 (CEP3) in Somitogenesis Fabian K. Blacksheep Mentors: Marianne Bronner and Shuyi Nie Somites, which give rise to vertebrae and ribs, musculature of the back and limbs, and dermis of the skin, are formed by paraxial mesoderm through a segmentation and epithelialization process. The signals that developing somites receive determine what derivatives they form. In addition to the processes of specification and differentiation, somitic cells also undergo dramatic rearrangements and migration during their development. To understand how these cell and tissue movements are regulated, my project focuses on studying the actincytoskeleton machinery that is essential for cell morphology and motility. Previous work done by my co-mentor identified an actin-cytoskeleton regulator, CEP3, which is expressed in pre-somitic mesoderm and somites. My hypothesis is that CEP3 modulates the dynamic arrangements of actin filaments in paraxial mesoderm to control somite formation. To test this hypothesis, I first confirmed the expression of CEP3 transcripts in Xneopus laevis embryos by in situ hybridization. Next, I will manipulate the levels of CEP3 expression using both gain- and loss- of function approaches. I will then examine the effects on somite development by identifying the changes in the expression pattern of genes involved in somite formation, such as MyoD and Myf5. Engineered Bombyx mori to Suppress Female Population and Increase Efficiency in Sericulture Keneil K. Brown Mentors: Bruce Hay and Omar Akbari The silk industry is a multibillion dollar trade with the unit price of raw silk being almost 20 times that of raw cotton. Bombyx mori, the mulberry silk worm, is an economically important insect because it is used in silk production. B. mori males have greater resistance to disease, consume less food and produce better quality silk than females. It has been estimated that a male-only sexing strain of B. mori could save the farmer about 10% in rearing costs, increase silk production by about 15% and improve the silk quality by one or two grades which translates into greater profits per kilogram. However, at present, separating males from females is impractical; therefore, male-only rearing techniques are desirable and would greatly increase efficiency and profits in sericulture. By exploiting the sex determination machinery in B. mori we propose to engineer a strain in which expression of an antibiotic resistance gene is exclusive to males. After doing this, worms at the first instar stage of development will be fed with the antibiotic and this should lead to the death of all females in the population as females do not possess the antibiotic resistance gene. The Function of Ig Protein Dpr 11 in Drosophila NMJ Development Kelsey Carter Mentors: Kai Zinn and Robert Carrillo The immunoglobulin superfamily (IgSF) of proteins functions in a vast diversity of molecular pathways, including those involved in neural development. Studies in Drosophila revealed an IgSF “interactome” comprising the interaction between two IgSF subfamilies: the defective in proboscis response proteins (DPRs) and the Dpr interacting proteins (DIPs). Nothing is known about the function of these genes except for the founding member, dpr1. Recently, the Zinn lab found that mutating dpr11 results in a developmental defect at the larval neuromuscular junction (NMJ), highlighted by supernumerary satellite boutons. To begin teasing out the function of dpr11 in NMJ development, we performed genetic interaction studies with previously characterized genes that display the same phenotype when mutated. Persistence of the satellite bouton phenotype by the simultaneous partial knockdown of dpr11 and a candidate gene would suggest a common molecular pathway. We have tested six candidate genes and initial results propose that dpr11 genetically interacts with daughters against dpp (dad) and wasp (wsp). This implies that dpr11 may function in a BMP signaling cascade. Further experiments are needed to determine if dpr11 and the interacting genes indeed function together for proper development of the NMJ.

107

In vitro Study of the Effect of Mechanical Stress on the Cell Proliferation Rate and Migratory Behavior of Fibroblasts Using Time-Lapse Imaging Lisa Eshun-Wilson Mentors: Morteza Gharib and Hesham Azizgolshani Fibroblasts maintain the structural integrity of connective tissues by continuously secreting precursors of the extracellular matrix and collagen. Following tissue injury, fibroblasts migrate to the site of damage, where they deposit new collagen and facilitate the healing process. The proliferation and activities of fibroblasts are highly regulated by mechanical force. However, the analysis of how fibroblasts sense and respond to mechanical stress is limited by the availability of techniques that can apply controlled mechanical forces to living fibroblasts and simultaneously measure changes in cell and molecular distortion, including alterations of intracellular biochemistry. Time-lapse imaging provides one such tool to closely observe the spatiotemporal relationships of biomechanically manipulated fibroblasts. To be feasible for long-term observation of cellular changes, time-lapse imaging calls for a suitable microenvironment. Furthermore, mechanical stimulation of the cells needs to be done in a controlled fashion in order to prevent extreme strains and stresses that would damage the cells. To these ends, an incubation chamber is constructed around a stereomicroscope to control for temperature, CO2 content, humidity, and to provide a stably buffered culture medium for the fibroblasts. To induce regulated, measurable amounts of mechanical stress, a micromanipulator is used to move two rigid, cell-coated substrates apart with a goal of generating sheets of fibroblasts. Research, Thermal Analysis, Timeline, and Day-in-the-Life Analysis of ExoplanetSat CubeSat Concept Janan Ferdosi Mentors: Muthu Jeganathan and Louise Hamlin Almost every space mission undergoes extensive thermal analysis. In order to perform such analysis, there must be a clear pathway, and organized, sequencing of events which would take place during the launch, initial orbits, day-in-the-life, and the entire lifetime of a satellite. Depending on the orbital geometry, altitude, and other mission-specific aspects of the mission, decision would be made regarding the type and extent of analysis. Of specific interest to the proposed ExoplanetSat project, a spreadsheet was developed of the basic, static state (nontime dependent) thermal analysis of a simple flat plate in the orbit around the Earth, along with basic thermal analysis of the spacecraft’s focal plane radiator. This paved the way for more in depth thermal analysis of the entire proposed mission. A work in progress, the team is now converging on technical design details of the mission concept. Simultaneously with the thermal analysis, research was performed on the potential spacecraft timeline, commissioning and day-in-the-life analysis. The timeline was crafted as a high level spreadsheet including every event and phase Exoplanetsat would undergo immediately after launch up to the point of establishing a communications link with the Earth, along with the listing of each of components, its state (on/off), and power mode, at each phase after launch. The day-in-the-life is a work in progress which will eventually list all the events (and phases) that the proposed ExoplanetSat mission would undergo in an ordinary day in the orbit. Solar-Driven Thermochemical Fuel Production: Model Development for Response Rate of Thermochemical Materials as a Function of the Oxygen Diffusion Coefficient and Surface Reaction Constant Michael Forsuelo Mentors: Sossina Haile, Chirranjeevi Balaji Gopal, and Lincoln Collins Solar energy is in relative abundance compared to modern societal energy needs. Converting solar energy into transportable and storable chemical energy can be achieved by utilizing oxygen absorption and de-absorption of non-stoichiometric ceria. By coupling oxygen absorption at lower temperatures and oxygen de-absorption at higher temperatures in the presence of water or carbon dioxide, the water or carbon dioxide molecules are split, yielding hydrogen, a renewable fuel, or processable carbon compounds. The efficiency of such a two-step thermochemical cycle is governed principally by the kinetics of the material. In order to calculate the response rate, defined as the rate at which re-equilibration is approached, a model has been developed founded upon oxygen transport in the bulk and across the gas-solid interface of the material, described by the oxygen diffusion coefficient and surface reaction constant, respectively. The model considers variability in the oxygen diffusion coefficient and assesses boundary conditions. Several numerical algorithms have been developed to calculate oxygen concentration profiles throughout space and time. These variable diffusivity algorithms deliver material response and suggest further assessment of relationships governing oxygen transport.

108

Outcome Devaluation Effects Among a Neurotypical Population Examination of a Novel Paradigm for Assessing Reward Devaluation in Humans Anthony Gómez Mentors: Ralph Adolphs, Justin Feinstein, John O’Doherty, and Mimi Liljeholm Outcome Reward devaluation is a branch type of goal-directed learning that entails showing a decreased preference for the outcome of an associated action after satiation or aversion. Studies using functional magnetic resonance imagingfMRI in humans suggest that this process occurs inthat the ventromedial prefrontal cortex (O’Doherty, 2011vmPFC) plays an important role in reward devaluation. Consistent with these data, lesions to the vmPFC in rats and non-human primates have been shown to disrupt reward devaluation. To date, however, there have been no reward devaluation studies conducted in humans with vmPFC lesions. This project aimed to create a novel paradigm for assessing reward devaluation in humans, with the long-term goal of testing this paradigm in patients with vmPFC lesions. However, those who have lesions in this region of the brain may not exhibit devaluation effects. In this study, researchers developed aWe developed a behavioral task in which neurotypical participants received food rewards from a gumball machines by learning stimulus-response contingencies usingfor specific different fractals. Following training, pParticipants are then satiated with one reward and given the opportunity to earn rewards againthen undergo extinction. It is . Researchers hypothesized that participants will devalue the satiated reward while upholding the value of the non-satiated reward. Upon achieving a satisfactory effect size among a neurotypical population, this paradigm will be used to test human lesion patients at the University of Iowa, with the prediction that patients with focal bilateral vmPFC lesions will show an impoverished reward devaluation effect. Researchers predict they will not observe a devaluation effect among lesion patients, suggesting a correlation between with their lesion and this behavioral deficit. Dark Matter Search Using Razor Kinematic Variables at the CMS Detector at CERN Natalie M. Harrison Mentor: Maria Spiropulu The objective of this research is to look for the production of theoretically favored dark matter (DM) particle candidates in proton-proton collisions at the Compact Muon Solenoid experiment at CERN. We use the Razor Kinematic variables developed by researchers at Caltech to focus on events where DM is produced in association with multiple jets, leaving behind large missing transverse energy in our detector. We look at the distribution of the razor variables for data and compare this to Monte Carlo expectations for the background and Monte Carlo generated signal events for various possible models. We perform an inverted hypothesis test using the modified Frequentist method, called CLs, to determine the 95% confidence level upper limit on the cross section for the production of DM particles. We use this same analysis to examine a simplified supersymmetry (SMS) model. For both analyses, we calculate limits using only 5 fb-1 of data (a quarter of the 8 TeV dataset). In future, is pertinent to translate the cross sections for dark matter production into dark matter nucleon cross sections to compare to direct detection results. The method described above could also be applied to a search for Higgs decaying into two invisible particles. Preparation and Chiral Resolution of the Cyclobutane Synthon in the Total Synthesis of Psiguadial B ShuMing Huang Mentors: Sarah Reisman and Lauren Chapman

Stereochemistry plays a significant role in the bioactivity of natural products. The cyclobutane synthon of Psiguadial B, a bioactive natural product, was thus prepared with the focus on setting the correct stereochemistry at the alpha position. The synthetic route of preparing the key intermediate was optimized, and a single enantiomer of the cyclobutane synthon was obtained by classical resolution. This technique employed chiral amines, such as (S)-1phenylethylamine, to formed diastereomeric salts. The different solubilities of disastereomeric salts enabled the separation of the enantiomeric cyclobutane synthon through recrystallization.

109

Characterizing the Effect of Growth Rate on Gene Transcription Arman Mohsen Nia Mentors: Rob Phillips and Robert Brewster The division time of E.coli varies between 20 minutes up to a few hours depending on the growth conditions. These changes in the growth rate are accompanied by major physiology changes. One significant change is in the abundance of RNA polymerases (RNAP), the protein responsible for transcription. A previously developed model of transcription, based on the principles of thermodynamic, successfully predicts gene expression as a function of molecular parameters such as binding strength of RNAP to the promoter. However, the model does not explicitly consider global parameters like the growth rate. To explore this relationship, using a combination of fluorescence and enzymatic activity measurements, we simultaneously measure the RNAP level and reporter gene expression of four strains each with a unique promoter binding affinity. Our data suggest that the number of RNAP is inversely proportional to the growth rate except in a few cases. In addition, we confirm that the reporter’s expression rate is consistent with the previously developed statistical model for a given growth rate. In order to extend the application of the statistical model to different growth rates, ultimately we hope to boil down the complex global effect of changes in growth rate to a simple set of measurable parameters such as the RNAP level. Pushover Analysis for Steel Frame Buildings Ryan Ortiz Mentor: John Hall Pushover analysis is a non-linear analysis of a structure under vertical loads and steadily increasing lateral loads, which closely represent forces caused by earthquakes. A pushover curve is a plot of the total base shear versus the roof displacement of a structure. The base shear force is the sum of horizontal components of the shear and axial forces in the first story columns. It is an estimate of the maximum expected lateral force that will occur due to strong seismic ground motion at the base of a structure. Such a curve indicates how ductile the building is, and more importantly, it tells us the yield strength (the point at which the structure no longer behaves elastically) and ultimate (maximum) strength of the structure. Through the application of a program written by Dr. John F. Hall called “FRAME,” we seek to show that the exclusion of panel zones, interior frames, vertical loads, concrete slabs, and the usage of perfect connection welds in place of fracture-prone welds greatly effects the pushover strengths of 6 and 20-story steel frame buildings. Thus far we have determined that the strength of a structure is greatly decreased when fracture-prone welds are applied in place of perfect welds. The connections between the beams and columns are allowed to break and consequently enable the building to deform more rapidly. Current analysis of the data retrieved from the other design parameters is underway. Development of an in vitro Blood-Brain Barrier Model for Evaluating Nanoparticle Transcytosis Angel M. Reyes Mentors: Mark E. Davis and Andrew J. Clark The blood-brain barrier (BBB) formed by brain capillary endothelium presents a significant challenge to the treatment of neurological disorders by preventing passage of most neurotherapeutics from the blood to the brain parenchyma. Nanoparticles provide a potential avenue for drug delivery across the BBB by utilizing transport mechanisms already used by bloodstream components. bEnd.3 mouse cerebral capillary endothelioma cells were grown on transwell supports to prepare an in vitro model of the BBB in order to test the ability of various gold nanoparticle formulations to cross the BBB. The presence of tight junctions between cells was measured using transepithelial electrical resistance, and permeability of the cells to targeted/nontargeted nanoparticles was determined using Nanoparticle Tracking Analysis (NTA). Further assays were developed to accurately quantify the amount of a targeting ligand (transferrin) on the surface of nanoparticles, using either Enzyme-linked immunosorbent assay (ELISA) or transmission electron microscopy (TEM). The in vitro model and nanoparticle detection/characterization assays will subsequently be used to determine the ability of gold nanoparticles containing cleavable targeting ligands to cross the BBB model compared to nanoparticles containing non-cleavable ligands. Probing the Binding of Anti-Freeze Proteins to Trehalose Crystals Alejandra Rios Mentors: William A. Goddard, III, and Soo-Kyung Kim Antifreeze Proteins (AFPs) refers to a class of polypeptides that allow cold-adapted organisms, such as fish and insects, to survive in subzero environments by lowering the freezing point of water without affecting the melting point. This project focuses on finding the binding modes of AFPs, particularly of DAFP-1 an AFP from the beetle Dendroides canadensis, and trehalose. The binding process of this system can give a better understanding of the physiological process on how these organisms can survive at temperatures below zero. Experimentally it has being demonstrated that the most favorable binding site of DAFP-1 is with the (-110) plane of trehalose. Using bio software programs and molecular dynamics simulation, an approximation can be made to model and validate the system of interest as well as calculations of the free energies to confirm the experimental results.

110

An Optogenetic Model for Parkinson’s Disease Victoria Servin Mentors: Viviana Gradinaru and Bin Yang Parkinson’s disease (PD) is a prevalent neurodegenerative disease. It is characterized by difficulty initiating and executing movements, and caused by degeneration of dopaminergic (DA) neurons in the substantia nigra pars compacta (SNc). Many Parkinsonian models have been used to study the disease. Current models employ irreversible techniques in order to obtain Parkinsonian symptoms or do not address the pathogenic mechanism of PD. We have used optogenetics in an attempt to develop a model of PD that will allow for a better understanding of the underlying neuronal circuitry and how it contributes to PD symptoms. We selectively silenced dopaminergic neurons with fast inhibitory opsins- halorhodopsins and Archaerhodopsins. Selective expression of opsins in the SNc DA neurons was achieved via direct delivery of adeno-associated viral vectors encoding a doubly-invertedopsin into the TH-Cre transgenic lines. Light stimulation (500-600nm) of the SNc resulted in dyskinesia, freezing behavior, and gait abnormality. We established a fast light-switchable model that can be used to study PD’s pathogenic mechanism, screen for drugs, and develop specific therapeutic approaches. Characterizing the Spatiotemporal Expression and Function of Cadherin Proteins During Neural Crest EMT Karla Terrazas Mentors: Marianne Bronner and Crystal Rogers The epithelial to mesenchymal transition (EMT) is a series of events which converts adherent epithelial cells into individual migratory cells. Although this process is most well-known for its role in cancer metastasis, it is also a normal developmental event that occurs when neural crest (NC) cells emigrate from the neural tube, thus providing an excellent model to study the principles of EMT in a nonmalignant environment. It has been shown that cadherin proteins, which are a family of calcium-dependent cell adhesion molecules, play important roles in mediating EMT. The modulation of these cadherins is crucial for the proper coordination of the transition from adherent cells to individual migratory cells. Previous studies show that the dysregulation of cadherin proteins results in the failure of NC cells to begin or complete the process of EMT. In addition, it has been shown that Sip1 and Snail2 transcription factors play a key role in mediating EMT during metastasis by binding to and repressing epithelial cadherin (Ecad). Other cadherins such as Ncad, Cad6B, and Cad7 have also been found to play key roles in EMT and NC cell migration. Here, we characterize the gene and protein expression patterns of four cadherin proteins in the avian embryo using immunohistochemistry and in situ hybridization. With high magnification imaging of whole mount and transverse sections, the data gathered show that neural cadherin (Ncad) transcript and protein are expressed in the developing neural tissue throughout the developmental stages assayed. Additionally, the protein, but not transcript, is down regulated in the dorsal neural tube and NC cells just prior to EMT. In contrast, cadherin 7 (Cad7) protein is primarily expressed in migratory NC cells. Our data also suggest that epithelial cadherin (Ecad) is highly expressed in the ectoderm and in the early migratory NC cells. Finally, cadherin 6B (Cad6B) is strongly expressed in the midline of the dorsal neural tube, supporting previous studies which found that Cad6B was expressed in this region just before NC emigration and then down regulated to allow for the process of NC delamination. Our results indicate that the cadherin proteins are expressed in overlapping and distinct tissues suggesting cooperative and independent roles for each protein during development. Prediction of 3-Dimensional Structures for ORs in Cluster 7 Using Homology-Based GEnSeMBLE Franklin Woode Mentors: William Goddard, III, and Soo-Kyung Kim Olfactory receptors (ORs) are a class of G protein-coupled receptor (GPCR). ORs are responsible for recognizing odorant molecules and are integral to our ability to smell. Most GPCRs are tuned to respond only to a specific ligand and related agonists. The question is whether this hypothesis holds true for ORs, each of which seem to respond to several odorants. The goal of this project was to determine the 3D structures of the 63 human ORs in Cluster 7, paying attention to interactions that affect binding. Proteins were grouped into clusters based on their similarity to the template hOR1G1. The family head is the OR most homologous to the template. 3D structures were generated from the template using Homology-based GEnSeMBLE, a program which consists of four steps: 1) PredicTM, 2) Helix-based Homology, 3) Bihelix/Combihelix, and 4) SuperBihelix/SuperCombihelix. A stable N1.50-E2.50-N7.49 hydrogen bond network was found, which keeps the protein rigid during activation. However, a strong D3.49-K6.30 or E3.39-H6.40 interaction has not been found. The lack of a strong 3-6 interaction may suggest that hO51J1 is easily activated. Docking of hO51J1 with simple ligands will clarify this. Further analysis of Cluster 7 ORs is necessary to ascertain whether or not these interactions are conserved in the cluster.

111

AMGEN AMGEN SCHOLARS PROGRAM

A M G E N

  Elucidation of Commensal Microbial Regulation of the Mammalian Serotonin (5-HT) Metabolism Phoebe Ann Mentors: Sarkis Mazmanian and Elaine Hsiao Humans have established a mutualistic relationship with various collections of microbes that compose the body’s microbiota. Recent research indicates that the gut microbiota modulates host immunity, gastrointestinal (GI) integrity, and the nervous system. It has also been shown that alterations of gut microbiota in turn affect the activity of intestinal enteroendocrine cells, which produce over 90% of the body’s serotonin (5-hydroxytryptamine, 5-HT). 5-HT is a crucial neurotransmitter, involved in numerous fundamental biological processes; therefore, disrupted 5-HT metabolism is associated with a pleiotropy of neural and physiological disorders: depression, irritable bowel syndrome, and cardiovascular disease, among many others. That germ-free (GF) mice—mice which lack microbiota—have less than 40% of serum 5-HT levels than their wildtype counterparts also indicates the microbiota’s influence on 5-HT metabolism. To elucidate the specific bacteria which most influence 5-HT metabolism, GF mice were conventionalized with a limited microbial community (LMC) shown to restore SPF 5-HT metabolic levels. Mice were subject to various behavioral tests, and results showed that LMC conventionalization significantly restored wildtype gut motility, and mildly restored wildtype anxiety, and social interaction. Such effects of LMC and other promising bacterial collections will further understanding of microbial regulation of 5-HT metabolism, ultimately begetting novel microbial therapeutics for 5-HT deficient diseases. Regulation and Processing of Fibroblast Growth Factors During Drosophila Development Kenny Chen Mentors: Angelike Stathopoulos and Vince Stepanik Fibroblast growth factors (FGFs) are important developmental regulators, and misregulation of FGF signaling underlines developmental disorders, and diseases such as cancer. Within Drosophila, the FGF ligands Pyramus (Pyr) and Thisbe (Ths) activate the Heartless (Htl) receptor and have overlapping, yet distinct functions. This project seeks to explore the regulation of FGF signaling in Drosophila at the level of the ligand. In our model for FGF signaling, we have yet to determine how Pyr and Ths interact with Htl; one possibility is that signaling is initiated by homo- or heterodimerized ligand complexes. Our objective is to explore the possibility of heterodimerization of Pyr and Ths. Additionally, studies from our lab have shown that Pyr and Ths are secreted and detectable as cleaved forms with multiple isoforms that may result from differential cleavage, raising the possibility that alternative cleavage may affect the activity of the ligands. We are therefore examining the cleavage patterns of Pyr and Ths in other cells lines, including Kc167 and S2 cells (embryonic lines) and Dv-1 cells (larval wing disc line from a related species). These experiments will allow us to explore the differential processing of Pyr and Ths in specific tissue types. ULP-4 Desumoylates HMGS-1 to Regulate C. elegans Bioenergetics in Aging and Development Kaitlin Ching Mentors: Paul Sternberg and Amir Sapir Methods by which metabolic networks cope with aging and mitochondrial stress have yet to be well understood. We hypothesize that post-transcriptional modification of key proteins plays a role in this regulation. Here, we demonstrate that ubiquitin-like protease 4 (ULP-4) regulates the electron transport chain (ETC) by desumoylating HMG-CoA synthase (HMGS-1). Mitochondrial staining shows a diminished electrochemical gradient in ulp-4 mutants, suggesting that the ETC is impaired in the absence of ULP-4. Lifespan extension and decreased oxygen consumption similarly indicate mitochondrial stress in ulp-4 mutants. RNAi knock-down of hmgs-1 results in phenotypes similar to ulp-4 knock-out; hence the previously observed interactions between HMGS-1 and ULP-4 are likely positive. The mevalonate pathway, in which HMGS-1 plays a major role, produces coenzyme Q which shuttles electrons through the ETC. Perhaps by regulating HMGS-1 activity, ULP-4 impacts the ETC. This idea is supported by the preliminary observation that providing downstream products of the mevalonate pathway partially rescues the mutant phenotype. By unveiling the role of ulp-4 in mitochondrial regulation, we hope to better understand the dynamic nature of cells’ metabolic networks. Exploration of the Catalytic Conversion of N2 to NH3 by a Single-Site Fe Model Complex Bridget Connor Mentors: Jonas Peters and Jon Rittle Recently, Peters et al. have reported the first catalytic reduction of N2 to NH3 facilitated by a molecular Fe complex. However, little is known about the mechanism of this catalysis. We aim to synthesize compounds that may be intermediates or analogues of intermediates in the catalysis and test their intermediacy in the catalytic reduction. (TPB)Fe≡NAd+ has been prepared as a model complex for a proposed intermediate (TPB)Fe≡N-NH2+. Comparison of EPR data of the model complex and an intermediate formed during catalysis suggests that the intermediate is in fact (TPB)Fe≡N-NH2+. (TPB)Fe≡N+ may be an intermediate in the catalysis and will thus be independently synthesized. To investigate how protons add to the putative (TPB)Fe-N=N-H species, (TPB)Fe-N=N-Ar will be synthesized and subsequently protonated.

115

  Engineering Control Opsins (COPs) for Use in Optogenetics Research Christopher S. Finch Mentors: Viviana Gradinaru and Claire Bedbrook Optogenetics, which allows researchers to control brain activity with light, is rapidly becoming the standard for neurobiology research. Hundreds of projects using optogenetics to understand the mammalian brain, normal and diseased, have been published since the technique was developed in 2005. Despite its rapid growth and broad application, the current optogenetic toolkit lacks proper negative controls: inactive opsins that mimic the size, expression level, and membrane localization of currently used opsin tools. Such controls would guarantee that the behavioral changes observed in optogenetic studies occur solely because of optical recruitment of targeted neurons, and not, for instance, displacement of key endogenous membrane proteins. We have constructed variants of the commonly used opsins hChR2, Arch3.0/ArchT3.0, and eNpHR3.0 with the retinal-binding lysine residue mutated to glutamine, arginine, or serine. The expression, membrane localization, and activity of these mutant opsins were assayed in both HEK cells and cultured hippocampal neurons. For each opsin, a mutant with expression and membrane localization comparable to wild type, shown by confocal microscopy, but no response to light, revealed by electrophysiology, was successfully built. If in vivo testing confirms these results, these mutants will constitute a set of control opsins (COPs) for all future optogenetics research. Structural and Functional Analyses of the Adaptor Nucleoporin Network in the Symmetric NPC Core Jonathan Herrmann Mentors: André Hoelz and Tobias Stuwe The Nuclear Pore Complex (NPC) is the sole gateway in the nuclear envelope and is one of the largest and most complex structures in the eukaryotic cell with a mass of ~120 MDa (~10 million atoms). Due to its sheer size and great flexibility, determining an atomic resolution structure of the fully assembled NPC is not possible with current X-ray crystallography technology. An alternative approach is the determination of the atomic structures of individual nucleoporin complexes and their subsequent assembly into a pseudo-atomic model of the entire NPC. Of the four subcomplexes that compose the symmetric NPC core, the adaptor nucleoporin subcomplex is perhaps the least understood and consists of five nucleoporins in Chaetomium thermophilum. The two largest of these proteins, Nup192 and Nup188, are evolutionarily conserved from yeast to human and share a striking structural similarity; yet, Nup192 is preferentially incorporated in the adaptor subcomplex. We have shown that these two adaptor nucleoporins form alternative interactions within the adaptor subcomplex by employing an interdisciplinary approach combining electron microscopy, x-ray crystallography, and various biochemical techniques. These data indicate a significant conformational plasticity and variability in the adaptor nucleoporin interaction network, supporting its versatile function in the symmetric NPC core. A Computational Study of the Microrheological Properties of Colloidal Gels Daniel Jacobson Mentors: John F. Brady and Charles G. Slominski In recent years, microrheology has proven to be an effective method for probing the viscoelastic properties of a wide range of materials. The development of a theoretical picture of microrheological techniques has not yet caught up with this experimental success. The purpose of this project is to help understand the link between a material’s viscoelastic moduli, as obtained from microrheology, and its microscopic properties by examining colloidal gels. Colloidal gels offer a versatile model system for a material that exhibits viscoelastic behavior. To investigate this link, a dynamics simulation of a passive microrheology experiment was utilized. After refining and testing a previously developed Brownian dynamics simulation of a colloidal gel, free probe particles were introduced into the gel lattice. The viscoelastic properties of this system were then obtained from the averaged mean square displacements of the probes. Continued work on this project will allow for the quantitative relationship between the viscoelastic moduli and gel properties like bond strength, network configuration and colloidal volume fraction to be determined. Further, the development and analysis of this gel system will also lay the foundation for the future introduction of hydrodynamic effects. Structure-Guided Rational Design of Novel Vaccine Candidates Against HIV Siduo (Stone) Jiang Mentors: Pamela Bjorkman and Anthony West The HIV pandemic continues to burden more than 30 million individuals worldwide, while the quest for a global antibody-based vaccine remains elusive. In the past decade, clinical investigations of a rare subpopulation of HIVpositive individuals have isolated antibodies capable of potently neutralizing many strains of HIV. A recent study has identified the IgVH germline origin of a subclass of these antibodies. Induced clonal expansion starting from this germline is the first step through which neutralizing antibodies may be elicited. Unfortunately, naturally occurring gp120s (the HIV antigen) fail to bind and activate the B cell receptor of this germline. This project applies computational interface modeling and high-throughput screening strategies to design and evaluate novel gp120 candidates for binding to the germline receptor. So far, we have reconstructed a complex structure of a gp120 with the germline antibody, and using Rosetta, generated and executed scripts for design of multiple gp120s

116

  ensembles. The data then provided basis for the construction of a semi-random library, followed by binding screens. We have identified several candidates that demonstrate binding to the germline, and crystal structures will be determined to yield fresh starting points for further rounds of design. We are also currently working to simultaneously evolve the germline antibody toward a potent HIV neutralizer, and in the process generate a series of gp120s with optimized binding to each stage of antibody evolution. Metal Chelation and Hydrogen-Bonds as Studied by 1H-NMR Spectroscopy Holden W. H. Lai Mentors: John D. Roberts, Bright U. Emenike, and William R. Carroll Metal chelation and hydrogen bonds are crucial for biological processes including protein folding, enzymatic catalysis, molecular recognition, and signal transduction. Two model systems were designed to study these two non-covalent interactions using 1H-NMR spectroscopy. N,N-dimethylsuccinamic acid (DMSA) and its salts were employed to probe the influence of various metal cations on the conformational of organic molecules. By analyzing the conformational preference of DMSA as a function of its salts, it was determined that metal cations with smaller ionic radii generally have stronger influence DMSA’s conformation in polar aprotic media. 1 was used to probe the strength of hydrogen bonds as a function of solvent. The H---O bond strength was determined by comparing the pKa of the carboxylic acid of 1 with that of 2, where no hydrogen bonding between the amide and carboxylic acid is possible.

1:

2:

Investigating the Effects of Substrate Topology and Hydrogel Composition on 3-D Cell Migration Yulan Lin Mentors: Julie Kornfield and Amy Fu The cornea’s disorderly post-traumatic cellular response disrupts the collagen packing and composition of the cornea, leading to corneal haze. Previous in vitro studies have shown that electrospun nanofiber mats with aligned nanofibers can guide corneal fibroblast migration in a 2-D setting, restoring orderly packing. However, the mechanisms of cell movement are different in three dimensions, which more closely mimics an in vivo setting. Because hydrogels mimic the consistency of the corneal stroma, they may serve in various biomedical applications without disrupting the native environment. The goal of this project is to investigate the effects of hydrogel composition and nanofiber orientation on corneal fibroblast migration in a 3-D environment using confocal fluorescence microscopy. To this end, protocols to visualize cell viability and migration are being established. The Influence of Growth Factors on Suppressing the Myofibroblast Phenotype for Corneal Wound Healing Jacqueline J. Masehi-Lano Mentors: Julie Kornfield and Amy Fu The cornea is the clear, convex window covering the eye that is mainly responsible for the eye’s visual functionality. Transparency of the eye comes from the cornea’s stromal layer, which is populated by keratocytes and composed of an orthogonally organized extracellular matrix (ECM). In a corneal wound, keratocytes close to the wound differentiate into myofibroblasts, which contain stress fibers that distort the refracting surface of the cornea and leads to blindness. Here we study the effects of three growth factors (GFs) on suppressing the myofibroblast phenotype in a 2D setting: epidermal growth factor (EGF), which has been shown to increase the proliferation of keratocyte cells; platelet-derived growth factor (PDGF), which has been shown to induce keratocyte elongation without the formation of stress fibers; and fibroblast growth factor (FGF), which has been shown to encourage the fibroblast phenotype that promotes orderly wound repair. Transforming growth factor-beta (TGF-β) was used to induce rabbit corneal fibroblast cells to differentiate into myofibroblasts. The cells were then exposed to the GFs at 6 concentrations to find the optimal concentration that reduces alpha-smooth muscle actin (α-SMA), the cellular marker for myofibroblasts. Results reveal that lower GF concentrations (0.1 ng/mL to 10 ng/mL) display differing levels of α-SMA expression, while higher concentrations (100 ng/mL) successfully reduce expression by more than 90%, indicating the significant role GFs play in the corneal wound-healing process.

117

  Development of a High-Throughput Optogenetic Screening System for Freely-Moving Adult Drosophila Neeli Mishra Mentor: David J. Anderson High-‐temporal and cellular precision of stimulation of neurons in the brain is essential for understanding the neural mechanisms of behavior. Optogenetics, a method used to stimulate neurons via gene-‐encoded opsins, is a promising technique to help address these concerns. However, this technique had not yet been developed for use in adult, freely-‐moving Drosophila. Our goal was to develop a high-‐throughput system of optogenetic control in adult flies by screening a wide range of opsins. This system would prove to be useful for a variety of research in the fly field and we will use it to understand the neuromechanism of courtship song in adult flies. We were able to show that the recently described Red activatable Channelrhodopsin Receptor (ReaChR) allows for activation of CNS neurons in freely behaving adult flies at wavelengths that do not interfere with normal visual function. By using this tool, we are able to control neuronal activation with millisecond time resolution over a broad range of stimulation conditions. The Role of CIP2A in Neural Crest Development Pushpa Neppala Mentors: Marianne Bronner and Laura Kerosuo The neural crest is a population of pluripotent stem cells that give rise to various cell types including the peripheral nervous system, melanocytes and bone and cartilage of the face. Neural crest development comprises of several steps, induction at the neural plate border, specification in the dorsal aspects of the neural folds, EMT and emigration out of the neural tube, and finally migration into target destinations and differentiation into various derivatives. CIP2A (Cancerous Inhibitor of Protein Phosphotase 2A, PP2A) is expressed in the whole neural plate during neural crest induction and continues to be expressed throughout neural crest development and the migratory stage. CIP2A has been identified in cancer cells to be responsible for c-Myc stabilization via a competitive inhibition mechanism of the protein degradation cascade initiated by PP2A. So far, it has not been shown to bind other Myc family members. We have knocked down CIP2A in vivo by using morpholinos (MO) as well as in situ hybridization and immunostaining for the analysis. Our results show that CIP2A is required for neural crest induction at the neural plate border and knockdown of CIP2A also dramatically decreases the amount of neural crest cells during specification in the dorsal neural folds. These events ultimately lead to a decreased number of migratory neural crest cells. Interestingly, c-Myc is not expressed at the neural plate border, which led us to speculate whether CIP2A binds other targets during early embryo development. Based on our ongoing experiments we suggest that CIP2A has a dual role during neural crest development: It is first involved in the neural crest induction through regulation of active levels of N-myc and later in neural crest specification by affecting the active levels of c-Myc in the premigratory and migratory crest. We are currently performing rescues, overexpressions, and co-immunoprecipitation studies to prove our hypothesis. Promising Model in Understanding Metastasis: Uterine Seam Cell Development in Caenorhabditis elegans Sang M. Nguyen Mentors: Paul W. Sternberg and Srimoyee Ghosh Metastasis contributes to a low rate of survival in the majority of cancer patients. Changes in tumor cell shape occur during metastasis, and therefore understanding its mechanisms can shed light on the cause of spreading cancer. Using the Caenorhabditis elegans uterine seam cell (UTSE), we are identifying novel genes involved in cell shape change. The UTSE undergoes dramatic changes in cell outgrowth during development. One of the methods to identify new genes involved in UTSE cell outgrowth is by looking at genes that have been characterized in other migrating systems. By using RNAi to screen through a list of genes involved in C. elegans linker cell migration, we have shown that 10 of these genes are also involved in UTSE cell outgrowth. We also plan on looking at the role of meprin protease homologues in UTSE development. Meprins proteases are involved in cell shape change, and two C. elegans meprin homologues, nas-21 and nas-22 are highly expressed in the UTSE. According to our preliminary data, the treatment with nas-21 and nas-22 RNAi shows a decreasing in UTSE cell outgrowth. Therefore, we will also screen genes that regulate NAS-21 and NAS-22 functions to better understand how meprins affect cell outgrowth. Taste-Independent Hunger Control Is Mediated by Neuromodulators Dopamine and sNPF in Drosophila Ketaki Panse Mentors: David Anderson and Hidehiko Inagaki The brain translates changes in the internal metabolic state into changes in feeding behavior to maintain energy homeostasis, but how the brain does so is not well understood. Understanding this translation can lead to insight into human diseases such as obesity and cardiovascular disease. Specifically, the purpose of this project is to understand the neural mechanism of hunger control in Drosophila melanogaster. Here, we show that the fruit fly selects sorbitol, a caloric-rich food, after prolonged starvation without the activation of sweet-sensitive gustatory receptor neurons. These findings suggest the neural pathway for feeding amount is independent from sugar

118

  sensitivity. To characterize this neural pathway, we look at feeding behavior in both loss-of-function and gain-offunction transgenic flies. Neuromodulators--small molecules that mediate neuronal communication by acting on cell surface receptors-- change how neural circuits process information in different states. Thus, studying neuromodulation is integral to understanding how the brain functions. Using genetically based modifications, we show that the neuromodulators dopamine and sNPF are key components in nutrient content-based hunger control of Drosophila. An Optimization Study of Half-Life and Solubility of Broadly Neutralizing Monoclonal Antibodies Misha Raffiee Mentors: Pamela Bjorkman and Stuart Sievers The development of antiretroviral drugs has led to milestone achievements in therapies available to HIV-positive patients, however, the side effects of antiretroviral medications in combination with known patient drug resistance and distribution challenges demand further research into the development of more practical and effective alternatives. Two therapeutic techniques with the potential to deliver potent antibodies to protect against HIV infection, passive immunization and gene therapy, have emerged as novel alternatives to antiretroviral drug delivery. Here, we explore antibody reagents that may be utilized in such treatments, specifically broadly neutralizing antibodies (bNAbs), which have reported HIV-neutralization capabilities. Rather than binding to a single antigen, bNAbs bind to a wide range of antigens, expanding neutralization applications to include viruses within a single family that mutate rapidly. While the ability of bNAbs to accommodate binding to multiple antigens makes these antibodies especially effective against a large panel of rapidly mutating HIV viral strains, the elevated polyreactivity of engineered bNAbs introduces disadvantages by reducing half-life in vivo. In order to develop an optimized antibody, it is imperative to increase half-life and stability to preserve antibody HIV-neutralizing abilities not only under short-term laboratory conditions but also in extended clinical applications. By employing glycoengineering and recombinant PEGylation techniques, we investigate methods to increase half-life and stability of bNAbs while maintaining elevated neutralization levels to create optimized antibodies with effectual and comprehensive HIV-neutralizing abilities in vivo. Tuning Strand Binding Dynamics in Viscoelastic Hydrogels With Non-Canonical Amino Acids Jeff Shen Mentors: David A. Tirrell and Peter B. Rapp Leucine zipper P, known to form pentameric bundles, has been assembled into a triblock copolymer using two zippers connected by an elastin linker. The zipper P – elastin – zipper P triblock (PEP) forms hydrogels when dissolved in aqueous solutions, and Fluorescence Recovery after Photobleaching (FRAP) has been used to probe the dynamics of strand exchange in these physically crosslinked protein networks. We are currently attempting to tune the rate of strand exchange by incorporating non-canonical amino acids homoisoleucine and hexafluoroleucine into PEP through overexpression of Leucyl-tRNA synthetase. Fluorophore is introduced through click chemistry onto the modified PEP proteins and FRAP is performed to analyze the effects of non-canonical amino acids (ncAAs) on strand binding. The ability to tune properties of hydrogels using ncAAs would demonstrateof the broad utility of protein engineering. Studies on the Catalytic Asymmetric Reductive Acyl Cross-Coupling: Ligand Design, Synthesis, and Evaluation Steven Shuken Mentors: Sarah Reisman and Alan Cherney A mild, one-step synthesis of enantioenriched chiral α,α-disubstituted ketones from racemic benzylic chlorides and acyl chlorides using a chiral Ni(II) catalyst and stoichiometric Mn(0) reductant was reported in May 2013 by Reisman and coworkers. In order to expand the scope of this reaction, a protocol must be developed that tolerates more types of substrates. This study focuses on manipulating the identity of the ligand used in the reaction, specifically chiral bis(oxazoline) (BOX) ligands with phenyl substituents at the 4- and 4’-positions and cycloalkylidene linkers of various ring size and ring substitution. Attempts toward efficient, scalable syntheses of these reagents are described. The synthesized ligands give a wide range of yields and enantioselectivities when used in the cross coupling. It is suggested that the ligands that perform well with this model system be used in the attempted cross-coupling of otherwise inactive substrates. Phylogenetic and Phenotypic Diversity of Microbial Isolates From the Mars Exploration Rover and Phoenix Fairing Missions Garrett Smith Mentor: J. Nick Benardini NASA missions to bodies such as Mars are required to contain a bioburden under a given threshold in order to launch, which is assessed by the NASA Standard Assay. Viable microbial isolates from spacecraft components and clean rooms of the Mars Exploration Rover (MER) and Phoenix missions were cultivated and archived. Over 100 MER and 40 Phoenix Fairing (PF) isolates were revived from frozen stocks analyzed via 16s rDNA sequence

119

  alignment to understand diversity, and analyzed for phenotypic characteristics (for example, Omnilog biochemical tests) to discover carbon utilization to determine potential survivability on Mars. In past studies, species surviving Dry Heat Microbial Reduction often form spores and are generally found within the genus Bacillus or Paenibacillus. The organisms collected by the standard assay display a wide range of carbon utilization and biochemical potentials. Additionally, the isolates were preserved for both working stocks in a cryobead format and long-term frozen glycerol storage. The results of this study elucidate genetic and metabolic diversity of defined and novel organisms present on the MER and Phoenix space craft. This improved understanding can be directly used to enhance sterilization techniques that NASA uses for Mars-bound missions to prevent contamination as we explore the planet. Structural Characterization of SIRT1 by Substrate Interaction Casey Sondgeroth Mentors: André Hoelz and Ferdinand Huber The significant increase in human life expectancy over the past century and the concomitant demographic age shift in the population confront us with great medical challenges. To meet these challenges through therapeutic intervention, the mechanisms that determine healthy aging must be understood at the molecular level. Sirtuins are a large family of protein-modifying enzymes highly conserved throughout bacteria, archaea, and eukaryotes. Silent information regulator type 1 (SIRT1) was originally identified as a nuclear NAD+-dependent histone deacetylase, the activity of which is closely associated with lifespan under calorie restriction. The controversial life span increases associated with calorie restriction have been linked to the elevated activity of sirtuins in both rats and human cells, and the debated health benefits of red wine consumption are attributed to SIRT1 activation by transresveratrol, a plant polyphenol found in wine. Despite progress over the last decade, relatively little is known about the regulatory mechanism of SIRT1. We have carried out a comprehensive structural and functional characterization of various human SIRT1 activity states. By obtaining atomic resolution structural information, we have clarified the mechanisms through which SIRT1 selects various biological substrates. This information will be invaluable from a pharmacological perspective in the development of novel prospective therapeutics. Engineering of a Thermostable 1,8 Cineole Synthase Meaghan C. Sullivan Mentors: Frances Arnold and Thomas Heel Terpene synthases are enzymes that catalyze the formation of terpenes, the largest and most diverse class of natural small molecules found in all kingdoms of life. Within this enzyme family, monoterpene synthases cyclize a broad variety of industrially valuable molecules such as pinene, limonene and 1,8 cineole3. 1,8 cineole, also known as eucalyptol, is used in many applications from cosmetics to cough medicine. However, most terpenes are purified from plant extracts, using steam distillation1. A potentially more economical alternative would be recombinant production from a heterologous host such as E. coli. In order to do so, the stability of this enzyme must be improved. This would increase the evolvability of 1,8 cineole synthase, allowing the exploration of additional applications in the future. The Arnold lab has developed a high-throughput screen to enable directed evolution. For this screen, a surrogate was developed which mimics the natural substrate. This allows for easy identification of improved mutants of an error-prone PCR library by a colorimetric assay2. In addition to this method, semi-rational approaches have been examined as alternatives for more rapid engineering of thermostability. References 1. Herzi, N., Bouajila, J., Camy, S., Cazaux, S., Romdhane, M., Condoret, J.S. 2013. Comparison between supercritical CO2 extraction and hydrodistillation for two species of eucalyptus: yield, chemical composition, and antioxidant activity. Journal of Food Science. Vol. 78, 5,C667:C672. 2. Lauchli, R., Rabe, K., Kalbarczyk, K., Tata, A., Heel, T., Kitto, R., Arnold, F., 2013. High-throughput screening for terpene-synthase-cyclization activity and directed evolution of a terpene synthase. AngewadteChemical International Edition. V52:21:5571-5574. 3. Oldfield, E., Lin, F. 2012. Terpene biosynthesis: modularity rules. Angewandte Chemiscal Internation Edition. 51: 1124-1137. Developing a Novel Platform to Evaluate Immune Cell Functionality for Cancer Immunotherapy Julia B. Sun Mentors: James Heath, Jing Zhou, Jing Yu, and Alex Sutherland Metastatic melanoma is the most dangerous form of skin cancer. In the U.S., the number of melanoma cases diagnosed each year is increasing faster than any other cancer. One promising therapeutic strategy is adoptive cell transfer (ACT) of genetically engineered T cells. While the initial results for this approach were encouraging, Phase II clinical trials discovered that patients exhibit therapeutic resistance to ACT therapy despite the persistence of activated antigen-specific T cells. It is hypothesized that time-dependent functional changes of these T cells lead to the variable responses to treatment. In order to uncover these changes, the single-cell barcode chip (SCBC) was developed. This high-throughput microfluidic device allows for rapid single cell detection of up to 20 different secreted proteins. To better understand the functional complexity of the transferred immune cells, we have

120

  expanded the protein panel to study cytokines associated with Th17 cells, which are a subset of T helper cells important in autoimmune diseases. New DNA-antibody conjugates were synthesized that target cytokines related to Th17 cells including IL-21, IL-22, and IL-23. Furthermore, to increase the sensitivity of the DNA-antibody conjugates for patient testing, the conjugation process was optimized with respect to reaction time and temperature. Synthesis of Capture Agents Specific for the E17K Mutation of AKT1 Jeremy Work Mentors: James R. Heath and Kaycie M. Deyle Protein-catalyzed capture agents (PCCA’s) are a novel class of molecules that serve as a drop-in replacement for monoclonal antibodies. PCCA’s improve upon the function of monoclonal antibodies with greater thermal stability, lower batch-to-batch variability, and higher selectivity and specificity for the target protein. Due to their specific binding capabilities, PCCA’s can act as protein inhibitors. An enticing protein target for inhibition is Akt1(E17K), a mutation that can induce cancer formation. I have worked to synthesize three capture agents of varying complexity whose ligand sequences have been screened for strong specificity for Akt(E17K), and low affinity for wildtype Akt1. Once synthesized, I can test their ability to bind and inhibit Akt1(E17K) in cells. If effective, these capture agents could serve as a non-invasive and non-toxic means of chemotherapy. Maternal Infection Perturbs Fetal Brain Development Through IL-6 Signaling Zihao Yan Mentors: Paul Patterson and Wei-Li Wu Epidemiologic studies identify maternal infection as a risk factor for autism and schizophrenia in the offspring. To understand the mechanism of how maternal infection alters fetal brain development, we use a mouse model in which injection of the viral mimic poly(I:C) at mid-gestation elicits maternal immune activation (MIA). MIA offspring exhibit autistic- and schizophrenia-like behaviors as well as neuropathology characteristic of each of these disorders. Prior work demonstrated that maternal interleukin-6 (IL-6) is a key cytokine that mediates the effects of MIA on the fetus. The receptor for IL-6, IL-6Rα, is expressed in brain regions associated with pathology in autism and schizophrenia. To examine where this cytokine acts, we are studying patterns of activation of IL-6 signaling in the fetal brain and the placenta. First, we demonstrate that IL-6 mRNA and protein increase in fetal brain 3 hours following maternal poly(I:C) injection. Second, we find that highly selective regions of the embryonic brain respond to MIA: phospho-STAT3 (pSTAT3) immunostaining is elevated in the hindbrain and cerebellar primordium, and laser capture micro-dissection analysis of the JAK/STAT signaling pathway yields results consistent with the immunostaining. To further examine the role of IL-6, we have generated brain-specific and placenta-specific IL-6Rα knockout mice. We can study whether IL-6 activation and autistic- and schizophrenia-like behaviors can be prevented by knockout of IL-6Rα in either of these two specific regions. Gender Effects on Bargaining With Asymmetric Information Lucie Yang Mentors: Colin Camerer and Gideon Nave The project addresses the neuroeconomics question of whether men and women bargain differently. Women have been shown in previous studies to be more risk-averse, less overconfident, and more likely to sacrifice selfinterests to maintain harmony – qualities that should have obvious effects on economic behavior. The present study utilizes a simple economic game paradigm simulating bargaining with private information between an “informed” player (player who knows the size of the endowment) and an “uninformed” player (player who does not know the size of the endowment) to examine differences in behavior and related brain activity via two experimental templates: (1) electroencephalography and Vicon camera motion tracking and (2) Z-tree. Minimal research has been done on the bargaining strategies of men and women, so this study looks to expand on the current intra-gender bargaining research.

121

LIGO THE LASER INTERFEROMETER GRAVITATIONAL-WAVE OBSERVATORY

L I G O

  Spectral Line Monitoring Tool Evan H. Anders Mentors: Greg Mendell and Rick Savage We have created a 'Spectral Line Monitoring Tool' (SLM) in the Python programming language which tracks amplitude, phase, and power spectral density at specified frequencies in LIGO data channels. The program uses Fast Fourier Transforms to cast time domain data into the frequency domain to determine characteristics of sinusoidal variations within the data. The uncertainty of SLM's amplitude measurements for real LIGO data has been studied. Additionally, LIGO's noise has been determined to reasonably follow a Gaussian Distribution; frequency-dependent variations from such a distribution are shown. After development, SLM was utilized to investigate calibration lines in LIGO data from the Sixth Science Run. Intensity Stabilization of NPRO Lasers in the LIGO 40m Lab Charles Blakemore Mentors: Rana Adhikari, Nicolas Smith-Lefebvre, and Jenne Driggers Interferometric gravitational wave detectors such as the Laser Interferometer Gravitational Wave Observatory (LIGO), are some of the most precise instruments used today, detecting gravitational strains as small as 10-22. However, there are many significant sources of noise when measuring such small displacements, including fluctuations in the output power of LIGO’s cavity lasers. In an effort to reduce these fluctuations, this project aims to design and construct an Intensity Stabilization Servo (ISS), a feedback control system that reduces laser intensity noise. By comparing the unstabilized laser power spectrum to the noise spectrum set by the thermal fluctuations in the mirror coatings used on the interferometer’s test masses (considered a theoretical limit), it was possible to calculate an open-loop gain requirement for the servo. To meet this requirement, an analog circuit was designed and prototyped on a breadboard while its transfer function was determined using a spectrum analyzer. With the filtering behavior of the analog servo verified, the design schematics were transferred to a printed circuit board (PCB) layout. The completed board will be tested, and its behavior optimized, in both LIGO’s 40m prototyping interferometer as well as other LIGO facilities on the Caltech campus. Modeling Mirror Shape to Reduce Substrate Brownian Noise in Interferometric Gravitational Wave Detectors Emory A. Brown Mentors: Yanbei Chen and Matthew Abernathy This is a report on the effect of varying mirror shape upon Brownian noise is the test mass substrate. Using finite element analysis, it was determined that by using frustum shaped test masses with a ratio between the opposing radii of about 0.7 the frequency of the principle real eigenmodes of the test mass can be shifted into higher frequency ranges. For a fused silica test mass this shape modification could increase this value from 5951 Hz to 7210 Hz, and in a silicon test mass it would increase the value from 8491 Hz to 10262 Hz, in both cases moving the principle real eigenmode to a frequency further from LIGO bands, reducing slightly the noise seen by the detector. The Brownian thermal noise in the optimized substrate was calculated in order to demonstrate that it is lower than in a cylindrical substrate. Extraction of Gravitational Wave Signatures From Highly Magnetized Core-Collapse Supernovae Adam Bruce Mentors: Christian Ott and Philipp Moesta An extreme class of core-collapse supernovae, so-called "hypernovae", are hypothesized to be driven magnetorotationally by a combination of rapid rotation and ultra-strong magnetic fields and may explode extremely asymmetrically. The extreme conditions in the engine of such hypernovae are expected to lead to copious gravitational wave emission, which may perhaps be detectable out to megaparsec-distances with the upcoming generation of Advanced LIGO interferometers. We analyze gravitational wave signatures of magneto-rotationally driven core-collapse supernovae, which have proposed links to obseved hypernova, taking into account magnetohydrodynamic effects, and compare these results to extractions where no such effects are present. Modeling the Calibrated Response of the Advanced LIGO Detectors Charles L. Burks Mentors: Alan J. Weinstein and Jameson G. Rollins The goal of LIGO is to detect and study gravitational waves, which are as yet undetected fundamental predictions of general relativity. Specifically the work focused on calibrating the gravitational wave strain data channel on the new Advanced LIGO detectors. This project is to develop a model focusing on the optical response part of the sensing chain element. These models are created with high attention to detail from real world disturbances. LIGO deals with gravitational waves that produce movement several orders of magnitude smaller than an atomic nucleus in the detectors. For precision of these measurements to be possible, understanding of the effect of external

125   

  sources is of paramount importance. The end result of the project was a thorough examination of how real world changes to the detectors will affect the calibration in an error propagation study. This will improve the performance of the detector when brought into operation as well as ensure that real world changes are measured well enough to achieve accurate calibration. Development of a Real Time Front-End Time-Domain Calibration System for the Advanced LIGO Detectors Craig Cahillane Mentors: Alan Weinstein and Jameson Rollins The Advanced LIGO gravitational wave detectors are currently under construction along with new, more precise calibration control systems. Our goal is to understand the Advanced LIGO detector’s response to incident gravitational waves. During detector operation, parameters characterizing the detector can change in real time, altering the detector’s response to gravitational waves. These changes in the detector response must be tracked and taken into consideration when calibrating gravitational wave strain. We propose instituting a real time frontend time-domain calibration system in order to better track the changes of the detector response during science runs. A prototype of this calibration system is underway at the Caltech 40m interferometer. Simulating the Advanced LIGO Interferometer Using the Real Control Code Juan F. Castillo Mentor: Joseph Betzweiser The interferometer of the advanced LIGO requires complicated software controls to operate. To better understand both the hardware and software, we should apply the same software used to control the interferometers in a simplified simulation. This project will incorporate a simulation of the control code to the seismic isolation and suspension models. By using the actual control code, we will be able to better discover issues specific to that code and distinguish them more easily from issues that may arise in the hardware of the interferometers. The ability to troubleshoot the overall system by using a simulation model is critical in effectively dealing with present and future problems. Distributing Frequency Stabilized Light via Fiber Optics Erica Chan Mentors: Rana Adhikari and Tara Chalermsongsak A laser beam with low frequency noise is desired in LIGO labs to act as a reference signal. The purpose of the project is to distribute a laser that has been frequency stabilized to a Fabry-Perot cavity to various labs via fiber optics and characterize the noise added by the fiber. To distribute the laser, we have mode matched the laser beam into the fiber in order to obtain the highest power output. The noise from the fiber would first be measured using self-homodyne detection. If the noise level is satisfactory, we would distribute the beam and measure the noise using a phase-locked loop. If the noise level above the labs’ requirements, then various techniques will be used to reduce that noise. Here, we outline the techniques used to mode match and determine the noise from the fiber and laser beam .

Ly  Lx  Laser  BS 

Photodetector

126   

  Designing a Coating-less Optical Cavity Based on Total Internal Reflection Deep Chatterjee Mentors: Koji Arai and Matthew Abernathy Gravitational Wave astronomy requires ultraprecise measurements of the miniscule changes in length scales, far smaller than the size of an atom. Present day laser interferometry promises to deliver the desired accuracy; However, noise from various sources is a major barrier towards this goal. In fact, the precision required for detecting strong signatures is to the level where the spontaneous thermal fluctuation in the test masses result in ‘Thermal Noise’ in the interferometer output. This project aims towards noise reduction by developing an optical cavity without the high reflecting coating layers and resorting to Total Internal Reflection (TIR) instead to obtain high reflectivity. Finite Element Analysis (FEA) using COMSOL and MATLAB is employed to calculate the Thermal noise for specific geometries. The results are crosschecked with certain analytic results for the correctness of the model. It is expected that suitable design and materials brings cancellation effects between the Thermo Elastic (TE) and Thermo Refractive (TR) noise. An Automated Photodiode Frequency Response Measurement System for LIGO Alexander Cole Mentor: Eric Gustafson LIGO will detect gravitational waves using laser interferometers that will be quantum noise limited over most of the apparatus's operating frequency range. To build an interferometric gravitational wave detector that works at the limits set by quantum mechanics, one must ensure that the detector can be controlled and read out optically. In the LIGO interferometers, several photodiodes are used to sense various degrees of freedom and provide feedback signals so that the cavities are in optical resonance. In addition, the main interferometric gravitational wave signal is read out with a photodiode. It is thus necessary to treat the photodiode and its readout electronics as systems whose performances, including frequency response, can change over time and with changing operating conditions. This project’s purpose was to build an automatic frequency response measurement system for the interferometer’s photodiodes. We use a modulated diode laser coupled through a fiber optic distribution system to illuminate the photodiodes, and then automatically and quickly measure the frequency response of each photoreceiver using a network analyzer and an RF switch to select the photodiodes one after another. The experiment was carried out at Caltech on the LIGO 40m prototype interferometer and designed with Advanced LIGO scalability in mind. Testing Fully Dynamical Adaptive Mesh Refinement in the Einstein Toolkit Cutter A. Coryell Mentors: Christian D. Ott and Roland Haas The explosion mechanism of core-collapse supernovae is not yet fully understood. Two candidate mechanisms exist: magnetically-driven explosions and neutrino-driven explosions. Initial simulations with spherical (1-dimensional) and axial (2-dimensional) symmetry did not yield robust explosions, despite including neutrinoheating in their models. Instead these simulations saw the post-bounce shock stall, never making it to the surface of the star to cause explosions that would be visible from Earth. Current 3-dimensional general relativistic simulations, such as the one featured in Ott et al. 2013, study the explosion mechanism in greater detail to draw conclusions about the supernova from the observed signal. To do so, the shock must be simulated with highresolution, but increasing the resolution of the entire simulation is too computationally costly. We develop a routine in the Einstein Toolkit that finds the location of the shock at any time and applies adaptive mesh refinement to resolve the shock surface without resolving unimportant regions. Influence of Differential Rotation and Precollapse Angular Momentum on Stellar Collapse and Its Gravitational Wave Signature Alexandra M. DeMaio Mentors: Christian Ott and Ernazar Abdikamalov Core-collapse supernovae (CCSNe) are theorized to produce gravitational waves that can be detected by observatories such as Advanced LIGO. We present results from detailed 2D, axisymmetric, equatorially symmetric, general relativistic simulations of core-collapse supernovae in the bounce and post-bounce phases, investigating the effects of precollapse differential rotation and central angular velocity on post-bounce stellar dynamics and gravitational wave signals. We conclude that the distribution of angular momentum in CCSNe only plays a significant role in the dynamics and gravitational wave signals of CCSNe from rapidly rotating progenitor stars. Principle component analysis via singular value decomposition is used in conjunction with Bayesian Model Selection to assess the potency of observed disparities among the model waveforms.

 

127   

  Investigations of Consistency in LIGO Open Science Data Ashley Disbrow Mentors: Michele Vallisneri and Roy Williams The LIGO Open Science Center is preparing to release the LIGO archival data set to the public, and is developing software, tutorials, cookbooks, wikis, and teaching materials that guide the user through working with this data. We create and document a worked-out example of accessing and analyzing the LIGO Open Data Set to search for compact-binary-coalescence signals. To do so, we generate compact-binary-coalescence templates using wellmodeled waveforms and perform a matched-filter search to identify LIGO hardware injections. We test for accuracy by comparing the injection signal-to-noise ratio we detect to the signal-to-noise ratio we expect to find. Characterising Behavior of LIGO Subsystems Through Noise in the Seismic Environment Scott S. Dossa Mentors: Alan J. Weinstein and Jameson G. Rollins ALIGO's seismic isolation system aims to attenuate seismic noise that affects the sensitivity for gravitational wave detections. This system needs to be tested and to have each of the installed systems characterized in such a way that problems can be quickly found. By tracking coincident non-Gaussian transients (glitches) through the subsystem's chambers, various statistical analysis including coincident glitch rates and average signal to noise ratios can be compared to determine which chambers are glitching most and which are not successfully attenuating noise. Problematic areas can also be found by finding which chambers are creating glitches most often and how these glitches propagate through the system. This information can be then used to localize problems for aLIGO's commissioners at the detectors in order to aid in their mitigation. Improving the Sensitivity of Advanced Gravitational Wave Detectors With Time Dependent Squeezing Mariel Freyre Mentors: Larry Price, Leo Singer, and Nicolas Smith-Lefebvre The endeavor to directly observe gravitational waves stands as one of the most exciting feats of engineering and data analysis of the century. Gravitational waves are products of incredibly high energy astrophysical events, but they are weak and require incredible precision and care during detection and analysis of collected data. Squeezed light is a potential avenue for increased sensitivity in future LIGO detectors. Filter cavities produce broadband squeezing, but they are costly and introduce potential for more optical noise. Time dependent squeezing, where the squeeze angle is adjusted in real time, may lead to better detector sensitivity to signals sweeping through the band. This work seeks to provide a proof of concept of the effectiveness of time dependent squeezing. Extracting Astrophysical Parameters From Gravitational-Wave Observations Karla Guardado Mentors: Alan Weinstein and Vivien Raymond LIGO (Laser Interferometer Gravitational-Wave Observatory) is engaged in the search for gravitational waves. Compact binary systems, those of black holes and/or neutron stars, lose energy through gravitational radiation during coalescence as predicted by Einstein’s theory of general relativity. We wish to identity these gravitationalwave signatures. I worked on data analysis for LIGO at the California Institute of Technology (Caltech), specifically on parameter estimation. We used a number of post-Newtonian based approximations to describe the coalescences. These approximants are methods of solving Einstein’s field equations, as there exist so few exact solutions. For this project, we focused on a simulated black hole-black hole binary and compared and extracted all approximants and necessary parameters. Determination of these parameters provided insight into the underlying physics of this compact binary system, as well as how important is choice of approximant on parameter recovery for spinning signals, keeping in mind known limitations of each one. Using inference software developed within the LIGO-VIRGO collaboration, we seek to evaluate its performances in regards to parameter estimation. Searching for Spinning Black Hole Binaries in Advanced LIGO and Virgo Deborah L. Hamm Mentors: Stephen Privitera and Alan Weinstein A promising source of gravitational waves (GWs), detectable by LIGO and VIRGO observatories, are coalescing binary black holes (BBHs). Searches for coalescing BBHs employ the technique of matched filtering in which theoretical predictions of the waveform signal are used to filter the data. Therefore, these searches rely crucially on an accurate GW model for detection. BHs in binaries may have significant spins and the effect of spin is encoded in the waveform. Here we explore the benefits of including the effects of spin aligned with orbital angular momentum in the search of GWs from coalescing BBHs. We show that for the advanced LIGO high power zero-detuned sensitivity in the mass range of [50,100]M and for effective spin range [0.0,0.85] the inclusion of spin-aligned effects can increase the visible volume by a factor of two or more.

128   

  Stochmon: A LIGO Data Analysis Tool Gabriela Hernandez Mentor: Eric Thrane Stochmon is a data-quality monitor for LIGO-Virgo stochastic analyses. It provides sensitivity estimates and diagnostic plots to track potentially problematic data-quality issues in real time. With Stochmon, LIGO-Virgo researchers will get a (blind) preview of the results of stochastic analyses, which will facilitate planning while providing useful feedback for detector characterization and commissioning efforts. Detecting Deviations From General Relativity Using Continuous Gravitational Waves Maximiliano Isi Mentor: Alan Weinstein The direct detection of a gravitational wave with the Advanced LIGO detectors provides the opportunity to measure departures from General Relativity. These departures can arise in the speed of the gravitational wave, existence of alternate polarizations and parity violation. To measure these, we can simulate a single detector measurement of a continuous gravitational wave from a well-defined pulsar source, for example the Crab pulsar, due to an asymmetry in the moment of inertia. The speed of a gravitational wave can be measured from the Doppler frequency modulation of the signal, with an accuracy that depends on the strength of the signal. We precisely quantify the achievable accuracy and compare it with other methods of measuring the “speed of gravity”. Furthermore, we extend our previous analysis of gravitational wave polarizations to include new pulsars and LIGO S6 data. Seismic Isolation and Suspension System Instrument Characterization for Advanced LIGO Commissioning Chase O. Kernan Mentors: Alan Weinstein and Jameson Rollins Reaching the intended sensitivity and astrophysical reach of the Laser Interferometer Gravitational-Wave Observatory (LIGO) will require significant effort in characterizing environmental and detector subsystem noise. Identifying the creation and propagation of non-Gaussian ‘glitches’ will aid onsite commissioners in fixing instrumentation problems and achieving nominal equipment performance. We focus on the installed seismic isolation (SEI) and suspension (SUS) subsystems and leverage existing event trigger generators (ETGs) to develop statistical analyses of glitch behavior. We create novel visualization tools for investigative work and detail the nature of glitches originating from the inner stages of the SEI and SUS subsystems. Quantum-Noise Reduction Schemes for Future Advanced Gravitational-Wave Detectors Mikhail Korobko Mentors: Yanbei Chen and Haixing Miao Future advanced gravitational-wave detectors, such as Advanced LIGO, Advanced VIRGO, and KAGRA will be limited by noise due to quantum fluctuations in the light, around the most sensitive detection band ~ 100Hz. There are two different approaches for improving the sensitivity: squeezing or canceling the noise by properly designing the input or output optics of the detector, and increasing the response to the gravitational-wave signal by modifying the test mass dynamics. In this project we take advantage of frequency-dependence of the optical spring to effectively reduce inertia of the test mass with multiple optical springs. Earth Tide Deformations: Prediction and Compensation for Advanced LIGO Noah A. Kurinsky Mentor: Kiwamu Izumi The solid earth tides, deformations in the earth’s shape due to tidal forces from the moon and sun, cause the arms of the aLIGO interferometers to be deformed by up to 200 microns over the course of a day, which far exceeds the locking condition of less than one nanometer of longitudinal deformation. We have developed a program which can predict the deformation introduced into the arms in real time, and have used these predictions to analyze whether existing feedback is capable of suppressing such a large effect. We have also developed a feed-forward system to work with the prediction program to compensate for the effect in real time, and have begun the implementation of such a system in the aLIGO control systems.  

129   

  Working Towards Finding an Upper Limit for Crackle in LIGO’s Maraging Steel Blade Springs Benjamin E. Levy Mentors: Rana Adhikari and Eric Quintero Crackle is a type of nonlinear, high frequency noise generated as individual molecules in a material shift when the entire system is driven at low frequencies. Such unwanted noise may occur in components of LIGO’s gravitational wave observatories. The steel blade springs that hold highly sensitive mirrors in place are of central importance to this project. An optical apparatus was created to determine the upper limit for the intensity of crackle that could occur in such springs. Several improvements were made to the apparatus for the purpose of amplifying the crackle, while decreasing the effects of other background noise sources such as acoustic and seismic vibrations. A detailed model of the data that could be expected from this experiment was created for use in testing the analysis process. New algorithms allowed the level of crackle noise present in a very noisy signal to be extracted. This study will help LIGO to ascertain whether provisions will be necessary to mitigate the effects of this nonlinear noise in its observatories. Development of a Low Noise External Cavity Diode Laser in the Littrow Configuration Chloe Ling Mentors: Rana Adhikari and Tara Chalermsongsak Laser diodes are a cheap way to produce laser light, but they have high noise levels. The noise level can be reduced via optical feedback by locking the laser diode to an external cavity. In this work, we aim to design and build a 1064 nm external cavity diode laser (ECDL) prototype in the Littrow configuration for use as a tabletop laser at the Laser Interferometry Gravitational-Wave Observatory (LIGO). In the Littrow configuration, the external cavity is formed with a diffraction grating and the reflective backing of the lasing material. The diffraction grating reflects first order 1064 nm light back at the laser diode generating optical feedback, and the output beam is formed by the zero order diffraction beam. We outline the factors taken into account during design of the ECDL and how these factors were modeled to affect the final noise levels. Then, we discuss the construction of our prototype ECDL. Finally, we present data demonstrating the noise reduction of the prototype ECDL. Further modifications will be made to this ECDL with lower noise components in order to refine the design for regular use at LIGO. Installation of Physical Environment Monitoring for Advanced LIGO Emily Maaske Mentors: Robert Schofield and Daniel Sigg LIGO uses an interferometer to try and detect gravitational waves. To detect these waves LIGO must be sensitive to as little as 10^-20m, because of this sensitivity there are many environmental aspects that can create noise. That noise can block signals on the gravitational wave channel. The installation of the sensors in the Physical Environment Monitoring system (PEM) will help the interferometer become more sensitive by disregarding signals from environmental aspects and to also help identify excess noise that can then be decreased before advanced LIGO becomes fully operational. We installed all the sensors that were needed and helped to identify problem signals for the half interferometer test (HIFO-Y). We investigated the problem of cross talking between cables in Endevco boxes and using the sensors to look at propagation speeds. We continued to update PEM website with relevant information for each sensor, such as sample spectrum, calibration, and grid location. Magnetic Levitation Approach Giorgos Mamakoukas Mentors: Rana Adhikari and Haixing Miao Seismic noise is one of the most important low-frequency disturbances that limit the sensitivity of advanced gravitational-wave detectors. For isolation, the usual approach has been the multiple-stage pendulum suspension; however an alternative seismic noise isolation scheme is investigated. Magnetic levitation utilizes the symmetry and local extremum of the magnetic force between disk magnets and—by incorporating the sensing and feedback control system—can in principle achieve stable low-frequency isolation of six degrees of freedom. To test this idea, a prototype has been built, though, calibration of the setup and characterization of the coupling among different degrees of freedom remains before acquiring lock for the setup. Dynamics and Gravitational Wave Signatures of Charged Black Hole Binaries Zachary R. Mark Mentors: Yanbei Chen and Huan Yang We analytically investigate the natural gravito-electric oscillations of the Kerr-Newman (KN) solution to the Einstein-Maxwell equations. The quasinormal modes (QNM) are deduced to the first order in Q, the charge of the black hole. Using the NP formalism and the phantom gauge, we obtain two coupled, partial differential equations for the gravitational and electromagnetic radiative degrees of freedom. A second perturbation analysis about the

130   

  small parameter Q is applied to obtain first order corrections to the QNM. We anticipate future work based on this analysis. We will check if future gravitational wave observatories, such as Advanced LIGO and eLISA, will be able to detect the presence of charge in black hole binary systems. We will also use our numerical results for the QNM to address the possibility that the “electric” and “magnetic” QNM are isospectral. Characterization of Second Harmonics Generators for Advanced LIGO Squeezing Andres Medina Mentors: Sheila Dwyer and Daniel Sigg The sensibility of Initial LIGO gravitational interferometers wave detectors was fundamentally limited above a few hundred Hertz by quantum noise, namely shot noise. This noise can be reduced by mainly increasing the laser power, or by implementing squeezed light. Squeezed light is a quantum phenomenon that results in the creation of two correlated photons. Having collated photons reduce the standard deviation in their photon-counting static, which results in a decrease in the shot noise. One essential component in creating squeezed light is a second harmonic generator, which produces green light at twice the frequency of the interferometer laser. The aim of this paper is to characterize the second harmonic generator that might be used in Advanced LIGO. We measure the intra-cavity losses, characterize the locking of the cavity, and characterize the efficiency of conversion from infrared to green light. Directly Comparing Ultra-Stable Lasers With Large Frequency Separations Jonathan H. Mishler Mentors: David Reitze, Rana Adhikari, and David Yeaton-Massey Frequency stabilized lasers for applications in metrology exist at many different wavelengths. Directly comparing the stability of different sources is often desirable. Two sources we are interested in comparing are at 1064 nm and 1550 nm. Since the direct detection of optical beat signals is limited to the GHz scale, we need to employ more sophisticated techniques. The focus of this project is to analyze and compare different techniques available, with the goal of choosing the optimal method to integrate with existing laboratory infrastructure, and to create an experimental layout. A prototype network that utilizes frequency combs, a broadband optical spectrum that can be used to measure optical frequencies, has been drafted. In the network, the stability of one of the sources is transferred to the frequency comb, after which an optical beat signal between a comb tooth and the other source can be detected. 3D Low-Latency Localization of Gravitational Wave Event Candidates Patricia A Porter Mentors: Larry Price, Vivien Raymond, and Leo Singer Advanced LIGO and Virgo sensitivities to gravitational waves enable the prospect of making joint electromagnetic observations of the mergers between binary systems containing two neutron stars or a neutron star and a black hole. Gravitational wave signals can be used to localize their source’s sky position, enabling telescopes to detect the corresponding electromagnetic counterpart. In this project, we assess the accuracy of our information from gravitational-wave triggers about the distance to the source and look at ways of combining this information with the source’s probable sky location for more effective electromagnetic follow-up campaigns. High-Fidelity Initial Models for Neutron Star Simulations Matthew J. Raives Mentors: Christian D. Ott and Mark A. Scheel Precision predictions of gravitational waves from pulsating and merging neutron stars rely on numerical relativity simulations, which, in turn, rely on physically and numerically accurate initial conditions. Specialized codes called initial data solvers are used to numerically integrate the general relativity and hydrodynamics equations to determine the matter and spacetime properties in and around the star. The current single star initial data solver used by our group, CST, is of limited accuracy; we would require a different code if we desired more accurate models. To that end, we have investigated two different initial data solvers in order to determine their suitability as a replacement for CST. These codes are evaluated against each other and against CST by the degree to which they satisfy several constraints, most notably a virial identity (GRV2) which vanishes for axisymmetric, stationary, and asymptotically flat spacetime, conditions which are met for single star models. Future work will entail comparing the different initial data solvers for binary stars and evolving single and binary star models in neutron star evolution code.

 

131   

  Calibration and Analysis of Nanoindentation Data Elaine L. Rhoades Mentor: Matt Abernathy The ultimate goal of this project is to measure the mechanical properties of various optical coating materials. At the California Institute of Technology, these calculations are done using nanoindentation data; at Embry-Riddle Aeronautical University, these calculations are done using acoustic measurements. In order to extract the mechanical properties from the nanoindentation data using the analysis of Oliver and Pharr, it is imperative that the machine compliance and indenter area function are accurately known. These quantities are calibrated by implementing an iterative procedure that calculates machine compliance and indenter area function alternatively and applies them to the data until both quantities converge to their true values. The calibration is performed on fused quartz because its material properties are well known. The Oliver and Pharr method is then used to calculate the Young’s modulus and Poisson’s ratio from the corrected data for alumina (Al2O3), hafnia (HfO2), and tantala (Ta2O5). Results of the calibration procedure are discussed. Values of the Young’s modulus and Poisson’s ratio are presented for each material examined and compared to previously published measurements. It is recommended that the results obtained here be compared with results obtained by the Embry-Riddle acoustics group when available. Controlling Transient Thermal Effects in High-Power Gravitational-Wave Detectors Arnaldo Rodríguez Mentor: Aidan Brooks Detecting gravitational waves from high-frequency emitters (such as black hole collisions) requires gravitationalwave detectors with high operating laser powers; these cause thermal effects in the test-masses that deteriorate control signals and decrease interferometer sensitivity. To correct these, a prototype control system model was developed for Advanced LIGO’s thermal compensation system to reduce optical wavefront distortion from thermal lensing and surface deformation. A finite element model of the heating in the test-mass was developed in COMSOL to interact dynamically with the control system developed in MATLAB, as well as a linearized time-efficient heating model using only MATLAB. We analyzed data from both models and from experiment to verify the system’s behavior and validate the control system model’s integrity. Higher-Order Gravitational Wave Emission in Core-Collapse Supernovae James St. Germaine-Fuller Mentors: Steve Drasco, Christian Ott, and Philipp Mösta Gravitational waves are important messengers that carry information about the multi-dimensional fluid dynamics in the central engines of core-collapse supernovae. So far, most simulations use the so-called quadrupole formalism to extract the waves from the matter dynamics, but recent results suggest that significant emission may also occur at higher than quadrupole order. In our simulation we calculate the gravitational wave signal including both the quadrupole and octupole order terms. By comparing the relative sizes of the first- and second-order terms we obtain a quantitative measure of how well the quadrupole moment represents the gravitational waves. Quality Factor of Crystalline Silicon at Cryogenic Temperatures Edward Taylor Mentor: Nicolas Smith-Lefebvre 3rd generation LIGO detectors will be limited by thermal noise at a low frequency band where gravitational wave signals are expected to exist. A large contribution to thermal noise is caused by internal friction of the mirror and suspension elements. In order to meet the quantum mechanical sensitivity limits of the detector, it will be necessary to further push down the contribution of thermal noise. Future detectors will require new materials with extremely high mechanical quality. Silicon at cryogenic temperatures shows the promise to provide the required mechanical quality due to its vanishing expansion coefficient at 120 K. The fluctuation dissipation theorem links thermal noise to mechanical dissipation which, in turn, motivates us to study the quality factor of silicon cantilevers. Utilizing a ring-down method in vacuum, we determine the quality factor of a silicon cantilever at cryogenic temperatures. Further experiments should be performed to determine how etched silicon affects its quality factor. Commissioning an Angular Sensing and Control (ASC) System for a 40m Prototype Interferometer Gautam Venugopalan Mentors: Manasadevi P. Thirugnanasambandam, Jenne Driggers, and Rana Adhikari Laser Interferometer Gravitational-Wave Observatory (LIGO) detectors are extremely sensitive to the angular orientation of the interferometer test masses. Fluctuations in the angular orientation of these test masses adversely affects laser power build-up in the various Fabry-Perot cavities that constitute the interferometer, and the beam centering on mirrors. In this project, an Angular Sensing and Control (ASC) system is commissioned for the auxiliary laser beam at a 40m prototype interferometer. The system consists of piezoelectrically controlled

132   

  steering mirrors that direct the auxiliary beam into the Fabry-Perot arm cavity. These actuators, once calibrated, are integrated with a software-based remote monitoring and control environment in which a software servo is implemented to feedback to the steering mirrors. This allows stable power build-up in the interferometer and precise input pointing of the auxiliary laser beam into the interferometer’s arm cavities.

133   

NASA/JPL Programs PGGURP Planetary Geology and Geophysics Undergraduate Research Program

SPACE GRANT The National Space Grant College and Fellowship Program

USRP Undergraduate Student Research Program

JPLSIP JPL Summer Internship Program

N A S A

  Sample Integrity and Sealing Methods for Mars Sample Return Katherine Acord Mentors: Paulo Younse and Charles Budney Although extensive progress has been made in the search for habitability on Mars, we have yet to make significant progress in looking for bio-signatures. A potential Mars Sample Return (MSR) campaign would deploy a four stage mission in order to core and store Martian rock cores and eventually return them to Earth for scientific study. The focus of this research project aims to explore the parameters for returning scientifically viable samples back to Earth during the MSR campaign. To address this, we propose: to assess the effects of a high gravity-force landing on Martian-like rocks through drop testing; to determine the limitations of proposed sample return hardware by analyzing results from vibe, pressure, and push-out tests as well as modify the designs for future success; and to verify which method of sample tube sealing would best preserve rock cores during transportation through the use of Helium leak tests. Thus far, we have successfully observed multiple push-out tests of the four sealing methods in question and extrapolated the amount of force and energy necessary for failure to occur; in addition to verifying the suitability of multiple cache drop mechanism designs for future shock tests. The combination of these results would effectively provide functional analysis for determining core selections, modes for sample storage, as well as Earth-return landing methods for MSR. Visualizing Model Based Systems Engineering Elena Agapie Mentors: Scott Davidoff, Hillary Mushkin, and Maggie Hendrie Systems Engineering is critical in planning and executing space missions. It is a long term process requiring the collaboration of large groups of people in updating complex sets of information. We are designing and implementing a system that supports engineers in understanding and interacting with the design and evolution of the data representing space mission states. We design prototypes and an interactive data visualization that supports interactions with a large data set of systems engineering data. We support fluid interactions with graph data and allow focus on relevant information. We assess the needs of users and test our designs with various engineers. SMAP Flight Software Tools Nicolas Ajalat Mentors: Tom Fouser and David Hecox The Soil Moisture Active Passive (SMAP) project is a satellite mission with the goal of gathering data on soil moisture and freeze/thaw state. Although the majority of the project's flight software (FSW) has been written and is in the process of being tested, there is still a need to develop support tools that will interact with the software. The goal of this project is to develop, test, and document some of these tools. Tools developed include decoders for two memory dumps that are created by FSW and sent as data products, as well as a tool that encodes parameter files for the spacecraft's attitude control system. Zip Code Mars: From Classic to Modern Joseph Anz Mentors: Carolina Martinez, Melody Ho, Michelle Viotti, and Sarah Marcotte Zip Code Mars is a website designed for the public to access information on mission personnel working on current and past Mars missions. It showcases the individual team member contributions to Mars Exploration. The purpose of the site is to provide career role modeling, and also make the Martians accessible and recognizable, so that adults, children, and teens can connect to the people behind the missions. My project objective was to modernize the site with engaging and interactive user experiences by utilizing the latest web development technologies. To do this, I used several web programming languages, including HTML5, CSS3, JavaScript, ColdFusion, MySQL, and jQuery. This project entailed a working knowledge of computer science and web design in order to deliver a functional and stylish website for its users. Comparison of XCO Measurements at European TCCON Sites 2

Elaine Arbaugh Mentor: Greg Osterman The Orbiting Carbon Observatory (OCO-2), a carbon dioxide-measuring satellite set to launch in July 2014, will use create a data product, Atmospheric CO2 Observations from Space (ACOS), which shows measurements of dry-air column-averaged mole fraction of CO2 (XCO2). To validate the data product, ACOS data from the Japanese satellite Greenhouse Gases Observing Satellite (GOSAT) using the OCO-2 retrieval algorithm is compared with data gathered from various carbon dioxide detectors, especially the Total Carbon Column Observing Network (TCCON), which is a network of detectors that measure column CO2 and other gas data accurately. In this project, data from TCCON sites in Europe located close together—Bialystock, Bremen, Garmisch, Karlsruhe, and Orleans—were studied. Plots of daily average XCO2 from each site and comparisons of average XCO2 from each site to the overall

137   

  average were created using Python’s Matplotlib library. Comparisons of these plots showed overestimations for certain sites and underestimations for others, and research was done to try to find reasons for these trends. Plots comparing TCCON data and ACOS data from the areas around the five sites were also created and studied. Further research in this area could be used to improve calibration of TCCON instruments and therefore provide better data sets for validating ACOS data. Telemetry-Based Ranging Shwan Ashrafi Mentor: Jon Hamkins The goal of ranging technology is to determine the distance between a ground antenna and a spacecraft. A common approach is PN-ranging where the round-trip-time is measured. We can achieve ranging accuracy within 1 meter. However, a dedicated signaling for the downlink is not desirable as the spacecraft power is a very precious resource. We propose to use telemetry-based ranging which takes advantage of existing telemetry signal and is more power efficient. The spacecraft measures the time between the start of the acquired PN-sequence and the next telemetry code-word. This delay is then embedded in the telemetry signal. In the ground processing unit, downlink delay is measured and combined with the delay measured at the spacecraft to compute round-trip-time. In order to achieve ranging resolutions less than 1m, it is necessary to evaluate the performance of the system and find bottlenecks. We developed an end-to-end simulation of telemetry-based ranging system using MATLAB with a GUI. This simulation tool provides mean-squared-error and transient behavior of tracking loops and plots of desired signals. According to preliminary simulation results, tracking loops are not performance-limiting. Incorporating other factors such as the accuracy of delay measurement at the spacecraft will be the next step to find the bottleneck of the system. Structural Analysis at JPL Patricia AuBuchon Mentor: Darlene Lee In order to test new spacecraft’s and their components a new test bed at JPL is necessary. The test bed environments must be able to simulate extraterrestrial factors that will be encountered by means of gravity, surface lighting, and weather. The structure not only must be able to simulate such environment but also be able to carry instrumentation necessary to study the behavior of the R&D projects. In order to help in this undertaking, I will be running FEM analysis on different components and designs. At least two test support structures will be built with one indoors and the other in the Mars Yard. I will perform structural analysis and simulations for these structures and also the soil mechanics for the system in the Mars Yard. The Mars Yard has loose soil that I will need to perform geotechnical checks on to ensure the test bed is properly anchored in place. Other analysis work was on the SMAP satellite, components for the LDSD testing, and a Matlab script to rewrite Fortran codes for dynamic analysis. Overhauling the Scarecrow Rover’s Flight Software Riley Avron Mentor: Matt Heverly The Mobility Systems Test Bed rover, unofficially called Scarecrow, has used driving software based on the TCL scripting language which implements the bare minimum of features. In an effort to add functionality and eradicate bugs, its software implementation has been ported to a framework based on the flight software of ATHLETE, known as ASAP. ASAP allows separate streams for uplink and downlink, runs as compiled C code, and supports true asynchronous communication. It should also be a less daunting challenge for future control systems to be used with ASAP, given that it communicates over standard Unix-style sockets. As a result, the system is more responsive and fault-tolerant, maintains functionality, and adds features, while conforming more closely to a labwide model for flight software protocols. High-Throughput Electrochemical Characterization of Fuel Cell Catalysts Saad Azam Mentor: Charles Hays Much effort has been invested into research on developing catalysts on low temperature polymer electrolyte fuel cell over the last two decades. Platinum metal is a widely used material for these catalysts, but due to high cost and insufficient availability, new catalysts are being developed. In an effort to reduce the platinum group metal (PGM) loading for the cathode in hydrogen-air fuel cells, JPL has examined a wide range of alloy compositions where Pt metal is alloyed with valve metals (Ti, V, and Zr) and the late transition metals (Co, Ni). The results of the electrochemical measurements, voltage swept cyclic voltammograms and potentiodynamic polarization curves, were investigated using the rotating disk electrode (RDE), and the novel multi-electrode electrochemical test system developed at JPL. The samples that were studied were Pt RDE, PtTF, PtCoZr, and PtNiTi. In each system, measurements of the active area, revealed by measuring the area associated with the hydrogen-oxidation-reaction (HER) at anodic potentials were conducted; along with polarization curves conducted to determine the oxygen-

138   

  reduction-reaction (ORR) current density at 0.9 V, vs. NHE). The Pt RDE data was analyzed as a function of vertex potential and bath temperature. As the temperature increased, cyclic voltammograms, higher active areas and more active structure reconstructions were observed. At 30 C, and low vertex potentials, the sample exhibited a reduced area, with only the (111) crystal crystal surfaces of the polycrystalline RDE specimen being observed, while at 60C, under the same driving potential, the surface had (110) and (100) reconstructed surfaces formed. Also, as a function of potential vertex, the highest potential (1.1 V) had the highest active area. The next sample studied was an array of PtCoZr thin films, with up to 30 atomic % Zr, and as little as 56 atomic % Pt, present in the alloy. This five year old sample was measured again for its electrochemical activity. This sample exhibited outstanding electrochemical activity in its original condition (110µA/cm^2 ORR, > 10X larger than (111) Pt thin films), and large surface areas (250µC/cm^2, ~ 2X > (111) Pt.). Previous measurements showed reproducible results with up to one year or more between measurements. X-ray photo-electron-spectroscopy (XPS) measurements showed that the electrochemically treated surface was dominated by Pt-Zr surface layers, with Co, being leached from the surface, and/or reconstructing below the upper surface. The ozone ashing treatment used to prepare the surface for XPS measurements resulted in a 3X loss of electrochemical ORR activity, due to the formation of complex oxides at the specimen surface. During five years of measurements, this sample underwent aggressive durability testing; e.g., thousands of cycles up to 1.6V NHE. Regardless of the extreme conditions, this sample still shows decent ORR activity (40µA/cm^2, still > 4X larger than (111) Pt), and active areas (230µC/cm^2), comparable to polycrystalline. From this we can conclude that this sample is still active and very durable to harsh conditions. The results of PtTF and PtNiTi will be shown in the presentation. Magnetic Shielding for the Cold Atom Laboratory Luci Baker Mentors: Anita Sengupta, Dave Aveline, and David Conroy The goal of the Cold Atom Laboratory (CAL) Mission is to provide a multi-user facility for studying Bose-Einstein condensates (BEC’s) in the microgravity environment of the ISS. Laser cooling and magnetic trapping techniques are used to form and maintain the BEC, and the CAL instrument will be available to research institutions to investigate the quantum nature of matter. Attenuating external magnetic fields is critical to extending the lifetime of the BEC. This is done with high-permeability mu-metal shields, which absorb and reroute the magnetic field around the instrument. The shields were designed by creating magnetic models in ANSYS Maxwell 3D. The shield geometry was optimized to attenuate the magnetic field strength at the most sensitive parts of the instrument to 1 mG while conforming to volume, mass, and assembly constraints. Lead Soldering Review and Surveys Yazmin Baltazar Mentor: David Lomeli Soldering is a technique used to join metal pieces and parts. This technique uses heat and filler metal to join the pieces of metal. The potential exposure for serious health effect is the filler metal that contains lead. The health effects may include…. Many workers at JPL use lead solder for soldering. Overexposure to lead is one of the leading causes of work related illness. My project consists of identifying and locating workers in soldering operations throughout the JPL facility and ensuring their exposure to lead is below Cal/OSHA standards. Through this survey, I am determining whether or not workers have the potential to be exposed to lead at or above Cal/OSHA permissible exposure limits (PEL) during soldering activities and if they require a need for personal exposure monitoring. Throughout my investigation, I have located several workers who warrant exposure monitoring. I conducted personal and area sampling using NIOSH and/or OSHA Methods to determine the amount of worker’s lead exposure. My samples were sent to an accredited lab to detect the level of lead. Previous exposure monitoring results from lead samples were none detect or below the Cal/OSHA PEL. The sampling results that I have conducted read levels below the Cal/OSHA PEL. If samples were to read above Cal/OSHA PEL, control measures will have to be implemented. Overall, my project consists of ensuring the safety of the workers in soldering operations by determining their level of lead exposure. Deep Space Network: Antenna Motion Protection Guadalupe Banales Mentors: Daniel Olson and Neil Bucknam The DSN is the largest sensitive scientific telecommunications system in the world. It supports selected Earthorbiting missions, interplanetary spacecraft missions, and radio and radar astronomy observations for exploring the universe. It’s crucial for antennas to receive constant mechanical and structural design support. This internship focused on documenting the sector switch assemblies for 34-M Beam Waveguide (BWG) antennas, which would help identify the best design for a new antenna. Sector switches located near the mechanical wrap in the antennas are used to determine azimuth rotation orientation as related to neutral positioning and with cable wrap final limits. Issues arise if the sector switch fails, which requires trained individuals to manually evaluate the failure mode and assess position. If a failure occurs in an ambiguous position, then the antenna's clockwise (CW or CCW) hardware

139   

  limits may be deactivated. Diagnosing correct antenna positioning may be complicated due to absent correct direction indication labels. This has previously resulted in technicians turning the cable wrap in the wrong direction, which led to cable wrap damage. A redesign of the sector switch assembly and new labels are part of the overall recommendations by the review board that identified methods to prevent future cable wrap damages. Visualizing Turbulent Fluid Dynamics Daniel Barella Mentors: Scott Davidoff, Hillary Mushkin, and Maggie Hendrie The study of turbulent fluid flow is a complex and developing field of interest in the realm of theoretical and experimental physics as well as applied sciences and engineering. However, in lieu of advanced computational modeling techniques and the advent of readily-accessible supercomputing, we have only a minimal understanding of the mechanics underlying turbulent fluid flow. Working with Caltech's McKeon Research Group, we created an interactive visualization of a novel mathematical model of turbulent flow developed by Dr. Beverley McKeon et al. The visualization may be used both to demonstrate and to explore the consequences of this theory by observing fluid flow deconstructed into individual constituent modes, and then by recombining modes into a full turbulent model. Prototype Analytics Tools Bianca Barr Mentors: Robert Witoff and Gabriel Rangel Within the last 10 years, we have gained the ability to accrue and store unprecedented amounts of (big) data at a very low cost. In order to make JPL an increasingly competitive and data-driven research center, it is necessary to have the correct tools to analyze and visualize this big data. Appropriate analytics tools were identified for use inside JPL through a trade study. Next, the data visualization tool Tableau was prototyped by creating an interactive workbook that gives end users and non-technical persons the ability to explore and analyze institutional charging data. Tableau Server, a business intelligence application, was then utilized to provide business administrators and a director access to the interactive tool through the web and iDevices. Furthering this investigation of tools, self-hosted web analytics tools were then studied in order to understand the usage of numerous internal webpages. Through the trade study it was found that Piwik, an open-source and self-hosted option, would be appropriate for use in JPL. Piwik was then prototyped in Amazon’s GovCloud and implemented for the NASA Engineering Network. MSL Tool Development for Data Analysis Cyrus-Jan Batino Mentor: Matthew Heverly Mars Science Laboratory (MSL) rover, Curiosity, is currently exploring the surface of Mars. There are many groups working in association with the rover, such as the rover planners. The rover planners use the Robot Sequencing and Visualization Program (RSVP), a software suite, to build and simulate robot drive sequences for the Mars Science Laboratory. If the simulated sequences result positively, the sequences are sent to the vehicle allowing the rover to drive and record data of its performances. The rover planners must submit a full report of the uplink data sent to the rover at the end of each sol, a Martian solar day, to the Mars Science Laboratory Reports page, viewable for all teams affiliated with the Curiosity project. The current uplink reports are cluttered with unnecessary information and are organized in a way that it is not aesthetically pleasing when read by a human or a computer. In order to maintain the success of the Curiosity project, the uplink reports page will be revamped containing tables of useful information, maps of drive path, videos of drive path, etc. This will be done through the parsing of data outputted by RSVP. The project, developed under Matthew Heverly, will support the continuing success of the Mars Science Laboratory. Design, Implementation, and Verification of iSwitch Core for the ISAAC Technology Porter Beus Mentor: Yutao He With the ever increasing complexity of demands on Earth and Space Exploration instrumentation comes a greater cost in development. Due to the unique, individualized nature of these instruments much of this increased cost comes in the form of nonrecurring expenses and repeated efforts. ISAAC was envisioned to remedy this situation by providing a standardized development process along with a library of tested and verified modules to be used in advanced FPGA designs. The iHub is one of these modules or cores and has been designed to provide a common and reusable method for dynamically transferring data between otherwise incompatible instrument protocols. The dynamic switching comes via the iSwitch Core and allows effective inter-system communication by creating

140   

  dynamic, nonblocking, point to point connections post-synthesis and by avoiding useless connection logic which many switching methods imply. The prototype of iSwitch module has been implemented, verified by simulation and through synthesis, and found to effectively carry out its design objectives and has been used as part of the iHub project. Web Development for the Cassini Mission Samantha Bielli Mentor: Diane Conner There is a large framework of web pages and online tools that support the Cassini satellite’s mission to explore Saturn. This project consists of several smaller parts, all centered around the goal of improving these online operations. One featured aspect of this project is the refactoring of an existing web-based file exchange system to increase security and comply with regulations. Another part is the creation of new tools to make previously tedious tasks much easier and user-friendly, including a website to monitor the connection status of important servers, and another to monitor the CPU and memory usage of several different workstation computers. In its totality, this project is designed to employ applications that make useful data accessible, while continuing to maintain security standards. Teleoperation of Simulated Robotic Systems Luis Bill Mentors: Steven Myint and Bob Balaram Remotely operated robots can complement humans during exploration missions. The main concept being explored in this project involves efficient methods to teleoperate robotic systems in and/or from outer space. Advances in 3D graphics, and software libraries to develop graphical user interfaces, allow for the development of interfaces to better teleoperate and monitor robotic systems. One type of such robotics systems may include robotic spacecrafts located at Lagrange points, which will interact with near-Earth asteroids that have been captured and brought to a Lagrange point. These robotic systems will be teleoperated from a remote operator station, located either at a Lunar Lagrange point (L2) halo orbit or on Earth. The goals of this project involve developing alternative teleoperation interfaces, while also extending to a client-server architecture to support low/high bandwidth operations. The current architecture will allow remote clients to visualize telemetry data to analyses spacecraft condition, and will also allow clients to compare and verify outcomes obtained from software/physical simulations. Crater Hazards in the InSight Mission Landing Site Area Colin Bloom Mentor: Matt Golombek The InSight Mission is a NASA Discovery mission to Mars carrying a geophysical science package including a seismometer, heat flow probe and precision tracking station designed to better understand how Mars formed and differentiated. The InSight Mission has a number of landing site safety requirements that result from the spacecraft design and entry, descent and landing process. Requirements range from elevation and latitude to rocks and slopes. Twenty-two potential landing ellipses were evaluated in Elysium Planitia. Using HiRISE imagery (25 cm/pixel), a high resolution visible camera, and thermal infrared imaging (THEMIS IR Day/Night at 100 m/pixel) with ArcGIS, crater hazards from secondary crater rays and regolith thickness were constrained. Secondary crater rays were identified by dark streaks in THEMIS IR Night and were verified using HiRISE images. Secondly, by identifying and calculating the area of fresh rocky ejecta craters on different terrains, the depth of bedrock was determined and the amount of regolith, loose sediments above bedrock, was inferred. A Low-Temperature Digital 8b10b Encoder and Serializer for Next-Generation Integrated Camera Dylan Bloom Mentor: Colin McKinney The Next-Generation Integrated Camera project (NIC) aims to create a fully integrated 1 Megapixel (MP) CMOS imager chip, operational down to -150°C for Martian and Lunar applications. The motivation for a new design includes a tenfold reduction in size, weight, and power consumption over previous generation imagers, such as the MER CCD cameras. The NIC project plans to achieve low temperature functionality through using an older, but well documented, IC fabrication process, TSMC 0.35 um. Because of its larger feature size compared to other more current nanometer processes, TSMC 0.35 um resists the negative effects of cold temperature on electronics and allows the use of +5 Volt logic levels, both of which equate to better imager performance in the target extreme environments. One key component of NIC is an 8b10b serializer. The 8b10b serializer converts bytes of raw data from the imager’s photodiode pixel array to a DC balanced serial stream of digital bits using a special algorithm. The NIC2013 test chip, planned for fabrication in August 2013, implements the 8b10b encoding algorithm as a block of combinational logic paired to a shift register/serializer architecture laid out using Cadence automated place and route tools. The benefit of using the 8b/10b circuit is that, combined with a low voltage differential signaling circuit (LVDS), it allows for high bandwidth transmission of imager data using only one output differential pair, dramatically reducing the wire counts and interfacing requirements on spacecraft for future cameras.

141   

 

Soil Moisture Active Passive: Project Verification and Validation of Parameters and Spacecraft Checkout MST Michael Bonnici Mentor: Carolina Barltrop The Soil Moisture Active Passive mission, or SMAP mission, is a Flight Project at JPL that is scheduled to launch in the fall of 2014. Currently in Phase D, SMAP will provide global measurements of the soil moisture as well as the freeze and thaw states of boreal landscapes. The verification and validation of the observatory is critical to mission success to assure that the space vehicle will operate as expected and achieve its science mission on orbit. The first task assigned for the internship was to scope and define the verification and validation program for the spacecraft parameters. There are thousands of parameters on the spacecraft for each subsystem that govern day to day activities as well as fault protection functions. Each parameter needs to be verified and evaluated for proper functionality. The second task was to write the spacecraft checkout procedure, which details the step by step activities that need to be completed to verify spacecraft systems are functioning properly before the spacecraft can transition into its primary science mission. These two tasks are a part of the total project system verification and validation. Performance Assessment of Cloud Deployments Jared Borner Mentor: Costin Radulescu Cloud computing services offer scalable, on demand systems for both long-term and transient computational needs. In particular, they offer solutions for large scale, short-term infrastructure deployments without the need to invest in physical hardware. In order to determine how much of a benefit these services provide, several factors need to be considered. Among these factors are how well the deployed software can scale in a distributed system, data movement within the cloud environment itself, and data movement between the cloud and local realm. For this project, these issues were addressed with respect to the deployment of AMMOS-PDS Pipeline Services (APPS) (i.e. validation and transformation) in a cloud environment. Statistical Risk Analysis and the Development of Methodological Failure Handling Routines for NASA’s Deep Space Network (DSN) Ian Bornhoeft Mentor: Silvino Zendejas DSN equipment failures can cripple NASA’s telecommunication service capabilities and irreversibly result in data losses during mission nominal and critical events. The continuous effort to make the DSN more operationally efficient and less costly to operate, introduces challenges in the ground data system software architecture and the ability to manage risk effectively. The concepts behind relational databases and the tools provided by the SQL query language in Oracle, provide a means for quantifying the risk of data loss under various conditions involving different levels of automation. The concepts behind the development of the Erlang programming language, designed to address the needs of robust, reliable, and efficient telecommunication software solutions, can be applied to specific functional areas of the DSN Service Management System (DSMS). Using a combination of predictive methods based on analyses of historical DSN error data and more robust distributed software architectures developed with Erlang, it is possible to make the DSN more reliable. Brain Visualization Abdelwahab Bourai Mentors: Scott Davidoff, Hillary Mushkin, and Maggie Hendrie Recent developments in neuroscience and MRI have allowed researchers to identify correlations between functional regions using data called fMRI, or functional MRI. However, these correlations do not have a viable visualization tool. Using graph theory, we created a tool to abstract the brain's functional connectome to an undirected graph, with individual nodes representing widely accepted functional regions of the brain. The tool has three different views to visualize the connections: a 3-D model of the brain, a 2-D flattened version of the brain, and a fully abstracted map. These visualizations are compiled into a publicly available web application for neuroscientists to access and upload their data for viewing or analytical purposes. Uranus Radiation Model Emma Bradford Mentors: Henry Garrett and Michael Kokorowski The purpose of this summer’s task was to develop a complete model of the Uranus radiation belts. Planetary radiation is caused by high-energy particles becoming trapped by a planet’s magnetic field. This radiation can cause detrimental effects on a spacecraft by penetrating the spacecraft’s shielding which can then lead to damage to its electrical systems. By making a model for the radiation belts of Uranus one can find the intensity and location

142   

  of the high-energy particles. With this information the spacecraft can then avoid the radiation belts or be built with shielding material in order to withstand the radiation’s effects. The model is being created by taking data from Voyager’s LECP and TET particle detectors during the flyby of Uranus in 1986. Many aspects, however, need to be taken into account when creating the model such as the effects of the solar wind and the orbiting moons. The model I am developing will include a complete map of the radiation belts of Uranus at different energies, locations, and magnetic pitch angles for both protons and electrons. If all is successful, this model will be used on future missions to Uranus to design the spacecraft shielding and its trajectory. Expanding the Capabilities of Cassini's Planning Software and Graphics Martin Brennan Mentors: David Mohr and Jeffery Boyer The legendary Cassini spacecraft has been orbiting Saturn for nine years, observing the planet, its moons, and complex ring system. In order to coordinate Cassini’s many instruments, software was developed to give a basic visualization of the science observations and instruments’ fields of view (FOV). An investigation was performed to expand these capabilities with an accessible and adaptable graphics software package. Some of the software packages analyzed were Celestia, Cosmographia, Satellite Orbit Analysis Program (SOAP), and Eyes on the Solar System. The software package with the most potential for mission and science planning is Cosmographia. It is currently in development and beginning collaboration with JPL to enhance its capabilities. Cosmographia’s predecessor, Celestia was created by the same software developer and has a similar architecture. While Cosmographia development is underway, Celestia has been adapted to demonstrate its rich graphical detail and potential to aid in Cassini planning. The various Cassini instrument FOV were displayed in visualizations of the Wave at Saturn mosaic, Titan flybys, and other science opportunities. The experience gained in exploring these visualizations with Celestia will expedite the development of Cosmographia for mission and science planning. Lean Six Sigma: Cryogenic Pump House, Building 150 DeAnna Brown Mentor: Andrew D. Rose The Cryogenic Pump House contains equipment servicing the 25 ft. space simulator. The Lean Six Sigma process was applied to the room in order to increase efficiency of tasks performed there. Lean Six Sigma is a process for continual workplace improvement. The goals were to decrease set up time by 30%, reduce inventory by 33%, and to increase floor space by 30%. The Lean Six Sigma process seeks to sustain these goals by changing the overall workplace culture. This ensures that the new standards of efficiency reached will not degrade upon completion of the Lean Six Sigma event. Mars 2020 Percussive Drill Design James Burdick Mentor: Avi Okon The percussive subsystem of the drill on Mars Science Laboratory operates using a straightforward reciprocating mass driven by a voice coil. It is known that when percussively drilling brittle materials (such as rock), high velocity impacts are usually most effective for extracting regolith. Due to the velocities required of the drill on Curiosity, the single mass/voice coil device required a somewhat massive voice coil structure to withstand the collisions with the anvil. In an attempt to create a lighter, more efficient drill for Mars 2020, a reciprocating mechanism with two interrelated masses has been physically modeled. This drill design promises to be lighter and more efficient than its predecessor simply by the use of a separate, smaller striker mechanism. Additionally, by using specific orientations of spaces, springs and mass sizes, the output velocity of the new striker device can be multiplied higher than the velocity of the input voice coil by as much as a factor of two. This presentation will discuss the limitations and results of the preliminary modeling and optimization process for the percussive subsystem of the Mars 2020 drill design. Asteroid Family Identification and Modeling Their Evolution Using WISE/NEOWISE Physical Properties Mario S. Cabrera Mentor: Joseph Masiero We have used the Hierarchial Clustering Method (HCM) to be able to associate the 76 known asteroid families in the main belt based on their proper orbital elements. Converting the difference in the proper orbital elements between the center body (also referred to as the Parent of the family) and surrounding objects into a pseudovelocity, objects that are within a certain velocity will be added into the family or may be rejected if they are determined to be background objects. Plots consisting of orbital elements such as diameter, albedo, proper semimajor axis, eccentricity, and inclination where carefully analyzed at different cutoff velocities to reject possible background objects or to add those that may belong to the family. The size, albedo, and orbital distribution representative of the whole family will be used in modeling the behavior of these families forward in time, which will then be used to investigate the behavior and the evolution of these objects within our solar system.

143   

  Automating the Comparison of Software Dictionary Requirements to Tested Verification Items Norman Cao Mentor: Elizabeth Deems Properly verified flight software is key to safe and effective testing, integration, and operation of flight hardware. In the Soil Moisture Active Passive (SMAP) Project, exposed software functionality is tracked using an online tool called the Dictionary Management System (DMS), but the verification testing of items is tracked by hand in Verification Item Tracking Spreadsheets (VITS). The process of comparing DMS items to what is tracked in VITS was automated using the freely available scripting language PowerShell and the .NET Microsoft Office Interop. Regular expressions (regex), used in the unix command-line tool grep, were employed to parse the semi-regular format of the VITS. This approach was successful in finding discrepancies between DMS and VITS while automatically generating useful summary statistics. Crash Proof Vertical Surface Climbing Robots Kalind Carpenter Mentor: Aaron Parness Current robotics relies on fairly even flat terrain or a fluid medium for movement. Many areas which are desirable for science or remote presence are therefore out of reach. This project seeks to create a class of crash-proof robot vehicle for stair climbing, curb mounting and vertical rough surface mobility. Preliminary development has led to the use of microspine technology in a compliant rotary implementation. To advance this technology an iterative rapid prototyping method is utilized, incorporating shape deposition manufacturing, a molding process utilizing machinable wax and a CNC mill. This allows insert molding and the use of multiple urethanes to create the desired compliant properties. The products are assembled, tested, analyzed then the results are synthesized into an improved iteration. With multiple design strains, improvements are rapid in inception and implementation. The robot design and configuration plays a large role in the success of the traversal, the current platforms are a tricycle stair, curb, wall climber and under 20 gram micro climbers. Current designs can climb six story concrete block buildings, curbs even when painted, and overhung stairs. Mockup Star48 Adapter Design for the LDSD Integration Training Activity Leonard Carrier Mentors: Arbi Karapetian, Paul Lytal, and Mark Yerdon To prepare for the flight tests for JPL’s Low Density Supersonic Decelerator (LDSD), the test vehicle will be undergoing an integration training activity, which will fit the vehicle together to verify the assembly. However, as some subassemblies will not be ready for use by the date of the activity, mockup versions will be made to replace those subassemblies, redesigned to reduce manufacturing time and cost while retaining essential functionality. One such subassembly is the Star48 Adapter, which I have redesigned under the guidance of Paul Lytal and Mark Yerdon. With the requirements and restrictions of the integration training activity in mind, the mockup adapter is designed with mostly Aluminum and has complex sections mimicked by simplified parts. The mockup adapter is expected to be completed and verified in dynamic proof testing in time to be shipped for the activity. Interior Exploration Using Seismic Investigations, Geodesy, and Heat Transport: Grapple Support Makai A. Cartman Mentor: Randall Foehner The Interior Exploration using Seismic Investigations, Geodesy and Heat Transport (InSight) Mission is set to launch to Mars in 2016. Onboard the lander is the Seismic Experiment for Interior Structure (SEIS) to measure the seismic activity of Mars, and the Heat Flow and Physical Properties Package (HP3) to measure the internal temperature of Mars at different depths. I have been supporting the engineering staff responsible for developing the release mechanism (grapple) used to pick up and place SEIS and HP3 on the surface of Mars. Specifically my assignment has been developing test plans for a frangibolt actuator that will be used to; 1) restrain the grapple mechanism to the flight deck of the lander where it is to remain until after landing and 2) release the grapple after InSight lands. Testing of this mechanism is important because it is vital to understand the way that temperature and voltage affects the actuator to ensure the success of the grapple being restrained and released. SMAP Flight Operations Team Notification Filter Aaron Castellanos Mentors: Brian Hammer and Wallace Hu Soil Moisture Active Passive (SMAP) is one of four first-tier missions recommended by the National Research Council’s Committee on Earth Science and Applications from Space and JPL is assigned the overall responsibility of the SMAP project. Its prime objectives are to provide global measurements of soil moisture and its freeze/thaw state. The multi-mission AMMOS Mission data Processing and Control System (AMPCS) software is responsible for sending out email notifications to spacecraft operators every time a hardware or software anomaly is detected from the spacecraft telemetry. Since the AMPCS software is re-started for each downlink pass, and since AMPCS does

144   

  not retain alarm state across runs, the operators can receive duplicate notifications for alarms that persist. Since notifications are often sent via text messages on a 24x7 basis, this is undesirable behavior. In order to mitigate this, a new software tool called AMPCS Notification Filter (ANF) has been created to intercept alarm notifications from AMPCS, apply filtering criteria to remove duplicates, improve the formatting to be more compatible with mobile devices, and forward notifications to the operators. Software for the Cassini Mission Delvison Castillo Mentor: Diane Conner The success of the Cassini mission relies heavily on the ability of its team to easily carry out their tasks. One manner in which this is done is through the use of software. The Cassini mission uses sophisticated web-based applications in order to accomplish certain goals. Many of these applications were initially written at the start of the mission and need to be updated in order to fit certain standards and to provide more useful features. My role in this effort was to write new and refactor pre-existing software applications used by the Cassini team. One task was to develop a framework in which individual teams of the mission can use in order to dynamically produce webpages that are generated appropriately for the accessing user. This will increase security and bring the existing webpages up-to-date. I have also worked on writing an application that will allow for the automatic generation of Microsoft Office files, which will be a useful tool when regarding report and presentation automation. Collectively, my work here at Cassini will in turn provide the tools necessary that members of the team need in order to maintain the success of the mission. Teleoperation of Robot Arm Using a 2D Interface Rachel G. Champoux Mentors: Victor Luo and Alex Menzies At NASA’s Jet Propulsion Laboratory, the Human Interfaces group is determined to find the best interface to remotely control or teleoperate robots. With immersive software, we can engage the teleroperator in a 3D virtual environment that feels almost real. However, we rarely use these kinds of interfaces in our everyday lives; we are used to 2D screens and using a mouse to simply point and click. Do we really need a 3D interface to teleoperate a robot or could a 2D interface be just as good, if not better? The purpose of this project is to explore that question by creating a 2D interface where the user controls the end effector of a robot arm. To show three dimensions, the user can see multiple camera views on the two dimensional computer screen and place the end effector with the mouse. User tests conducted had participants use the interface and attempt to stack blocks. This report then analyzed the findings from the user studies to identify whether 2D interfaces are viable for teleoperation. An Autonomous Testing System for NASA’s Eyes Visualizations Nicholas W. Chan Mentors: Kevin Hussey and Douglas Ellison NASA’s Eyes on the Solar System is an award winning visualization software which gives the public the opportunity to explore the entire solar system using real NASA mission data. With the success of Eyes on the Solar System, the Eyes engine has since branched off to other NASA interests such as the Earth, Exoplanets, and the Deep Space Network. Since new features are constantly added, the Eyes programs have become increasingly difficult to test manually. Eye’s new autonomous testing system utilizes free and open source software to comprehensively ensure that new revisions of Eyes are fully operational. Developers can easily modify the test for their own objectives and append new parameters to the test as Eyes as new features are added in updates. When run, the testing programs will simulate a user interacting with the Eyes software. The system generates images with a screenshot of the test on one side and a screenshot of how the program should behave to allow developers to effortlessly identify and diagnose issues in Eyes. CO2 Jet Cleaning of Mars Sample Return Hardware Nicole Chen Mentor: Shirley Chung Flight hardware in contact with Martian samples must meet stringent cleanliness requirements to minimize cross contamination. My internship involved testing the cleaning efficiency of particle removal (spores and sub-micron microspheres) using the new CO2 SnoPen device. I will introduce the Planetary Protection concepts, requirements, and its goals. I will describe the main goal of my project, including a short introduction to the CO2 SnoPen. The processes used to test the cleaning efficiency of spore removal from surfaces are: coupon preparation (precleaning, sterilization), CO2 cleaning with varying key parameters, biological assay, and data analysis. The cleaning process and analysis method of sub-micron fluorescent particle removal will also be discussed. These particles are deposited onto coupons and analyzed using a fluorescent microscope. Images are taken using a stitching method and analyzed in a particle counting program. I will conclude the presentation with an overview of possible next steps such as cleaning surfaces with different roughness values and cleaning nonlinear surfaces (i.e. a screw as opposed to a flat metal coupon).

145   

 

Limb Profile Analysis of Ridges on Enceladus Heather Chilton Mentor: Robert Pappalardo Ridge and trough terrains on Saturn’s moon Enceladus are key to understanding the deformation and heating of the body, but remain poorly understood. Utilizing the Cassini spacecraft’s rich imaging data set, we use limb profiles and other supporting images to identify and characterize the morphologies of ridges apparent in limb skylines. Limb images and derived elevation profiles, provided by Dr. Peter Thomas of Cornell University, are compared with the Enceladus global imaging basemap and supplemental reprojected images to corroborate the locations of the ground traces. Prominent features are then linked among all images and the derived skyline profiles, guided by the most discernible features. We then identify those individual ridges that are unobstructed by nearby features in either the foreground or background. The derived elevation profiles of those unobstructed ridges indicate characteristics such as symmetry and slope angle. The basemap and supplemental images detail the associated surface features, ridge and trough linearity, and other relevant characteristics, which can be integrated with the cross-sectional ridge morphology to help constrain origin. Use of derived elevation profiles from Enceladus limb images provides a penetrating look into the mechanisms forming the ridge and trough terrains on this moon. Migration of Ground Test Software to a 64-bit Architecture Jared Christen Mentors: David Henriquez and Keith Martin Simulation software for testbed and workstation-based tests was developed for the Mars Science Laboratory (MSL) mission to allow comprehensive evaluation of the flight hardware and software. This software was successfully reused on the Soil Moisture Active Passive (SMAP) mission, and the Simulation and Support Equipment group (SSE) predicts that the code base will be used again for future projects. However, the code was written for 32-bit versions of Linux and the Real Time Application Interface (RTAI) extension, while 64-bit architectures are becoming increasingly dominant and offer advantages such as access to increased amounts of memory. AvSim, Hydra, and ASE, the core libraries developed by SSE, were examined for weaknesses that would cause errors on a 64-bit system. These errors were corrected when possible and flagged otherwise, and the core libraries were recompiled into 64-bit binaries. The successes and challenges of migrating code to a 64-bit architecture are discussed for reference in future projects. Data Visualization Sarah Churng Mentors: Scott Davidoff, Hillary Mushkin, and Maggie Hendrie Scientists and engineers face an increasingly difficult challenge: to see and extract insights from complex data. The goal of the Data Visualization project is to create visual interactive tools across three projects that allow scientists and engineers to explore complex data sets through creative expressions of their scientific inquiries. 1) Brain fMRI data offer groundbreaking insights on correlation networks, but these currently lack visual representation. We take insight from graph theory to portray complex brain networks as streamlined structures. We also identify three layers of abstraction for multiple views: a 3D representative model; a highly abstract mapping; and an intermediate, distorted 2D representation inspired by Harry Beck’s 1931 redesign of London subway routes as partially abstracted maps. The final project is a publicly available web application available through Caltech. 2) Aeronautic scientists need to demonstrate the complexities of fluid dynamics in order to study drag. Fluid flow is an invisible phenomenon that is as much a part of our daily interactions as it is a powerful force in engineering feats. When fluid flow becomes visible, such as with smoke plumes or water streams, it is enthralling to watch. Our solution is an interface that lets scientists explore flow decomposition, while demonstrating flow through dynamic, visually engaging animations. 3) JPL is world-renown for systems engineering, a complex process with many cross-dependent relationships. We are tasked to pursue a model-based approach that aids in streamlining cross-systems collaboration. We identify this challenge as signal-to-noise problem. But, the signal can be highly variable (What changed in the collaboration? What gets altered indirectly?) And the noise (What remains unaffected? What information should be hidden?) has varying degrees of reducibility. We propose a solution that is completely user-task dependent and flexible, and we get insight from graph theory, filtering algorithms, and existing conventions in interactions over network structures.

146   

  Data Querying Practices for an Extended Mission Kyle Cloutier Mentor: David Mohr The Mars Exploration Rover (MER), Opportunity, has been in operation on the Martian surface since January 2004. In nearly 3,400 Sols, Opportunity has traveled over 37 kilometers exploring the diverse, cratered terrain of Meridiani Planum. With tactical planning occurring almost every day, there is an emergent need to quickly gather key information regarding mission status and rover condition from diverse sources, perform analysis and checking, and present the information in a standard and efficient format for human interpretation. With the extended length of the mission, the information needed to support it has evolved and data querying has become far more complicated and demanding than previous surface operations. The purpose of this project was to perform an intensive review of the data querying, presentation, and visualization process. Along with an increase in efficiency, this project also provides an improvement in data security, with the consolidation of project data, removal of dependencies on user directories, and closeout of orphaned files. High Precision Doppler Limited Line Center Determination for the First Overtone 2 Brennan Coffey Mentors: Brian Drouin and Greg Osterman

0 of HCl

Precise and accurate positions of the line centers of HCl have not been determined for quite some time, however, using CH4 as a transferrable standard, we are able to report those positions to an order of magnitude better than previously characterized. The line centers for the first overtone of (2 0) in the spectral region 5500-5900 cm-1 are calibrated in a dual scan process using a Bruker IFS. The positions are determined through regression analysis to 1x10-5 cm-1. Further analysis using multi spectrum fitting software allows us to determine the pressure broadening parameters to high precision as well. This new standard is then compared to the HITRAN 2012 line centers of HCl for the same transition. Designing Presentations as a Model Ben Companeitz Mentor: Cin-Young Lee The development of system architecting tools is an endeavor that demands perhaps the most foresight of any engineering field. Not only is it necessary to plan paths to immediate and long-term software design goals, the systems engineer must frequently engage in meta-planning – the act of abstracting future engineering abstractions into a comprehensive descriptive framework. Currently, the SysML model ontology is quite adaptable at a diagrammatically atomic level insofar as each block or relationship can be described within nearly any structure imaginable via stereotypes and their constituent meta-classes. However, the language provides little to no support for analysis beyond categorization; more advanced model manipulation requires a more algorithmic approach with a basis in constraints. These emerging language characteristics will hopefully enable modelers to develop not only descriptively apt and consistent models, but also model-digesting processes that can create elements, modify connections, generate presentation material, run unit tests, or any combination thereof. By approaching the presentation aspect of this paradigm via development of a model-defined syntax for table generation, I endeavor to make the benefits of MBSE more tangible to systems engineers. Putting emphasis on straightforward design and organization, this new table-making functionality in DocGen provides a clean graphical representation of in-document data representation that can be included alongside algorithmic model parsing. In the future, this concept could be extended to any kind of embedded presentation element. AUV Path Planning Matt Cook Mentors: Steve Chien and Christopher Clark Oceans are highly dynamic environments with information that change in both time and space. Ocean models can produce forecasts for the ocean, which include information on currents, temperature, salinity, and more. With this modeling, there is a degree of uncertainty in the forecasts. This project seeks to optimize Autonomous Underwater Vehicle (AUV) path planning to produce a path that best satisfies a given objective. For this work, we use the ocean predictions from the Regional Ocean Modeling System (ROMS) in the Southern California Bight. The objective we seek to maximize is the reduction in the overall ocean temperature uncertainty. We use an A* algorithm, with an objective function of maximizing temperature uncertainty gathered. This planner is evaluated against random walks and a greedy algorithm. Underwater path planning has an extra variable to consider, and that is timevarying currents. Strong currents can affect the optimal path of the AUV significantly, and some paths can be considered infeasible if the AUV cannot overcome the opposing currents. Further data analysis will be done after deploying the Iver2 AUV in Southern California Bight, but preliminary results through simulated missions indicate that heuristic search methods offer promise in outperforming random and greedy exploration methods.

147   

  DESDynI Mission Science Value Function: Using Maple to Assess System Sensitivity Suzanne Craig Mentor: Jennifer Rocca Complex space missions require extensive modeling in order to accurately assess cost, requirements and launch dates to name a few key factors. Legacy Excel spreadsheets were the tool of choice along with some Matlab programs. My and my fellow intern’s job was to use the Maple math environment to pull together disparate calculations and subsystem models to create a master spreadsheet which would give the DESDynI design team a comprehensive model of their system. To do this we will connect previously written Maple worksheets to one another as well as to some specialized Matlab programs through a master Maple worksheet. This will enable designers to observe how any proposed change will affect the overall mission with ease, leading to easily obtained quantitative assessments of design changes. Synthesis and Characterization of Icy Planetary Salts Parker Crandall Mentors: James Dalton and Corey Jamieson The surface composition of icy planetary bodies contains clues to understanding the geological and geomorphological processes at play beneath thick layers of ice. Due to the extreme cost and high demand for lander missions, NASA scientists have employed remote sensing techniques with orbiting spacecraft to identify surface materials. These techniques, however, require that apposite reflectance spectral data of potentially discovered compounds be carried out in the laboratory first in order to interpret data from the spacecraft. The Planetary Ice Characterization Laboratory is currently developing a robust spectral reference library by collecting the visible to near infrared spectrum of various substances likely to exist on the moons of Saturn and Jupiter. Recently, methods for synthesizing several magnesium sulfates (meridianiite [MgSO4•11H2O], pentahydrite [MgSO4•5H2O], sanderite [MgSO4•2H2O]) and hydrohalite [NaCl•2H2O] have been explored in preparation for spectral analysis. For spectral characterization, trihydrite [MgSO-4•3H2O] was measured at temperatures ranging from 300 K to 50 K at pressures on the order of 10-3 Torr to simulate icy satellite conditions. Samples were also sorted into at six grain sizes to yield adequate optical constants after applying an improved Kramers-Kronig method combined with iterative calculations of synthetic spectra using a Hapke reflectance model. Planetesimal Versus Gas Driven Planetary Migration in Circumstellar Disks Joseph Cullen Mentors: Wladimir Lyra and Neal Turner The discovery of 926 exoplanets has led to a Renaissance in planet formation and migration theory. The two most accepted theories of planet formation, core accretion and gravitational instability, require large gas giants, much like Jupiter, to form far from their central star. However, observations of such planets tell otherwise; many orbit within 1 AU of their central star. Models indicate that both gas and planetesimals near the protoplanet exchange angular momentum through their gravitational field, causing migration. It has recently been claimed that planetesimals can have a great enough effect on the planetary embryo to initiate outward migration, and reduce the relatively well-understood inward migration caused by waves launched at Lindblad resonances in the gas disk, known as Type-I migration. This claim was made using sophisticated N-body simulations, yet implemented an inconsistent treatment of hydrodynamics. In this talk I will show hydrodynamic simulations of planet-disk interactions with a high-order code (The Pencil-Code), accurate to sixth-order in space and third-order in time. The hydrodynamics is solved in a fixed grid, while the planetesimals are treated as numerical particles, interacting with the gas via drag forces. We find that the aerodynamics in the original work was not in the correct drag regime, which has a significant impact on the direction of migration. We outline the correction, also identifying possible solutions to the posed problem. CASH: A Framework for Streamlined, Daily Updates Andrew Darwin Mentor: Diane Conner Secure, efficient, and effective communication is a critical challenge for the Multinational Cassini Mission to Saturn. Cassini’s teams of developers share information amongst themselves through the use of internal webpages. Ideally, these webpages would utilize dynamic technologies to provide a custom experience for each user. Unfortunately, the challenge of developing dynamic webpages permeates into the daily tasks of content updates and access modifications. The site administrator must understand programming principles to be able to update content and control access for various user groups. The Cassini Structured Homepages Framework drastically simplifies this operation by supplying the site administrator with a small set of XML files that specify both content and associated access control. The system administrator simply modifies these files to achieve the desired content and access control updates. The web framework then fabricates a customized experience for each user, allowing the site administrator to skirt the complexities of HTML, CSS, and PHP.

148   

  Finding Fields Fast: A Novel Approach for Mapping Agricultural Fields Across Continents Stephanie R. Debats Mentors: Thomas J. Fuchs and David R. Thompson Food production in sub-Saharan Africa is dominated by smallholder rainfed agriculture, which is vulnerable to increasing climatic variability. Efforts to boost food security in this region would be greatly assisted by quantifying the distribution of smallholder agriculture. The goal of this research is to develop a statistical machine-learning algorithm to map individual agricultural fields, mirroring the accuracy of hand-digitization. For the algorithm, a random forest pixel-wise classifier distinguishes between fields and non-fields. These classifications can then be smoothed into coherent regions corresponding to agricultural fields. Our training data set consists of hand-digitized boundaries of agricultural fields in South Africa, commissioned by its government in 2008. Working with 1 km x 1 km scenes across South Africa, the hand-digitized field boundaries are matched with satellite images extracted from Google Maps. To highlight different information contained within the images, several image-processing filters are applied. After successful training and testing of the algorithm in South Africa, we aim to expand our mapping efforts across sub-Saharan Africa, thereby providing previously unavailable data at an unprecedented level of detail on the largest and most vulnerable group of farmers in sub-Saharan Africa. Improving Slip Interpretation for the Mars Science Laboratory Stephen Decker Mentor: Jeng Yen The Mars Science Laboratory (MSL) is an active rover that has been operating on Mars for nearly an entire Earth year. In operation of MSL, rover planners use a suite of surface operation tools collectively known as the Rover Sequencing and Visualization Program (RSVP) to plan out and simulate the rover’s path across the Martian landscape. Estimated rover location is essential for proper science targeting and accurate predictions of rover location becomes more difficult in high slip areas and currently visual odometry (VO) is the on-board tool for tracking rover slip. In order to account for slippage, RSVP contains a slip prediction tool that utilizes a “slip-factor” variable that is used to taken into account numerous factors not explicitly defined in the algorithm. The proposed tool’s goal is to provide better insight for predicting rover slip by calculating slip factor with downlink visual telometry data of mobility sols. In this way, on the following sol an easily interpreted snapshot of potential slip in the immediate area around the rover is available. Gravity-Anomaly Investigation of a Recent Crustal Model for Greenland Rupert H. Deese Mentor: Erik R. Ivins Hidden underneath the world’s second-largest ice sheet, Greenland’s crust is a relative mystery. It is characterized by quite limited geological field-work in areas where rock is exposed on the coast, and by seismic and gravity data. Improving our understanding of Greenland’s crust is important for a variety of reasons, one being that its thickness, composition, and dynamics determine the amount of heat that flows through the crust into the base of the ice sheet. Variation in basal heat transfer causes melting at the boundary between the crust and the ice, causing the ice sheet to lose mass, and reducing the friction holding the ice in place. It also determines the temperature within the ice. The flow rate of the ice depends exponentially upon temperature. CRUST1.0, a new 1º global model of the earth’s crust, could be used make a better estimate of basal heat transfer. The accuracy of CRUST1.0 is investigated by using it to construct a finite element model of Greenland’s crust, from which the crustal gravity anomaly due to each layer of the model is calculated, along with the total crustal anomaly. Possible model shortcomings are identified via a comparison to the well-determined free-air anomaly map. Kinetics of HO2 Self-Reaction Erin Delaria Mentors: Stanley Sander and Fred Grieman The reactions of HO2 radicals are important in controlling ozone concentrations in both the troposphere and the stratosphere. Because a significant sink of these radicals in the troposphere are reactions of HOx with organic peroxy radicals, understanding the kinetics of reactions between hydroperoxy and organic peroxy radicals is important for constructing accurate models of the atmosphere. The ultimate goal of the project is to conduct temperature dependence studies of the HOx + acetylperoxy radical reaction at conditions relevant to the mid to upper troposphere using Infrared Kinetic Spectroscopy (IRKS). However, before studying the HO2 + acetylperoxy radical reactions, it is necessary to optimize the signal of the IRKS apparatus, test and model an experimental mechanism of HO2 formation, and ultimately determine the rate constant and temperature dependence of the HO2+HO2 self-reaction. The sensitivity of the IR signal in the IRKS apparatus was analyzed by investigating the absorption of acetylene at 6636.886 cm-1. The IR signal was then optimized through room temperature kinetics experiments in which HO2 radicals were created though pulsed laser photolysis of either Cl2/C2H2/N2/O2 or Cl2/CH3OH /N2/O2 gas mixtures using a 351nm XeF excimer laser. The loss of HO2 was monitored in the near IR

149   

  using a frequency-modulated (FM) diode laser at 6628.6 cm-1 and in the UV using a deuterium lamp at 220nm. Room temperature studies indicate using methanol as an HO2 precursor is a more reliable method for conducting kinetics experiments because the reaction mechanism is well understood and models more accurately fit experimental results. Advanced Micro Sun Sensor Testing and Novel Sun Sensor Development Based on Rayleigh Scattering Kiichiro DeLuca Mentors: Sohrab Mobasser and C. Christian Liebe Sun sensors are widely used in spacecraft attitude determination subsystems to provide a measurement of the sun vector in spacecraft coordinates. At the Jet Propulsion Laboratory, California Institute of Technology, there is an ongoing research activity to utilize Micro Electro Mechanical Systems (MEMS) processes to develop a smaller and lighter sun sensor for space applications. A prototype Advanced Micro Sun Sensor (AMSS) has been designed and constructed. The AMSS is a state of the art miniaturized digital sun sensor that is expected to provide sun vector determination accuracies that are an order of magnitude greater than current flight qualified analog class sun sensors. The AMSS was tested at the coelostat solar test facility at JPL’s Table Mountain Observatory to validate the sun vector determination accuracy. Results from this performance test implied that the implementation of the AMSS in a flight system could dramatically increase its attitude determination capability. Planned procedures for future environmental qualification tests of the AMSS are presented. Additionally, a pathfinder method for determining the sun vector based on observations of Rayleigh scattering in the daytime sky was developed. The theory behind the Rayleigh scattering based sun sensor is outlined and preliminary results are discussed. LEGACY: Creating a Sustainable Martian Presence Chrishma Derewa Mentors: Kevin Barltrop, Wafa Alidwan, and Luke Duboard For centuries the distant red hue or Mars has captured the imagination of humanity. Now mankind is ready to embrace whatever lies beneath the rough surface. The NASA Jet Propulsion Laboratories at the California Institute of Technology have created the world’s first legacy suite of exploration techniques and devices designed to explore the Martian environment. To preserve and disseminate this valuable knowledge, a living model is being created which will facilitate the sharing and explanation of these unique competencies. The Legacy system chose the Mars 2020 mission as a template for this interactive data repository. Using model based systems engineering language Sysml and implementation tool Magicdraw Legacy hopes to not only provide a living database of information, but an advanced behavioral model with responsive connectivity capable of simulation and test. Legacy links systems and assemblies in project models with a diverse array of Lab-wide databases incorporating basic traits such as physical parameters, requirements and cost with more complex tools such as problem reporting and critical analyses. Visual, intuitive and synchronized modeling makes Legacy an essential tool for the future of exploration on Mars and beyond. Preserving Rock Cores From Effects of High-G Landing and Environmental Factors for Mars Sample Return Emma Dodd Mentors: Charles Budney and Paulo Younse The goal of the proposed Mars Sample Return (MSR) mission is to collect rock cores from Mars and return them safely to Earth where in-depth analyses can be performed. The current proposed landing requires the cores to experience a high-gravity hard landing that could fracture the cores, decreasing their scientific value. This research project is focused on determining the limits of force Martian-like rocks can withstand before reducing scientific value. The force is determined by dropping a shock block apparatus, filled with Martian-like rocks of different strengths, at approximately 3500G to replicate the shock experienced during a hard landing. Analyses are performed before and after shock of the cores to determine fractionation and mass loss. Future research would consist of testing additional terrestrial Martian-like rocks along with rock cores within a caching system. Another concern upon returning the cores from Mars is maintaining hermetically sealed sample tubes. Four different seals have been assessed using helium leak testing to determine the seal with the smallest leak rate. Seals have been tested on clean, abraded, and dusted tubes to verify their limitations. As the research continues, seals will be subjected to more flight like loads and tests to validate the best seal.

150   

  Chemical Fuel Cells: Laboratory Simulations of Prebiotic Hydrothermal Chimneys Ivria Doloboff Mentors: Michael Russell and Laura Barge Submarine hydrothermal systems, and their associated semipermeable membrane “chimney“ formations, have been proposed as a possible environment for the emergence of life. In the alkaline origin-of-life model, interaction between sulfide-containing alkaline hydrothermal fluid and acidic iron-containing Hadean ocean precipitated inorganic iron-sulfide membranes in the hydrothermal mound across which thermal, ionic, and electrochemical gradients were present. Using simulated hydrothermal membranes formed in out-of-equilibrium fuel cell experiments to simulate the ambient pH-gradient between the prebiotic ocean and the hydrothermal fluid, we tested the hypothesis that the interfacing of the two contrasting solutions across the precipitated membrane generates voltages and facilitates oxidation/reduction reactions; reactions which in situ could have given rise to a “proto-metabolism” by reducing CO from the ocean, thereby kick-starting the chemical pathway to a chemiosmotic origin of life. In our setup, a synthetic membrane separates the acidic and alkaline solutions which sit in separate reservoirs on either side of the membrane. The chemical disequilibrium between two reservoirs generates energy across the ion-selective membrane in the form of a potential gradient. Our project characterizes the electrochemistry of simulated hydrothermal membranes using voltage measurement, Environmental Scanning Electron microscopy (ESEM), Energy Dispersive X-Ray (EDX) spectroscopy, and analysis of membrane template materials. Dawn Telemetry Processing Tools Nadia Dubovitsky Mentors: Kenneth Starr and Roger Klemm The purpose of the Dawn mission is to explore the asteroids Vesta and Ceres to better understand the origins of our solar system. The spacecraft has already visited Vesta and is currently cruising towards Ceres. Telemetry from the spacecraft is received in large chunks every few weeks and is examined to make sure that the spacecraft is healthy, functioning properly, and on track. Anomalies are usually highlighted by alarms that have been programmed into the ground data system. However, when there is a problem, it is often difficult to compare current data to previous telemetry, because the information is stored in many different sources in different formats. To help with this, I extracted the commands that were executed on the spacecraft at the time that certain problems occurred into an excel format. Within the spreadsheet, the commands were compared to previously executed ones, and differences were noted. Furthermore, I worked on a tool that collects all the relevant information about temperature sensors and their respective limits from different sources. The tool then outputs all of the data in a separate spreadsheet and converts raw values to engineering units so that the numbers can be easily compared. Starshade Mission Planning Tool David Dyrda Mentor: Doug Lisman The task of efficiently stringing together a sequence of external occulter, or starshade, observations for the purpose of finding and characterizing exoplanets rests on several key parameters, and is made difficult by several complicating factors. The most important of these parameters is the search completeness, or the likelihood of being able to detect an exoplanet that is present within the habitable zone. The desired goal is to observe as many high completeness stars as possible within constraints on time, sun avoidance and launch and propellant mass. Retargeting requires propulsive ∆V to maneuver the occulter into position, along the telescope line of site to the next star. The work done this summer was to develop a spreadsheet based Starshade Mission Planning Tool to formulate observational sequences with a visual representation of those stars with high search completeness. The tool is capable of taking numerous input parameters, calculating and displaying the search completeness of each star, and computing the required retargeting propellant. The tool also supports propulsion trade studies by allowing the propulsion type to be varied and automatically applying the appropriate set of performance parameters and masses. Error Analysis of the Miniature Inertial Measurement Unit (MIMU) Andrew Edwards Mentors: Erik Bailey and Miguel San Martin JPL is currently developing technology to support a future proposal for a mission to autonomously retrieve and return sample from the surface of a comet to Earth. One critical component to the autonomous performance of the spacecraft is its inertial measurement unit. The purpose of this task is to characterize the accelerometer and gyroscope errors from a Honeywell MIMU to inform architectural decisions for the guidance, navigation, and control subsystem by predicting the unit’s performance during autonomous proximity operations. The results of this characterization are weighed against comet environmental factors. Accelerometer and gyroscope data were collected over one hour from a stationary unit, which was then analyzed to characterize the expected the bias,

151   

  drift, and noise level of the unit. The results showed that, even under the most ideal conditions, the observed error as exhibited by random walk exceeded allocated mission landing accuracy requirements of fifty meters, so additional navigation instruments will be necessary to reduce error, such as a camera and laser altimeter. Interactive Data Visualization Conrad Egan Mentors: Scott Davidoff, Hillary Mushkin, and Maggie Hendrie Scientists and engineers face an increasingly difficult challenge: to see and extract insights from complex data. The goal of the Interactive Data Visualization project is to create visual interactive tools for 1) Brain resting-state network research, 2) Decomposition of fluid turbulence along walls, and 3) Model based systems engineering for the Europa Clipper concept. These tools allow scientists and engineers to understand complex data sets through illustrative expressions of their data. 1) Brain fMRI data offers groundbreaking insights into which functional regions of the brain activate simultaneously, but these currently lack visual representation. We take insight from graph theory to portray complex brain networks as streamlined structures. We also identify three layers of abstraction for multiple views: a 3D representative model; a highly abstract mapping; and an intermediate, 2D representation that hybridizes the structural precision of the 3D model and the readability of the abstract model. The final project is a publicly available web application available through Caltech. 2) Aeronautic scientists need to understand the complexities of turbulence in order to study how to reduce drag. Dr. Beverly McKeon’s research group is developing a mathematical model based on Navier-Stokes and concepts from electrical engineering to “build up” turbulence through its primary signals known as “modes.” Our solution is an interface that lets scientists explore turbulence decomposition and recomposition, while demonstrating flow through dynamic, visually engaging animations. 3) JPL is world-renowned for systems engineering, a complex process with many cross-dependent relationships. We are tasked to pursue a model-based approach that aids in streamlining cross-systems collaboration. We identify this challenge as signal-to-noise problem. But, the signal can be highly variable (What changed in the collaboration? What gets altered indirectly?) And the noise (What remains unaffected? What information should be hidden?) has varying degrees of reducibility. We propose a solution that is completely user-task dependent and flexible, and we get insight from graph theory, filtering algorithms, and existing conventions in interactions over network structures. Solder at Cryogenic Temperatures: Correlating Macroscale and Nanoscale Deformation Mechanisms Using Electron Microscopy Techniques Alanna Eilenberg Mentors: Andrew Shapiro, Amanda Lupinacci, and Andrew Minor Exploration of Saturn’s moons is desired to look for evidence of life. First, development of materials that can withstand the cryogenic temperatures on these moons is needed. Ongoing studies of Pb-based and Pb-free solders aim to better understand ductile to brittle transitions in solder, which can lead to failure of solder joints at cryogenic temperatures. This study focused on three solder compositions: 70-30 Pb-Sn, 80-20 Pb-Sn, and 96-4 Sn-Ag. Lead is not susceptible to embrittlement thanks to its crystal structure, whereas tin has a structure that becomes brittle at temperatures below -125◦C. The lead rich solders should exhibit less embrittlement compared to previously studied lower lead content solders. The Sn-Ag solder is of interest as it is currently used in industry at low temperatures. For macroscale analysis, these solders were broken using an impact tester at temperatures ranging from -180◦C to 25◦C in order to study the relationship between temperature and energy required to break the solder. Additionally, mechanical polishing was used to prepare 10μm thick solder samples for future in situ cryogenic nanoscale deformation testing. Such testing will allow for analysis of the relationship between solder structure, mechanical behavior, and deformation mechanisms over the given temperature range. Chlorophyll Fluorescence as an OCO-2 Product Dillon Elsbury Mentors: Gregory Osterman and Christian Frankenberg Light energy absorbed by plants is used for photosynthesis, but this energy can also be dissipated as heat or reemitted at a longer wavelength, which is known as fluorescence. Because these processes occur in competition with each other, fluorescence observations can indicate levels of photosynthetic efficiency. Measurements of photosynthetic efficiency can be used to understand plant vitality and plant stress. Atmospheric Carbon Observations from Space (ACOS) data contains observations made in the O2 A-band, which overlaps the chlorophyll fluorescence spectrum. Using ACOS data, this project aimed to study fluorescence on a regional, global, and temporal scale. Timeseries plots were made to study fluorescence in 16 different biomes. Additionally, gridded

152   

  fluorescence maps were created to understand the magnitude of average fluorescence values on annual, monthly, and even daily scales. Specific areas of interest for this study included the United States Midwest, the African Savannah and South American rainforest. Plots indicate seasonal variations in fluorescence and that the magnitude of values observed are dependent on the state of the region the plants occupy. Mars Science Laboratory As-Run Database: A Compilation of As-Run Data Ryan Endres Mentor: Gregorio Villar III Everyday data is downlinked from the Curiosity rover. This data provides insight into Curiosity’s activities on Mars, which is useful for operations. Curiosity’s prime mission is 687 days, but the rover may live much longer than that. As a result, the data will become overwhelmingly large and difficult to search through. Therefore, a database was proposed to consolidate this information, providing the operations team the capability to quickly search for and find the data they are looking for. Operations team members were questioned about possible use cases and sources of relevant data. As information was collected, the database took the form of three different tools with unique capabilities. The first is a Mars Science Laboratory wiki page with several use cases that map to sources of data. The second tool, located on the operations workstations, prompts users for inputs and outputs relevant information from different sources. The last tool, developed by the Sequence Revitalization group, allows for queryable searches. These tools will allow for more efficient planning of Curiosity now and well into the future. Analysis of Network and Infrastructure Monitoring Tools and Logs Stefan Eng Mentor: Robert Witoff Working closely with the infrastructure and network engineers we investigated an aggregation of server monitoring tools. The team of engineers we were working with wanted to investigate the potential for success of a monitoring tool that drew data from a variety of other data providing monitors. This class of a tool is a Manager of Managers (MoM) and its purpose is to help intelligently reduce the outage time across servers and components at JPL by gaining visibility into more data. Using the strategy of low hanging fruit to quickly find areas of interest in the data sources we set out to create high visibility prototypes of interesting trends in this aggregated data to help the engineers perform their jobs more efficiently. Throughout our prototyping effort, we routinely attempted to access systems and data sources in new, direct ways for external analysis. While expertly supported, the extradition of data outside of its intended applications blazed new territory and exposed many hurdles to data access. Mars Reconnaissance Orbiter Overflight Analysis Tool: A Development of User Friendly Software for Long Term Predictive Relay Opportunities Between MRO and Rovers at Mars Kevin Ferrant Mentor: Shin Huh The Mars Reconnaissance Orbiter (MRO) acts as the main relay asset for rovers on the surface of Mars. As MRO continues its second extended mission not only does the satellite need to provide relay support to current rovers, it also needs to analyze potential landing sites for future missions and continue to study the Martian environment. With the desire to accomplish multiple objectives comes the decision of which task to perform at specific times. Currently this decision process is constrained by the ability to produce overflight analysis in a fast and reliable way. The Overflight Analysis Tool (OAT) being developed looks to alleviate this restriction by providing the user with an easy method to determine when future relay opportunities occur. Characterizing the overflight by maximum elevation and duration gives the user the ability to easily determine whether or not a specific future overflight might be a poor choice for relay tasks. Knowing that a future overflight does not lend itself to relay tasks allows MRO to be used for other scientific opportunities. Prime Focus Instrument: Torque Measurement and Analysis of Piezoelectric Motors Christopher Ferro Mentor: Mohammad Mojarradi The Prime Focus Instrument, currently in development for use in the Subaru Telescope on Mauna Kea, Hawaii, will provide scientists and researchers with important information regarding the formation of galaxies, dark matter, and interstellar mapping. A key feature of the telescope is its Prime Focus Instrument, which collects reflected light from outer space and redirects it to a spectrograph for scientific analysis. The instrument consists of 2400 piezoelectric motors assembled in a hexagonal close-packed fashion. A fiber optic cable is connected to each one and collects light while the motor rotates. Each motor contains two sub-motors (phi and theta) that take turns rotating during a full cycle. Torque analysis of each sub-motor was accomplished by using two miniature de Prony brakes (dynamometers) manufactured in a machine shop. Each de Prony brake was attached to a sub-motor and helped to counteract the torque generated by each motor as it rotated. Small masses were gradually hung on one end of the brake until the shaft of the motor either stopped or slowed significantly. Repeated testing showed that maximum torque occurs at the resonant frequency of the motor. Operating the piezoelectric motors at this frequency allows for maximum efficiency when gathering light from galaxies and stars.

153   

 

Next Generation High-Performance Spaceflight Computer Chris Flatley Mentor: Raphael Some Looking forward to future NASA missions, engineers noticed that currently available spaceflight computers would not be able to handle the strain that these new missions would place on them. This realization made apparent the need for a new computer that could efficiently handle the large amounts of data that new missions will generate. This project is focused on determining the capabilities that this next-gen computer must have, as well as creating a set of benchmarks that will thoroughly test several potential computer architectures. The successful completion of this project will yield a new spaceflight computer that sets the standard for years to come. Updating the MDL Gallery Veronica Flesch Mentor: Harish Manohara The galleries in the Microdevices Laboratory need updated displays for both the Microelectromechanical Systems and Nanotechnology walls. This is important because the lab is often visited by high-profile visitors from various organizations, and the most common way these people learn of MDL activities is through the galleries; they play a significant role in attracting new business to the lab. So, information and images have been collected for the newest technologies at MDL, including fused silica microgyroscopes, miniature stereo cameras, black silicon, damage detection sensors, deformable membrane mirrors, and carbon nanotube vacuum electronics. Photoshop was used to compile images, data, and information about these technologies into panels that can replace outdated technologies on the gallery walls. In addition, two new panels were designed: one displaying the developement of microgyroscopes at MDL, and the other showing the history of carbon nanotubes. MISR Smoke Plume Height Project Daniel Fong Mentor: David Nelson Wildfire smoke plumes are of interest to scientists in a wide variety of fields because smoke has significant effects on the atmosphere, climate, and public health and safety. The Multi-angle Imaging SpectroRadiometer (MISR) Instrument aboard the Terra satellite allows us to view smoke plumes stereoscopically, and to determine their height and associated wind speeds. When used in conjunction with data from the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument, which is also aboard the Terra satellite, we can also determine the radiative power of each fire. The two variables height and radiative fire power should be closely correlated, and with this information climate models can predict where the smoke will be transported, and what effects it will have on the atmosphere and surface albedo. We have been digitizing smoke plumes world wide in order to construct a dataset encompassing all of the global data collected by MISR throughout the year of 2008. In order to determine the heights and radiative power of these smoke plumes, we have been using software called MINX. Our end goal is for our data to be used in the upcoming Intergovernmental Panel on Climate Change (IPCC) report. OPALS: Optical Payload for LAsercomm Science Robert Forsyth Mentor: Bogdan Oaida This presentation will discuss the development of the Optical Payload for Lasercomm Science (OPALS) project and review my contributions during the summer of 2013. OPALS is a mission whose goal is to demonstrate laser communication capabilities from Low Earth Orbit to a ground station. OPALS will demonstrate these capabilities by flying a payload on the International Space Station, ISS, by downlinking video data to a ground station via laser communication. OPALS will focus on characterizing atmospheric turbulence, link availability, and ground station pointing performance. The OPALS payload is currently at Kennedy Space Center undergoing verification and validation (V&V) to ensure that it meets requirements. OPALS will launch as a payload on the SpaceX commercial resupply mission. This flight is scheduled for 2014. OPALS has an operational lifetime of three months. It will transmit to a ground receiver at the Optical Communications Telescope Laboratory in Wrightwood, California. This summer I have been supporting verification and validation efforts for OPALS. My main tasks this summer have been: creating closure documentation for verification activities, aiding and documenting flight hardware shipping, and developing an ISS telemetry visualization tools using the Systems Tool Kit and the Telemetry Resource Kit.

154   

 

Supersonic Retro-Propulsion: Understanding the Effects of Rocket Thrust Against a Supersonic Free Stream Environment Jonathan Gary Mentors: Steve Sell, Devin Kipp, and Adam Steltzner In pursuing human-scale missions to Mars, it is necessary to move beyond the exclusive use of aerodynamic drag to slow entry vehicles to subsonic speeds. Supersonic retro-propulsion is the next advancement in the suite of entry, descent and landing technologies. Initially studied in the 1950s and 1960s with favorable results, the proceeding decades saw little expansion, and the technology currently lacks the maturity required for mission integration. A series of tests designed to assess the capabilities of this advanced technology is under development, the first of which will take place on a rocket sled. To assist in the tests’ planning and execution, the supersonic retro-propulsion team developed a computer simulation that models the sled’s characteristics, the selected rockets’ propulsive capabilities, and the environmental conditions to produce useful and relevant output. By using normally distributed input parameters spread across numerous iterations, a certain degree of confidence in the tests’ success can be achieved. NUMIT 2.0: The Improved Version of the JPL Internal Charging Analysis Code Gilbert Ghang Mentor: Wousik Kim Spacecraft are exposed to a variety of radiation environments in which energetic particles can penetrate protective spacecraft surfaces and deposit charge on non-conducting materials. As the electrons accumulate within the nonconductor over time, an arcing or discharge can occur, damaging the local circuit or subsystem. This occurrence, known as internal charging, is one of the leading causes of spacecraft anomalies and failures. Internal charging can be modeled using the powerful 1-dimensional code Numerical InTegration (NUMIT). However, the limitations of the current version of NUMIT necessitate an improved, more generic internal charging evaluation tool, NUMIT 2.0; updates include the extension of the low energy limit, the smoothing of an artificial cusp created by the energy deposition algorithm, and the automatic calculation of previously user-inputted values. Testing of an Application-Specific Integrated Circuit Branden Ghena Mentor: Ryan Stern Computer systems for use in space-based applications need to be not only powerful but also reliable, low-power, radiation hardened, and capable of functioning over a wide temperature range. To this end, an Application-Specific Integrated Circuit (ASIC) has been created containing a fault-tolerant configuration of two ARM Cortex M0 processors. The chip is designed to be extremely robust to temperature variations and high-radiation environments. Verification of the ASIC requires several steps, starting with the creation of memory and I/O for the processor on an attached Field-Programmable Gate Array (FPGA). Software must also be created to thoroughly test all paths through the silicon. Additionally, testing will determine the maximum clock speed of the chip as well as nominal power usage. The specific process used to create the ASIC is new and its characteristics and limitations are currently unknown. If testing can successfully verify the device’s functionality, it will be a large step forward towards the future use of similar ASICs in flight missions. DESynI Science Value Function: Assessing Mission Parameters With Maple Cajer Gong Mentor: Jennifer Rocca In the past Excel worksheets were primarily used to assess changes in mission parameters and science value; however they proved to be difficult to use and understand, limiting use by non-experts in the respective fields. By implementing existing Excel functions in a Maple environment and creating a master worksheet which collates existing subsystem worksheets into a master worksheet, we hope to give the entire DESynI design team a comprehensive and easy to use tool which models the spacecraft. Doing so will streamline the design process by allowing an engineer working on an individual subsystem to visualize his or her effect on other subsystems and the spacecraft as a whole.

155   

  Development of a Test Apparatus for Micro Robotic Legs Stephen Goodwin Mentors: Rudranarayan Mukherjee and Aaron Parness In order to design micro robots for optimal perforce, it is necessary to understand the dynamic interactions between the robot’s leg and the ground. It is thought that the coefficient of friction is dependent on translational velocity of the robot as well as rotational velocity of the leg. In order to test this relationship, a small test apparatus capable of accepting user inputs for translational and rotational velocities and measuring normal and shear forces was developed. Originally the unit was designed for OctoRoach, an eight-legged micro robot developed at the University of California, Berkeley. After initial testing, multiple leg designs were prototyped to examine their effect on traction. In parallel, a hybrid technology wrist was developed in an effort to combine two developing technologies: Micro Spines and Gecko Dry Adhesives. Existing robots have successfully implemented each these technologies individually, but there has yet to be an effort to combine them into a single mechanism. Designing a Mobile Photoplethysmograph for Microwave Cardiography Ground Truth Data Comparison Robert Gougelet Mentor: Jim Lux FINDER (Finding Individuals for Disaster and Emergency Response) is a microwave radar developed by Jet Propulsion Laboratory to detect victims buried in disaster rubble. FINDER uses a form of microwave cardiography (MCG) to measure cardiac and respiratory activity by detecting the phase shift of a reflected continuous microwave radar signal caused by small motions of the victim’s skin surface. Performance testing of the prototype FINDER device requires ground truth data collected from human subjects acting as victims in rubble test sites. These data must be acquired in a field environment using portable, unobtrusive equipment. Photoplethysmography (PPG) detects changes in infrared and visual spectrum reflectance of tissue in response to changes in blood volume and oxygenation that are associated with cardiac and respiratory activity. PPG’s simple and lightweight design implementation makes it an ideal mobile measure for FINDER’s ground truth data comparison. A readily reproducible and inexpensive mobile PPG device was created for this project, using a microcontroller for time synchronization, data collection, time stamping, and storage. The newly accessible mobile PPG data permits validation and testing of FINDER life detection algorithms and source separation techniques. Test Implementation of Signal Chain Electronics for the James Webb Space Telescope James Kelly Griffin Mentor: Brian Franklin The signal chain electronics (SCE) control and digitalize data gathered from infrared detector arrays inside the MidInfraRed Instrument (MIRI) that will be onboard the James Webb Space Telescope (JWST). The specific circuits that underwent testing are voltage regulators via digital to analog converters under low noise conditions. The regulators supply different voltages for various detectors by converting signals from low-voltage differential signaling (LVDS) and field-programmable gate array (FPGA) controllers to DC voltages. Through numerous designs techniques, the voltage regulators have to produce their respective voltages while be subjected to extreme conditions. In space, electronics experience many extreme environment conditions such as temperature change, radiation exposer, and drift due to life expectancies for specific parts. In order to design and test wide-temperature radiation-tolerant circuits, we use a worst case analysis database created for the MIRI project. Comparing results from different topologies allows designers to acquire the most robust designs necessary for flight missions. Deep Space Atomic Clock Characterization: A Sensitivity Study of Stochastic Filters and Noise Models Erin Griggs Mentors: Jill Seubert and Todd Ely Precise timing is of upmost importance for spacecraft navigation. Traditionally, ground-based atomic clocks are used for deep space missions as space-based clocks lack the high-caliber accuracy and stability necessary for navigation. The Deep Space Atomic Clock (DSAC) mission will enable one-way tracking of signals from satellites traveling beyond Earth orbit with utilization of a small, low-mass atomic clock based on mercury-ion trapping technology. Measurements from the GPS satellites allow for precision orbit and clock determination to assess the performance of DSAC. Stochastic filters optimized to capture the clock behavior were developed through precise error modeling and simulation. A clock filter model consisting of the summation of two Markov noise processes was optimized to follow the dynamics of DSAC. This filter was tested with simulated data from various calibers of clocks. A sensitivity study was conducted to assess the orbit and clock estimation capabilities of both optimal and sub-optimal filters. Contributions from individual noise components were found in terms of Allan deviation, which lead to the development of a hierarchy of error sources, including clock noise, GPS temperature noise, and attitude errors.

156   

  Electron Impact Studies of Molecules Relevant to Planetary Atmospheres Brandon Grisanti Mentors: Paul Johnson and Charles Malone Collision processes, such as excitation and ionization between secondary photoelectrons and photons with neutral and ionized species of molecules occur in the atmospheres of Jupiter and Saturn. In modeling these collision processes, the cross sections for various electron collisions occur as parameters in corresponding rate equations. Many of these parameters are currently unknown or are unreliable. In order to better understand ultraviolet emission measurements collected from spacecraft, more accurate parameters must be measured experimentally. Electron-hydrogen collision cross sections were measured with a low resolution ultraviolet spectrometer inside a vacuum chamber over an energy range of 10 to 300 electron volts. Numerous diagnostic procedures were employed to improve the accuracy of results. Development of Statistical Weather Model of Ground Station for Use in Optical Communication Pointing, Acquisition, and Tracking Systems Engineering Tool David Hanley Mentor: Oscar Alvarez-Salazar Deep space optical communications technology promises to increase the data transfer rate of current space missions by one or two orders of magnitude. The Jet Propulsion Laboratory is heavily invested in the development of this technology. Due to the increased frequency of the optical signal, however, the atmosphere is more likely to occlude a signal. This project focuses on the development of a tool which models the atmosphere’s tendency to obstruct optical signals at a given location. The intent of this project is for this model to be used in a systems engineering tool which describes the pointing, acquisition, and tracking needs of a deep space optical communications system. ISS-RapidScat: Systems Engineering in a Low-Cost Science Mission Involving the International Space Station Aaron Harris Mentors: Jennifer Rocca and Stacey Boland ISS-RapidScat is a current science mission to launch a scatterometer instrument in April 2014. This instrument is being created using heritage hardware from QuickSCAT and SeaWinds, and will launch aboard a Falcon 9 vehicle inside of an unpressurized Dragon trunk with the SpaceX 4 manifest. ISS-RapidScat will provide a Radar Scatterometer mounted as an external payload on the Columbus Module of the International Space Station. The payload will then remain in orbit for two years and collect valuable scientific data on ocean vector winds in the tropical regions of planet Earth. The mission will have to consider its viewing angle as the attitude of the station changes in orbit. During my internship, I have supported the Project Systems Engineering team and the Mission Systems team to better understand the many interfaces between ISS-RapidScat and the ISS. Most significantly, I have created a program to cache real-time telemetry on the ISS to measure the stability of the ISS as an Earth-viewing platform. I hope this effort will assist the mission planning and operations elements of the project manage the complex interface to the ISS and maximize the science return for the mission. GRACE-FO Bakari Hassan Mentor: Robert Sharrow Gravity Recovery and Climate Experiment Follow On (GRACE-FO) is a 5-year earth science mission headed by NASA’s Jet Propulsion Laboratory (JPL). Set to launch in 2017, it will map the Earth’s gravity field. This mapping will continue the science from the original GRACE mission. The spacecraft subcontractor, Astrium, composes two documents that define spacecraft and instrument interface requirements. JPL needs to treat these as one interface control document. The process of linking and merging these documents was led through collaboration with the JPL Dynamic Object Orientation Requirements System analysts. JPL receives hundreds of documents from Astrium and also needed a DocuShare folder structure to organize these deliverables. A Python script was developed to replace the manual process with a 96% more time efficient method, reducing the required time from 3 weeks to 5 hours. Additionally, a requirement waiver and an engineering change request (ECR) were drafted and discussed with GRACE-FO flight systems engineers, resulting in an approved ECR for the accelerometer environment temperatures.

157   

  Rheology and Thermal Property Measurements of Icy Cryovolcanic Slurries; Insight Pressure GSE Emily Hawkins Mentor: Fang Zhong Studying the rheology and thermal properties of icy slurries will contribute to a detailed portrait of the nature of cryovolcanism in the outer solar system. The objective of the experiment is to step towards a deeper understanding of how crystal size, orientation, and its three dimensional network in cryogenic slurries affect its rheology, i.e. the relationship between strain and stress. Crystallization in cryovolcanic lava flow is hypothesized to play an important role in land formation on icy bodies such as Titan. The dual probe method is adapted to measure thermal conductivity and thermal diffusivity in icy slurries. An emphasis on the thermal transport properties of cryogenic slurries is necessary because crystallization follows a cold temperature gradient that depends on such thermal transport properties. During the SIP summer internship I performed various tasks that contributed to the completion of the experimental setup to study the thermal and rheological properties of the icy slurries. Such tasks included troubleshooting unforeseen components in both software and hardware. After an initial experiment test execution in June, other unforeseen problems arose in which the discovery of a malfunctioning dual probe needle was made. The decision was made to move forward in the remaining summer months to reproduce past experimental results without the dual probe thermal needle but with optical visualization of the crystal. Data will be obtained and presented, first at a constant temperature and next at a constant shear rate. During the month of July, 2013, focus was shifted to another project to validate a concept to test a pressure sensor with high stability for the Insight 2016 Lander to Mars. This project, titled, “Insight Pressure GSE,” was to design a mechanism and a system to verify that a pressure sensor and its low-pass inlet for the Mars Insight Lander will meet their requirements. Instrumentation was developed and a central monitoring and recording program was created to adhere to specific measuring components for this project. Communication was established for new instruments. Creating a functioning LabView monitoring and recording program for this project involved a significant amount of troubleshooting, as such unfamiliar instruments proved to contain difficulties in establishing proper communication. Data will be presented for several different tests, including a rough-pumping test to study overnight temperature drift, as well as tests to conduct sinusoidal movement at varied frequencies. Over the month of July, data was obtained and analyzed, leading to a report for a significant event in Div 38. Such data and data analysis as well as the development of programs to obtain the data will also be presented. iLab: Development of an Extensible Computing System Infrastructure for the ISAAC Lab Spencer Hawkins Mentor: Yutao He During the development of Instrument ShAred Artifact for Computing (ISAAC) project, engineers use Linux workstations to create and work with large code bases collaboratively. Certain software tools help engineers create, maintain, understand, and collaborate theirs and others' work. This project is based on creating, updating, maintaining, and integrating those tools to build a cohesive and productive computing environment, and creating sufficient documentation such that the system can be easily understood and maintained. The infrastructure behind the ISAAC project includes multiple workstations and a local server which engineers use daily. The process of development for ISAAC projects can be compacted and streamlined by creating and maintaining a functional local network between these machines. By properly documenting the setup and installation procedure, the process can also easily be repeated for new labs that could use a similar infrastructure. Model-Based Systems Engineering: Development of a Sequence Analysis Toolset for MagicDraw Jim Henrickson Mentors: Alek Kerzhner and Michel Ingham By approaching systems engineering tasks from a rigorous, modular, and model-based standpoint, model-based systems engineering provides increased efficacy in analyzing and developing complex systems while also providing robustness to on-the-fly change. The system models created in this process are correspondingly complex and necessitate unique tools that allow adequate analysis for verification and validation purposes. This project developed and applied an initial prototype for one such tool that focuses on the analysis of a sequence of activities specified for a system. The tool, a NoMagic MagicDraw™ plugin, allows its user to generate a sequential list of all states a particular system component passes through for the entirety of a modeled sequence. This includes not only every state explicitly included in the sequence diagrams, but also includes states along transition paths described by relevant state machines. With this information, the user can rapidly verify whether or not the state transitions described in the model are valid, and can quickly identify transitions that require modification. As an illustration, this tool is applied to check consistency in specified sequences for a concept cellularized space architecture.

158   

  Prototyping Gecko Adhesive Grippers for Use in Orbital Applications Tyler Hilgendorf Mentors: Aaron Parness and Matt Heverly The work described in this paper deals with the field of synthetic materials which imitate the microscopic structure on a gecko’s foot, or “gecko adhesives”. These materials, like the foot of a gecko, are able to adhere to very smooth surfaces such as glass. They detach quickly, easily, and cleanly, and can sustain many cycles of attachment and detachment without diminishing in adhesive strength. This material is being developed for future use in anchoring to objects in the zero-gravity environment of space. A gecko-gripper prototype was developed which can be activated with one hand, and which could one day serve as an easily-moveable anchoring point which attaches to an interior wall or window of the International Space Station and holds a laptop or other object in place. This gecko gripper was also adapted for use as an end-effector for the legs of a microgravity-climbing robot called LEMUR. LEMUR will use its gecko-gripper feet to adhere to smooth surfaces such as glass, plastic, or metal. This is the first step towards a long-term goal of creating an inspection robot capable of climbing freely around on the outside of the International Space Station. The Origin of Life Through Serpentinization and Alkaline Hydrothermal Vents Adam Hoffmann Mentor: Steven D. Vance Serpentinization is a reaction between ocean crust and water that generates energy, hydrogen gas, and ions. Hydrothermal effluent from serpentinizing systems emerges at the Earth’s surface precipitating alkaline hydrothermal vents. Today, these vents harbor some of the most extreme forms of life and are theorized to have been where life originated during the Hadean (Russell and Hall 1997). The absence of plate tectonics and magmatic activity makes these systems attractive analogues to similar systems that may exist and host life on other wet rocky planets (i.e. Europa and/or Enceladus). A unique hydrothermal reactor (Mielke et al. 2010) was built to test the capability of ancient alkaline hydrothermal systems to synthesize organic molecules towards the earliest emergence of life, as theorized by (Russell and Hall 1997). Experiments conducted using this hydrothermal reactor simulate serpentinization and alkaline hydrothermal vents under conditions that would have been present on the early Earth. Analysis of the solid materials from the hydrothermal reactor was performed before and after each experiment to determine detailed geochemistry in ancient serpentinization, which would have contributed to organic synthesis. Findings of minerals associated with serpentinization and some catalytically active species of iron sulfides offer evidence that serpentinization-like conditions are successfully reproduced using this novel instrument. Real-Time Landing Point Tracking and Relative Landing Point Vector Generation for Unmanned Arial Vehicle Rooftop Landing Procedure Utilizing Computer Vision Methods Christian Howard Mentor: Roland Brockers To operate unmanned aerial vehicles (UAVs) safely in complex environments, human operator control can be facilitated significantly if the UAV can execute simple maneuvers autonomously. One example that is particularly important for reconnaissance scenarios is rooftop landing. For autonomous landing on a flat rooftop surface, the UAV needs to detect a sufficiently flat rooftop landing area, pick a safe landing spot and then track the landing spot during a descent maneuver for precision landing. While the first two parts were covered by other tasks, the goal of this work was to develop a real-time algorithm for tracking the desired landing point during descent, while simultaneously generating a three dimensional vector from the camera to the landing location. Landing point tracking was achieved by applying a keyframe based homography scheme which allows to track unmarked points on flat surfaces. The 3D vector to the landing spot was generated using pose information froma secondary pose estimation framework that runs on-board the UAV. The resulting output allows for precision control of the aircraft during descent and landing. Hardware Interface Testing for the INSPIRE CubeSat Devin Hupp Mentors: Allen Kummer and Andrew Klesh The Interplanetary Nano-Spacecraft Pathfinder In a Relevant Environment, or INSPIRE, will be the first CubeSat to travel beyond low Earth orbit. A primary mission objective for INSPIRE is to demonstrate communications with the Deep Space Network. Therefore JPL is developing the first CubeSat form factor X-Band radio. This radio is called the Iris radio, and has a control board called Marina-2, which draws from previous JPL CubeSat missions COVE and GRIFEX. The Marina-2 board communicates with the INSPIRE flight computer using general purpose inputs and outputs (GPIO) and a serial interface, called SPI. The Marina-2 interface must be tested in a stand-alone state before delivering to the INSPIRE team. To perform this testing, a graphical interface (GUI) was developed in python to run on a Raspberry Pi. Raspberry Pi is an inexpensive Linux Box popular with hobbyists: it has well developed SPI and GPIO libraries that can be controlled from simple python scripts. The Pi uses the GUI to control

159   

  the GPIO and SPI needed to interface with the Marina-2 board, thereby acting as the flight computer. This GUI allows the Marina-2 board to be powered and then can communicate with either the PCM memory module or the FPGA that controls the radio. The Pi and GUI will be delivered as ground support equipment for the Iris payload and will be used to test future JPL missions that draw from the same heritage. Characterization of the JPL INSPIRE 3D-Printed Cold Gas Thruster for Attitude Control Travis Imken Mentors: Allen Kummer and Andrew Klesh The JPL INSPIRE mission is launching two 3-Unit CubeSatellites beyond Earth orbit to demonstrate functionality, communication, navigation, and payload hosting in interplanetary space. To maintain X-Band communication with Earth, the satellites require an Attitude Control System (ACS) to properly orient the antennas. The ACS uses a nonhazardous and non-explosive cold-gas propellant that is stored as a liquid, expanded to a vapor, and expelled through nozzles to create torques on the spacecraft. The thruster unit features a 3D-printed main manifold, embedded temperature and pressure sensors, a heater, and high-performance valves. Thrust determination is done with a ballistic pendulum in a thermal vacuum chamber to characterize the ACS prototype. This research is being conducted by The University of Texas at Austin Texas Spacecraft Lab through the JPL INSPIRE project. Model-Based Systems Engineering: Advancing the Functionality of the Model-Based Engineering Environment Benjamin Inada Mentor: Cin-Young Lee Model-Based Systems Engineering (MBSE) formalizes the means by which missions and other systems are formulated during the design phase through the time of actual implementation. MBSE produces system models that contain descriptive views that represent their corresponding systems. My team is building an environment for MBSE called the Model-Based Engineering Environment (MBEE). The MBEE makes use of the Model Development Kit (MDK), a set of tools designed to make modeling efficient and highly intuitive. One of the MDK’s key features is a functionality to generate documents from system models and organize them in a way that is relevant to project stakeholders. The majority of my work this summer has been focused on adding a style application capability to the MDK that allows users to save graphical styles from a pattern and load them onto an instance. I have also explored the promises of Cloud Computing in the context of porting our development environment to the Amazon AWS GovCloud. The end goal of moving to the Cloud is to realize a consistent environment across development, testing, staging, and production that could be instantiated quickly and reliably among many developers. An Accurate Approach to Complex Dielectric Constant Determination Aurya Javeed Mentor: Martin Barmatz The scarcity of known dielectric properties of liquid cryogenic hydrocarbons at 90 kelvin has limited the interpretation of Cassini’s 13.78 gigahertz radar and radiometric measurements. For example, the depths of Titan’s lakes are indeterminable due to a lack of reliable measurements of the imaginary dielectric constant of hydrocarbons such as methane and ethane. The dielectric constants of liquid cryogenic hydrocarbons are currently being measured by way of an accurate resonant cavity approach. The approach is applicable to the general class of TM0n0-resonating cavities coaxially loaded with an arbitrary number of concentric cylindrical regions of different dielectric properties. A prescription for the dielectric constant of a particular region follows from cavity resonant frequency and quality factor measurements. I will present an introduction to this accurate approach. Progress achieved in optimizing the computational infrastructure employed in the application of the approach will be summarized. Thus far, programs have been made more user-friendly and computation times have been reduced by a factor of 60. Progress in the preparation of a publication manuscript detailing the approach and its effectiveness will also be presented. Particulate Deposition: A Near-UV Spectroscopy Study on Steel Coupons Sam Johnson Mentor: Cambria Logan A myriad of particulate contaminants, invisible to the naked eye, are present in even some of the cleanest of environments. Though ideally human exploration of space would not carry any native microscopic organisms to foreign planets or planetoids, a compromise was reached that mandated that certain thresholds of contamination were not to be exceeded. In my project, I scanned sterilized, stainless steel coupons at increasing time intervals in order to determine the rates of deposition, as well as to study several factors affecting these rates. These factors include scratched surfaces, and monolayers of known carbonate crystals. In the near future, additional work will be conducted on coupons with silicate monolayers. These monolayer experiments serve to simulate conditions that may be experienced by a lander or rover that is exposed to sandy or otherwise rocky environments. In all

160   

  experiments, the coupons themselves serve to simulate any flat, stainless steel surface on a space bound machine. It is expected that particulate contamination (organic or inorganic) will occur in higher frequency at sites of defect, such as scratches or indentation. If these mechanisms can be better understood, our methods of preventing contamination (or likelihood of contamination) can progress correspondingly. The Workman-Reynolds Effect: A Study of the Ice-Water Interface of Dilute Salts Travis Johnson Mentor: Kevin Hand As a dilute aqueous solution of salt rapidly freezes, ions in solution become preferentially differentiated at the icewater interface. The Workman-Reynolds effect (WRE) is the resulting potential that forms between the ice and its constituent dilute salt solution. This potential can range from a few to ~215 volts with the magnitude and sign of the potential being primarily dependent on the type of salt, concentration of ions, and freezing rate. Little work has been done on the WRE, primarily focusing on dilute NaCl solutions of 10-6 to 10-4 molar. The results published by a few researchers are greatly inconsistent ranging from ~20 to 43 volts for NaCl. Preliminary experiments with 10-4 molar solutions of sodium chloride, potassium chloride, and ammonium chloride validates the WRE, yet the variability in our results conflicts with those previously published receiving readings of only ~6-8 volts for NaCl. Differences in pH between the ice and water have also been observed which could lead to significant H2 and O2 gas production. Further exploration into the WRE applied to icy worlds and Snowball Earth could pose as a possible driver for important prebiotic chemistry and origins of life. USSV Enhanced Simulation Matthew Jones Mentor: Michael Wolf With advances in autonomous capabilities in robotics, seaworthy vessels may no longer require a human operator. Steady progress in the development of maritime autonomy is bringing the operation of Unmanned Sea Surface Vehicles (USSVs) from a dream to reality. Because implementation requires the certainty of safe operations, the fast-moving USSV must be able to recognize other ships in the water and react in accordance with the international “rules of the road”. Such a requirement involves intensive testing both on and off of the water. Off-water testing takes place in a virtual simulation environment. Through successive modifications, this simulator was improved to address both user-access and functionality. Input was built to be quicker and more comprehensive. Additionally, avoidance behaviors were added to the simulated ships so that they would better emulate the international “rules of the road” in the presence of the USSV. Upon completion of modifications to the simulator code, an outside program was designed and built to generate thousands of randomized scenarios to test the USSV’s operational capabilities in each case. A scoring method was used to examine the results of each scenario and analyze the USSV’s overall motion planning performance. Design and Development of a Climbing Gecko-Adhesive Inchworm Robot Simon Kalouche Mentor: Aaron Parness There currently exists a void in the capabilities of servicing areas inaccessible to astronauts on board the International Space Station (ISS). A robotic platform used for inspection would be advantageous for its ability to be both autonomous and physically capable of navigating in tight spatial conditions. Due to the absence of significant gravity forces a directionally controllable fabricated gecko-adhesive, turned ON and OFF by application of a shear force (increases real area of contact which in turn increases van der Waals forces by orders of magnitude), is used to ensure constant contact between the robot and the surface of the ISS that is being climbed. A 6 degree-offreedom robot was kinematically designed to be able to climb both vertical and inverted planes as well as transition from plane to plane. In addition to the size constraint, the design of the robot was dependent on the performance capabilities of the gecko pads. Utilizing a controllable mechanism for actuating the gecko adhesives proved crucial to successful and consistent climbing. Because gecko adhesives have limited performance with respect to applied normal forces, shear forces, and moments, the robot’s mass needed to be minimized, the location of the center of mass needed to be optimized, and various mechanical advantages needed to be employed. Several design iterations were fabricated via rapid prototyping techniques and consequently tested. This design methodology combined with analysis converged the design to a final solution. Power System Testing for INSPIRE So-Hee Kang Mentors: Allen Kummer and Andy Klesh CubeSat projects have been proven to be cost and time effective missions that provide real science data in Earth orbit. In order to expand the use of CubeSats, Interplanetary NanoSpacecraft Pathfinder In a Relevant Environment (INSPIRE) will be taking two CubeSats into deep space for demonstration of survivability, navigation and communications. The prototype boards of INSPIRE have been delivered to JPL and are being tested and characterized. One such system is the electrical power system (EPS), which comes from the University of Michigan

161   

  with flight heritage from a previous CubeSat, Rax-2. EPS for INSPIRE takes the energy from the solar cells to charge the battery and provide the rest of the spacecraft with regulated power. In order to prepare the system for integration with the other subsystems of the spacecraft, procedures have been written to test and characterize the EPS hardware. Some of these tests include calibration of on-board sensors, input and output regulator efficiency and various battery characterizations. The procedures were then used to characterize and prove functionality of the prototype hardware. Some automated testing scripts in Python were updated to make the testing process quicker for future revisions of the board. Future work for the EPS system includes integration with the rest of the satellite, and other integrated tests such as environmental testing. Experimental Validation of Gravity Independent Anchoring Methods in Microgravity Conditions Jonathan P. King Mentor: Aaron Parness The goal of this research is to present the design of experiments and procedures to validate and evaluate the performance of current state-of-the-art methods for gripping and anchoring to objects of variable properties (mass, shape, roughness, etc.) in microgravity conditions. This work will utilize the design principles of two existent anchoring methods; microspines and gecko adhesives for rough and smooth surfaces respectively. To achieve microgravity, the experiments will be carried out during a reduced gravity flight aboard a modified Boeing 727. The objectives of the experiments will be to evaluate the performance of these systems in carrying out several microgravity maneuvers deemed to be valuable to potential future space missions including: supporting relatively large loads, grappling and capturing stationary and moving objects, dexterous manipulation of captured objects, drilling and sampling of captured objects, and object release. Performance evaluations will be conducted by characterizing the system behaviors for each test condition, and by analyzing the data acquired through the use of sensors monitoring force and torque, voltage and current, and accelerometer data and visual odometry. It is the desired result of these experiments that the collected data and lessons learned will allow anchoring technology to mature toward a flight ready design. Stereoscopic Image Processing for Autonomous Robotic Behavior Katherine Kirby Mentor: John Wright The All-Terrain Hex-Limbed Extra-Terrestrial Explorer (ATHLETE) is a six-sided rover that consists of two threesided tri-ATHLETE rovers with a cargo pallet between them. The rover is tele-operated, but it is useful to automate the tracking and manipulating of objects in the rover's vicinity. The ATHLETE rover interacts with cargo containers and tools, and the tri-ATHLETE rovers dock together with the cargo pallet. To facilitate tracking and manipulating objects, each object will have fiducial markers placed on it. Fiducials are high-contrast images with easily found features, and these features enable fiducial identification as well as calculation of the fiducial's position and orientation. The software takes in images from the rover's onboard stereoscopic cameras and detects the fiducials that are within view using either stereo images or a single image. Based on fiducial identification, objects are identified. The fiducial position and orientation are then used to determine the object's position and orientation. Once an object is identified and its location is known, commands are generated for the rover to interact with the object. This report will discuss the fiducial detection and identification parts of this process. Testing of Flight Software in a Simulated Environment Andrew Kirkham Mentor: Katie Weiss When flight software is developed, it needs to be tested rigorously to ensure expected execution in all test cases. Testing on hardware can be dangerous to equipment if something goes wrong and flight hardware resources are limited. Testing in a simulated environment allows the engineers to quickly test changes made to the software with no penalties for fatal execution. As more changes are made and more components developed, it is important to regression test in order to expose any errors that may have been introduced. By taking regression tests from the Mars Science Laboratory Flight Software, new tests were developed that are able to better exercise the JPL Flight Software Product Line Core and EPS components. These tests simulate ground commands to the flight software and analyze the results to determine if successful execution was achieved. The results showed that the current flight software build was able to handle both normal and abnormal commands successfully. Simulations of Titan’s Atmosphere Mitchell Kirshner Mentors: Karen Willacy and Mark Allen Of all the moons in our solar system, only Saturn’s Titan has a uniquely dense atmosphere. It is the only solar system body (apart from Earth) where there is known to be active organic chemistry occurring today. Analyzing Titan’s atmosphere can provide more information about the processes that occur on the surface of this moon and

162   

  possibly shed some light on the history of Earth’s atmosphere. A coupled transport and chemistry model known as KINETICS received values for reaction rates of various chemical interactions to produce output data pertaining to the flow of various gases in the atmosphere. This data was compared to that recorded by the Cassini-Huygens spacecraft. KINETICS provided information to aid further analysis of Titan’s atmosphere. Mars Science Laboratory Drilling Database Development and Drill Performance Analysis Douglas Klein Mentors: Daniel Limonadi and Avi Okon Extensive drilling performance information for the Mars Science Laboratory drill system currently exists in several formats and locations. The task for this period was to create a drill database that combines activity descriptions, pre-contact target knowledge, sample acquisition parameters, and performance data for all drilling activity conducted at Mars or in the testbeds on Earth. This report details the database development process and its intended usage. The database is operated from a MATLAB generated graphical user interface. Its design focuses on flexibility, functionality, user friendliness, and ease of maintenance. The database is used to search for and organize information, as well as perform analysis on the information. Analysis tools allow the user to visualize data to evaluate testbed consistency, explore drilling anomalies, compare flight results with test results, and predict drilling performance expectations on Mars for upcoming activities. Control and Automation Software for Xenics Infrared Cameras Ravi Kumar Mentor: William Johnson This project aims to develop software to automate the control of Xenics infrared cameras. The central task being automated is to rotate through the five filters in the camera, saving an image from each. The software supports a continuous capture mode as well as a single full rotation, and supports using different calibrations and exposure times for each filter. This program will enable us to more efficiently capture in situ images used for calibration and validation of data from orbital/airborne cameras such as HyTES. Mars 2020 Sampling and Surface System Field Testbed Kearney Lackas Mentors: Brett Kennedy and Tony Ganino Proper testing of complex systems is required. This is especially true for flight hardware where one is unable to fix hardware problems after launch. Furthermore, testing throughout the entirety of a project aids in the exploration of system-wide implications of potential hardware configurations. Thus, the task of this project was to develop a functional prototype testbed for the Mars 2020 Surface System, specifically the arm, drill, and containerization elements required for testing in laboratory conditions. The MSL Brassboard arm was the starting point for this testbed. Therefore, several components were designed and/or fabricated. Specifically, these components included backlash mechanisms to study the controllability and drilling performance between various actuator types, drill interfaces to accommodate both the TRL4 drill designed at JPL and the RANCOR drill designed by Honeybee Robotics, and cable harnesses to power both drills. Further testbed integrations would include integration onto a functional rover prototype for field testing. Temporal Analysis of Forest Cover Disturbance in the Laurentides Forest Reserve Kristofer Lasko Mentors: Marc Simard and Jennifer Corcoran The Normalized Difference Vegetation Index (NDVI) is an important indicator of vegetation health. Taking the difference between one year and a later year (i.e., “change map”) is a method that researchers have used to find disturbed and deforested land. A time-series of NDVI change maps was used in a custom python program that finds pixels in each change map which correspond to NDVI change values above a threshold of 0.2 NDVI. The pixels that correspond to a change in the NDVI value above the threshold were considered to be deforested or severely disturbed. The output raster values from the custom program were then assigned a new value corresponding to the year the change was detected. The end result is a single thematic map, depicting areas of deforestation and the year of deforestation. The resulting data can be incorporated with phased-array synthetic aperture radar (SAR) imagery to model several other factors, such as canopy height, forest regrowth rate, and estimates of aboveground biomass. NASA CloudWatch: Comprehensive Cloud Computing Usage Analysis and Management Kenny Lei Mentor: Darren Dao Cloud computing services are used internally at NASA to support its science and engineering efforts. However, an issue is posed with the cloud management since there is little visibility into how NASA actually uses these resources. The specific interest of this tool is to provide a way to visually analyze and manage the usage of cloud

163   

  computing resources with granularity at the account and service level. With real-time usage analysis, this tool aims to provide intelligent detections of abnormal usage and promptly alert account subscribers of this occurrence. Additionally, this information will help NASA gain a better understanding of how it uses the cloud and make adjustments to reduce spending (e.g., reserved instances). Usage history will also assist users in projecting their future cloud computing expenses. AWS billing data was parsed and stored inside a Mongo database for fast and compact querying. Backend source code was written for connecting with the AWS API, calculating statistics, storing data, and sending alert messages. A Grails web application was created to handle MVC and data requests. The frontend usage graphs were rendered using the Flot chart library. At completion, this tool will bring greater control and visibility of NASA’s cloud computing assets. Small Walk Efficiency: Characterizing Gait From Contact Forces of Small Legged Robots Nikhil Lele Mentor: Rudranarayan Mukherjee Investigation of the walking behavior of small legged robots using the OctoROACH from UC Berkeley to better understand the efficiency of these systems. Better models for the gait of cm-scale robots will lead to more efficient designs in the future. A test setup is built to measure shear and normal contact forces from a single leg walking under different configurations, using a capacitive sensor from Stanford University. Leg conditions such as speed, resting height, geometry, and compliance, as well as environmental conditions such as surface roughness are varied for different tests. From there, estimates of the coefficient of static friction are built as a measure of the efficiency of the leg under these conditions. Continued work will use a model-based controller to add robot properties such as mass to the system. The Porosity of Lunar Soil and Its Effects on Surface Temperature Erin Leonard Mentors: Benjamin Greenhagen, Paul Hayne, and Matthew Siegler The surface temperature of the Moon is a complex function of a number of physical properties including soil texture, solar insulation, and composition. The aim of this project is to study whether the porosity of lunar soil has an effect on its surface temperature by holding insolation and composition constant. We used a thermal camera to measure the thermal emission of lunar soil simulant samples of different density inside JPL’s Simulated Airless Body Emission Laboratory (SABEL) chamber. SABEL simulates a lunar environment by heating samples from below and/or from above using a solar simulator in a vacuum and a thermal shield cooled with liquid nitrogen. We also used a one-dimensional thermal model programmed with the same conditions to compare with the experimental results and investigate the thermal behavior at depth. We find that less porous samples heat and cool slower than the more porous samples. This results generally matched Diviner Lunar Radiometer observations of eclipse cooling and identification of “fluffy” areas on the Moon that cool to lower nighttime temperatures. Database of System Level Parameters for Historical, Current, and Developing Space Technologies Julia Levy Mentors: Pezhman Zarifian, John Elliott, and Sarah Bairstow JPL's A Team conducts many low Concept Maturity Level (CML) studies to determine the feasibility of new concepts and ideas. These studies oftentimes examine past missions as a reference for the capabilities of certain space technologies. It was determined that A Team is in need a searchable database that includes system level parameters of previously flown missions in order to improve the effectiveness of low CML studies. After initial research of system level parameters for various missions was performed, it was discovered that a team of workers in JPL Mission System Concepts is already in the process of developing a web database—called Integration of Missions, Programs, and Core Technologies (IMPaCT)—that documents elements of historical, current, and developing space technologies. The IMPaCT team was approached about enhancing their database to suit the A Team’s needs, and the two groups are now collaborating to develop a mutually agreeable final product. ISS-RapidScat Instrument Integration, Interface, and Safety Engineering Eric Lew Mentors: Jennifer Rocca and Stacey Boland I had the opportunity to spend my summer as a systems engineer on the ISS-RapidScat, a radar scatterometer payload traveling to the ISS in April of 2014. Since the project has only existed since 2012, it has moved on an accelerated schedule, allowing me to see how systems engineering works at several critical junctures of the project not normally accessible to an intern due to time constraints. Thus, my internship experience has been a flurry of meetings and paperwork. While writing papers and sitting in meetings does not sound especially daunting, the human aspect behind them certainly is. Any interactions are part of a large cycle: create the appropriate science and engineering aspects of a project; present to NASA and any other contractors associated with any stock

164   

  equipment given to the payload and/or sending it into space; get input on what paperwork is required through approvals, revisions, and assigning new tasks; repeat from step one. It is a very delicate balancing act that requires a systems engineer to be able to contribute any aspect of a project, for example, designing test suites for parts of the instrument, if the need arises. Reduction and Analysis of Data on Thermal Emission From Saturn Matthew Lewis Mentor: Glenn Orton Ground- and spaced-based telescopes have shown zonal variability in Saturn’s atmosphere through the presence of long trains of waves in the temperature field. Cylindrical maps revealing these waves from 1996-to 2010 are collected over time and across different instruments. To compare ground and space-based instruments, limbdarkening corrections, sigmoid weighting for combining images, and Fourier characterization according to a power series fit of the waves are all used. The location and strength of the waves are found to differ according to the latitude and wavelength observed. There is little variance of waves in the stratosphere that is sensed by images taken in ethane emission at 12.2-12.5μm or methane emission at 7.8-7.9μm, with waves centered near 0-30°N and 30-45°S. In the troposphere sensed by images 17.8-18.0μm, waves are originally centered near the equator but migrate north over time, eventually reaching 30°N or remaining near 30-45°S. A phase of 0.5°W per day retrograde is found to fit to the waves in both the stratosphere and troposphere. General agreement is found between observations from ground-based telescopes and the Cassini thermal infrared instrument (CIRS). Smart Adapter: A Study on Implementing a Protocol Decision Making Pattern on FPGA Barry Yunliang Li Mentors: David Henriquez, Yutao He, and Rick Graves The growth in gate count on commercial FPGAs provides the opportunity to implement protocol decision making processors within an HDL device to identify the type of protocol at the hardware interface and simplify the software complexity. Of specific interest for this project, Universal Asynchronous Receiver/Transmitter (UART) protocol with unspecified baud rate and Instrument Command Control (ICC) protocol are first used for developing the overall structure as well as testing the decision making ability of the adapter. The method of sending software commands with known op-codes and using multiple preset frequency UART modules and an ICC module to sample the input data in parallel is used. Based on the correct outputs of Modelsim and ChipScope, the outer decision making core successfully identifies the type of protocols by collecting result from each input sampling module and arbitrating at a higher level. Once the framework is set, four more protocol interfaces, TZ DownLink, Rover Inertial Measurement Unit (RIMU), Descent Inertial Measurement Unit (DIMU), and NonVolatile Memory Camera (NVMCAM), have been incorporated to the adapter. The positive result of the research shows the feasibility and the potential of building a powerful universal adapter by adding more protocol interfaces with more complex decision making logic. Automatic Visual Contact Detection Using Wavelet-Based Fractal Signature Analysis Andrew Liang Mentor: Michael T. Wolf This paper describes the detection and classification of particular boats using the camera system of an autonomous sea vehicle (ASV). An omnidirectional camera head atop the ASV feeds grayscale images of the environment into the central perception system. In this work, we aimed to identify a rigid-hulled inflatable boat (RHIB) at various ranges. The detection of the RHIB is challenging due to its small size and its visual similarities to the water. This problem was addressed with a wavelet-based fractal signature algorithm. Horizontal, diagonal, and vertical signatures were extracted from a training set of ground-truthed images to examine important features of the RHIBs and to calculate a signature bandgap range of values for testing. Analysis of both positive and false positive hits on the testing dataset revealed that horizontal features were not strong representatives of target boat characteristics, but vertical features were strong indicators. In addition, studies revealed that the clustering algorithms on wavelet hits need to be reexamined and improved upon to better detect the RHIB. The identification and detection algorithms are demonstrated via image replays and, in the future, via on-water tests. Investigating the Merits of FORMULA for the Formal Verification of MBSE Ontologies David Lindecker Mentor: Nicolas Rouquette Formal verification can be intractable in many practical situations and as such there are different tools that emphasize varying aspects of the problem. This work deals with investigating the benefits of applying such a tool, FORMULA from Microsoft Research, to the formal verification of JPL systems engineering ontologies. We focus on the application of FORMULA’s model finding features to verifying properties of the specification itself. One recurring issue with formal specifications of MBSE ontologies is that they tend to include many details that are irrelevant to reasoning about many properties. It is desirable to create a simplified version of the specification and show that it is equivalent to the original with respect to the properties to be proven. We demonstrate FORMULA’s ability to

165   

  accomplish this with a case study involving SysML and the restrictions imposed on it at JPL. We use transformations and model finding to discover inconsistencies between the two representations. We have found that the model finding capabilities of FORMULA are somewhat limited but still promising for addressing formal verification issues in MBSE. Automated Unit Test and Validation for Operating Curiosity on Mars Michelle Liu Mentor: Jeng Yen Before the rover planners send commands to Curiosity, the sequences are tested in simulation. When applicable they may also be tested in hardware. Rover Sequencing Visualization Program (RSVP) is a suite of software tools that gives rover planners the ability to generate command sequences and visualize the results in a fast high fidelity simulation. The simulation in RSVP uses an embedded rover flight software which controls Curiosity’s arm and mobility systems. To improve the efficiency of validating the simulation results provided by rover flight software, a script is created to automatically run test cases and check the results provided by RSVP. Baseline state variables from the same test case run on hardware are compared against simulated state variables from RSVP and a summary of any violations detected is provided. NASA’s Earth Science Satellites: The Characterization and Calibration of HyspIRI/HyTES Support Systems Christopher Martin Mentors: William R. Johnson and Gerardo Rivera Currently, over 20 functioning satellites operate under NASA’s Earth observation program, with 25 more somewhere in the design stage. Together they form a network of eyes in the sky, giving scientists valuable information on our planet’s ecosystems. Satellites such as HyspIRI will carry advanced multispectral imagers capable of providing information on natural disasters such as volcanoes, wildfires and droughts. However, scientific use requires the characteristics of an instrument’s system be outlined in painstaking detail. The conception and creation of the mechanics and procedures for characterizing an MCT array with a long wave infrared cut-off formed the basis for my research. To provide SI traceability, a blackbody was used as a standard temperature metric to take thermal data for various filters across a range of temperatures and integration times. The resulting calibration pipeline was successful in correcting filtered images and mitigating error. Future research might focus on constraining sources of error to further improve the accuracy of measurements. This research provides a baseline foundation for future airborne and space sensors that will be using similar MCT technology. Fluid Flow Behavior of Organic Solvents and Materials Under Zero G Steven C. Martinez Mentor: Murray Darrach The study and understanding of the properties of commonly used materials onboard the International Space Station when exposed to different immiscible fluids is important for the development of future hazard prevention instrumentation for the ISS and the future of long duration manned space flight. There is a need to understand how different chemicals act, as a means of developing detection procedures which will separate any foreign chemicals which contaminate the water supply in order to safeguard future spacecraft and crew life support systems. Under normal earth gravity (1g) trace hydrophobic organics inside water will separate, forming either layers on top or bottom of the water, depending upon their density. Under microgravity the behavior of these hydrophobic organics (e.g. aldehydes, alkanes), especially their interaction with surfaces and under variable flow, is poorly understood. This has dramatic implications for the design of future miniature sample handling and delivery systems for water quality monitors. An experiment has been design, developed, built and tested to be flown on the International Space Station using a 3U Ardulab module to be interfaced with the current Nanoracks equipment onboard. The experiment, which was developed at the Jet Propulsion Laboratory, in Pasadena CA, will test the nature of immiscible liquids/chemicals in a microgravity environment along with how they interact with each other and how they behave when introduced to several different surfaces. Electromagnetic Compatibility Support and Automation for Radiated Susceptibility Timothy Mastny Mentor: Pablo Narvaez Radiated Susceptibility testing for Electromagnetic Compatibility is a time-consuming process that requires tedious manual real-time control of instruments and data recording of real-time results. Automated testing, however, resolves this issue by instantly recordings data while automatically controlling instrumentation, thus minimizing errors and time. Therefore, the goal of the project was to update automation software and provide complete and detailed documentation of the application. Work on the automation process was started by Electromagnetic Compatibility engineer Nelson Green. Using TILE, a commercial instrument control system, he created a basic template that a TILE programmer could use to design an automated test for a specific Radiated Susceptibility test. Unfortunately, TILE limits the input from the user and requires knowledge of TILE programming to be able to

166   

  change any testing parameters. In an effort to increase the flexibility and usability of TILE automated tests, new detailed code and TILE templates were written to facilitate this goal. The document provides basic step-by-step information that teaches the software to a new user. In addition to a basic operating procedure, the document explains advanced features for the benefit of users interested in the entire feature-set of TILE. The automation program, along with accompanying documentation, ensures that a new user can successfully use the software to run Radiated Susceptibility tests. SWOT Loop Heat Pipe Testbed: Data Acquisition Setup, Integration, and Preliminary Testbed Characterization Robert McBride Mentor: Ruwan Somawardhana As a joint U.S.-French mission between oceanographers and hydrologists, the Surface Water and Ocean Topography (SWOT) satellite plans to generate the first global survey of Earth’s ocean and inland surface water. The Ka-band Radar Interferometer (KaRIN), the primary payload, is being designed to acquire high-resolution elevation measurements of Earth’s lakes, rivers, reservoirs and oceans. Precise measurements require a hardware temporal stability of 0.05°C/min. To meet this requirement, a versatile heat transfer technology known as a loop heat pipe (LHP) is being tested and characterized to carry waste heat (~300 W) from KaRIN to an external radiator. This summer project involved the design, setup, and testing of a data acquisition system to conduct experiments on the SWOT LHP testbed. Initial setup required constructing a LabView VI, tuning a chiller/heater, and testing the inherent accuracy of type K thermocouples as well as the overall measurement system. Test procedures along with a GUI for data post-processing were developed to assist in characterizing the LHP’s temporal stability. Experimental results and the knowledge gained from testing will be used to improve future SWOT LHP tests. Modeling Solar Radiation Pressure in Solar Probe Plus Sarah Elizabeth McCandless Mentors: Allen Halsell and Troy Goodson Solar Probe Plus (SPP) is a mission that will explore the solar atmosphere in great detail for the first time. SPP will use in situ measurements and imaging to understand how the solar corona is heated and how the solar wind is accelerated. The Jet Propulsion Laboratory will provide navigation for SPP during operations, which requires a faithful trajectory model that includes all disturbances. A significant disturbance is solar radiation pressure, which increases to 500 times that experienced at Earth upon SPP’s closest approach to the Sun. This presentation documents the development of and improvements to the solar radiation pressure model. This includes development of time-based solar panel area models, spacecraft shape modifications, and extensive trajectory comparisons. Additionally, the effect of solar plasma on Doppler data is investigated, particularly as it pertains to orbit determination. Risk Minimization of Innovative Detector Payload in Lab and on Orbit Using Simulation and Hardware Casey Meehan Mentor: Paula Pingree Both lab and orbit environments harbor a variety of risks that may damage electrical flight hardware and lead to mission failure. In lab, voltage transients from any bench source can damage hardware in a matter of microseconds. On orbit, an instrument may overheat if oriented towards the sun for an extended period of time. To minimize both the lab and orbit risks of an innovative detector payload flying on the Geo-Cape Read-OutIntegrated-Circuit Flight Validation Experiment (GRIFEX) CubeSat, orbit simulations were created and hardware designed that help avoid these unnecessary risks. Simulating a range of likely orbits with passive magnetic alignment revealed expected risk of instrument sun exposure. In expected orbits, GRIFEX’s payload should not experience excessive sun exposure, though likelihood of entering higher risk orbits should be further investigated. To protect hardware in the lab, circuitry was designed to provide a redundant overvoltage protection from a lab voltage source. Using SPICE simulation, the circuit’s programmable overvoltage set point precision and reaction time were optimized to improve safety. Circuit testing and characterization should be continued for approval of lab integration. SWOT Deployment Testing: NAMS Validation of ACT10 SWOT Hinge Repeatability Herbert Mehnert Mentor: Samuel Case Bradford The Surface Water and Ocean Topography mission developed by JPL and slated for launch in 2020, makes use of twin Microwave arrays deployed on hinged booms to conduct research on the globe's in-land and ocean water reserves. The quality of the scientific data is dependent on various factors including the tolerances and precision of the mechanical subsystems that support the science package. This paper addresses the accuracy of deployment of

167   

  the hinge element in the SWOT booms developed under ACT10 (Advanced Component Technologies), and covers the test systems used to validate the hinge deployment including the NAMS (Nano-meter Accuracy Measurement System). The document also compiles the researcher's work in conjunction with the ALPS (Advanced Large Precision Structures) Laboratory at JPL from June to August of 2013. Electron Precipitation Events in the Ionosphere and Geomagnetic Activity Daniel Miladinovich Mentors: Tony Mannucci and Mark Butala The objective of this study is to understand the effects of electron precipitation in the Earth’s ionosphere. We focus on high latitudes in the northern and southern hemisphere. The Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) satellite system provides electron density measurements, using the Abel transform inversion method. The peak electron density and its altitude are plotted for each occultation event during January 1, 2008 and November 8 – 16, 2012. Here we compare electron density profiles between summer and winter, for magnetically quiet and active days in the northern and southern hemisphere. This study presents seasonal and diurnal variations between electron density profiles. Also, correlations between the AE index, the magnitude of the peak electron density and its altitude reveals relationships or lack thereof between these variables, which gives us a perspective of the underlying physics in the polar regions. Python-Based Automation for Detector Characterization Alex Miller Mentors: Shouleh Nikzad and Timothy Goodsall I present the design of an automated system to increase the efficiency of experiments conducted in the MDL's Advanced Detector Array and Characterization Lab. The Python-based system involves a series of stepper-motors, a picoammeter, and a visible-to-near-infrared grating and monochromator which all serve as a setup for detector characterization. With the automation in place, a researcher simply needs to run the program in order to obtain a range of wavelength detector response with resolution as low as 0.1nm/step. The new setup will not only save time overall, but will allow for more characterization runs per detector in order to form a better analysis of each chip. In addition, the error margin from run to run and detector to detector will be reduced, allowing for better understanding of overall detector characteristics. This better understanding will in turn provide guidance for future detector development. The automation techniques created and utilized for this project will be used as a guide for creating better experimental detector setups in the future both at JPL and Arizona State University. Design and Analysis of Low Gravity Soil Mechanics Hardware and Software for Comet Surface Sampling Chris Miller Mentor: Brian Trease Collecting and analyzing samples has long been an important way to gather concrete scientific information about materials and environments. The systems that acquire these samples must often be specifically designed for each new environment to ensure their successful operation, and an important method of identifying an appropriate design is through mathematical modeling. To generate these models, empirical data is first required. Using an iterative design/build/test approach, an apparatus was developed to collect both qualitative and quantitative information about soil particle interactions and motion during multiple penetrometer testing scenarios. Not only will observations made in a laboratory setting prove useful in further understanding soil mechanics on Earth, but the apparatus was also designed for operation onboard a Reduced Gravity Aircraft (RGA) to collect information about how these particular soil behaviors change in low gravity environments. The results of these tests and the accompanying generated computer models will support future missions requiring sampling in low gravity conditions that exist on comets and asteroids. Marsviewer iPad Application: A Case Study in Client and Server-Side Solutions for Mobile Deployment Nolan Miller Mentors: Bob Deen and Nicholas Toole The Marsviewer application is the primary image-rendering tool for the image products generated by the Mars rovers. With recent advances in mobile device technology, there is demand for a mobile application with the same functionality as the current Marsviewer utility. Development of such an application has provided insights and solutions to the problems that arise from content heavy mobile deployment. In order to provide the same functionality as the desktop application, server-side processing became necessary and different divisions of labor were investigated. In particular, Webification (W10N) was implemented to expose many of the core functions of Marsviewer to arbitrary client-side consumption. Mobile platforms also present different interaction opportunities than conventional computing, such as multi-touch panning, pinching and tapping. A variety of user interfaces were tested over the course of the project and a suitable interface was created. There is still an opportunity for further testing with more exotic interfaces that may provide more natural interaction with the data. Ultimately, through the fusion of server and client processing, the Marsviewer iPad application has demonstrated the feasibility of mobile platforms for consuming data-heavy technical resources.

168   

 

Multi-Mission SSE Software: Automated Reconfiguration and Build System for Supporting Multiple Flight Projects Steven Morad Mentors: David Henriquez and Derek Adams Currently, the ground support equipment software, and its associated build system, are customized for every mission. This leads to forked codebases, build system obfuscation, and unused code residing in the ground support equipment software source tree. Rebuilding the Soil Moisture Active Passive (SMAP) project using a dynamic build system was a way to mitigate this problem. Moving from Autotools to CMake has yielded a more transparent and dynamic build system. Almost all scripting is now self-contained within CMake build files, and the build system no longer requires external dependencies to run. This new build system can easily be adapted to support future missions, and it allows easier refactoring and restructuring due to a much higher level of abstraction. Developers can spend more time writing code, and less time learning and maintaining the build system. Cold Atom Laboratory: Ground Test Bed Development and Magnetic Shield Design David Newill-Smith Mentors: Anita Sengupta and David Aveline Bose-Einstein Condensate (BEC) is a state of matter achieved at extremely cold temperatures, far less than a millionth of a degree above absolute zero. In these conditions, quantum mechanical effects can be seen on a macroscopic scale, and the study of matter in this state will lead to a better understanding of the fundamental laws that govern our universe. The Cold Atom Laboratory (CAL) is an instrument being built for the International Space Station capable of studying BEC’s in microgravity for lengths of time not attainable on Earth. An important step in the design of the instrument is the development of CAL’s Ground Test Bed laboratory. Over the course of this project, a variety of tasks were completed in order to further this development and characterize different aspects of CAL, including the design and construction of an interlocked laser enclosure, optical collimator mounts, a computerized Magnetic Field Plotter, and enriched potassium dispensers. The work also included experimental characterization of the alignment tolerances of the Thorlabs Fiberport system and programming arbitrary waveforms for magnetic field control. In addition, a study was co-conducted with another intern, Lucia Baker, to design magnetic shielding for CAL’s science module. Mars Rover Caching Architecture Testing and Analysis Rowland O'Flaherty Mentor: Paul Backes The proposed Mars Sample Return (MSR) mission campaign has the ultimate goal of bringing Mars samples to Earth for examination and analysis in Earth-based laboratories. To achieve this goal it is proposed to split the campaign into three flight missions: the caching mission, the orbiter mission, and the lander mission. The caching mission, now planned for a 2020 launch, would carry a rover similar to the Mars Science Laboratory mission’s Curiosity rover which would acquire and cache the samples on the surface of Mars. Several architectures have been proposed for the acquisition and caching process. The Minimum Scale Sample Acquisition and Caching (MinSAC) architecture employs the 5 degree-of-freedom robotic arm on the front of the rover to perform the tube transfer tasks for caching. Currently it is unclear if the arm on the rover has the capability to perform the required tasks. Algorithms were developed that demonstrate the caching procedure with a prototype robotic arm with similar kinematic structure as the proposed robotic arm on the rover. This demonstration and analysis has produced confirmation that the MinSAC architecture may be feasible for the caching task in the caching mission. Mission Operations Wiki Design: The Development of a Communications and Operations Tool Megan O’Sullivan Mentor: Patrick J. Guske The project for this summer was to develop a user-friendly system that could be employed by the Mission Operations System Team (MOS) of the OCO-2 Project to manage and organize their documentation. The main problems faced during this project were finding the right wiki development software that met all of the system requirements, and determining the organizational structure of the resulting wiki. The wiki development software that was chosen was Confluence 5 due to its user-friendliness and its ability to fulfill all communication and operational requirements. Another added benefit of Confluence 5 was that during this summer the decision was made to purchase a lab-wide license for Confluence, meaning the wiki that is developed for OCO-2 will have on-lab tech support. The resulting product is an OCO-2 Mission Operations Tool that is composed of a team communication system and has an organizational structure that is easy to navigate. The structure has been reviewed and approved by fellow team members.

169   

  Numerical Modeling of the Boundary of a Molecular Dust Cloud in the Taurus Dust Complex Matt Orr Mentors: Paul Goldsmith and Jorge Pineda Molecular clouds are the nurseries of star formation, and understanding their boundaries and how they form is critical to knowledge of stellar formation, and stellar formation rates in our galaxy. Within the Taurus molecular cloud complex, our group has previously observed several boundary regions, where the present species transition from predominantly atomic to molecular in nature. One of these regions, a linear edge boundary, was modeled using the Meudon PDR code, and then compared to spectral line data taken with the Herschel Space Observatory (HSO) and the Five College Radio Astronomy Observatory (FCRAO), using the Ratran software suite. We have found that the cloud complex is consistent with a cylindrically symmetric density profile, following a function proposed by I. King (1962), with an interstellar radiation field (ISRF) between 0.5 and 0.05 times the local average value, as found by Habing (1968). Additionally, we varied the “packing fraction” of the dust and gas in the cloud, in order to better fit the observed spectral features for CI, CII, CO and 13CO. Though we have significantly restricted the values of our physical parameters, principally the density, value of the ISRF, and packing fraction, subsequent work will focus on finding more exact values. Investigating Lakes in the Northern Alaskan Slope Andrew Park, Jr. Mentor: Kevin Hand Many of the lakes in the northern Alaskan slope offer an interesting perspective of the complex chemistry that occurs at an ice-water interface during freezing. In order to better understand this chemistry, it is necessary to employ both scientific methods and engineering applications. The first part of this summer project focuses on the extensive research done on the Workman-Reynolds effect, which describes a phenomenon across the freezing front of dilute aqueous solutions, where depending on freezing rate, present ions, and the concentrations of those ions, a potential difference develops due to the advancing ice matrix selectively including some ions while rejecting others. The second part of this research project discusses the science objectives that can be achieved with a small-scale underwater ice drill that is able to core ice samples and capture them for laboratory analysis. By combining these two research efforts, the Planetary Ices Group hopes to develop a more comprehensive understanding of the chemical processes that occur in icy environments. This knowledge may then be applied to the generic research of icy moons such as Europa, which many scientists and astrobiologists believe to be habitable for extraterrestrial life. Web Development in the Cassini Project Phillip Peralez Mentor: Diane Conner Cassini has many web services and sites that are spread across many workstations and servers. Several of these services and sites were flagged to be improved or fixed. Parts of the project that were worked on increased security by modifying the file sharing framework and the code that it would generate for the user. Several monitoring services for network connections and workstation statuses were overhauled to be easily accessible and more user friendly. In addition, fixes for lost dependencies of a status tool for Titan were made to allow the tool to be functional again. Porting PlotOhMatic to iPad Alicia Pixton Mentor: David Oh The current user interface for JPL’s Mission Data Processing and Control System is somewhat antiquated. This project seeks to demonstrate that tools employing touch-sensitive interfaces could potentially replace many of the existing command line interfaces without losing functionality, and while becoming significantly more intuitive to use. By porting basic functionality from the existing software tool PlotOhMatic to iOS, and by altering PlotOhMatic’s user interface to take advantage of the iPad’s innate user-interactive capabilities, this project demonstrates that existing software tools can be updated to provide a more intuitive user experience as well as markedly increased tool portability. DARPA Anti-Submarine Warfare Continuous Trail Unmanned Vessel (ACTUV) Alexa Poulton Mentors: Ryan Mackey and John Day The ACTUV project is working to create a completely autonomous vessel which will be able to find and track enemy submarines in open water during sixty to ninety day missions. JPL, in support of prime contractor Science Applications International Corporations (SAIC), is developing a useful, testable prototype reasoner to demonstrate Integrated System Health Management (ISHM) capabilities of the vessel. The prototype reasoner processes data similar to that produced by the ACTUV C4 (Command, Control, Coordination, and Communication) Room, using the best available information provided by sensors and computer performance monitors. The ISHM prototype executes

170   

  rules created within the Spacecraft Health INference Engine (SHINE) reasoning system, which are compiled and delivered in C++. These SHINE rules are used for global and local reasoning, providing detection and recovery of faults, mode estimation, and diagnosis of vessel hardware. The ISHM prototype is sufficiently realistic to permit real testing on the ACTUV Surrogate test ship, requiring only minor modifications to the SHINE rules to accommodate the final configuration. TTC Central Linux Application Testing Automation David Powell Mentor: Carol Glazer Telemetry, Tracking and Command (TTC) Central Applications Task is part of the Deep Space Network (DSN) Project for support of all missions using DSN services. This includes both NASA missions from JPL and other centers and international missions (e.g. ESA, JAXA, ISRO). Reliability is very important to the mission customers of the TTC Central software. Newer missions have requested the delivery of a small subset of applications on a Linux platform rather than the Solaris platform that has been the standard for many years. Testing on this newer platform is key to being able to deliver the subset of applications. Automated tests are used to improve repeatability and efficiency in testing. The TTC Central software follows a 6-month development/test/delivery cycle. It is important to be able to sufficiently test the deliveries in a short time period. For this project, a script was written to run the applications in an identical fashion on a Solaris and Linux machine. All output is collected and compared for any differences, and any tools that do not behave identically are reported. Flight Software Functional Integration Testing Adam Randall Mentors: Tom Fouser and Cindy Oda Flight software controls all aspects of a spacecraft. Because of this, it has great responsibility to be reliable. Therefore, software must undergo extensive testing in order to ensure that it will be able to work properly, as well as successfully recover from the inevitable errors that occur in space. Test scripts are written to create situations that the software must be able to recover from. The success of these tests ensures functionality, as well as robustness. Tests scripts, like the software they evaluate, must be developed and maintained. As changes in flight software are made, it is the tester’s job to ensure that the test scripts still function properly. My task is to improve the ability of test scripts to recognize received messages from the spacecraft. As the software is developed, the format of these messages changes and the tests must be modified to work correctly with new flight software versions. This is essential to allow the continued feedback from testing that helps the developers create more resilient software. Creating Visual Media to Aid Communication of Research Objectives to a Non-Scientific Audience Maria Raykova Mentor: Marc Simard Remote sensing instruments such as UAVSAR are being used to map the structure of mangroves, coastal wetland forests which are among the most endangered ecosystems on Earth. At various points in this process, it is necessary for researchers to communicate with members of the non-scientific community. The community includes grant donors who read the proposals, forest rangers and local officials who use the results to inform their decisions, and the citizens who will be affected by those decisions. It is important that these people all understand what the objective and the value of the research is. The aim of this project was to convey that information in a clear and concise manner. The form of an animated video was chosen because communicating with visual aids is more effective than using words alone to address a wide non-expert audience. The video provides a quick lesson on mangrove ecology, threats to mangroves, and NASA’s use of radar remote sensing to study these. When it is completed, the video will be posted on the Internet, and will also be shown during presentations in multiple languages. Calibration Methods Jacob Richards Mentor: Emilio Vazquez The process of calibration is used to verify and ensure accuracy of measurement systems. The measurement instrument in question is typically compared to a known working standard, and then possibly adjusted to account for error. Error is the sum of systematic and random error, and the systematic errors are reduced by adjustments. There exist a variety of different calibration methods one may use, which yield different levels of accuracy. Here, we consider the Lakeshore Model 340 Temperature Controller along with a temperature sensor as an example of a measurement system that requires calibration. The temperature controller utilizes a calibration curve that maps input quantity to output temperature values. There are several ways to calibrate a Temperature Controller that will produce a certain degree of accuracy. These calibration methods are (in order of accuracy from low to high):

171   

  Generic Curve System Calibration, Electrical Calibration, Manufacturer Curve System Calibration, and Custom Curve System Calibration. We carried out a case study, comparing four sensors with a known standard and using each of the four calibration methods, to determine the errors resulting from each type of calibration. We found that General Calibration is the least accurate and Custom Curve System Calibration is the most accurate. Overall, different calibration methods are useful for different applications, due to their varying strengths and weaknesses. Covering the SMAP Ground Data Systems Automation Gaps With a Pass-Aware Time-Based Automation Scheduler Samuel Richerd Mentor: Antonio Sanders The Soil-Moisture Active Passive (SMAP) mission is intended to be JPL's first fully automated operations mission. The Ground Data Systems (GDS) team has identified three areas of automation to be addressed by the GDS software: file-based, pass-based, and time-based automation. The Time-based Automation Scheduler (TAS) provides the ability to schedule and manage miscellaneous reoccurring GDS directives driven by timing (e.g. log file cleanup, system maintenance) and allows these directives to be configured to avoid conflicts with spacecraft SN/NEN passes. Relying on the Unix-standard cron daemon for GDS directive execution, the TAS provides a graphical user interface to manage the scheduled directives, creates a static crontab file containing time-datedirective entries, and autonomously re-schedules directives and re-creates a crontab as necessary when a new or updated pass list is available. Developing a Process for Making 3D Prints of Extraterrestrial Terrain Jacob Rodeheffer Mentor: Oleg Pariser 3D printing technology can be used to make physical models of extraterrestrial terrain features. Such models could be used by NASA scientists and engineers in their work or for advancing the knowledge of the general public. In theory, a 3D digital model can be generated from any object or scene that has been sufficiently photographed, and this digital model can be printed by a 3D printer to make a (possibly scaled) copy of the original object. The fidelity of the copy depends on the three elements of the 3D printing process: gathering 3D data of the object, a method for producing a 3D-printable 3D digital model from the data, and the technology of the 3D printer itself. This paper will present and evaluate different variants of these second and third elements. Simulation of a Marine Vehicle Miguel Rodriguez Mentor: Abhinandan Jain The Dynamics And Real Time Simulation (DARTS) Laboratory at JPL has developed software, known as Dynamics Simulator for Entry, Descent & Surface landing (DSENDS), to simulate the behavior of rigid bodies. The overall goal is to sufficiently expand DSENDS to allow for the simulation of marine vessels. Such a simulation generally requires a coupling of Euler's Equations for rigid bodies and the Navier-Stokes Equations for fluid flow, making this a set of coupled ordinary and partial differential equations. An alternative and common approach is to account for the hydrodynamical effects that a vessel experiences through an added mass, reducing the problem to a set of ordinary differential equations. Routines are developed to model a surface marine vessel in DSENDS with the latter approach. These routines are then tested by applying them to a cannonball, a cylinder, and a marine vehicle. Automation for Browser Testing Using PHP and Selenium Jose Rodriguez-Salinas Mentor: Joseph C. Hunt Functional testing for web applications can be a time consuming, manual process. In most cases this testing requires long procedures that must be repeated manually by testers for various supported browsers. This presentation will describe the research and implementation of an automated solution. We built a comprehensive suite that simulates user actions (i.e. clicking, dragging or submitting data) and verifies that they match expected behavior. We used current standards and practices to create tests that are robust, easily extended and cover all the displayed data and user interactions in our web application. In the future this functional test suite will be added to the Continuous Integration server to run automatically on code changes and generate reports for the development team. Assessing Land Cover and Land Use Change at a Watershed Scale in Central America Melanie F. Rosenberg Mentors: Marc Simard and Jennifer Corcoran Mangroves are tropical and subtropical salt-tolerant trees that provide a nourishing habitat for a biologically diverse environment as well as natural protection from hurricanes, tsunamis, and other natural disasters. Land cover maps of the Yucatan Peninsula, Mexico and the Gulf of Fonseca, Honduras were developed for the years 1986, 2000, or

172   

  2010. The remotely sensed data used to create the land cover and watershed maps included: Landsat 30m optical imagery, Normalized Difference Vegetation Index (NDVI), and elevation data from the Shuttle Radar Topography Mission (SRTM). Based on the land cover results, change detection maps were created to illustrate the amount of deforestation in mangrove forests between the years of 1986 and 2010. Watershed maps were generated using the SRTM and land cover and land use was summarized within all contributing watersheds connected to each of the three field sites. The analyses performed at a watershed scale connect contributing upland land cover and land use change to the coastal mangrove ecosystems. This work compliments ongoing data collection of socioeconomic information within the contributing watersheds. Future work will focus on how land cover and land use change influences socioeconomics of the region, such as whether an increase in agriculture, including shrimp farming, contributes to economic successes at the cost of mangrove deforestation and if mangroves continue to thrive in harmony with local populations. Mangroves are a vital part of the community and are recognized as a valuable resource to be protected. This research will significantly contribute to the understanding of mangrovesocioeconomic dynamics at the watershed scale. Verification and Validation of the Orbital Carbon Observatory #2 (OCO-2) Satellite: Assembly, Testing, and Launch Operations Phase #2 (ATLO-2) Juan Felipe Ruiz Mentors: Pavani Peddada and Ben Solish The Orbital Carbon Observatory satellite is an earth observing Jet Propulsion Laboratory spacecraft designed to study atmospheric carbon dioxide (CO2) from a polar, sun synchronous orbit. OCO-2 will provide global, in-situ carbon dioxide measurements with a high level of precision and resolution, and will be able to detect seasonal changes and variations of this atmospheric carbon dioxide. By characterizing the distribution of atmospheric carbon dioxide in a discrete, time dependent, and geographical manner, as opposed to holistic measurements, the OCO-2 satellite will help scientist further understand the natural regulation of carbon dioxide, and increase the accuracy of models that predict future CO2 growth and its impact on global climate. The Project Verification and Validation team for OCO-2 is tasked with ensuring that all the system level and subsystem level verifications (Level 2 thru Level 4) of the requirements are complete and accurate. A key factor in this flow as the project approaches launch is Observatory testing (Instrument & Spacecraft) and certifying that the verification is complete. In addition to verification, another extremely important factor is the validation (the other “V” in the VnV) of the entire OCO-2 Observatory. Validation of the Observatory is completed in the final test of the entire system to see if the system was built properly and if it will accomplish the overall mission objectives (i.e. acquire photons and transmit the data down to earth for the scientists to analyze). Besides the normal flow of VnV, some additional tasks completed within this capacity include the creation of contingency plan recommendations for the observatory, compiling and submitting a Test As You Fly exception list, and updating and revising OCO-2 nominal and contingency procedures for the spacecraft instrument. These tasks support the work of the PSE V&V team, and help move the project closer to full launch and operations readiness. DARPA F6 Trade Studies and Margin Analysis (Part 1) Tyler J. Ryan Mentor: Steven L. Cornford After three years of the F6 project, it has developed from a simple simulation to a robust spacecraft design tool capable of creating, simulating, and analyzing the construction and operation of fractionated spacecraft clusters. Upon arriving at JPL for my third summer, I spent a couple weeks familiarizing myself with the progress over the last year. After this process, the F6 tool was analyzed and debugged, and its outputs examined for verification and validation in preparation for the completion of the project. In doing so, several python modules used in payload and bus selection were constructed that have added flexibility and functionality to the tool. The model also underwent a complete renaming to provide consistent and understandable naming convention. Furthermore, a Margin Analysis Module was implemented, allowing for users to account for design margin in their spacecraft bus selection. It is of great value to analyze how margining on payload design parameters can influence the cost of the bus that must house the payloads, the cost of the launch vehicle, and the mission itself. This paper will summarize the work over the summer but focus on showing that our Margin Analysis Module can allow for a robust frontend analysis on how design choices influence mission cost as well the amount of data, and therefore value, returned by the mission. A New Technology Report has been filed for the Margin Analysis Module (NTR#49251). Developing Force Models for Full Body Contact and Soft Tire-Solid Ground Interaction Adam Ryason Mentors: Abhinandan Jain and Calvin Kuo This research develops force contact models to be implemented and used in real-time dynamic simulations. There currently exists a need for full body contact modeling and soft tire on solid ground interaction. The Collision Forces model enables rigid bodies to interact with external objects by creating collision constraints between intersecting bodies and using penalty method, applies a spring-damper force based on the penetration distance. The Tyre Collision model simulates vehicle applied forces using Pacejka’s “Magic Formula”, creating soft tire on solid ground interactions, favorable for High-Mobility Multipurpose Wheeled Vehicle (HMMWV). The Tyre Collision applies

173   

  longitudinal and lateral forces based on the vertical load, camber angle and slip angle while avoiding divergence at low velocities. The contact models are applied to a bipedal robot and rover vehicles assembly for testing. Various regression tests are performed and a GUI was used to ensure that the simulations respond naturally and results are compared to known methods. The Design, Synthesis, and Characterization of Novel Thermoelectric Materials for Use in Cooling Systems Aboard Spacecraft Robert Salazar Mentors: Jean-Pierre Fleurial and Charles Hays Scientific data received from detectors aboard spacecraft is limited by the detector’s dependence on an exhaustible supply of liquid helium for their cooling systems. Cooling systems that utilize novel thermoelectric materials offer an alternative. Thermoelectric materials, being electronic solid state components, are free of the dependence on liquid cryogens, and offer ultimate long term utility as a cooling system. Our objective is to develop a thermoelectric material that is optimized for the temperature regimes encountered in space, one that can assist in cooling detectors near absolute zero and maintain structural integrity throughout the environment of its service. By measuring the thermoelectric material’s thermal conductivity, electrical conductivity, and Seebeck coefficient, across a broad temperature regime, we will be able to determine its figure of merit. The figure of merit, or ZT value, is a measure of the energy efficiency of these materials at a certain temperature and a means of comparing various thermoelectric materials at a common standard. The highest ZT values for a thermoelectric material are achieved by decoupling the electron and phonon transports in a material. By varying crystal structure and composition we strive to achieve this. Semiconductors based on the skutterudite crystal structure show promise in this regard and are prime candidates. Our ultimate goal is the successful development of a novel thermoelectric material that can meet or exceed the cooling requirements needed in space. Conversion of Historical SEASAT Data to NetCDF Alessandro Sanchez Mentors: Jessica Hausman and Ed Armstrong SEASAT, launched in 1978, was a NASA proof-of-concept mission to demonstrate the feasibility of monitoring oceanographic parameters from space. It measured ocean sea-surface wind, sea-surface temperature, wave height, ocean surface topography, and more. The data were stored in a flat binary tape format typical of 1980, which made accessing and reading the data a difficult task. In order to make these data more accessible to oceanographers and scientists, they have been reformatted into NetCDF format with CF metadata standards via MATLAB. The creation of the read software was highly dependent on existing documentation detailing the structure and format of these geophysical data records. The SEASAT data are organized primarily with respect to which instrument collected the data (ALT, SMMR, SASS) and the nature of the record types (Basic Geophysical, Basic Sensor, Supplementary Geophysical, and Supplementary sensor). Once the data were successfully converted, subsequent regression tests were performed to ensure its quality. This included a bit by bit comparison of the original binary files to those converted by MATLAB. These combined SEASAT datasets are particularly important because they contain a significant amount of information about the global oceans from the 1970’s, a time when such data were scarce or non-existent. Mapping, Characterizing, and Interpreting Mineral Fabrics in Mafic and Ultramafic Rock Samples From Mars Analog Sites in Samail, Oman, Using the Ultra-Compact Imaging Spectrometer (UCIS) Cecilia B. Sanders Mentors: R. Glenn Sellar and Bethany Ehlmann The Ultra-Compact Imaging Spectrometer (UCIS) is a hyperspectral instrument that enables an autonomous geologist, such as Mars 2020 or future missions of planetary exploration, to determine the mineral fabric and composition of rock samples. In this investigation, UCIS was used in a micro-imaging configuration in a laboratory environment to image the reflectance spectra of solid rock samples in the visible through shortwave-infrared wavelengths (0.50 to 2.50 µm.) These data were used both to evaluate the capabilities of UCIS at the micro-scale and to analyze the spectral and mineralogical diversity of rocks from Mars-analog sites. The primary site of interest for this investigation was the Samail Ophiolite in Oman, where subsurface serpentinization and subsurface/subaerial carbonate deposition mimic some of the processes undergone by a Noachian Mars. Data were processed with the IDL-based image analysis software ENVI to generate detailed parameter maps distinguishing carbonate and serpentine minerals in varying modes of aqueous alteration. Close inspection of these maps yielded new spectral parameters, including the strength of absorption bands at 1.39, 1.90, 2.12, and 2.34 µm, continuum shapes including slopes about 1.08 µm, and feature shifting, which reliably identify and map serpentines and carbonates with distinct cation contents, free waters, and interlocking textures.  

174   

  Extending Interfield Analysis of Tumuli on Terrestrial Inflated Lava Flows to Mars Simran Sangha Mentors: Serina Diniega and Suzanne Smrekar To better understand lava flow emplacement history and dynamics, we focus our study on tumuli—small-scale, positive topographic features (~10 meters in width) found both on terrestrial and Martian inflated lava flows. Inflated lava flows are aptly named for their domed, rigid upper crust that insulates and is lifted by a fluid interior. Locally high magmatic pressures arising from the transport of lava through a network of subsurface lava pathways within these flows can cause a section of the upper crust to tilt upwards and outwards on opposite sides of an axial fracture and produce a tumulus. As tumuli form over lava tubes, we believe that tumuli can be used as records of a flow’s interior structure. We aim to quantitatively investigate hypothesized relationships between tumuli morphometrics (such as tumuli sizes, shapes, and orientations) and larger-scale lava flow emplacement structure (flow directions, placement of flow boundaries, and scale of the flow). In particular, we hypothesize that: (1) Tumuli form predominantly over very low slopes (1605 tumuli within six diverse terrestrial fields generally support our hypotheses. Subsequent measurements of >1160 tumuli in seven fields within the Elysium Planitia region of Mars also support our hypotheses. Furthermore, basic tumuli dimensions (length, width, elongation, etc.) are strikingly similar between both planets—suggesting commonality in tumuli formation. Additionally, we have not yet found tumuli on Martian lava flows outside of very young flows within Elysium Planitia (