12023 research outputs found
Sort by
Between Seismic Speed and Glacial Pace: Cryoseismic Observation of Intermediate-Scale Processes at Lemon Creek Glacier, Alaska
Get PDFIn this thesis, I present three studies in environmental seismology. First, I present an analysis of seismic tremor generated from subglacial water flow during the rapid drainage of an ice-marginal supraglacial lake, collected by an on-ice nodal seismic array. I find that seismic tremor indicates a partial pressurization of the subglacial hydrologic system that was not accompanied by the expected change in glacier surface velocity, suggesting that factors like glacier geometry play a significant role in whether pressurization necessarily leads to velocity change. Using seismometers in this way allows remote observation of active subglacial hydrologic systems as they vary over space and time, a vital parameter for understanding how liquid water affects glacier motion, melting, fracture, and hazards. Second, I present observations of glacier surface crevasse development over space and time, as detected by a dense array of seismometers atop the glacier. I find that icequakes associated with surface crevassing have a magnitude distribution that is swarm-like, rather than aftershock-like, and that the spatiotemporal distribution of events indicates that crevasses regularly widen, deepen, reactivate, and trigger activity at nearby crevasses through cryoseismicity. Understanding surface crevassing activity is valuable for constraining the degree to which glacier surface velocity measurements represent ice flow as a whole, and for interpreting how glacier flow responds to changes in forcing over time. Third, I present an investigation of changes in anthropogenic urban seismic noise in Los Angeles associated with changes in community behavior. I find that changes in human activity from the scale of hours to the scale of months create distinguishable differences in ambient seismic noise power that correlate well with other measures of community behavior. Characterizing anthropogenic seismic noise is beneficial for accurately interpreting measurements of transient seismic wave data collected in urban areas toward goals such as hazard mapping.</p
The Development of the Research Journal and the Prehistory of Peer Review in the Philosophical Transactions of the Royal Society
No full text[Introduction] Modern science descends primarily from the ancient Greeks, when the body of scientific knowledge was small enough that they were able to favor the oral method of passing on knowledge. As the need for international scientific communication expanded, alternate methods of communicating science arose. The development of European public postal networks starting in the fifteenth century allowed communication by letter between scientists. Long-distance correspondence networks allowed researchers to share short-form information or solicit feedback. However, letters were generally problematic as means to disseminate the results of research. Although letters provided an effective way to transmit information, they did not allow channels for a scientist’s work to be broadly and easily reviewed by his peers. Moreover, they were expensive to send and could not reach a wide audience.</p
Stories in Single Cell RNA Sequencing
Get PDFThis thesis describes the projects I have worked on since starting the Caltech bioengineering program in fall 2017. The general theme of my projects is that they are all about single cell RNA sequencing (scRNA-seq), spanning the experimental and computational realms.
Chapter 1 is an introduction explaining the essential concepts and is meant to be readable by a wide audience. For the other chapters, each one describes a separate project in a succinct manner, including links to the related preprint, published paper or code repositories at the start of each chapter.
Chapter 2 describes the scVI generative model for scRNA-seq data and the scvi-tools framework, which forms the basis of many of my computational projects.
Chapter 3 describes an open source 3D printable syringe pump system that was developed envisioning facilitating many kinds of experiments, in particular droplet based scRNA-seq.
Chapter 4 describes a new way of fabricating hydrogel beads with unique DNA barcodes that are used for scRNA-seq experiments.
Chapter 5 describes a database listing most published scRNA-seq studies that I helped create, and provides a useful overview of the state of the field.
Chapter 6 describes the kallisto bus workflow, which is used for pre-processing scRNA-seq data, going from FASTQ file to gene count matrix in a very efficient manner.
Chapter 7 describes a new way of using scVI to quantify the trade- off in the quality of scRNA-seq of a given dataset when surveying more cells or sequencing more reads per cell.
Chapter 8 describes tools developed for the WormBase users to leverage scRNA-seq data on C. elegans, and which can be deployed with any other scRNA-seq dataset.
Chapter 9 describes a remarkably successful offshoot of the devel- opment of these tools: a simple scVI based analysis and visualization strategy for finding candidate marker genes using C. elegans scRNA-seq data, which was experimentally validated by members of the Sternberg lab.</p
Synthesis and Applications of Terpenoid Natural Products: Total Synthesis of Scabrolide A and Havellockate, and Synthesis of Pinene Oxidation Products for Atmospheric Investigations
Get PDFThe total synthesis of complex natural products remains one of the enduring challenges in organic chemistry. Whether motivated by the biological activity of the target, or its structural complexity, total synthesis continues to serve as a proving ground for synthetic methodology as well as strategic planning. Herein is described the first total synthesis of the polycyclic furanobutenolide-derived (nor)cembranoid diterpenoids scabrolide A and havellockate. Our total synthesis of scabrolide A involves an intramolecular [4+2] cycloaddition, which forges a fused [5–5–6] tricycle possessing two of the three carbocycles that characterize the natural product. Next, a photocycloaddition/fragementation sequence is employed to forge the final seven-membered ring and complete the total synthesis. Using a similar strategy, we could forge the [5–5–6] tricyclic core of the related diterpenoid havellockate with an intramolecular [4+2] cycloaddition. This is followed by a challenging enone allylation which installs the final carbon atoms of the target. Finally, elaboration of the allyl group, followed by a Cu/TEMPO-catalyzed oxidative lactonization furnishes havellockate. The final chapter of this thesis describes the synthesis and characterization of several pinene oxidation products and their dimers, which have been observed in pinene-derived secondary organic aerosol samples, as standards for atmospheric studies. This research uses the power of chemical synthesis to confirm (and in some cases reassign) the structures of naturally occurring, yet difficult to characterize chemical species found in the atmosphere.</p
Regret-Optimal Control
Get PDFOptimal controllers are usually designed to minimize cost under the assumption that the disturbance they encounter is drawn from some specific class. For example, in H₂ control the disturbance is assumed to be generated by a stochastic process and the controller is designed to minimize its expected cost, while in H∞ control the disturbance is assumed to be generated adversarially and the controller is designed to minimize its worst-case cost. This approach suffers from an obvious drawback: a controller which encounters a disturbance which falls outside of the class it was designed to handle may perform poorly. This observation naturally motivates the design of adaptive controllers which dynamically adjust their control strategy as they causally observe the disturbance instead of blindly following a prescribed strategy.
Inspired by online learning, we propose data-dependent regret as a criterion for controller design. In the regret-optimal control paradigm, causal controllers are designed to minimize regret against a hypothetical optimal noncausal controller, which selects the cost-minimizing sequence of control actions given noncausal access to the disturbance sequence. Controllers with low regret retain a performance guarantee irrespective of how the disturbance is generated; it is this universality which makes our approach an attractive alternative to traditional H₂ and H∞ control. The regret of the causal controller is bounded by some measure of the complexity of the disturbance sequence; we consider several different complexity measures, including the energy of the disturbance sequence, which measures the size of the disturbance, and the pathlength of the disturbance, which measures its variation over time. We also consider the alternative metric of competitive ratio, which is the worst-case ratio between the cost incurred by the causal controller and the cost incurred by the optimal noncausal controller. This metric can also be viewed as a special case of data-dependent regret, where the complexity measure is simply the offline optimal cost. For each of these complexity measures, we derive a corresponding control algorithm with optimal data-dependent regret. The key technique we introduce is an operator-theoretic reduction from regret-optimal control to H∞ control; each of the regret-optimal controllers we obtain can be interpreted as an H∞ controller in a synthetic system of larger dimension. We also extend regret-optimal control to the more challenging measurement-feedback setting, where the online controller must choose control actions without directly observing the disturbance sequence, using only noisy linear measurements of the state.
We show that the competitive controller can be arbitrarily well-approximated by the class of disturbance-action-controller (DAC) policies. The convexity of this class of policies makes it amenable to online optimization via a reduction to online convex optimization with memory, and this class has hence attracted much recent attention in online learning. Using our approximation result, we show how to obtain algorithms which achieve the "best-of-both-worlds": sublinear policy regret against DAC policies and approximate competitive ratio. These performance guarantees can even be extended to the "adaptive control" setting, where the controller does not know the system dynamics ahead of time and must perform online system identification.
We present numerical experiments in a linear dynamical system which demonstrate how the performance of regret-optimal controllers varies as a function of the complexity of the disturbance. We extend regret-optimal control to nonlinear dynamical systems using model-predictive control (MPC) and present experiments which suggest that regret-optimal control is a promising approach to adapting to model error in nonlinear control.</p
Structural Basis of Antibody Recognition of Viruses
Get PDFThe Zika epidemic in 2015-2016 and COVID-19 pandemic in 2019-2021 are the latest reminders of the enormous impact of viruses on the world. Zika, a flavivirus transmitted by mosquitos, can cause severe neurodevelopmental abnormalities including microcephaly in the newborns of the infected mothers. Vaccine design is complicated by concern that elicited antibodies may also recognize other epidemic-causing flaviviruses that share a similar envelope protein, such as dengue virus, West Nile Virus, and yellow fever virus. This cross-reactivity, if non-neutralizing, may worsen symptoms of a subsequent infection through antibody-dependent enhancement (ADE). To better understand the neutralizing antibody response and risk of ADE, we compared germline and mature antibody binding to Zika and other flaviviruses. We showed that affinity maturation of the light chain variable domain is important for strong binding of VH3-23/VK1-5 neutralizing antibodies to Zika virus envelope domain III (EDIII) and identified interactions that contribute to weak, cross-reactive binding to West Nile Virus EDIII. These findings informed our design of EDIII-conjugated mosaic nanoparticles as a pan-flavivirus vaccine candidate. Sera from immunization trials with nanoparticles displaying EDIIIs of Zika and dengue serotypes 1-4 showed cross-reactive binding to Zika, dengue 1-4, and West Nile Virus, a promising step towards the development of safe and effective flavivirus vaccines.
Coronaviruses are another group of viruses responsible for widespread morbidity and mortality, including the severe acute respiratory syndrome coronavirus (SARS-CoV) and Middle East Respiratory Syndrome coronavirus (MERS-CoV) epidemics and current SARS-CoV-2 pandemic. Given concerns regarding new SARS-CoV-2 variants and the possibility for additional zoonotic betacoronaviruses to cause future outbreaks, we investigated how the epitopes on the SARS-CoV-2 receptor binding domain (RBD) targeted by VH3-30-derived antibodies correlate with their neutralization potency and breadth of betacoronavirus recognition. Analyses showed how variations in antibody light chains and CDRH3 lengths facilitate the diverse RBD epitopes, cross-reactivity, and neutralization profiles of VH3-30 Abs, illustrating their importance for vaccine design and therapeutic antibody development.</p
Exploration of Materials and Mesostructures Accessible via Inorganic Phototropic Growth
Get PDFBiological systems have evolved complex methods to interact with and adapt to a given environment to optimize fitness, and replication of these natural mechanisms in artificial systems has been a long-standing area of research interest with significant potential utility. Phototropic growth is a natural phenomenon wherein an organism spatially orients biomass addition to optimize light collection. An artificial analog, inorganic phototropic growth, has been demonstrated and relies on a similar process: semiconductor mass is selectively added near regions of high light absorption, thus optimizing light collection and instructing further localized material addition. Inorganic phototropic growth effected via light-mediated electrodeposition has been used to generate anisotropic Se-Te mesostructures with optically-defined morphologies by capitalizing on inherent asymmetries in light absorption at structured, semiconducting interfaces to direct anisotropic growth. This thesis broadens the previous understanding of inorganic phototropic growth via a series of investigations that expand both the material library and complex morphologies accessible and includes detailed analyses of associated structural evolutions and the underlying optical phenomena. First, inorganic phototropic growth of highly ordered and periodic PbSe and CdSe mesostructures with optically-defined morphologies is demonstrated. Second, deposition using temporally varying illumination inputs to generate Se-Te mesostructures with tunable morphological complexity in both the in-plane and out-of-plane directions is examined. Third, the use of single, static, short wavelength (green to ultraviolet) inputs to simultaneously define two orthogonal sets of periodic features in Se-Te deposits is explored. A suite of optically-based simulations is used throughout to model the growth processes and elucidate the fundamental light-matter interactions which defined the empirically observed morphologies.</p
Bidirectional Brain-Machine Interfaces for Modulating Stimulation and Neural Plasticity
Get PDFIn prosthetics, tactile feedback can let us feel how we interact with the environment. Without this, it is extremely difficult to perform a motor task with fine control. The same idea can be applied in the brain-machine interface (BMI), which is an interface that directly connects external devices such as prosthetic limbs to the brain. Bidirectional BMI can deliver a stimulation to the brain as a sensory feedback, which can improve the performance of motor tasks. Such a bidirectional BMI can also serve a different role, if the stimulation encodes different information: if it encodes neural activity from another brain area, for example, then bidirectional BMI can provide a bypass for a damaged neural circuit. This may also affect the neural connectivity, strengthening or weakening the underlying neural connections. In this thesis, we present experiments that explore such applications of bidirectional BMI. First, we describe an experiment for characterizing neural connectivity between different brain areas. We found neural connectivity between supramarginal gyrus (SMG) and PMv (ventral premotor area), and also between anterior intraparietal (AIP) and Brodmann’s area 5 (BA5), characterized by field-field, spike-field, and partial spike-field coherence. Through partial spike-field coherence, we also revealed that the spikes in PMv may drive the activity in SMG, which is obscured in ordinary spike-field coherence. Next, we provide evidence of changes in neural connectivity caused by stimulation in S1. With spike-triggered stimulation, which delivers stimulation in S1 in response to spikes recorded in a selected channel in SMG, we could significantly increase the correlation between SMG and S1, measured by the spike time tilling coefficient (STTC) to avoid dependencies of the correlation on firing rates. Furthermore, we found that not only spike-triggered stimulations, but also random stimulations on multiple channels in S1, can vary partial spike-field coherence in theta and alpha bands within S1; such changes mostly occurred in channel pairs with zero phase difference in partial spike-field coherence. Finally, we demonstrate the possibility of volitional control on stimulation pattern in bidirectional BMI. It is shown that the participants could not only increase or decrease a single-channel firing rate, but also hold the firing rate in a given range, demonstrating a fine control over firing rate. These findings would begin to establish a framework for closed-loop modulation of neural activity with bidirectional BMI and could be used to develop new treatments for neurological damage, such as to promote plasticity in or bridge brain areas affected by stroke.</p
Three Problems in the Design and Specification of Biomolecular Circuits
Get PDFProgramming biological materials is a daunting challenge. Although part of this challenge is practical -- cloning is difficult, synthesizing DNA is expensive at scale, etc. -- a number of the challenges of bioengineering (and synthetic biology in particular) are problems of design and specification. If we could place arbitrary molecules on a surface with perfect precision, what should we place and where? If we could arbitrarily change the genetic content of a cell, even with perfect knowledge of the function and action of every component, what changes would actually enact the functions we want that cell to have? In this thesis, we explore three specific design and specification challenges at three different levels of abstraction, and demonstrate methods for overcoming them. On the level of design language, we use a specialized class of cellular automaton to probe what chemistry can do when restricted to a surface. On the level of \textbf{part specification}, we use several models of CRISPR/Cas9-based transcriptional regulators to understand what dynamic functions those regulators can perform and why, and provide some some suggestions for how to engineer such regulators to more robustly perform those functions. On the level of module design, we consider an easy-to-encounter trap in when modeling a replicating DNA species in a CRN-based biocircuit simulation, for which we suggest a simple, flexible, biologically-plausible workaround.</p
Investigation of Some Small Molecule-Protein and Protein-Protein Interactions in Nicotine Addiction, Opioid Use Disorder, and COVID-19
Get PDFNicotine addiction, opioid use disorder, and COVID-19 have made lasting impacts on every aspect of society. These are complicated conditions, and studies in these fields will likely continue for decades, if not centuries. Here, we make contributions to each of these issues using electrophysiology and microscopy. The first chapter goes into the motivation behind this thesis and the major experiments I used in my graduate career. In the second chapter, we introduce a new amino acid into the mouse muscle nicotinic acetylcholine receptor in an attempt to understand the dynamics of receptor activation. In the third chapter, we continue the Lester lab’s work on the neuroscientific effects of menthol and how it plays a role in nicotine addiction. We found the binding site for menthol on the α4β2 nicotinic acetylcholine receptor, which continues our hypothesis that the neuroscientific effects of menthol are detrimental to cigarette smokers. Fortunately, partly because of our studies, mentholated nicotine products are being phased out of the United States. The fourth and fifth chapters investigate μ-opioid receptor trafficking, both the trafficking from the endoplasmic reticulum and endocytosis from the plasma membrane. Both of these events play a role in inducing opioid use disorder and increasing the danger of using opioids. We hope that these studies will help other researchers understand opioid use disorder and fight the opioid epidemic. Finally, we studied the effects of SARS-COV-2 proteins on epithelial sodium channels. These channels are important for regulating lung fluid levels where their improper function may cause pulmonary edema. Pulmonary edema has been observed in COVID-19 patients. Altogether, we believe that we have made meaningful impacts on these important health concerns in this thesis. We look forward to how the scientific communities continue to build on our results.</p