Contemporary deep-learning models for the life-sciences have outpaced the tooling that lets experimentalists compose them. Three contributions are presented in response, a software platform, exemplary tasks built on it, and a predicted structural proteome of a defined gut microbiome. The underlying argument is that for experimentalists who use rather than build deep-learning methods, difficulties with composition and usability now outpace availability.
\r\n\r\nTRILL, a platform for AI-based protein engineering and analysis, is open-source, runs locally, and wraps models/methods behind a uniform vocabulary of thirteen top-level commands. Furthermore, TRILL is scalable, ranging from parallel fine-tuning of large models on a supercomputer to democratized, parameter off-loading in compute-limited scenarios. Models can be swapped with a one-argument change rather than a pipeline rewrite, and fast predictions can be paired with physics-based validation where overconfidence costs most.
\r\n\r\nProtein language models were fine-tuned using a homology-aware strategy, decreasing data leakage when evaluating generated proteins. Classifiers for cellulase, antimicrobial, and toxin activity were trained and applied to a scan of over two hundred million proteins from the NCBI non-redundant catalogue. An end-to-end pipeline carried seventeen predicted toxins of unknown function through structure prediction, binder design, and molecular dynamics on nearly nine hundred designed complexes.
\r\n\r\nThe third contribution targets hCom2, a defined synthetic gut consortium. We present a structural resource, where roughly four hundred thousand structures of its proteome were predicted using TRILL, segmented into eight hundred thousand domains, and assigned CATH designations. A case study demonstrating the utility of this structural database identifies nineteen carriers of the Helicobacter pylori virulence-factor TIPalpha fold across fourteen strains where sequence-only annotation fails.
", "doi": "10.7907/b5ye-jy33", "publication_date": "2026", "thesis_type": "phd", "thesis_year": "2026" }, { "id": "thesis:18716", "collection": "thesis", "collection_id": "18716", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:05312026-053755214", "primary_object_url": { "basename": "LFC_thesis_v4.pdf", "content": "final", "filesize": 27174849, "license": "other", "mime_type": "application/pdf", "url": "/18716/1/LFC_thesis_v4.pdf", "version": "v4.0.0" }, "type": "thesis", "title": "Engineering Immunological Solutions for Pandemic-Level Threats", "author": [ { "family_name": "Caldera", "given_name": "Luis Fernando", "orcid": "0009-0005-3254-6563", "clpid": "Caldera-Luis-Fernando" } ], "thesis_advisor": [ { "family_name": "Bjorkman", "given_name": "Pamela J.", "orcid": "0000-0002-2277-3990", "clpid": "Bjorkman-P-J" }, { "family_name": "Mayo", "given_name": "Stephen L.", "orcid": "0000-0002-9785-5018", "clpid": "Mayo-S-L" } ], "thesis_committee": [ { "family_name": "Rothenberg", "given_name": "Ellen V.", "orcid": "0000-0002-3901-347X", "clpid": "Rothenberg-E-V" }, { "family_name": "Bjorkman", "given_name": "Pamela J.", "orcid": "0000-0002-2277-3990", "clpid": "Bjorkman-P-J" }, { "family_name": "Mayo", "given_name": "Stephen L.", "orcid": "0000-0002-9785-5018", "clpid": "Mayo-S-L" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "Pandemics remain among the most serious threats to global public health, especially when viruses can efficiently spread and persist in human populations. Two major examples are the AIDS pandemic caused by HIV-1, which emerged in the early 1980s, and the COVID-19 pandemic caused by SARS-CoV-2, which began in late 2019. Despite their differences in timescale and transmission, HIV-1 and SARS-CoV-2 exhibit high antigenic diversity that drives an ongoing arms race between viral escape and the development of vaccines and therapeutics. This thesis presents three molecular engineering strategies to confront that problem. For SARS-CoV-2, we utilized computational tools to generate mosaic nanoparticle vaccines displaying engineered SARS-CoV-2 or selected sarbecovirus receptor-binding domains (RBDs) sharing conserved antigenic features to drive immunity toward eliciting cross-reactive responses. In na\u00efve and pre-vaccinated mice, our lead candidate, mosaic-7COM, elicited broader and more potent cross-reactive antibody responses compared to our prior mosaic-8b candidate. For HIV-1, we engineered a stabilized CD4-Ig reagent specific to HIV-1 envelope (Env) glycoprotein with greatly reduced off-target recognition of MHC class II. The final CD4 design had markedly increased thermostability (>20\u00b0C) and a nearly 50-fold improvement in mammalian expression compared to the wild-type construct, highlighting its potential as a therapeutic biologic. Additionally, we developed a yeast-display screening platform to isolate nanobody (VHH) domains against the caldera, a conserved, glycan-free epitope exposed on Env upon engagement with CD4. This campaign yielded five VHH-Fc candidates that bind receptor-bound Env and established a foundation for future biologic development targeting a previously inaccessible site on the virus.", "doi": "10.7907/wgps-rp81", "publication_date": "2026", "thesis_type": "phd", "thesis_year": "2026" }, { "id": "thesis:18528", "collection": "thesis", "collection_id": "18528", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:04302026-154155078", "primary_object_url": { "basename": "phd_thesis-45.pdf", "content": "final", "filesize": 25789805, "license": "other", "mime_type": "application/pdf", "url": "/18528/2/phd_thesis-45.pdf", "version": "v5.0.0" }, "type": "thesis", "title": "Illuminating the Regulatory Dark Matter of E. coli with Massively Parallel Reporter Assays", "author": [ { "family_name": "R\u00f6schinger", "given_name": "Tom", "orcid": "0000-0002-4900-3216", "clpid": "R\u00f6schinger-Tom" } ], "thesis_advisor": [ { "family_name": "Phillips", "given_name": "Robert B.", "orcid": "0000-0003-3082-2809", "clpid": "Phillips-R" } ], "thesis_committee": [ { "family_name": "Thomson", "given_name": "Matthew", "orcid": "0000-0003-1021-1234", "clpid": "Thomson-M-W" }, { "family_name": "Rothenberg", "given_name": "Ellen V.", "orcid": "0000-0002-3901-347X", "clpid": "Rothenberg-E-V" }, { "family_name": "Elowitz", "given_name": "Michael B.", "orcid": "0000-0002-1221-0967", "clpid": "Elowitz-M-B" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Phillips", "given_name": "Robert B.", "orcid": "0000-0003-3082-2809", "clpid": "Phillips-R" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "All cells respond to changes in their environment through the regulation of their genes. Despite decades of effort in Escherichia coli, huge gaps remain in our knowledge of both the function of many genes - the so-called y-ome - and how they are regulated. For roughly 40% of genes, no function has been assigned, and for the majority we do not know which, if any, transcription factors control their expression. Here we describe a joint experimental and theoretical dissection of the regulation of 117 promoters in E. coli across 39 diverse environments, enabling us to identify the binding sites and transcription factors that mediate regulatory control at base-pair resolution.
\r\n\r\nUsing Reg-Seq - a combination of saturation mutagenesis, massively parallel reporter assays, mass spectrometry, and tools from information theory - we go from complete ignorance of a promoter's environment-dependent regulatory architecture to detailed models of its behavior. We first develop the theoretical framework for interpreting the information footprints that are the primary readout of Reg-Seq, using toy models to establish the expected scale of mutual information at binding site positions and how the noise floor scales with sequencing depth. We then describe improvements to the genome-integrated Reg-Seq protocol, including redesigned constructs and an expanded condition panel spanning carbon source shifts, antibiotic stress, anaerobiosis, osmotic shock, and stationary phase.
\r\n\r\nAs proof of principle, we chose a combination of gold standard promoters with well-characterized regulation, genes from the y-ome, toxin-antitoxin pairs, and genes hypothesized to be part of regulatory modules. At well-characterized promoters, Reg-Seq recovers known binding sites for LexA, CpxR, MarA, Rob, MprA, and CRP under the expected conditions, while also revealing previously unreported features: new transcription start sites, a novel CRP binding site at the uncharacterized gene yadI, and condition-specific activation patterns not predicted by existing annotations. Extending the method to 34 promoters with no prior regulatory information, we discovered a host of new insights into the regulatory landscape of the y-ome. A cluster of osmotically induced promoters shares a conserved binding motif for an unidentified transcription factor, and genome-wide scanning with this motif identifies a putative regulon spanning osmoprotectant transport, trehalose metabolism, and envelope modification. At the anaerobic gene ybiY, we identify YciT as a repressor - correcting the existing database annotation - and reveal an unidentified activator that does not correspond to any characterized anaerobic regulator. At the cryptic prophage gene yagB, mass spectrometry identifies both XynR and H-NS, and the overlapping architecture of the repressor and sigma^S binding sites explains the stationary phase specificity of expression.
\r\n\r\nA systematic survey of single-mutation effects across the library reveals that many promoter regions harbor latent sequences one base change away from creating a functional sigma^70 promoter. These de novo promoters are strongly enriched at loci that require a specific activator, consistent with the signal only being detectable against a silent background. In a complementary set of results, we find that several loci with multiple annotated transcription start sites resolve to fewer active sites under physiological conditions.
\r\n\r\nTogether, these results demonstrate that Reg-Seq can systematically annotate the regulatory architecture of uncharacterized genes, correct existing annotations, and generate testable hypotheses about transcription factor identity and condition-specificity, bridging the gap between single-gene studies and the largely uncharacterized regulatory landscape of E. coli.
", "doi": "10.7907/3qy4-em46", "publication_date": "2026", "thesis_type": "phd", "thesis_year": "2026" }, { "id": "thesis:18394", "collection": "thesis", "collection_id": "18394", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:02262026-191401810", "primary_object_url": { "basename": "markarian_nicholas_thesis.pdf", "content": "final", "filesize": 55194097, "license": "other", "mime_type": "application/pdf", "url": "/18394/3/markarian_nicholas_thesis.pdf", "version": "v6.0.0" }, "type": "thesis", "title": "Mixtures of Latent Variable Models for Interpreting Gene Expression Covariation from Pathways to Transcriptomes", "author": [ { "family_name": "Markarian", "given_name": "Nicholas", "orcid": "0000-0003-1347-2392", "clpid": "Markarian-Nicholas" } ], "thesis_advisor": [ { "family_name": "Sternberg", "given_name": "Paul W.", "orcid": "0000-0002-7699-0173", "clpid": "Sternberg-P-W" } ], "thesis_committee": [ { "family_name": "Pachter", "given_name": "Lior S.", "orcid": "0000-0002-9164-6231", "clpid": "Pachter-L" }, { "family_name": "Sternberg", "given_name": "Paul W.", "orcid": "0000-0002-7699-0173", "clpid": "Sternberg-P-W" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Thomson", "given_name": "Matthew", "orcid": "0000-0003-1021-1234", "clpid": "Thomson-M-W" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "This thesis centers on interpretable subspace learning and latent variable models for characterizing covariation modulated by categorical variables in the context of biology. First, it introduces a probabilistic model with ties to Principal Component Analysis and k-means clustering, k-spaces, which has implications across different biological analyses through its interpretations as a subspace learning technique, a latent variable model, and a dimension reduction technique. Second, it establishes the problem of simultaneously characterizing gene covariation and expression level in known pathways in human tissue samples and applies k-spaces to GTEx data to lay the foundations for this line of research. Finally, it outlines a path forward to being able to use such data as references for clinical samples from patients.", "doi": "10.7907/dcbd-wy35", "publication_date": "2026", "thesis_type": "phd", "thesis_year": "2026" }, { "id": "thesis:17771", "collection": "thesis", "collection_id": "17771", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:11242025-191228061", "type": "thesis", "title": "Naturally-Inspired Circuits for Microbial Composition Control and Biosensing", "author": [ { "family_name": "Kratz", "given_name": "Matthieu Francois", "clpid": "Kratz-Matthieu-Francois" } ], "thesis_advisor": [ { "family_name": "Murray", "given_name": "Richard M.", "orcid": "0000-0002-5785-7481", "clpid": "Murray-R-M" }, { "family_name": "Elowitz", "given_name": "Michael B.", "orcid": "0000-0002-1221-0967", "clpid": "Elowitz-M-B" } ], "thesis_committee": [ { "family_name": "Demirer", "given_name": "G\u00f6zde S.", "orcid": "0000-0002-3007-1489", "clpid": "Demirer-G\u00f6zde-S" }, { "family_name": "Hay", "given_name": "Bruce A.", "orcid": "0000-0002-5486-0482", "clpid": "Hay-B-A" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Murray", "given_name": "Richard M.", "orcid": "0000-0002-5785-7481", "clpid": "Murray-R-M" }, { "family_name": "Elowitz", "given_name": "Michael B.", "orcid": "0000-0002-1221-0967", "clpid": "Elowitz-M-B" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "When considering the design of gene circuits, there are many possible sources of inspiration. Many early synthetic gene circuits used nature as an inspiration, seeking to recreate biological behaviors with non-native components. As the field grew, alternative approaches sourcing designs from adjacent engineering fields and computational approaches emerged and grew in prominence. Despite this shift, there remains a great deal of naturally-inspired circuits that provide useful functions for biotechnology. Indeed nature has often been uniquely capable of exploiting typically undesirable phenomena, e.g. noise to create biologically useful function. This thesis presents two projects directly inspired by natural systems. Each project aims to replicate a behavior or circuit topology found in nature, leveraging its unique dynamics to address key challenges in biotechnology. Chapters 2 and 3 will cover the development of a circuit emulating the microbial behavior of phase variation, whereby individual cells reversibly and stochastically transition between distinct phenotypes. We recreate this behavior using serine recombinases and demonstrate how it can enable stable, bulk control of phenotype composition\u2014a task of great relevance to biotechnology. Chapter 4 lays the groundwork for applying the biologically-relevant feed-forward loop topology to the problem of spurious biosensor activation. We realize this topology in a modular manner using small transcription activating RNAs (STARs) and provide a preliminary characterization of its dynamical properties. Finally, we discuss alternative implementations that may provide more directly applicable properties than the current STAR implementation", "doi": "10.7907/d42b-jh46", "publication_date": "2026", "thesis_type": "phd", "thesis_year": "2026" }, { "id": "thesis:17389", "collection": "thesis", "collection_id": "17389", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:06032025-002120461", "primary_object_url": { "basename": "Thesis.pdf", "content": "final", "filesize": 25197916, "license": "other", "mime_type": "application/pdf", "url": "/17389/1/Thesis.pdf", "version": "v5.0.0" }, "type": "thesis", "title": "A Biophysical Approach to Normalization and Trajectory Inference in Single-Cell RNA Sequencing Data Analysis", "author": [ { "family_name": "Fang", "given_name": "Meichen", "orcid": "0000-0002-8217-0710", "clpid": "Fang-Meichen" } ], "thesis_advisor": [ { "family_name": "Pachter", "given_name": "Lior S.", "orcid": "0000-0002-9164-6231", "clpid": "Pachter-L" } ], "thesis_committee": [ { "family_name": "Thomson", "given_name": "Matthew", "orcid": "0000-0003-1021-1234", "clpid": "Thomson-M-W" }, { "family_name": "Pachter", "given_name": "Lior S.", "orcid": "0000-0002-9164-6231", "clpid": "Pachter-L" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Chong", "given_name": "Shasha", "orcid": "0000-0002-5372-311X", "clpid": "Chong-Shasha" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "Single-cell genomics assays, particularly single-cell RNA sequencing that enables genome-wide profiling of gene expression, have been driven forward by a combination of technological and computational advances. While producing extraordinary large amounts of data for biological discovery, methods for mining results currently rely heavily on heuristics and lack of modeling has resulted in limited mechanistic biological insight. This thesis presents two models for normalization and trajectory inference in single-cell RNA sequencing analysis to demonstrate how biophysical modeling, when combined with principled statistical inference, can yield interpretable insights grounded in rigorous theoretical frameworks.
\r\n\r\nWe begin by explaining the two cultures in single-cell RNA sequencing analysis. Next, we present the chemical master equation, which forms the theoretical foundation for biophysically informed stochastic models of gene expression, and explore an existing gap in developing uniform approximations over time under the large-volume limit. Returning to single-cell RNA sequencing data analysis, we introduce two mechanistic models for normalization and trajectory inference, which are essential components of single-cell RNA sequencing analysis.
", "doi": "10.7907/asek-t904", "publication_date": "2025", "thesis_type": "phd", "thesis_year": "2025" }, { "id": "thesis:16459", "collection": "thesis", "collection_id": "16459", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:06012024-054725051", "primary_object_url": { "basename": "240531_PB_thesis_final.pdf", "content": "final", "filesize": 44817586, "license": "other", "mime_type": "application/pdf", "url": "/16459/1/240531_PB_thesis_final.pdf", "version": "v4.0.0" }, "type": "thesis", "title": "Modeling and Design of Synthetic Biochemical Circuits for Biological Phenotypes", "author": [ { "family_name": "Bhamidipati", "given_name": "Pranav Subramanyam", "orcid": "0000-0002-6199-6505", "clpid": "Bhamidipati-Pranav-Subramanyam" } ], "thesis_advisor": [ { "family_name": "Thomson", "given_name": "Matthew", "orcid": "0000-0003-1021-1234", "clpid": "Thomson-M-W" } ], "thesis_committee": [ { "family_name": "Elowitz", "given_name": "Michael B.", "orcid": "0000-0002-1221-0967", "clpid": "Elowitz-M-B" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Barr", "given_name": "Alan H.", "clpid": "Barr-A-H" }, { "family_name": "Thomson", "given_name": "Matthew", "orcid": "0000-0003-1021-1234", "clpid": "Thomson-M-W" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "Biological behaviors arise from the dynamical interactions of biochemical networks. For example, the various immune responses to damage are manifestations of signaling networks between immune cell types. A central goal in systems and synthetic biology is to elucidate the design principles of these networks, or circuits, both in the sense of dissecting how function arises from structure in the natural context and in the sense of understanding the guidelines for optimal engineering of synthetic biological systems. The study of design principles in both senses is aided by mathematical modeling and simulation, which provide a self-consistent framework for evaluating the theoretical implications of biological hypotheses as well as a testbed for the development of novel circuits for desired biological phenotypes. This thesis pertains to two related challenges in this field, namely the scaling of computational design to larger circuits and the engineering of global phenotypes that emerge nonlinearly from local interactions.
\r\n \r\nThe first section of this thesis presents a novel design platform for biological circuits, called CircuiTree, that uses a game-playing paradigm to overcome the combinatorial complexity of \\textit{de novo} circuit design. This platform treats circuit design as a game of circuit assembly and traverses the tree of possible assemblies using Monte Carlo tree search (MCTS). Borrowed from artificial intelligence (AI) agents that have mastered complex games, MCTS is a reinforcement learning (RL)-based search algorithm that efficiently searches for the most effective design strategies and naturally discovers design principles in the form of network motifs, which appear as clusters of solutions in the search tree. Finally, when tasked with designing fault-tolerant oscillators with five components, CircuiTree finds a novel design strategy, which we call motif multiplexing, in which multiple sub-oscillators are interleaved so as to render the circuit highly resistant to deletions and knockdowns. This design principle, which may be responsible for the multiple oscillatory loops observed in eukaryotic circadian clocks, opens the possibility of engineering synthetic circuits at a larger scale and suggests that larger biological circuits contain yet-unknown design features that are not simply extensions of smaller circuits.
\r\n\r\nThe second section describes a novel mechanosensitive property of the SynNotch synthetic chimeric receptor and uses a multicellular modeling framework to show how it can be used to control spatiotemporal patterning \\textit{in vitro}. Modified from the endogenous juxtacrine receptor Notch, SynNotch binds to an arbitrary extracellular ligand and, in response, releases an arbitrary transcription factor, thus acting as a user-defined signal transducer. We show that, in mouse fibroblasts, a simple sender-receiver SynNotch circuit ceases to transduce a membrane-bound GFP signal at high cell densities in 2D culture. Because of this feature, a lawn of cells expressing a signal-relay circuit, which we call the transceiver circuit, can undergo spatially limited activation, where the signal propagates in a wave outward from a GFP-expressing sender cell until, due to cell division, the cell density crosses a threshold value and the signaling system shuts down. Using a multicellular lattice-based model combined with experiments, we demonstrate that perturbations of growth parameters can be used to control the size of activated spots. Finally, we achieve spatiotemporal patterns of activation by seeding the growth dish nonuniformly, creating a wave of activation at the millimeter scale that recapitulates the kinematic wave patterning phenomenon observed during vertebrate somitogenesis.
\r\n\r\nTogether, this body of work represents an advance in the use of computational methods and mathematical modeling to guide the design and control of complex biological phenotypes. Advances in these methods promise to catalyze the development of more advanced cell-based therapies and engineered tissues.
", "doi": "10.7907/gpc6-hb40", "publication_date": "2024-06-14", "thesis_type": "phd", "thesis_year": "2024" }, { "id": "thesis:16185", "collection": "thesis", "collection_id": "16185", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:09212023-182422125", "type": "thesis", "title": "Acquiring Enzyme Sequence-Fitness Data at Scale Toward Predictive Methods for Enzyme Engineering", "author": [ { "family_name": "Johnston", "given_name": "Kadina Elizabeth", "orcid": "0000-0002-2214-3534", "clpid": "Johnston-Kadina-Elizabeth" } ], "thesis_advisor": [ { "family_name": "Arnold", "given_name": "Frances Hamilton", "orcid": "0000-0002-4027-364X", "clpid": "Arnold-F-H" } ], "thesis_committee": [ { "family_name": "Yue", "given_name": "Yisong", "orcid": "0000-0001-9127-1989", "clpid": "Yue-Yisong" }, { "family_name": "Mayo", "given_name": "Stephen L.", "orcid": "0000-0002-9785-5018", "clpid": "Mayo-S-L" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Arnold", "given_name": "Frances Hamilton", "orcid": "0000-0002-4027-364X", "clpid": "Arnold-F-H" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "The emergence of machine learning methods for expediting directed evolution via protein fitness prediction has recently shed light on the need for more, high quality sequence-fitness data from which to learn the mapping from sequence to fitness. Enzymes specifically are highly selective catalysts and engineered enzymes are becoming increasingly important for human applications such as pharmaceutical synthesis. This thesis thus focuses on the collection of enzymatic sequence-fitness data to enable both development and validation of emerging approaches. Chapter 1 describes the process of traditional directed evolution as well as ways that machine learning methods have been used to accelerate it. It also discusses the experimental considerations for applying machine learning to the various steps of protein engineering campaigns, as the experimental constraints are not always obvious to the machine learning community. One of the major constraints for the application of machine learning methods is the requirement to sequence all variants required for model training, a step that is often skipped by traditional, lab-only directed evolution due to it not being worth the time and cost. Chapter 2 introduces a solution to this problem with \u201cevery variant sequencing\u201d (evSeq), which enables higher throughput collection of sequencing data for a similar time and cost as commonly used Sanger sequencing methods. This method not only enables implementation of ML methods such as machine learning-assisted directed evolution (MLDE) and focused training MLDE (ftMLDE) by sequencing variants during an evolution campaign, but also offers promise to fill existing protein sequence-fitness databases with protein engineering datasets. This type of data collection can enable the development of newer, more accurate ML methods, and was an inspiration for the work presented in Chapter 3, which details the collection of a combinatorially complete, epistatic sequence-fitness landscape in an enzyme active site. Oftentimes, the effects of mutations on protein fitness can be considered largely independent and laboratory recombination of them can find an optimal variant. This general principle breaks down when the effects of mutations are not independent, termed epistasis, and sequence-fitness landscapes with these interactions are difficult to traverse. Thus, collection of this dataset provides a challenging task for the development of both ML and physics-based models and pushes the boundary of predictive methods for protein engineering.", "doi": "10.7907/xjz4-k217", "publication_date": "2024", "thesis_type": "phd", "thesis_year": "2024" }, { "id": "thesis:16355", "collection": "thesis", "collection_id": "16355", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:04152024-205944516", "primary_object_url": { "basename": "JAC_thesis_FINAL.pdf", "content": "final", "filesize": 16699403, "license": "other", "mime_type": "application/pdf", "url": "/16355/1/JAC_thesis_FINAL.pdf", "version": "v4.0.0" }, "type": "thesis", "title": "The Bioenergetics of a Low-Power, Phenazine-Dependent Maintenance Metabolism in Pseudomonas aeruginosa", "author": [ { "family_name": "Ciemniecki", "given_name": "John Alan", "orcid": "0000-0003-2789-6700", "clpid": "Ciemniecki-John-Alanlan" } ], "thesis_advisor": [ { "family_name": "Newman", "given_name": "Dianne K.", "orcid": "0000-0003-1647-1918", "clpid": "Newman-D-K" } ], "thesis_committee": [ { "family_name": "Phillips", "given_name": "Robert B.", "orcid": "0000-0003-3082-2809", "clpid": "Phillips-R" }, { "family_name": "Orphan", "given_name": "Victoria J.", "orcid": "0000-0002-5374-6178", "clpid": "Orphan-V-J" }, { "family_name": "Ruby", "given_name": "Edward G.", "orcid": "0000-0002-4112-4830", "clpid": "Ruby-Edward" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Newman", "given_name": "Dianne K.", "orcid": "0000-0003-1647-1918", "clpid": "Newman-D-K" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "A common feature of all life is the metabolic transformation of energy from the environment to biochemical energy in the organism. While this process is well-characterized in molecular detail for fast-growing or otherwise fast-metabolizing organisms such as humans, many microorganisms subsist in the environment around us with little to no exogenous energy for extended periods, and we have only vague ideas how. Questions about the metabolic mechanisms and rates underpinning these astounding survival capabilities speak to the fundamental question of the lower energetic limits of life. Motivated by this big-picture question in biology, this thesis represents one line of physiological inquiry into a specific anaerobic survival metabolism of Pseudomonas aeruginosa, an opportunistic bacterial pathogen. Pseudomonas is perhaps best known for its characteristic production of colorful, redox-active, secondary metabolites called phenazines that allow a metabolic process called extracellular electron transfer. Phenazine extracellular electron transfer has been previously shown to unlock a slow, anaerobic glucose catabolism that facilitates the survival of energy-limited populations of cells. My thesis work has elucidated the predominant membrane-bound protein complexes involved in phenazine reduction and the predominant subcellular location of reduction for each of the main phenazines produced by Pseudomonas. I show that the survival metabolism powered by these phenazines places them in a true maintenance state where there is no detectable growth in the population at the single-cell level. The metabolic rate of this maintenance was measured and found to be 1,000 times slower than when the cells are growing in aerobic culture, 100 times slower than estimates of maintenance rates made in continuous culture, and 10 times slower than the mean basal metabolic rate estimated across all life on the planet. These results open the door to investigations of metabolic attenuation, a physiological state that underpins microbial survival in nature and disease. In pursuit of these discoveries, various new experimental assays that allow further investigation into the bioenergetics and biochemistry of phenazine metabolism were developed. Finally, intellectual frameworks are presented that, in conjunction with the discoveries made and methods developed, collectively bring us steps closer to understanding the bioenergetic basis of microbial resiliency.", "doi": "10.7907/n992-ey51", "publication_date": "2024", "thesis_type": "phd", "thesis_year": "2024" }, { "id": "thesis:16503", "collection": "thesis", "collection_id": "16503", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:06042024-163846436", "type": "thesis", "title": "Implementing and Modeling Gene Drives for Population Modification and Suppression", "author": [ { "family_name": "Ivy", "given_name": "Tobin William", "orcid": "0000-0002-9116-3854", "clpid": "Ivy-Tobin-William" } ], "thesis_advisor": [ { "family_name": "Hay", "given_name": "Bruce A.", "orcid": "0000-0002-5486-0482", "clpid": "Hay-B-A" } ], "thesis_committee": [ { "family_name": "Sternberg", "given_name": "Paul W.", "orcid": "0000-0002-7699-0173", "clpid": "Sternberg-P-W" }, { "family_name": "Elowitz", "given_name": "Michael B.", "orcid": "0000-0002-1221-0967", "clpid": "Elowitz-M-B" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Hay", "given_name": "Bruce A.", "orcid": "0000-0002-5486-0482", "clpid": "Hay-B-A" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "Gene drive as a technology has immense potential for modifying species from the local to the global population level. While the advent of the CRISPR-Cas9 system has paved the way for many previously untenable gene drives, it has also illuminated two substantial pitfalls: homing based gene drives are particularly susceptible to generating drive breaking resistance alleles and many if not most gene drives are too powerful to be regionally contained. Lack of confinability makes such gene drives impractical for real world application where international law would be violated by their usage. We developed a new general form of gene drive known as cleave and rescue (ClvR), then built and simulated the potential of multiple variants of this drive which are capable of modifying or suppressing target populations, including variants with and without introduction thresholds for drive. We also developed scripts in Python capable of generating population dynamics simulations of a user-defined gene drive, either as a deterministic, population proportion model or a stochastic, discrete individual model. This tool provides a very useful first pass answer about a given gene drive\u2019s ability to modify or suppress a population under varying fitness costs, drive activity rates, and release proportions.", "doi": "10.7907/10pa-x574", "publication_date": "2024", "thesis_type": "phd", "thesis_year": "2024" }, { "id": "thesis:15184", "collection": "thesis", "collection_id": "15184", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:05172023-011031610", "primary_object_url": { "basename": "20230516-David_Goertsen_PhD_Dissertation.pdf", "content": "final", "filesize": 8086677, "license": "other", "mime_type": "application/pdf", "url": "/15184/1/20230516-David_Goertsen_PhD_Dissertation.pdf", "version": "v5.0.0" }, "type": "thesis", "title": "Expanding Adeno-Associated Viral Capsid Engineering to Multiple Variable Regions for Diversified Tropism", "author": [ { "family_name": "Goertsen", "given_name": "David Gerald", "orcid": "0000-0001-7138-1697", "clpid": "Goertsen-David-Gerald" } ], "thesis_advisor": [ { "family_name": "Gradinaru", "given_name": "Viviana", "orcid": "0000-0001-5868-348X", "clpid": "Gradinaru-V" } ], "thesis_committee": [ { "family_name": "Van Valen", "given_name": "David A.", "orcid": "0000-0001-7534-7621", "clpid": "Van-Valen-D" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Wang", "given_name": "Kaihang", "orcid": "0000-0001-7657-8755", "clpid": "Wang-Kaihang" }, { "family_name": "Gradinaru", "given_name": "Viviana", "orcid": "0000-0001-5868-348X", "clpid": "Gradinaru-V" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "Adeno-associated virus research is critical for the advancement of gene therapy and treatment of myriad debilitating genetic disorders. Targeted delivery of genetic components to a tissue or cell population remains a bottleneck for gene therapy, but the selection of AAV capsids through directed evolution can yield vectors that target desired tissues or cells. This thesis details the engineering of the AAV capsid to acquire desired tropism, namely reduction in liver transduction or increased transduction of the lung. Chapter I chronicles the history of AAV engineering, provides useful information about the AAV capsid proteins, and describes how AAV has been engineered for altered tropism in works preceding this thesis. Chapter II describes the development of AAV9.452sub.LUNG1, an AAV variant that is enriched in the lung of mice after systemic injection. Chapter III details the engineering of variants with attenuated tropism in the liver while maintaining previously acquired brain transduction after systemic injection. Two of these variants, AAV.CAP-B10 and AAV.CAP-B22, display similar tropism in the marmoset after systemic injection. Chapter IV describes the parallel engineering of prominent variable regions of the AAV capsid. Overall, the work presented in this thesis expands the toolbox available for gene therapy and represents an advancement of methods for AAV capsid engineering.
", "doi": "10.7907/7x3a-g504", "publication_date": "2023", "thesis_type": "phd", "thesis_year": "2023" }, { "id": "thesis:15231", "collection": "thesis", "collection_id": "15231", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:05302023-162053583", "primary_object_url": { "basename": "McGehee_James_2023_thesis.pdf", "content": "final", "filesize": 34895132, "license": "other", "mime_type": "application/pdf", "url": "/15231/1/McGehee_James_2023_thesis.pdf", "version": "v4.0.0" }, "type": "thesis", "title": "Optogenetic Approaches for Determining the Temporal Role of Morphogen Inputs on Target Gene Expression", "author": [ { "family_name": "McGehee", "given_name": "James Mitchell", "orcid": "0000-0002-9353-1235", "clpid": "McGehee-James-Mitchell" } ], "thesis_advisor": [ { "family_name": "Stathopoulos", "given_name": "Angelike", "orcid": "0000-0001-6597-2036", "clpid": "Stathopoulos-A" } ], "thesis_committee": [ { "family_name": "Bronner", "given_name": "Marianne E.", "orcid": "0000-0003-4274-1862", "clpid": "Bronner-M-E" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Goentoro", "given_name": "Lea A.", "orcid": "0000-0002-3904-0195", "clpid": "Goentoro-L-A" }, { "family_name": "Stathopoulos", "given_name": "Angelike", "orcid": "0000-0001-6597-2036", "clpid": "Stathopoulos-A" }, { "family_name": "Zinn", "given_name": "Kai George", "orcid": "0000-0002-6706-5605", "clpid": "Zinn-K-G" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "The Dorsal transcription factor and morphogen is important for patterning the Dorsal- Ventral axis of Drosophila melanogaster and while it has been extensively studied, the temporal dynamics of Dorsal are not well understood. There are many processes that contribute to Dorsal nuclear concentration levels, including Toll signaling and Cactus degradation, interactions with other proteins, shuttling of Dorsal to the ventral side, DNA binding, and nuclear spacing. Dorsal nuclear levels are known to activate or repress target gene expression in a concentration or threshold dependent manner. To test how Dorsal dynamics and changes to the Dorsal gradient over time affect target gene expression, we added two optogenetic tags to Dorsal at the endogenous locus to control Dorsal nuclear levels: Blue Light Inducible Degradation (BLID) and Light Inducible Nuclear Export System (LEXY). We found that upon degradation of Dorsal using blue light and BLID that a downstream ratchet was able to maintain the expression of high threshold target genes. Using blue light and LEXY to export Dorsal, we identified an important window where Dorsal activity is required to allow activation of high threshold target genes at later stages. In comparing BLID and LEXY in conjunction with mutations to a nuclear export sequence, we also identified how rapid nuclear import and export of Dorsal is sufficient for low threshold target gene expression but actively disrupts high threshold target gene expression. We conclude that not only are final concentration levels, but also the dynamics leading to those levels are important for proper gene expression.", "doi": "10.7907/c610-za20", "publication_date": "2023", "thesis_type": "phd", "thesis_year": "2023" }, { "id": "thesis:15090", "collection": "thesis", "collection_id": "15090", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:01202023-073514292", "primary_object_url": { "basename": "LTsypin_Thesis.pdf", "content": "final", "filesize": 12781413, "license": "other", "mime_type": "application/pdf", "url": "/15090/1/LTsypin_Thesis.pdf", "version": "v6.0.0" }, "type": "thesis", "title": "The Discovery and Biological Mechanisms of a Widespread Phenazine's Oxidation", "author": [ { "family_name": "Tsypin", "given_name": "Lev Maximovich", "orcid": "0000-0002-0642-8468", "clpid": "Tsypin-Lev-Maximovich" } ], "thesis_advisor": [ { "family_name": "Newman", "given_name": "Dianne K.", "orcid": "0000-0003-1647-1918", "clpid": "Newman-D-K" } ], "thesis_committee": [ { "family_name": "Parker", "given_name": "Joseph", "orcid": "0000-0001-9598-2454", "clpid": "Parker-J" }, { "family_name": "Orphan", "given_name": "Victoria J.", "orcid": "0000-0002-5374-6178", "clpid": "Orphan-V-J" }, { "family_name": "Leadbetter", "given_name": "Jared R.", "orcid": "0000-0002-7033-0844", "clpid": "Leadbetter-J-R" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Newman", "given_name": "Dianne K.", "orcid": "0000-0003-1647-1918", "clpid": "Newman-D-K" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "During the 2017 Microbial Diversity course at the Marine Biological Laboratory in Woods Hole, MA, Scott Saunders and Yinon Bar-On started enrichment cultures in hopes of dis-covering biological oxidation of phenazine-1-carboxylic acid (PCA). I took these enrich-ment cultures and described their PCA oxidation activity. From one of the mixed cultures, I isolated a bacterial strain that recapitulated the behavior of the enrichment. I identified it as a strain of Citrobacter portucalensis via a whole-genome analysis and called the strain \"MBL\" in reference to the Marine Biological Laboratory. Using a combination of analytical chemistry, quantitative fluorescence measurements, and genetic engineering, I showed that C. portucalensis MBL couples PCA oxidation to each mode of anaerobic respiration it employs with nitrate, fumarate, dimethyl sulfoxide (DMSO), and trimethylamine-N-oxide (TMAO) as terminal electron acceptors (TEAs). I further found that most of the PCA oxidation activi-ty depends on electron flux through the quinone/quinol pool but can be driven by certain terminal reductase complexes when no quinones are available, particularly in the case of ni-trate reductases. Every bacterial strain I tested catalyzed PCA oxidation when provided the appropriate TEA. My described mechanism for bacterial PCA oxidation is generalizable and implies that this previously undocumented phenomenon should occur wherever PCA is produced in rhizosphere environments.
", "doi": "10.7907/rmsf-e465", "publication_date": "2023", "thesis_type": "phd", "thesis_year": "2023" }, { "id": "thesis:14552", "collection": "thesis", "collection_id": "14552", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:04182022-222416729", "primary_object_url": { "basename": "Andre\u0301s'_PhD_Thesis.pdf", "content": "final", "filesize": 4762405, "license": "other", "mime_type": "application/pdf", "url": "/14552/2/Andre\u0301s'_PhD_Thesis.pdf", "version": "v5.0.0" }, "type": "thesis", "title": "Combinatorics and Stochasticity for Chemical Reaction Networks", "author": [ { "family_name": "Ortiz-Mu\u00f1oz", "given_name": "Andr\u00e9s", "orcid": "0000-0003-1824-3230", "clpid": "Ortiz-Mu\u00f1oz-Andr\u00e9s" } ], "thesis_advisor": [ { "family_name": "Winfree", "given_name": "Erik", "orcid": "0000-0002-5899-7523", "clpid": "Winfree-E" }, { "family_name": "Fontana", "given_name": "Walter", "orcid": "0000-0003-4062-9957", "clpid": "Fontana-Walter" } ], "thesis_committee": [ { "family_name": "Pierce", "given_name": "Niles A.", "orcid": "0000-0003-2367-4406", "clpid": "Pierce-N-A" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Phillips", "given_name": "Robert B.", "orcid": "0000-0003-3082-2809", "clpid": "Phillips-R" }, { "family_name": "Winfree", "given_name": "Erik", "orcid": "0000-0002-5899-7523", "clpid": "Winfree-E" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "Stochastic chemical reaction networks (SCRNs) are a mathematical model which serves as a first approximation to ensembles of interacting molecules. SCRNs approximate such mixtures as always being well-mixed and consisting of a finite number of molecules, and describe their probabilistic evolution according to the law of mass-action. In this thesis, we attempt to develop a mathematical formalism based on formal power series for defining and analyzing SCRNs that was inspired by two different questions. The first question relates to the equilibrium states of systems of polymerization. Formal power series methods in this case allow us to tame the combinatorial complexity of polymer configurations as well as the infinite state space of possible mixture states. Chapter 1 presents an application of these methods to a model of polymerizing scaffolds. The second question relates to the expressive power of SCRNs as generators of stochasticity. In Chapter 2, we show that SCRNs are universal approximators of discrete distributions, even when only allowing for systems with detailed-balance. We further show that SCRNs can exactly simulate Boltzmann machines. In Chapter 3, we develop a formalism for defining the semantics of SCRNs in terms of formal power series which grew as a result of work included in the previous chapters. We use that formulation to derive expressions for the dynamics and stationary states of SCRNs. Finally, we focus on systems that satisfy complex balance and conservation of mass and derive a general expressions for their factorial moments using generating function methods.", "doi": "10.7907/s9mc-3d59", "publication_date": "2022", "thesis_type": "phd", "thesis_year": "2022" }, { "id": "thesis:14435", "collection": "thesis", "collection_id": "14435", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:11282021-042001335", "primary_object_url": { "basename": "Rachel_Caltech_PhD_Thesis_V2-6.pdf", "content": "final", "filesize": 93199721, "license": "other", "mime_type": "application/pdf", "url": "/14435/11/Rachel_Caltech_PhD_Thesis_V2-6.pdf", "version": "v6.0.0" }, "type": "thesis", "title": "Experimental and Theoretical Studies of Non-Equilibrium Systems: Motor-Microtubule Assemblies and the Human-Earth System", "author": [ { "family_name": "Banks", "given_name": "Rachel A.", "orcid": "0000-0003-2028-2925", "clpid": "Banks-Rachel-A" } ], "thesis_advisor": [ { "family_name": "Phillips", "given_name": "Robert B.", "orcid": "0000-0003-3082-2809", "clpid": "Phillips-R" } ], "thesis_committee": [ { "family_name": "Van Valen", "given_name": "David A.", "orcid": "0000-0001-7534-7621", "clpid": "Van-Valen-D" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Thomson", "given_name": "Matthew", "orcid": "0000-0003-1021-1234", "clpid": "Thomson-M-W" }, { "family_name": "Phillips", "given_name": "Robert B.", "orcid": "0000-0003-3082-2809", "clpid": "Phillips-R" } ], "local_group": [ { "literal": "div_chem" } ], "abstract": "Systems out of equilibrium are pervasive around us. In fact, being out of equilibrium is a key property of life, as described by Erwin Schrodinger in his series of essays \"What is life?\". Through the consumption of energy, i.e. food, living organisms achieve ordered states that would be very unlikely to occur at equilibrium, such as the mitotic spindle during cell division, swarms of bacteria, or flocks of starlings. The Earth system is another example of a non-equilibrium system. The state of the Earth has been evolving for billions of years, often under the influence of life. Today, humanity is a dominant influence forcing the Earth system to new states. Understanding these non-equilibrium systems has posed many challenges; in this thesis, we work towards quantitatively dissecting and gaining an intuition for the functioning of both a molecular scale and planetary scale non-equilibrium system.
\r\n\r\nUnderlying many cellular functions such as cell division and transportation of organelles is the cytoskeleton composed of motor proteins and their constituent filaments. One of the key components are kinesin motors, which consume chemical energy to walk along and reorganize microtubules. Collections of these motors and microtubules are able to form organized structures. Understanding how these structures are formed has remained an open question. In Chapter 2, we develop a system of kinesin motors and microtubules wherein motor activity is controlled by light, thereby gaining spatiotemporal control over the formation of motor-microtubule assemblies. We demonstrate the creation of a variety of structures of different sizes and geometry, and measure how length and time scales of these assemblies depend on the activated region.
\r\n\r\nA remaining question was how the microscopic details of the interaction between motors and microtubule affect the dynamics and steady-state structure formed. With our scheme for light-control in hand, we extended the system to a variety of motor proteins that have different speeds, processivities (how many steps they take before unbinding from the microtubule), directionalities (which end of the microtubule they walk towards), and forces they are able to exert in Chapter 3. We found that the size of steady-state structures, distribution of motors within assemblies, and rate of contraction of networks depend on motor properties. Further, we demonstrate that various structures can be formed by combining different motors. This work begins to build a connection between the detailed microscopic interactions of cytoskeletal components to the larger scale structures they form.
\r\n\r\nChapter 4 begins our work on understanding the state of the human-Earth system. A major hurdle to quantitatively understanding this system is the difficulty of finding and parsing the relevant data, which is often within long, complicated reports. In order to facilitate access to this data, we created the Human Impacts Database, which houses a collection of $>$ 300 carefully curated values related to human impacts on the Earth, introduced in Chapter 4. In this chapter, we describe the format of the database as well as demonstrate how it can be harnessed to gain a more holistic perspective on humanity's influence on the Earth.
\r\n\r\nHaving this data is only a starting point towards deciphering the ways that humans are altering the state of the Earth, though. In Chapter 5, we combine these quantitative measurements with simple order-of-magnitude estimates to gain an intuition for the magnitude of several of the values. In this way, we show that many of the ways humanity is affecting the Earth can be tied back to how much land, water, and power we use. We further contextualize the magnitude of human influence by comparing human activities to natural analogs, finding that humans currently rival natural processes in influencing the state of the Earth system.
", "doi": "10.7907/5ee6-j454", "publication_date": "2022-06-10", "thesis_type": "phd", "thesis_year": "2022" }, { "id": "thesis:14467", "collection": "thesis", "collection_id": "14467", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:01042022-184525578", "type": "thesis", "title": "Development and Scaling of Differentiation Circuit Architectures for Improving the Evolutionary Stability of Burdensome Functions in E. coli", "author": [ { "family_name": "Williams", "given_name": "Rory Logan", "orcid": "0000-0003-2605-5790", "clpid": "Williams-Rory-Logan" } ], "thesis_advisor": [ { "family_name": "Murray", "given_name": "Richard M.", "orcid": "0000-0002-5785-7481", "clpid": "Murray-R-M" } ], "thesis_committee": [ { "family_name": "Leadbetter", "given_name": "Jared R.", "orcid": "0000-0002-7033-0844", "clpid": "Leadbetter-J-R" }, { "family_name": "Goentoro", "given_name": "Lea A.", "orcid": "0000-0002-3904-0195", "clpid": "Goentoro-L-A" }, { "family_name": "Ismagilov", "given_name": "Rustem F.", "orcid": "0000-0002-3680-4399", "clpid": "Ismagilov-R-F" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Murray", "given_name": "Richard M.", "orcid": "0000-0002-5785-7481", "clpid": "Murray-R-M" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "With advances in the sequencing and synthesis of DNA, automation, and computation, we are increasingly able to rapidly and reliably program functions into cells. However, because the functions we engineer cells to perform are often both unnecessary for the cell\u2019s survival and burdensome to cell growth, mutation and natural selection can rapidly lead to loss of function. Though numerous strategies have made headway, improving the evolutionary stability of engineered functions remains a goal of the synthetic biology community. To address this problem generally, we developed a strategy relying on integrase-mediated recombination which allows non-producing progenitor cells to differentiate at a tunable rate, thereby continuously replenishing producer cells expressing the orthogonal T7 RNAP. While this strategy removes selective pressure for mutations inactivating the function of interest in the progenitor cell population, a strategy of terminal differentiation, in which the capacity of differentiated cells to grow is limited, was necessary to prevent the expansion of such mutations in the differentiated cell population. To experimentally implement terminal differentiation, we co-opted the R6K plasmid system, using differentiation to simultaneously activate expression of T7 RNAP, and inactivate expression of \u03c0 protein (an essential factor for R6K plasmid replication), thereby allowing limitation of differentiated cell growth through antibiotic selection. Critically, we demonstrated computationally that terminal differentiation endows the circuit with robustness to mutations which disrupt T7 RNAP driven expression, and to plasmid instability effects that result in decreased expression. Intuitively and computationally identifying the category of mutations which disrupt the differentiation process as the Achilles's heel of terminal differentiation, we developed a redundant architecture using a novel split-\u03c0 protein system which required 2 mutations to break the circuit. We experimentally demonstrated a trade-off between rate of production and duration of function as the differentiation rate is tuned, an increased benefit of terminal differentiation with higher-burden expression, and that redundancy improves the evolutionary stability of the terminal differentiation architecture. Specifically we achieve a maximum of ~2.8x (single-cassette terminal differentiation) and ~4.2x (redundant terminal differentiation) the total fluorescent protein production achieved from comparable high-burden naive expression in which all cells inducibly express T7 RNAP. We further demonstrate differentiation can enable the expression of even toxic functions, and develop a terminal differentiation circuit architecture which will allow the degree of redundancy and therefore the evolutionary stability of the architecture to be scaled to arbitrary degrees.
", "doi": "10.7907/5k67-b636", "publication_date": "2022", "thesis_type": "phd", "thesis_year": "2022" }, { "id": "thesis:13824", "collection": "thesis", "collection_id": "13824", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:06212020-111155395", "type": "thesis", "title": "Expanding the Scope of Metalloprotein Families and Substrate Classes in New-to-Nature Reactions", "author": [ { "family_name": "Knight", "given_name": "Anders Matthew", "orcid": "0000-0001-9665-8197", "clpid": "Knight-Anders-Matthew" } ], "thesis_advisor": [ { "family_name": "Arnold", "given_name": "Frances Hamilton", "orcid": "0000-0002-4027-364X", "clpid": "Arnold-F-H" } ], "thesis_committee": [ { "family_name": "Shapiro", "given_name": "Mikhail G.", "orcid": "0000-0002-0291-4215", "clpid": "Shapiro-M-G" }, { "family_name": "Arnold", "given_name": "Frances Hamilton", "orcid": "0000-0002-4027-364X", "clpid": "Arnold-F-H" }, { "family_name": "Clemons", "given_name": "William M.", "orcid": "0000-0002-0021-889X", "clpid": "Clemons-W-M" }, { "family_name": "Reisman", "given_name": "Sarah E.", "orcid": "0000-0001-8244-9300", "clpid": "Reisman-S-E" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "Heme proteins, in particular cytochromes P450, have been extensively used in biocatalytic applications due to their high degree of regio-, chemo-, and stereoselectivity in oxene-transfer reactions. In 2013, it was shown for the first time that engineered heme proteins can also catalyze analogous carbene- and nitrene-transfer reactions. Research in this field has since grown dramatically, with emphasis on developing new heme protein variants to increase the scope of biotransformations accessible through these new transfer reactions. This thesis details the expansion of these new-to-nature carbene and nitrene-transfer reactions to include new substrate classes previously unexplored with iron-porphyrin proteins, the use of non-heme metalloproteins for these transformations, and steps toward improving the robustness of the new-to-nature biocatalytic platform. Chapter 1 introduces the steps the field of biocatalysis has taken toward engineering enzymes with new catalytic functions and the process by which these activities are discovered and enhanced. Chapter 2 details the discovery and engineering of heme proteins which catalyze the stereodivergent cyclopropanation of unactivated and electron-deficient alkenes via carbene transfer, expanding the substrate classes beyond styrenyl alkenes. Chapter 3 shows the development of engineered variants of a heme protein (Rhodothermus marinus nitric oxide dioxygenase) for the diastereodivergent synthesis of cyclopropanes functionalized with a pinacolborane moiety, enabling product diversification through standard cross-coupling reactions. In Chapter 4, a collection of non-heme metalloproteins is curated, and a non-heme iron enzyme (Pseudomonas savastanoi ethylene-forming enzyme) is shown to be both amenable to directed evolution and non-native ligand substitution to enhance its nitrene-transfer activity. Chapter 5 describes the expansion of sequence space targeted for screening in the serine-ligated cytochrome P411 from Bacillus megaterium (P411BM3) biocatalytic platform to enhance the mutational robustness of these remarkable enzymes. Overall, this work provides a framework for bringing model new-to-nature reactions to their full potential in synthetic biocatalytic reactions.
", "doi": "10.7907/7qh5-5130", "publication_date": "2021", "thesis_type": "phd", "thesis_year": "2021" }, { "id": "thesis:14111", "collection": "thesis", "collection_id": "14111", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:03262021-160841703", "primary_object_url": { "basename": "thesis.pdf", "content": "final", "filesize": 3545856, "license": "other", "mime_type": "application/pdf", "url": "/14111/1/thesis.pdf", "version": "v7.0.0" }, "type": "thesis", "title": "Signal Amplification in Synthetic Bacterial Communication", "author": [ { "family_name": "Parkin", "given_name": "James Michael", "orcid": "0000-0002-4058-2338", "clpid": "Parkin-James-Michael" } ], "thesis_advisor": [ { "family_name": "Murray", "given_name": "Richard M.", "orcid": "0000-0002-5785-7481", "clpid": "Murray-R-M" } ], "thesis_committee": [ { "family_name": "Winfree", "given_name": "Erik", "orcid": "0000-0002-5899-7523", "clpid": "Winfree-E" }, { "family_name": "Leadbetter", "given_name": "Jared R.", "orcid": "0000-0002-7033-0844", "clpid": "Leadbetter-J-R" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Thomson", "given_name": "Matthew", "orcid": "0000-0003-1021-1234", "clpid": "Thomson-M-W" }, { "family_name": "Murray", "given_name": "Richard M.", "orcid": "0000-0002-5785-7481", "clpid": "Murray-R-M" } ], "local_group": [ { "literal": "div_bbe" } ], "abstract": "Synthetic biology will one day enable embedded control of a variety of chemical and biological contexts, from the human gastrointestinal tract to crop roots. Groups of engineered organisms, also known as synthetic consortia, can inhabit niches of interest while monitoring and intervening according to their genetic design. However, the spatial structure of the deployment environments can obstruct coordination between cosortia members. The mechanisms engineered bacteria use to communicate must contend with these adversarial conditions to maximize group performance.
\r\n\r\nCoordination between synthetic bacteria is typically achieved using small molecules that can traverse cell membranes through passive transport. Cell communicate by producing and sensing these small molecules. In cell-cell signaling relationships composed of a sender population and a receiver population, the concentration of signaling molecule sensed by the receiver cells depends on the spatial patterning of the two groups, the geometry of the diffusive environment, and the sender population\u2019s signal secretion rate.
\r\n\r\nTo make sender-receiver communication more robust to these environmental features, we introduce a third consortium strain that transiently amplifies local signaling molecule concentrations. These amplifier cells employ a synchronized pulse-generating circuit built using Lux-type quorum sensing components and an IFFL transcriptional architecture. When applied to sender-receiver consortia growing on semi-solid media, these amplifier cells respond to sender-secreted signaling molecules by contributing a small amount themselves. The support of amplifier cells enables communication over longer distances than can be achieved by sender cells alone and can partially recover coordination in small consortia where the sender population is too small to successfully signal its receiver population alone. We extend these results using simulation to investigate the benefit that amplifier cells confer to consortia of varying complexity.
", "doi": "10.7907/50p8-bd89", "publication_date": "2021", "thesis_type": "phd", "thesis_year": "2021" }, { "id": "thesis:11726", "collection": "thesis", "collection_id": "11726", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:06082019-034706928", "primary_object_url": { "basename": "einav_tal_2019_thesis.pdf", "content": "final", "filesize": 40337719, "license": "other", "mime_type": "application/pdf", "url": "/11726/1/einav_tal_2019_thesis.pdf", "version": "v6.0.0" }, "type": "thesis", "title": "Taming the Molecular Dance: Harnessing Statistical Mechanics to Quantitatively Characterize Allosteric Systems", "author": [ { "family_name": "Einav", "given_name": "Tal", "orcid": "0000-0003-0777-1193", "clpid": "Einav-Tal" } ], "thesis_advisor": [ { "family_name": "Phillips", "given_name": "Robert B.", "orcid": "0000-0003-3082-2809", "clpid": "Phillips-R" } ], "thesis_committee": [ { "family_name": "Hsieh", "given_name": "David", "orcid": "0000-0002-0812-955X", "clpid": "Hsieh-David" }, { "family_name": "Phillips", "given_name": "Robert B.", "orcid": "0000-0003-3082-2809", "clpid": "Phillips-R" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Eisenstein", "given_name": "James P.", "orcid": "0000-0001-5460-0464", "clpid": "Eisenstein-J-P" }, { "family_name": "Motrunich", "given_name": "Olexei I.", "orcid": "0000-0001-8031-0022", "clpid": "Motrunich-Olexei" } ], "local_group": [ { "literal": "div_pma" } ], "abstract": "The pace of biological research continues to grow at a staggering pace as high-throughput experimental techniques rapidly increase our ability to sequence DNA, quantify cell behavior, and image molecules of all types within the cellular milieu. Given this surge in experimental prowess, the time is ripe to examine how well our conceptual cartoons of biological phenomena can not only recapitulate the data but also successfully predict the outcomes of future experiments.
\r\n\r\nOne of the fundamental challenges in biology is that the space of possible molecules is overwhelmingly large. The number of variants of a moderately-sized protein (20^300) is larger than the number of atoms in the universe, as is the space of possible bacterial genomes, protein interaction networks, and effector functions; progress in any of these fronts requires a theory-experiment dialogue that can extrapolate our small drop of data to explain large swaths of parameter space.
\r\n\r\nMy thesis strives towards this goal by analyzing a number of central molecular players in biology including enzymes (biological catalysts that accelerate chemical reactions), transcription factors (proteins that bind to DNA and regulate its expression), and ion channels (signaling proteins that regulate ion transport). I develop a quantitative description in each context by harnessing the statistical mechanical Monod-Wyman-Changeux model of allostery which coarse-grains the behavior of a multi-state system into two effective states, demonstrating that these seemingly diverse molecules are all governed by the same fundamental equation.
\r\n\r\nWrit large, there are two overarching goals encompassed by these projects. The first is to translate our biological knowledge into concrete physical models, enabling us to quantitatively describe how the key molecular components in each system interact to carry out their function. The second goal is to analyze how mutations can be mapped into the fundamental biophysical parameters governing each system. In my opinion, predicting the effects of mutations remains one of the great unsolved problems in biology, and it has been incredibly exciting to make progress on this front.
\r\n\r\nLooking back at my amazing graduate school experience, one of the most surprising aspects of my PhD was how closely each of my projects revolved around experiments. I entered graduate school as a theoretical physicist expecting to work on esoteric mathematical models, yet the direct connection with data provided a window into the exhilarating world of biology. While I have never physically manipulated these biological systems in the lab, my models allow me to push and prod and examine their behavior from the most mundane to the utterly extreme limits. Through modeling, I test our assumptions of how these systems work and tease out insights into their underlying biophysical mechanism. Most importantly, these models enable me to harness the incredible wealth of hard-won data to weave a few more threads of understanding into our tapestry of how these incredible living systems operate.
", "doi": "10.7907/S4CV-T162", "publication_date": "2019", "thesis_type": "phd", "thesis_year": "2019" }, { "id": "thesis:11281", "collection": "thesis", "collection_id": "11281", "cite_using_url": "https://resolver.caltech.edu/CaltechTHESIS:11262018-100422941", "primary_object_url": { "basename": "Noah_Olsman_Thesis_Final.pdf", "content": "final", "filesize": 3608771, "license": "other", "mime_type": "application/pdf", "url": "/11281/1/Noah_Olsman_Thesis_Final.pdf", "version": "v5.0.0" }, "type": "thesis", "title": "Architecture, Design, and Tradeoffs in Biomolecular Feedback Systems", "author": [ { "family_name": "Olsman", "given_name": "Noah Andrew", "orcid": "0000-0002-4351-3880", "clpid": "Olsman-Noah-Andrew" } ], "thesis_advisor": [ { "family_name": "Goentoro", "given_name": "Lea A.", "orcid": "0000-0002-3904-0195", "clpid": "Goentoro-L-A" }, { "family_name": "Doyle", "given_name": "John C.", "orcid": "0000-0002-1828-2486", "clpid": "Doyle-J-C" } ], "thesis_committee": [ { "family_name": "Murray", "given_name": "Richard M.", "orcid": "0000-0002-5785-7481", "clpid": "Murray-R-M" }, { "family_name": "Bois", "given_name": "Justin S.", "orcid": "0000-0001-7137-8746", "clpid": "Bois-J-S" }, { "family_name": "Doyle", "given_name": "John C.", "orcid": "0000-0002-1828-2486", "clpid": "Doyle-J-C" }, { "family_name": "Goentoro", "given_name": "Lea A.", "orcid": "0000-0002-3904-0195", "clpid": "Goentoro-L-A" } ], "local_group": [ { "literal": "div_eng" } ], "abstract": "A core pursuit in systems and synthetic biology is the analysis of the connection between the low-level structure and parameters of a biomolecular network and its high-level function and performance. Elucidating this mapping has become increasingly feasible as precise measurements of both input parameters and output dynamics become abundant. At the same time, cross-pollination between biology and engineering has led to the realization that many of the mathematical tools from control theory are well-suited to analyze biological processes.
\r\n\r\nThe goal of this thesis is to use tools from control theory to analyze a variety of biomolecular systems from both natural and synthetic settings, and subsequently yield insight into the architecture, tradeoffs, and limitations of biological network. In Chapter 2, I demonstrate how allosteric proteins can be used to respond logarithmically to changes in signal. In Chapter 3, I show how control theoretic techniques can be used to inform the design of synthetic integral feedback networks that implement feedback with a sequestration mechanism. Finally, in Chapter 4 I present a novel simplified model of the E. coli heat shock response system and show how the the mapping of circuit parameters to function depends on the network's architecture.
\r\n\r\nThe unifying theme of this research is that the conceptual framework used to study engineered systems is remarkably well-suited to biology. That being said, it is important to apply these tools in a way that is informed by the molecular details of biological processes. By combining structural and biochemical data with the functional perspective of engineering, it is possible to understand the architectural principles that underlie living systems.
", "doi": "10.7907/DGPY-1679", "publication_date": "2019", "thesis_type": "phd", "thesis_year": "2019" } ]