71446 research outputs found
Sort by
Understanding Disease States Through Transcriptomics
No full textThe enclosed thesis encompasses three principal topics: a single-cell transcriptomic characterization of cellular states in Microscopic Colitis; an examination of the cellular profile in human breast milk; searching for targetable vulnerabilities in chordoma. Single-cell RNA sequencing (scRNAseq) was used to profile the cells in Microscopic Colitis (MC), a gut inflammatory disorder, and place it in the spectrum of commonly studied inflammatory colitides. The cellular profile of MC presents similarities with other colitides, such as ulcerative colitis (UC). Importantly, overactive inflammatory pathways in UC, with existing targeted therapies, are implicated in MC pathogenesis. This study suggests options for a disease (MC) where therapeutics are lacking.
scRNAseq was also used to characterize the secreted cells in human breast milk. Maternal cells in breastmilk were previously shown to take up residence in various tissues in the infant, yet the cellular profile of these cells is poorly understood. The presented data show that a significant proportion of these cells are immune and have a highly muted inflammatory potential. Further, secreted cells appear selected to prime the infant's own immune system against potential exogenous threats. Analysis of the chemokine and receptor repertoire on secreted epithelial and immune cells, respectively, suggests an alternative form of cell trafficking that does not rely on currently documented chemokines.
Chordoma is a rare tumor of the spine or base of the skull thought to arise from remnants of embryonic notochord. Notochordal cell profiles were generated and compared to chordoma and other body tissues to provide the first whole-transcriptome molecular evidence linking the two. The transcriptomic profiles helped identify a therapeutically targetable gene network with a dependence on TGFβ. Brachyury is unique to both notochord and chordoma, and is essential to chordoma survival. However, Brachyury is a transcription factor, and challenging to therapeutically target. An expression cloning high throughput screen using a Brachyury transcriptional reporter helped identify other Brachyury pathway members that could be easier to target. The screen also led to the discovery of an inherent overactive interferon response in chordoma cells. Results point to a link between the overactive interferon response and Brachyury. The overactive interferon response can lead to apoptosis in chordoma cells, a therapeutically exciting option for a rare disease where surgery and adjuvant radiotherapy are currently the only options.Medical Science
The Role of the Motor System in Speech and Language in Autism
No full textCommunication challenges in autism are often attributed to social and cognitive factors, but emerging research suggests that motor processes also play a critical role. Despite growing interest in this area, the brain and behavioral underpinnings of speech-motor differences in autism, and their relationship to language, remain poorly understood. The present dissertation examined how motor system differences in the brain contribute to speech and language challenges in autistic individuals who use speech to communicate. This work integrated studies that bridge brain and behavior, while also incorporating the perspectives of autistic people.
The first study examined how three key language-relevant brain networks – speech perception, working memory, and speech production networks – related to language challenges in autistic children. FMRI results revealed that autistic children showed atypical and reduced engagement of the speech production network, but not more traditional language processing regions, during nonword repetition. This atypical motor engagement correlated with poorer task performance, higher autism traits, and lower social communication skills. The second study built on this finding to explore how motor system differences impact speech and language in autism beyond nonword repetition. Research in nonautistic individuals, including lesion-based work, has suggested the importance of key motor regions in both the production and perception of prosody. Thus, we turned to prosody as an area of language that may be impacted by brain-based motor differences. Prosody – the use of pitch, duration, and loudness in speech – plays a key role in communication, and atypical prosody is commonly noted in autism. Study two revealed that autistic individuals reported greater challenges with prosody, and that these challenges were strongly correlated with autism traits in both autistic and nonautistic adults. Acoustic analyses showed that the autistic participants produced speech with atypical timing, with longer pauses and a slower speech rate, during a spontaneous picture description task. Notably, diadochokinetic and articulation rate did not differ between groups, suggesting that these prosody differences were unlikely due to an underlying speech-motor disorder (e.g., apraxia of speech), but may stem from broader motor system differences.
The third study further examined the motor-based brain differences in autism, by first precisely localizing each participants’ speech production regions, and then examining how these functionally-defined regions were engaged during prosody perception. Results from this fMRI study revealed that nonautistic individuals engaged several of the same functional regions for speech production as prosody perception (i.e., listening to sentences with dynamic versus flat pitch contours). In contrast, autistic individuals showed a reduced activation in these regions. Significant group differences emerged across several motor regions, including the supplementary motor area. Moreover, activation in motor regions during prosody perception was positively correlated with participants’ accuracy on an out-of-scanner prosody perception task, highlighting the behavioral relevance of this motor activation.
Taken together, these studies provide evidence that atypical engagement of the motor system contributes to language differences in autism. By reframing language differences in autism through the lens of speech motor system function, this dissertation offers new insights into the neurobiological mechanisms underlying language challenges. These findings lay a foundation for future research aimed at supporting autistic communication.Speech and Hearing Bioscience and Technolog
Dual Species Atom Arrays for Quantum Simulation and Computation
No full textOptical tweezer arrays of neutral atoms have emerged as a promising platform for studying quantum physics. These individual atoms can be carefully prepared in single quantum states, and easily manipulated with microwaves and lasers. MHz scale interactions are accessible at several micron scale distances by exciting these atoms to Rydberg states. These arrays can also be created in arbitrary 2D geometries with the use of a spatial light modulator. Adding the capability of a second species opens new possibilities in non-destructive measurement and quantum simulation of bipartite systems with widely tunable parameters.
In this thesis, we describe our efforts to build a flexible platform for studying quantum phenomena utilizing two species of atoms, sodium and cesium. In particular, we discuss the laser technology and setups required to create dual species optical tweezer arrays. These arrays are made defect-free with real-time rearrangement of atoms using a separate set of tweezers created with acousto-optical deflectors. We then describe how to excite these atoms coherently to the Rydberg state with coherence times in Cs as long as 20 microseconds. With these techniques established, we are able to probe an Ising critical point in 1D and 2D systems with up to 81 atoms. In particular, we measure and confirm the value of the universal critical exponent, , via adiabatic preparation of the ground state at the critical point. Then, we measure the interactions between sodium and cesium atoms setting the stage for future experiments. Lastly, a theoretical proposal is presented where Rydberg atoms can be used to enhance the interaction rates of ultracold polar molecules, as well as measure their states non-destructively.Physic
Foundations for Genome-Scale Artificial Intelligence
No full textModeling whole genomes—the complete sequence of base pairs, including both coding and non-coding regions organized within a three-dimensional architecture—represents a grand challenge in bioinformatics. Achieving this goal could transform our ability to understand complex polygenic diseases and develop treatments for both rare and common conditions. However, whole-genome modeling presents two fundamental challenges: the vast size of genomic data and the intricate complexity of genetic interactions. Artificial intelligence (AI) offers a powerful approach to processing large datasets and extracting meaningful patterns. However, for AI to effectively model whole genomes, it must be (1) generalizable, capable of predicting the effects of novel, unseen mutations; (2) capable of reasoning across sequences and biological scales, as genetic function emerges from the multi-level interactions of sequences; and (3) multimodal, integrating information from diverse sources, including scientific literature. While this dissertation does not yet achieve full-scale genome modeling, it introduces four novel AI methodologies—SPECTRA, Phyla, RLDIF, and Fleming—each addressing essential components necessary to enable this vision. Together, these contributions establish a conceptual and methodological framework that lays the groundwork for the future development of genome-scale artificial intelligence. SPECTRA is a framework for evaluating model generalizability beyond conventional dataset splits. By systematically varying train-test similarity, SPECTRA reveals that existing biological foundation models fail to generalize to sequences dissimilar from their training data. In response, Phyla is designed to explicitly learn how to compare sequences by leveraging evolutionary relationships. Trained on protein phylogenies, Phyla not only excels at sequence comparison but also reconstructs phylogenetic trees with high accuracy, revealing both known and novel evolutionary insights. Additionally, this dissertation presents RLDIF, a categorical conditional diffusion model for protein inverse folding, which leverages reinforcement learning to improve multi-scale modeling. By optimizing sequence design with respect to structural recovery, RLDIF achieves state-of-the-art performance in generating diverse sequences that accurately fold into a specified target structure. Beyond sequence information, AI models must incorporate knowledge from external sources to avoid rediscovering established principles. To address this, Fleming is developed as an AI agent for antibiotic design in tuberculosis, integrating scientific literature with machine learning tools to generate novel antibiotic candidates. The same multimodal approach can be extended to genome analysis, enabling AI to reason over diverse data modalities. Together, these innovations establish a foundation for genome-scale AI, providing key tools for understanding and reasoning over whole genomes.Biomedical Informatic
Examining the Interrelationships of Child Infections, Growth, And Development towards an Integrated Intervention Agenda in Low and Middle-Income Countries
No full textChildren in low- and middle-income countries (LMICs) face multiple social, economic, and environmental challenges that increase their risk of mortality, morbidities, poor growth, and suboptimal development outcomes. Considering the shared risk factors as well as the inter-relationships between child infectious disease morbidities, growth, and development, integrated interventions may be more effective in improving child well-being in resource-limited settings. However, critical evidence gaps remain in designing and delivering such interventions. First, a better understanding of the relationship between common infections and child development is needed. Second, while nutrition is identified as a critical link between child survival, growth and development, there is limited evidence on mediators that could be targeted to potentially maximize benefits across multiple domains. For example, diet and care may mediate the effects of parenting interventions on childhood infections, an area that has not been previously explored Finally, little is known about whether early child development interventions can also affect broader outcomes beyond the primary focus, such as child growth and infectious disease morbidity. To address these gaps, the overarching research question answered in this thesis was “How is early childhood infectious disease morbidity exposure associated with child development, and can the analysis of early childhood intervention effects on broader outcomes beyond the primary focus help inform integrated approaches for reducing child infectious disease morbidity and improving child growth and development?”
Using longitudinal cohort data from a sequential randomized controlled trial in rural Niger, Paper 1 found negative relationships between common child infectious disease morbidities and development outcomes in the first two years of life. Paper 2 used data from a four-arm factorial-designed cluster-randomized controlled trial of a home-based responsive stimulation and nutrition intervention in the first 1000 days of life integrated into an existing community health worker program in rural Pakistan. Paper 2 showed that responsive stimulation intervention reduced the rates of common childhood infections while a nutrition intervention including iron-containing MNP increased the risk of fever among children under two years of age. Using data from a stepped wedge cluster-randomized controlled trial conducted in rural Pakistan, Paper 3 found that a center-based ECCE intervention without any nutrition input positively affected child linear growth while negatively affected weight-based anthropometric indicators among children 4.5 to 5.5 years of age. Collectively, these findings inform the need for integrated approaches for child infectious disease morbidity, growth, and development in the first and next 1000 days of life.Population Health Science
In vitro and in vivo studies of two essential MRSA cell wall synthesis enzymes
No full textMethicillin resistant Staphylococcus aureus (MRSA) is a growing healthcare threat worldwide. The mechanism of antibiotic resistance in MRSA relies on the acquired cell wall synthesis enzyme PBP2a, which rescues transpeptidase (TP) activity upon inhibition by β-lactam drugs. Previous evidence reveals a functional coordination between PBP2a and PBP2, an essential S. aureus enzyme with glycosyltransferase (GT) and TP activities.
Excitingly, recent data have demonstrated that PBP2 and PBP2a form a physical complex, which can be isolated via tandem affinity purification and size exclusion chromatography. In Chapter 2 of this thesis work, we replicate the PBP2/PBP2a complex purification in an effort to further characterize the heterodimer. However, binding assays and structural studies present challenges that suggest a weak binding interaction in vitro.
To investigate the coordination between PBP2 and PBP2a within the S. aureus cell, we express PBP2a in a methicillin sensitive Staphylococcus aureus (MSSA) background and study its impact on PBP2 localization. In Chapter 3, we explore PBP2’s recruitment to the division site and the relevance of PBP2a for this recruitment under antibiotic treatment. Here, we find that PBP2a can rescue PBP2’s septal localization when it is disrupted by β-lactams. We further demonstrate that antibiotic inhibition of either GT or TP activity displaces PBP2 from the division site, but a catalytically inactive copy of PBP2 is still able to localize properly in the presence of other active PBP2 enzymes.
Ultimately, our data suggest a unique model of PBP2 septal recruitment, in which nascent septal peptidoglycan recruits PBP2 to the division site in a positive feedback loop, and other divisome components stabilize the enzyme’s localization once it arrives. These discoveries add important insights into how S. aureus is positioned for its essential role in cell division and how PBP2a acts to rescue MRSA strains from β-lactam inhibition.Chemical Biolog
Flirting with Boston: a Novel
No full textJennifer ‘JJ’ Kelsey is in her early 30s when she moves to Boston to work as a journalist for a small online magazine. She’s assigned mostly fluff pieces, so when a potential murder cover-up comes across her desk, she can’t help but chase the real story. As she combs through police records, social media profiles, and public databases, the reader realizes that JJ is being watched and hunted. Can she trust the people she is closest with? Or will she discover how very dark and twisted even the people you trust most can be?
Flirting with Boston is a 54,000-word, character-driven novel that explores the boundaries of friendship, trust, and obsession through two voices: the curious–and, at times, reckless–protagonist and the attentive stalker following her every move.Extension Studie
The Case for Semi-Sovereign Participation in International Relations
No full textInternational relations theory holds that the ability to participate in international relations is one of the fundamental abilities of a state, with several schools of thought in international relations, such as realism, using the state as the starting point for their theories. However, as there are semi-sovereign entities that participate in international organizations, rather than view these entities as exceptions, an examination of the role of these entities from the perspectives of both international relations theory and legal theory supports broader normalization for their continued participation. As states of widely different capacity co-exist within a decentralized international social order, the inclusion of entities with deferred sovereignty would not be inconsistent with current practice.
Rather than viewing sovereignty solely as a question of territorial integrity, or the authority of a state over its geographic domain, this thesis considers sovereignty as a spectrum. This approach allows for a more nuanced understanding of the potential powers of political entities, notably the capacity to enter into treaties or relations with states and other entities. Recognizing the capacity of non-state entities to enter into treaty relations explains the participation of semi-sovereign entities in international organizations today and points to new ways of addressing contemporary global issues.
Although states are nominally equal under international law, international relations scholars use the term hierarchy to describe a system where political, economic, and social status among actors is highly differentiated even as they retain a degree of formal sovereign equality. This thesis also questions the notion that states should be assumed to be considered legally equal and the role of sovereign equality as a foundational principle of the international legal order. Viewing sovereignty as a spectrum would allow for greater participation of semi-sovereign entities in international institutions without affecting the status quo. Consequently, the participation of semi-sovereign entities in international relations, including membership in international organizations, should be normalized.Extension Studie
Improving CAR T Cell Persistence in Pancreatic Cancer Using an In Vivo CRISPR Knockout Screen
No full textChimeric antigen receptor (CAR) T cell therapy has demonstrated remarkable success in treating a subset of hematologic malignancies. However, efficacy against solid tumors has remained limited owing, in part, to the poor CAR T cell persistence in the immunosuppressive solid tumor microenvironment. Preclinical CRISPR screening studies have improved our understanding of the mechanisms that regulate CAR T cell exhaustion. However, such work has primarily been conducted in vitro, failing to reproduce the complex challenge of the hypoxic, cytokine-depleted vivo tumor microenvironment. Here, we perform in vivo CAR T cell CRISPR knockout screens (14 or 28-day CAR engraftment) in pancreatic adenocarcinoma using a curated Mario guide library to identify genes that enhance CAR T cell persistence in vivo. Results suggest a temporal progression to the enrichment of gene hits. The short-term, 14-day Mario screen may have identified hits that improve the short-term enrichment of Mesothelin CAR T cells. However, the 28-day screen identified gene knockouts implicated in the JAK-STAT signaling pathway, such as SOCS1 and PTPN2, which demonstrate more durable tumor control.Medical Science
Accelerating Inference: Mitotic Stein Variational Gradient Descent for Bayesian Analysis of Dynamical Systems
No full textThis thesis introduces mitotic Stein variational gradient descent (mSVGD), a novel
enhancement to Stein variational gradient descent (SVGD) designed to improve speed,
convergence behavior, and robustness of particle-based variational inference. The research
focuses on addressing computational inefficiencies in manifold-constrained Gaussian
Process (MAGI) inference, a framework for Bayesian inference of ordinary differential
equation systems. By leveraging a structured particle expansion approach inspired by
mitotic cell division, mSVGD mitigates sensitivity to hyperparameter selection, accelerates
convergence, and enhances robustness against noisy and sparse data.
Empirical evaluations on the FitzHugh-Nagumo, Hes1 protein, and Lorenz models
demonstrate that mSVGD achieves a 30× to 50× speedup over the traditional MAGI
implementation that uses Hamiltonian Monte Carlo sampling, while maintaining or
improving inference accuracy. The proposed method also exhibits superior consistency in
convergence behavior and increased stability compared to SVGD. These results position
mSVGD as a scalable and efficient alternative to both traditional MCMC-based inference
techniques and standard SVGD.
This work contributes to ongoing advancements in variational inference, particularly for
dynamical systems with computational constraints, and highlights the potential of mSVGD
as a powerful tool for Bayesian inference in complex, real-world scientific applications.Computer Scienc