95038 research outputs found
Sort by
Characterizing the Cell-Autonomous Role of CHCHD2 in Neural Stem Cell Fate Decisions During Adult Hippocampal Neurogenesis
Get PDFAdult hippocampal neurogenesis (AHN) is the process wherein an endogenous pool of neural stem cells (NSCs) in the hippocampus makes fate decisions to become adult-born neurons (ABN). While many factors influence AHN, mitochondrial metabolism has become a central regulator of both NSC self-renewal and differentiation into ABNs. CHCHD2 has emerged as a fascinating gene to study in the context of elucidating the mechanisms underlying mitochondrial metabolism's role in regulating AHN. Despite CHCHD2 promoting mitochondrial oxidative phosphorylation through stabilization of mitochondrial Complex IV, NSCs, which rely on glycolysis, express elevated levels of CHCHD2 relative to ABNs, which rely on oxidative phosphorylation. Characterizing the role of CHCHD2 in NSC fate decisions is critical to leveraging NSC biology to enhance AHN
Score-Based Genome-Wide Association Analysis of High-Dimensional Multi-Omics Data
Get PDFHigh-dimensional sequencing data, such as RNA-Seq for quantifying gene expression, ATAC-Seq for assessing chromatin accessibility, and 16S rRNA sequencing for measuring microbiome abundances, are commonly used in systems biology research. Analyzing associations between different types of omics data provides valuable insight into underlying biological mechanisms. Accessible chromatin allows transcription factors and regulatory elements to bind to DNA, thereby regulating transcription through the activation or repression of target genes. The association analysis of RNA-Seq and ATAC-Seq data provides insights into gene regulatory mechanisms. Most existing analytic tools exclusively focus on cis-associations, despite regulatory elements being able to physically interact with distant target genes. Furthermore, conventional approaches often utilize Pearson or Spearman correlations, which ignore the count-based nature of RNA-Seq data. In the first project, we introduce PETScan, a computationally efficient genome-wide PEak-Transcript Score-based association analysis, utilizing negative binomial models to better accommodate RNA-Seq data. We leverage score tests and matrix calculations for improved computational efficiency, and combine an empirical permutation method with genomic control to ensure valid p-value calculations in studies with limited sample sizes. We further extend this framework to PETScan_Mixed, incorporating negative binomial mixed models to account for within-subject correlations. Imbalances in microbial composition, potentially regulated by host gene expression, have been associated with various diseases. Investigating the associations between the microbiome and host transcriptome can help uncover the mechanisms underlying human health and disease. In the second project, we introduce MITScan, a computationally efficient genome-wide MIcrobiome-host Transcriptome Score-based association analysis, employing zero-inflated negative binomial models to accommodate extra zeros commonly observed in microbiome data. Evaluating the model adequacy of negative binomial and negative binomial mixed models remains essential for RNA-Seq analysis, despite they are widely used. In the third project, we implement goodness-of-fit tests based on the cumulative sums of residuals over covariates. The asymptotic distributions under the null hypothesis can be approximated by zero-mean Gaussian processes, with realizations generated through simulations. The goodness-of-fit tests effectively control type I error and demonstrate power in assessing functional forms of covariates, checking a misspecified link function, and detecting outliers.Doctor of Philosoph
Mediation Analysis Using Saturated and Restricted Models
Get PDFMediation analysis is widely used to evaluate causal mechanisms, yet the impact of saturated versus restricted model specifications remains understudied. Saturated models freely estimate all possible paths, while restricted models impose theoretically driven parameter constraints. Despite the flexibility of both regression and Structural Equation Modeling (SEM) to incorporate restrictions, applied researchers often default to saturated models—even when theoretical hypotheses do not justify the estimation of all paths. This approach may serve as a safeguard against bias due to model misspecification but also introduces potential drawbacks, including reduced statistical efficiency and lower statistical power for detecting specific indirect effects. This study examines the implications of model specification choices for a common serial mediation structure through a Monte Carlo simulation, assessing parameter estimation, standard errors, confidence intervals, and statistical power across varying sample sizes and effect size magnitudes. Findings suggest that model specification meaningfully impacts the power to detect indirect effects, with restricted models having greater power than overparameterized models. Improperly restricted models may, however, produce biased estimates of indirect effects. Theoretical and practical considerations for aligning model specification with hypothesized mechanisms are discussed.Master of Art
Para Las Damas: Feminism and Social Change in Havana, 1918-1940
Get PDFThis dissertation examines feminist organizing in early republican Havana. Feminist activism flourished during the 1920s as women, particularly middle-class white women in Cuba’s capital, gained new freedoms in education, the workplace, and the domestic sphere. These developments contributed to the emergence of a robust feminist movement that sought to expand women’s rights and secure full civic, social, and legal equality with men. Feminists in Havana organized national congresses, founded clubs and associations, and promoted their cause through speeches and publications. While these efforts brought greater visibility to feminist demands, the movement’s priorities often reflected the social position of the elite white women who occupied leadership roles. As the 1920s progressed, deep ideological divisions emerged among feminist leaders. Moderate and conservative feminists often framed gender inequality as a moral issue, emphasizing the belief that women’s enfranchisement would help guide Cuban society and its citizens toward virtue. In contrast, radical leftist feminists demanded a revolutionary transformation that addressed the structural roots of sexism and an overhaul of the norms of white middle-class Cuban society. These fundamentally different visions for Cuba’s future fractured the movement’s initial sense of unity and led to increasing fragmentation along political lines. The rise of the dictatorship of Gerardo Machado (1925-1933) briefly united feminists once again. Confronted by state violence and authoritarianism, feminists joined students and workers in public protest and became some of the most visible members of the opposition. Feminists were instrumental in mobilizing resistance, but this period of unity proved fleeting. Once Machado fell, longstanding divisions resurfaced, and the movement struggled to maintain momentum. After the fall of Machado and the subsequent granting of suffrage the year after in 1934, feminism on the island began to wane. Many feminists turned their attention to formal politics, running for office or supporting other female candidates rather than sustaining independent feminist organizations. Radical women shifted to join the Communist Party and anti-fascist movements. While the feminist movement helped reshape public life in Cuba, it was ultimately hindered by its internal contradictions - particularly its failure to bridge divisions of class, race, and political ideology.Doctor of Philosoph
AT THE CROSSROADS OF CLIMATE, JOBS, AND JUSTICE: GREEN JOBS COALITIONS IN NORTH CAROLINA
Get PDFThis dissertation assesses institutional challenges facing coalition building between labor unions and environmental organizations. Labor unions occupy complicated and contingent political positions between expanding productive industries and supporting environmental protections. I build upon prior scholarships on industry sector contexts and a multi-institutional politics approach in three empirical studies with implications for labor-environmental coalitions: 1) U.S. public opinion on environmental concerns across industries and organizational leaders’ 2) green jobs frame variations and 3) preferences for cross-movement partnerships. Methodologically, I employ General Social Surveys representative of the U.S. public and semi-structured interviews of forty-five leaders and staff from twenty-seven North Carolina union and environmental organizations. Labor leaders’ positions on environmental issues are constrained by less favorable environmental attitudes from constituencies employed in extractive, construction, and manufacturing industries. Additionally, different institutional logics characterizing labor and environmental organizations – constituency servicing, social transformation, and technocratic development – shape leaders’ diverging cultural evaluations around green jobs and preferred labor-environmental partnerships. I conclude with policy implications and several research agendas on institutional change, social movements, and environmental politics.Doctor of Philosoph
The next generation of imaging genetics
Get PDFImaging genetics elucidates how genetic variations influence the human brain and links these variations to brain-related traits, diseases, and disorders. This field advances our understanding of the pathophysiological pathways underlying many brain conditions, with the potential to enable more precise and personalized treatments. Despite this promise, the computational challenges posed by high-dimensional imaging and genetic data remain substantial. For example, functional magnetic resonance imaging (fMRI) produces millions of measurements (e.g., voxels or vertices) per subject, while whole-genome sequencing in the UK Biobank (UKB) has captured approximately 1.5 billion variants across half a million subjects. Traditional methods are computationally infeasible to pinpoint genetic effects on the human brain at fine grids, never to say to investigate heterogeneity of genetic architecture across the brain regions. In this dissertation, we develop novel statistical frameworks that enable imaging genetic analyses at the voxel/vertex level. With particular emphasis on computational efficiency, we aim to create atlases of genetic associations at the highest possible resolution and share summary statistics of the entire image across the whole genome with the community for secondary analyses. In Chapter 2, we propose Representation learning-based Voxel-level Genetic Analysis (RVGA) for genome-wide association analysis (GWAS). RVGA decomposes raw images into smooth signals and random errors, which enhances statistical power as well as reduces computational and resource demands by 2-3 orders of magnitude. We propose a scheme to store and share a minimal dataset of GWAS summary statistics. We introduce a unified estimator for voxel heritability, genetic correlations between voxels, and cross-trait genetic correlations between voxels and non-imaging phenotypes. Moreover, we incorporate partitioned heritability analysis into RVGA. In Chapter 3, we extend RVGA for rare variants and introduce Representation learning-based Voxel-level Rare Variant Analysis (RVRVA). We first propose a framework to correct for sample relatedness at each voxel. We then address the unique challenge of controlling the type I error rate in rare variant analysis due to the failure of large-sample theory. Therefore, RVRVA is able to demonstrate genetic effects of variant sets containing as few as two singletons. We incorporate threshold-free cluster enhancement (TFCE) to evaluate spatial significance of associated brain regions. We also share summary statistics that are flexible to define any variant sets and incorporate various functional annotations. In Chapter 4, we propose a method to estimate the trajectory of heritability from longitudinal data. The approach relies on functional data analysis to reconstruct trajectory of phenotypes and estimate heritability using GWAS summary statistics from latent variables. We show that genetic influences on phenotypes are not consistent over time. We demonstrate the performance of our methods by comprehensive simulations and large-scale real data analysis from the UKB. We develop computationally efficient and user-friendly Python programs for RVGA and RVRVA in a toolbox called Highly Efficient Imaging Genetics (HEIG) at https://github.com/Zhiwen-Owen-Jiang/heig.Doctor of Philosoph
UNDERSTANDING THE EFFECTS OF INCREASED TRANSPARENCY ON DATA PREPROCESSING THROUGH IN-PROCESS VISUALIZATIONS
Get PDFMost work on evaluating bias in data science workflows tends to focus on the model. However, the training data fed into the model and the data preprocessing step that produces it can also have significant impact on model results. While there has been work on editing the data in data preprocessing to mitigate bias, the impact of conventional data preprocessing operations has been understudied. My dissertation delves into how the data preprocessing step can be improved to help analysts better understand the impact of the step and lead to smarter data science decisions. I first study the needs of data scientists when conducting data preprocessing through a small-scale interview study and compared the results with a literature survey of current preprocessing tools. The comparison analysis identified several key gaps between practice and theory. I utilized of result of the analysis to develop the Preprocess Analyzer (PPA) tool, which is designed to address some of the gaps by being integrated into existing data science work environments and provided users with a deeper insight into their data. I conducted a user study to evaluate the ability of PPA to aid with data preprocessing. The study results found that compared to existing popular tools, data scientists gained a better understanding of their data preprocessing workflow when utilizing PPA. Participants generally agreed that PPA included many helpful features such as the ability to quickly display useful statistics, highlight areas of concern, and integration into familiar work environments. I believe the results of this dissertation can guide the design of future data preprocessing tools to better meet the needs of the end user.Doctor of Philosoph
DRUGS, DEPENDENCY, AND DE-ADDICTION IN JAMMU CITY, JAMMU AND KASHMIR, INDIA
Get PDFThis dissertation examines the impact of the Indian state’s fraught relationship with the Indian territory of Jammu & Kashmir (J&K) and Jammu’s marginal positioning vis-à-vis Kashmir, focusing on how these regional and national political dynamics shape the state-sponsored de-addiction (the colloquial term used for drug use treatment) services for injectable heroin users in Jammu City. Through an ethnography of a de-addiction clinic, I demonstrate how Jammu’s undeniable political marginality in the larger equation between J&K and the Indian state, as well as Jammu’s inevitable political dependency on Kashmir, impact addiction and de-addiction across multiple scales: from practices of drug use, to the micro-politics of addiction management and treatment in the clinic, to the legal frameworks for deterring drug use and the implementation of state-sponsored harm reduction services. I conducted 18 months (2019 to 2021) of community-engaged ethnographic research with people who use drugs and their caregivers, addiction treatment professionals, law enforcement personnel, and government officials, during a turbulent political period in J&K (the abrogation of Article 370), further complicated by the COVID-19 pandemic.This dissertation shows how the administration of addiction treatment and substance users’ experiences of recovery are mediated by, and often reliant upon, multiple dependencies on kinship, state institutions, material substance, and complex political realities. These dependencies are simultaneously chemical, social, political, familial, and clinical. I theorize these multiple and intersecting relationalities as nirbharta. Through long-term ethnographic fieldwork, I illuminate how the recovery from substance use requires a wide range of dependencies that are more than the use of buprenorphine or any other drugs alone. The work of recovery is mediated by multiple kinds of nirbharta that include reliance on chemical substances, kin, interpersonal relationships within the clinic, the political stability of the region, and the public perception surrounding addiction treatment and recovery in India. Drawing on methods, including in-depth interviews, focus groups, survey, and participant observation, I show how nirbharta offers a wide-angle lens to foreground clients’, clinicians’, caregivers’, and government officials’ multi-layered and interdependent experience of everyday care, medical support and (in)visibility to the Indian state as they do the work of rehabilitation in Jammu City.Doctor of Philosoph
CHARACTERISTICS AND BIOMECHANICS OF BIRD FLIGHT OUT OF AND OVER WATER
Get PDFSuccessful bird flight depends on the interplay between the environment, morphology, and behavior. The shape and mass of a bird’s wing and body dictate the basic cost of flight. The environment could provide sources of energy that animals can exploit to power their flight, but it can also impose challenges that increase the cost of flight. Finally, the way birds fly (i.e. behavior) shape how the animal interacts with its surroundings. Here I explore these relationships to understand the effects of body size on takeoff performance, how birds exploit the energy from their environment to help meet the costs of flight while foraging, and how birds respond to foraging in high-risks environments. Water is a challenging environment to move in given its higher viscosity and density compared to air, and higher dissipation of energy compared to solids, making aquatic environments perfect to explore these relationships. I used three-dimensional (3D) field videography to capture the natural behavior of different species in the wild. I found that during aquatic takeoff, body size imposes challenges to larger species, while allowing smaller species to leave the water at a faster overall change rate, however the relationship between body mass, morphology, and takeoff performance is complex requiring the integration of environmental and behavioral factor to elucidate the whole picture. For foraging black skimmers, I found that they exploit the coastal wind gradient as they are ascending during their foraging flights, with the potential to save up to 2.5% in the cost of flight. Finally, I found that barn swallows foraging over water decrease their flight speed and increase the sharpness of their turns as they fly closer to the water surface indicating a shift to a strategy that keeps them safe but allows them to maintain maneuverability and respond to their fast prey. My dissertation uses non-invasive 3D motion capture methods to explore the natural behavior of birds in the wild for different movements out of and over water providing insight into the interaction between the environment, behavior, and morphology of birds and their effects on the biomechanics of movements in challenging environments.Doctor of Philosoph
Sex-specific effects in affective behavior, alcohol drinking, amygdala activation and the impact of the psychedelic psilocin in mice
Get PDFAlcohol use disorder (AUD) is a multifaceted, chronic condition characterized by a compulsion to drink despite the emergence of social, medical, and psychological consequences. AUD etiology and progression are complicated by factors such as sex, psychiatric conditions (e.g. anxiety, depression), and stress. Understanding the role these factors play when exploring therapeutic interventions is vital, especially as novel AUD treatments are being validated in their outcomes and efficacy. In the first study, sex differences in affective behavior, ethanol consumption, and amygdala activation were assessed in female and male mice. Results indicate that basal affective state did not predict subsequent drinking and ethanol intake did not change affective behavior in either sex. Females consumed more ethanol than males and ethanol reduced central amygdala (CeA) activation in females only. No change in basolateral amygdala (BLA) activation was observed in either sex. In the second study, sex-specific effects of the psychedelic psilocin on ethanol drinking and amygdala activation were examined. Findings demonstrate that psilocin reduced drinking across all groups in both sexes however these effects were not psilocin-specific. Additionally, ethanol-exposed females had increased CeA and BLA activation that were both decreased by psilocin with no change observed in males. In the third study, the effects of psilocin on CeA corticotropin-releasing factor receptor 1 (CRF1) neuronal activation and drinking were examined in females. Psilocin increased CeA activation and decreased CRF1 activation with subregion-specific differences in ethanol-naïve and ethanol-exposed mice. Psilocin also acutely decreased drinking in two paradigms that varied in ethanol severity. Psilocin did not significantly alter locomotion and increased corticosterone at 72hrs into ethanol withdrawal with no change at 24hrs or one-week post-injection. Collectively, these findings highlight how sex plays an important role in affective behavior, drinking, and amygdala activation. Considering these factors and how they can alter therapeutic outcomes may inform clinical therapeutic intervention strategies to validate and optimize novel AUD treatments.Doctor of Philosoph