Investigating Susceptibility to Tuberculosis Using Functional Genomics Approaches

A major goal of human genetics is to characterize the role of genetic variation on complex, polygenic phenotypes. With the discovery from genome-wide association studies (GWAS) that many associated variants have a small effect size and are located in non-coding regions of the genome, there has been a large effort to collect functional genomics data. The hope is that a better understanding of how the genome functions in diverse developmental states and environments will provide insight into the context-specific activity of associated non-coding variants. My research applies this paradigm to the complex phenotype of susceptibility to develop tuberculosis (TB). It has been estimated that 10% of individuals infected with Mycobacterium tuberculosis (MTB) progress to active disease. Despite being heritable, very few genetic variants have been associated with susceptibility to TB. For my studies, I use RNA sequencing (RNA-seq) to interrogate genome-wide transcript levels in in vitro cellular models. In Chapter 2, I use a joint Bayesian model to identify genes which are differentially expressed in macrophages only after infection with MTB and related mycobacteria, but not other bacterial pathogens. In Chapter 3, I build a support vector machine model to classify individuals as susceptible or resistant to TB based on the gene expression levels in their dendritic cells. In Chapter 4, I characterize the technical variation introduced by batch processing of single-cell RNA-seq (scRNA-seq) and propose an effective study design that accounts for technical variation while minimizing replication. In addition to providing insight into the genes important for the innate immune response to MTB infection, my work is informative for the design and analysis of future functional genomics experiments. (Note: Supplementary tables are provided in a .zip file available online. Captions for the tables are provided within the dissertation.)