Priya, Sambhawa2022-01-042022-01-042021-10https://hdl.handle.net/11299/225882University of Minnesota Ph.D. dissertation. October 2021. Major: Biomedical Informatics and Computational Biology. Advisor: Ran Blekhman. 1 computer file (PDF); x, 199 pages.While host genetics and gut microbiome have separately been identified as contributing factors to human health and disease, it is unclear how interactions between the two might drive disease risk. The modulation of host gene expression by the gut microbiome has been demonstrated as a potential mechanism by which microbes can affect host physiology. Therefore, understanding the molecular interactions between the microbiome and host gene regulation is critical for unravelling their contribution to the etiology of human diseases. Here, we comprehensively characterize functional interactions between the gut microbiome and host gene regulation across diverse human diseases to understand how these complex interactions might contribute to host pathophysiology. First, we characterized interactions between the gut mucosal microbiome and host gene expression in the colon of patients with cystic fibrosis to elucidate the potential role of host-microbiome interactions in the etiology of colorectal cancer in cystic fibrosis. Next, we developed a machine learning-based framework to jointly analyze host transcriptomic and microbiome profiles from colonic mucosal samples of patients with colorectal cancer, inflammatory bowel disease, and irritable bowel syndrome. We identified potential interactions between gut microbes and host genes that are disease-specific, as well as interactions that are shared across the three diseases, involving host genes and gut microbes previously implicated in gastrointestinal inflammation, gut barrier protection, energy metabolism, and tumorigenesis. We further adapted this integration framework to characterize multi-omic interactions between host gene expression, gut microbiome, and gut metabolome in irritable bowel syndrome. We also developed and applied supervised learning models to characterize patterns of host-microbiota interactions in diverse contexts to reveal microbial mediators of ethnic health disparities in the United States, and bacterial modulators of susceptibility to konzo in the Democratic Republic of the Congo. By identifying the host-microbiome interactions associated with human health and disease, results from our work can facilitate new insights into the molecular mechanisms by which microbiota impacts host health, and potentially lead to biomarkers for diagnostic and therapeutic interventions.engenomicshost gene expressionhuman diseasesmachine learningmicrobiomemulti-omicsThesis or Dissertation