| dc.description.abstract | Interpreting the consequences of the millions of genetic variants present in each human genome requires an understanding of human evolutionary history coupled with mechanistic knowledge of genome function. Given the emerging importance of chromatin folding to genome function and gene regulation, there is a critical need to quantify the role that three-dimensional (3D) genome architecture plays in evolution and disease. In this dissertation, I leverage techniques in human genetics, functional genomics, evolutionary biology, and machine learning to investigate the relationship between recent human evolution, 3D genome organization, gene regulation, and complex human disease. First, I quantify the contribution of archaic hominin (e.g., Neanderthal) ancestry to over 400 diverse traits. Synthesizing my results, I propose a model for using trait heritability and direction of effect to reveal how selection acted on different traits and how introgression may have facilitated adaptation to non-African environments. My work demonstrates that archaic ancestry influences certain traits in modern humans; however, elucidating the mechanisms through which variation contributes to traits remains difficult. To address this, I next consider the role of the 3D genome in common complex traits. Complementing previous work on rare disease, I find that common genetic variation in topologically associating domain (TAD) boundaries contributes more to complex-trait heritability, especially for immunologic, hematologic, and metabolic traits. Finally, synthesizing knowledge about 3D genome folding with outstanding questions about how human-Archaic genome differences contribute to species divergence, I apply deep learning models to reconstruct the 3D genome folding patterns of Neanderthals and Denisovans. Using the resulting chromatin contact maps, I demonstrate how differences in 3D genome folding between archaic and modern humans provide a putative molecular mechanism for some phenotypic differences. Additionally, contact map comparisons highlight that the 3D genome organization constrained sequence divergence and patterns of introgression in recent human evolution. In summary, I demonstrate that 3D genome folding provides functionally relevant context for studying links between genotype and phenotype, both within present-day humans and across hominins. This work establishes a foundation and proposes future directions to address previously unanswerable hypotheses about the role of the 3D genome in human evolution and disease. | |
