Population Genomics Across Global Populations: Flexible Methods for Inference and Detecting Rare Recessive Variation

University Of Southern California, Los Angeles CA

Investigators

Abstract

Over the past two decades, publicly available large-scale genome-wide sequence data catalyzed our ability to learn more about the human genome, health, and evolutionary history. Large datasets and computational resources provide the genomics research community with valuable insights about disease and complex traits. Our lab aims to continue this forward trajectory by actively bridging the fields of evolution, genetics, and statistics by developing and distributing novel computational methods. Our work will allow researchers to accurately infer complex admixture and demographic histories across human populations and deepen our understanding of deleterious variation that contributes to complex traits and lethality. To infer the timing and magnitude of admixture within a population’s history, our lab will utilize shared genomic segments inherited identical-by-descent (IBD) within populations from publicly available data from human populations. Previous research shows that IBD can be used to infer the demographic history of a population. Similarly, the lengths of runs of homozygosity (ROH) reflect the underlying demography of populations. Although ROH and IBD contain valuable information about a population's history, no existing methods jointly utilize them to infer demography or admixture. We will develop new mathematical models that incorporate empirical data and simulations to assess how population history influences the distribution of IBD and ROH. Then, we will leverage ancestry switches within IBD segments, which are typically excluded from analyses, to model the timing and magnitude of admixture events. Long ROH are a result of recent consanguinity and often harbor deleterious variants. Due to their recent formation, variants within ROH have not yet been substantially influenced by genetic drift, recombination, or selection. Thus, we will utilize ROH to develop and test new variant weighting schemes aimed at detecting rare recessive variations linked to complex traits and diseases. Then, we will analyze data from human biobanks and breed dogs to identify genomic regions with a deficit of ROH. Inbreeding during breed development has resulted in ROH in most genomic regions that can tolerate high homozygosity without negative effects. By examining the cross-species deficit of ROH, we aim to identify regions associated with recessive lethal mutations critical for population viability across species.