1Pathology, Brigham and Women's Hospital, 2Medical School, Harvard, 3Perelman School of Medicine, University of Pennsylvania, 4Department of Biology, Boston College, 5School of Human Evolution and Social Change, Arizona State University, 6Department of Human Genetics, University of Chicago
Friday 1:30-1:45, 200ABC
Copy number variants (CNVs, defined as polymorphic losses and gains of segments of genomic DNA) constitute a greater portion of human genomic variation as compared to single nucleotide variants, and play an important role in human disease and evolution. However, little is known about the impact of CNVs on nonhuman primate genomes and evolution. Therefore, we examined both within species CNVs and copy number differences (CNDs, changes in copy number of genomic segments between species) across the genomes of multiple human, chimpanzee, and rhesus macaque samples. In doing so, we delineated mechanisms in which such copy number changes can affect eventual phenotypes. Specifically, we present more than 2,000 polymorphic human CNVs that overlap with orthologous chimpanzee or rhesus macaque CNVs, confirming the presence of CNV formation hotspots in primates. Many of these hotspots are functionally relevant, with a bias toward genes involved in immune function. The genes in these primate CNV formation hotspots have significant differential expression levels between species (p<0.001) and show evidence for positive selection. We also identified 964 CNDs of conserved sequences across these primate species and demonstrated that samples with copy number different genes had significantly different expression than samples with neutral copy number (p<0.001). We postulate that cross-species CNV hotspots harbor signatures of ongoing, recent evolutionary pressures, such as response to pathogens, whereas CNDs reflect fine-tuning of developmental pathways by altering the levels of RNA. Overall, our results establish genomic copy number variation within and between species as a major driver of primate evolution.
This work was supported in part by National Institutes of Health Grants R01GM081533 and P41HG004221