Feng Gao, Alon Keinan
(Submitted on 22 Mar 2014)
Recent studies have shown that human populations have experienced a complex demographic history, including a recent epoch of rapid population growth that led to an excess in the proportion of rare genetic variants in humans today. This excess can impact the burden of private mutations for each individual, defined here as the proportion of heterozygous variants in each newly sequenced individual that are novel compared to another large sample of sequenced individuals. We calculated the burden of private mutations predicted by different demographic models, and compared with empirical estimates based on data from the NHLBI Exome Sequencing Project and data from the Neutral Regions (NR) dataset. We observed a significant excess in the proportion of private mutations in the empirical data compared with models of demographic history without a recent epoch of population growth. Incorporating recent growth into the model provides a much improved fit to empirical observations. This phenomenon becomes more marked for larger sample sizes. The proportion of private mutations is additionally increased by purifying selection, which differentially affect mutations of different functional annotations. These results have important implications to the design and analysis of sequencing-based association studies of complex human disease as they pertain to private and very rare variants.