View Program - 2022 Population, Evolutionary, and Quantitative Genetics Conference

Genomic time series offer us a chance to observe the interplay of various evolutionary forces, such datasets are increasingly common in both experimental evolution studies and ancient DNA datasets. The total variance in allele frequency change between two time points can be decomposed into the contributions of admixture, genetic drift, and linked selection. One promising use of such datasets is to obtain a genome-wide view of linked selection. In closed populations, the contribution of linked selection is identifiable because it creates positive covariances between time intervals, and genetic drift does not. However, natural populations are more complex than closed laboratory experiments and can be subject to another evolutionary force: migration. Indeed, repeated admixture between populations can also produce positive covariances between allele frequency changes in different time intervals. Here we lay out how to decompose the total variance in allele frequency change due to drift, linked selection, and admixture. We show how to accurately estimate the fraction of variance in allele frequency change due to these processes in the focal population experiencing gene flow from other source populations. Our approach should be widely applicable to the growing number of temporal population datasets. We apply these methods to ancient DNA datasets from Europe to characterize the contribution of these processes to allele frequency over thousands of years.

The contribution of admixture, selection, and genetic drift to allele frequency change in time series genomic data.