A method for detecting recent changes in contemporary effective population size from linkage disequilibrium at linked and unlinked loci

Abstract

Estimation of contemporary effective population size (N e) from linkage disequilibrium (LD) between unlinked pairs of genetic markers has become an important tool in the field of population and conservation genetics. If data pertaining to physical linkage or genomic position are available for genetic markers, estimates of recombination rate between loci can be combined with LD data to estimate contemporary N e at various times in the past. We extend the well-known, LD-based method of estimating contemporary N e to include linkage information and show via simulation that even relatively small, recent changes in N e can be detected reliably with a modest number of single-nucleotide polymorphism (SNP) loci. We explore several issues important to interpretation of the results and quantify the bias in estimates of contemporary N e associated with the assumption that all loci in a large SNP data set are unlinked. The approach is applied to an empirical data set of SNP genotypes from a population of a marine fish where a recent, temporary decline in N e is known to have occurred.