Recent and past effective population size in different horse breeds

Estimates of effective population size (Ne) are important in species conservation biology and in animal husbandry. This study evaluated the effective population size (Ne) from genome-wide linkage disequilibrium in three different horse breeds: Thoroughbred (n=17), Turkemen (n=11) and Persian Arabian (n=71). We used a recently developed method that estimates Ne from genome-wide linkage disequilibrium (LD) [SNeP software, Barbato et al., 2015]. The samples were collected from USA (Thoroughbred horses) and Iran (Turkemen and Persian Arabians) and genotyped using the Equine 670k Affymetrix SNP chip. The relationship between LD and Ne was estimated using the Haldane recombination rate modifier under the assumption of a genome-wide linear relationship such that 1 cM ~1 Mb. Genome-wide average LD (r2) decreased with increasing genomic distance for all breeds. Persian Arabian had the lowest level of LD. The LD decay of Turkemen was lower than that of the Persian Arabian and Thoroughbred. We observed a pattern of decreasing Ne with estimated values of 54, 62 and 245 at one generation ago for Thoroughbred, Turkemen and Persian Arabian, respectively. This pattern is consistent with artificial selection by breeders with increasing usage of some specific bloodlines in more recent generations. This result can be used in analyzing population structure and designing horse breeding strategies.

Keywords:

Effective population size , linkage disequilibrium , SNP , horse.