Evaluation of genetic differentiation and genome-wide selection signatures in Polish local sheep breeds

Since their domestication, sheep breeds have been a subject of artificial selection directed towards production of wool, meat and milk as well as natural selection associated with regional climates and nutritional conditions. These selection pressures created a large diversity of sheep breeds and the opportunity to study their genetic differentiation. Despite the existence of several studies on genetic variation and selection signatures in sheep, information on the genetic background and variability of local sheep populations, especially on a genome wide-level, is still incomplete. This is mostly due to the large number of local sheep breeds and populations in existence. Several of these populations are under conservative breeding because of their largely primitive character, which is not suitable for the modern intensive animal production. In this study we analysed population parameters and quantified genetic differentiation among seven Polish conserved sheep breeds or local sheep populations, namely: Black-headed (BH; n=104), Polish Merino of coloured variety (MPC; n=59), 'Old type' Merino (MPOT; n=50), Polish Mountain Sheep (PMS; n=103), Swiniarka (SW; n=100), Uhruska (UHR; n=69) and WrzosOwka (WRZ; n=90). The breeds represent mainly multipurpose production type with the emphasize on meat, wool and skin. Based on the data from OvineSNP50 BeadChip array (Illumina), we found that the genetic structures of the populations are not heavily affected, despite the bottlenecks in their recent breeding history. The only exception seems to be the SW breed in which we found very low effective population size (30.7) and high level of genomic inbreeding (17% as shown by F-ROH). Average MAF across breeds was high and ranged from 0.285 in MPC to 0.297 in PMS. The average observed heterozygosity per breed ranged from 0.378 in SW to 0.398 in UHR. The admixture patterns analysis showed that the analysed breeds have distinct genetic profiles with clear signs of common origin and admixture, similar to other worldwide sheep populations. By using F-ST statistics we also identified loci of the genome with the strongest divergence between the analysed breeds, presumably responsible for the fixed genetic differences among them which were created by artificial selection. The most pronounced of these results was selection signal at the locus of RXFP2 gene which is responsible for horn size/type in sheep. This analysis also allowed for genomic localization of genetic differences between the analysed breeds, providing a basis for further studies aimed at identification of sources of genetic variation in these sheep breeds.