Genome-wide copy number variation detection in a large cohort of diverse horse breeds by whole-genome sequencing

Abstract

Understanding how genetic variants alter phenotypes is an essential aspect of genetic research. Copy number variations (CNVs), a type of prevalent genetic variation in the genome, have been the subject of extensive study for decades. Numerous CNVs have been identified and linked to specific phenotypes and diseases in horses. However, few studies utilizing whole-genome sequencing to detect CNVs in large horse populations have been conducted. Here, we performed whole-genome sequencing on a large cohort of 97 horses from 16 horse populations using Illumina Hiseq panels to detect common and breed-specific CNV regions (CNVRs) genome-wide. This is the largest number of breeds and individuals utilized in a whole genome sequencing-based horse CNV study, employing racing, sport, local, primitive, draft, and pony breeds from around the world. We identified 5,053 to 44,681 breed CNVRs in each of the 16 horse breeds, with median lengths ranging from 1.9 kb to 8 kb. Furthermore, using Vst statistics we analyzed the population differentiation of autosomal CNVRs in three diverse horse populations (Thoroughbred, Yakutian, and Przewalski’s horse). Functional annotations were performed on CNVR-overlapping genes and revealed that population-differentiated candidate genes (CTSL, RAB11FIP3, and CTIF) may be involved in selection and adaptation. Our pilot study has provided the horse genetic research community with a large and valuable CNVR dataset and has identified many potential horse breeding targets that require further validation and in-depth investigation.