Genetic variation and domestication of horses revealed by 10 chromosome‐level genomes and whole‐genome resequencing

Abstract Understanding the genetic variations of the horse ( Equus caballus ) genome will improve breeding conservation and welfare. However, genetic variations in long segments, such as structural variants (SVs), remain understudied. We de novo assembled 10 chromosome‐level three‐dimensional horse genomes, each representing a distinct breed, and analysed horse SVs using a multi‐assembly approach. Our findings suggest that SVs with the accumulation of mammalian‐wide interspersed repeats related to long interspersed nuclear elements might be a horse‐specific mechanism to modulate genome‐wide gene regulatory networks. We found that olfactory receptors were commonly loss and accumulated deleterious mutations, but no purge of deleterious mutations occurred during horse domestication. We examined the potential effects of SVs on the spatial structure of chromatin via topologically associating domains (TADs). Breed‐specific TADs were significantly enriched by breed‐specific SVs. We identified 4199 unique breakpoint‐resolved novel insertions across all chromosomes that account for 2.84 Mb sequences missing from the reference genome. Several novel insertions might have potential functional consequences, as 519 appeared to reside within 449 gene bodies. These genes are primarily involved in pathogen recognition, innate immune responses and drug metabolism. Moreover, 37 diverse horses were resequenced. Combining this with public data, we analysed 97 horses through a comparative population genomics approach to identify the genetic basis underlying breed characteristics using Thoroughbreds as a case study. We provide new scientific evidence for horse domestication, an understanding of the genetic mechanism underlying the phenotypic evolution of horses, and a comprehensive genetic variation resource for further genetic studies of horses.