
The largest horse population genomics analysis ever by our team at Embryo Corp. Over 1500 genomes including our dataset of elite horse genomes that allowed us to find never before seen SNPs associated with performance Genomics is the most undervalued and underappreciated data in the world right now probably because it has lots of roadblocks when it comes to working with it. Its fairly expensive at $300-$500 per genome. It takes a good amount of computational power to process a genome in a reasonable time and size of genomic data is also huge, like 100GB per genome. Genomics is the foundation of understanding and building life though. Breeding without genetics is just guessing so its been a big part of what we do. We have been lucky enough to get exclusive contracts with some major breeders to get access to top horse genomes also, something that has never been done before. Understand that even the SNP analysis that we did is a population level statistic. Though I believe there is lots of room for using functional data on the individual level. We have developed our own way to look at individual horses by looking at how mutational variant affect the actual thermodynamics of proteins involved in performance. Maybe will show that some other time. We did an SNP analysis focusing on 156 performance related genes across breeds and recapitulated the Hill 2019 Myostatin (MSTN) mutation. Personal, I'm not a huge fan of this mutation. Its in a non-coding region that is suggested to be associated with a SINE element, maybe? While we don't know what exactly is going on with this mutation, it does seem to be associated with a high end haplotype for performance horses. The Hill analysis also suggested the non-coding MSTN C/C phenotype was associated with sprinting. Our elite dataset is mostly sprinting quarter horses and the dominant phenotype by far in these is C/C. Matching their hypothesis in Thoroughbreds. Don't you hate when you replicate a study you didn't believe? Besides MSTN, another interesting mutational variant we found enriched in elite horses was in COL5A2, a collagen fibril protein. First time it has been seen in literature. Honestly, this doesn't surprise me much as performance horses get worked hard they probably have some selection around joint/tendon/collagen and increased durability. Inbreeding percentages are much bigger in horse industries that have less permissive breeding practices, no surprise. Thoroughbred and Standbred horses are twice as inbred as Quarter Horses, a breed that allows artificial insemination, IVF and has a much more open studbook. If we look at the PCA plot we can also see this. Thoroughbreds have a tight genetic cluster while Quarter horses are one of the most diverse. If we look at the 2026 horses auctions for Quarter Horse we have 211 unique sires across 2,794 hips and in Thoroughbreds its 190 unique sires across 4,642 hips. About half the breeding diversity. Trying to relate inbreeding to performance is tough though. Outbred vigor is a legit thing but we have had a hard time quantifying it. In our quarter horse performance dataset, inbreeding as measured by pedigree alone without genetics indicates that performance drops at moderate to high inbreeding levels(same grandparent x great grandparent). However, there isn't much indication that current restrictive breeding practice are actually worse for performance in Thoroughbreds compared to Quarter horses when looking at change in performance over time. We even looked for accumulation of homozygous impactful variants because of inbreeding and couldn't find any indication this happens more. I think the question remains, how bad really is inbreeding? I would probably still ere on the outbred vigor side of breeding but there also seems to be data that points to some amount of inbreeding being good. Check out the data in the link below



















