Genomic research report that a catalog of genetic variants among Thoroughbred horses could one day help researchers and breeders use scientific precision to manage inbreeding while improving health and performance.

Currently composed of DNA data on 185 horses selected from a repository of 1,174 North American Thoroughbreds, the growing public database showcases different types of genes and their variants (known as alleles) among those horses. While it’s still too early to use the tool to determine how inbred the population of Thoroughbreds might be, the catalog can inform researchers about inbreeding at an individual level, said Jessica Petersen, PhD, associate professor of animal functional genomics in the Department of Animal Science at the University of Nebraska-Lincoln.

Using Genomics to Inform Horse Breeding Decisions

“We should be using genomics … to inform our management and breeding decisions, more so than just looking at broad measures of inbreeding overall,” said Petersen, during her presentation at the 2024 International Havemeyer Foundation Horse Genome Workshop, held in Caen, France, in May.

The project is a collaboration among scientists from the University of Kentucky, University of Nebraska-Lincoln, University of California-Davis, and University of Minnesota to characterize the diversity in the American Thoroughbred horse population. As they progress, the researchers plan to use the catalog to create computer models that predict inbreeding risks (such as health problems or conformation flaws) in specific mare/stallion choices, Petersen said.

The Thoroughbred breed has faced scrutiny in recent years due to concerns about inbreeding, because breeders often choose popular stallions and bloodlines to improve performance, Petersen explained. Recently, researchers suggested inbreeding could even contribute to pregnancy loss in Thoroughbreds. However, pedigrees don’t give a full picture of what’s going on in the genome, and previous research is often based on undisclosed genomic data, she said.

Genetic Variation Among Thoroughbred Horses

To make Thoroughbreds’ genetic variant situation transparent to the public, and to harness that information to monitor and protect the breed, Petersen and her fellow researchers developed this first-ever open Thoroughbred genome catalog. They obtained DNA samples from 1,091 Thoroughbreds residing in the U.S. and used samples from an additional 82 Thoroughbreds born between 1965 and 1986 that had been stored at the University of Kentucky since collection.

By comparing all these DNA samples, the researchers identified 14.5 million genetic variants, known as SNPs, among the horses. In other words, these are the genes that give horses their unique characteristics.

“Any one horse had about 4 to 5 million different variants compared to another,” Petersen said, adding that the results corroborate findings in a smaller Japanese study from 2021. “So the take-home message is that there is a lot of variation.”

Creating a Genome Catalog for Thoroughbred Horses

The combined data allowed the researchers to create a map of the Thoroughbred genome to diversity types, said Ted Kalbfleisch, PhD, professor at the University of Kentucky’s Gluck Equine Research Center, in Lexington, who collaborated on the project.

Then the team investigated each horse’s level of inbreeding based on a scale of 0 to 1 as a reference to homozygosity, or the number of alleles horses receive from each parent that are identical. Higher homozygosity suggests more inbreeding.

The homozygosity levels of the Thoroughbreds in their catalog ranged from 0.143 to 0.350, with an average of 0.262, Petersen said. Those born between 1965 and 1986 had lower inbreeding scores than those born after 2000.

“There are hints that there’s less diversity as you get more contemporary in these horses,” Petersen said at the workshop. “We do expect this if the breeders are doing a good job. This is a closed population under selection; breeders are getting rid of variants that aren’t contributing to their objectives.

Importantly, though, these younger generations didn’t have more longruns of homozygosity—meaning several homozygous areas of the genome in a row, Petersen explained. Short runs of homozygosity do not suggest significant recent inbreeding the way long runs do, meaning these horses might not have the kind of inbreeding that is disadvantageous, she said. The rate of inbreeding in the contemporary population is in line with the population from decades ago.

“Horses can have identical inbreeding coefficients, but they don’t necessarily share the same regions of inbreeding in their genome—and this is just an illustration of that,” Petersen said. “I believe the statistics show that there isn’t significant, intense, recent inbreeding in this contemporary population.” Including more horses into the catalog will shed more light on the question, she added.

Finally, the researchers compared their measures of homozygosity to the pedigree-based measures of inbreeding for the horses in their catalog. They found that the pedigree had only a moderate correlation with genomic findings about inbreeding. This is not surprising, she said, because pedigrees do not include inbreeding that occurred before establishment of the registry. Furthermore, pedigree predictions of inbreeding are based on random expectations for inheritance of genes and do not account for selection. 

Take-Home Message

Modeling breeding scenarios using the catalog should help estimate specific inbreeding risks by focusing on precise areas of the genome when selecting mares and stallions, Petersen explained. Ongoing work will also help determine inbreeding levels of the contemporary Thoroughbred population. “This is Step 1 of several steps that are going to allow us to do that, and in a very robust way,” she said.