The Simple Facts about Genetic Testing

Simply put, the 10% success rate of current industry evaluation methods alone is no match for the addition of genetic assessment. At ThoroughGen, we have been following and testing horses for over 3 years, and our data suggests that using genetics to complement traditional methods yields significantly greater results, correctly predicting more than 7 out of 10 horses that will have what it takes to achieve racing results. Whether you're a breeder, buyer or trainer, our safe, affordable testing is a must to help you stay ahead of the competition.

ThoroughGen Success

Ready to get started? Click here for ordering information, or click a link below for more details.


The ThoroughGen Performance Model

The model below is divided into two scales from 0 to 1: class rating and optimal distance. Since designing this model, we have tested over 1,500 horses (pre- and post-racing career), and the model (built in three stages) shows the reults of a significant sampling of those horses:

Thoroughbred model stage 1
Thoroughbred model stage 2
Thoroughbred model stage 3

This test appears to be very predictive. The closer to 0 on the scale are the true sprinters, the closer to 1 are the true route horses.


Horses that score above a 0.90 on the scale:
Slightly more than 85% of the horses that we've tested in this range have reached a peak Beyer speed of 90 or higher at some point during their two or three year old racing season. This is by far the most predictable group. One caveat is that the current crop of three year olds that we tested (as yearlings) with this model were not simply random samples. They were from the short list of our clients at the Saratoga and Keeneland yearling sales, thus they had already passed all their other tests (pedigree, conformation, vet etc). Our test does not assay for crooked legs or bad throats, just for the genetic background that seems to be common in good horses. The mean peak Beyer speed from this group of horses is 104.

Horses that score between 0.70 and 0.89:
These are still very decent horses. Of the horses we've tested to this score, 60% have reached a peak Beyer speed of 90. The mean peak Beyer speed of this group is 85. The main difference is the peak Beyer speed plus the percentage of stakes winning horses, which is lower in this group than in the top group.

Horses between 0.50 and 0.67:
About 40% of them reach the 90 in peak Beyer Speed. The mean peak Beyer speed for this group is 67.

Horses below 0.5:
These horses have the lowest % of individuals that reach a 90 Beyer speed. The mean peak Beyer speed from this group is 55. There are still horses scoring in this group that become elite racers, however the chances of finding one from this group is significantly lower than the other groups.
(back to top)

How the Model was Created

Association mapping is a powerful approach to identify genes affecting complex traits. It has dramatically improved our understanding of complicated human conditions such as Alzheimer’s disease, cardiovascular impairment and even traits as complex as behavior. The same approach can be used successfully in thoroughbred horses to identify genes that affect performance and therefore increase the power to pick winning horses. The logic behind association mapping was relatively simple; cluster individuals into ‘good’ and ‘bad’ performing groups, genotype all the individuals within those groups across the entire genome, and then test for allele (gene variant) frequency differences among the groups.

Two groups of thoroughbred horses with authenticated racing records were used to create the racing panels. We used 300 horses that were Grade I-winning colts and fillies with a minimum peak Beyer speed of 100+. These were compared to 350+ horses that were bred by the same level of stallion and broodmare, however these horses had peak Beyer speed of 79 or below, with no known injury or illness. By using multivariate analysis, the combinations of genes that are significantly associated with “elite performance” were identified to sites on 12 chromosomes. From this data we have generated a model that significantly increases the chances of identifying elite racehorses.
(back to top)

Broodmare Management

ThoroughGen customers have used this panel to decrease the number of broodmares on their farms to great success. For example, since using this strategy, a few of our clients who are primarily breed-to-race operations have decreased their overhead by 40% while increasing their success (purse $ + sales $) significantly.
(back to top)

Breeding Cloud

For decades, nicking has been the primary means to choose breeding pairs, however it is limited in the information it can provide a breeder because it is based on inferred genetic information primarily from the sire and broodmare sire. ThoroughGen’s innovative Breeding Cloud model gives breeders completely novel insights to Thoroughbred pair selection because the actual DNA sequence of each stallion and mare are known, and this information can be used to compile a “cloud” of every potential genetic offspring, including the probability of each outcome. The test takes into account the racing genotype of the mare and stallion, epistatic interactions between the mare and stallion, potential energy production of the mare and the score of reproductive potential of the stallion and the mare. This model represents the first genetics-based analysis of both the broodmare and sire in order to choose the mating pair with the best statistical chance of producing an elite racing prospect. For Breeding Cloud pricing and more information, please contact Dr. Steven Tammariello.
Phone: (607) 239-1366
(back to top)

Terms of Service | © ThoroughGen, LLC | ph: (607) 239-1366