We adopted a dog at the age of two. Our vet said it was a mix of an Australian Shepherd and Border Collie, which we've told everyone. Based on markings, I think it's likely. I am wondering, however, if there are any tests or other ways to know the mix of a dog if you don't know the specific parentage.

Does each dog breed have specific genetic markers that can be used to identify the breed? If so, has a dog breed test been developed or is there no commercial demand for such a test?


There are many commercially available mixed breed ID tests, as a quick Google search will tell you. However, I cannot vouch for their accuracy: this news article may be of interest. Also this.

The genetic markers used to identify breeds in a mixed-breed are microsatellite markers (sometimes SNPs): you can read more here, but from the abstract:

We used molecular markers to study genetic relationships in a diverse collection of 85 domestic dog breeds. Differences among breeds accounted for ∼30% of genetic variation. Microsatellite genotypes were used to correctly assign 99% of individual dogs to breeds.

If you're interested in dog genetics, I encourage you to check out Elaine Ostrander's work. She's written at least one book for laypeople, in addition to her numerous journal publications.

  • $\begingroup$ I'm not certain of the accuracy of these tests because dog breeds aren't commonly genotyped. I'm not certain if there is a central database of dog breed genotyping, or if there is, whether they have significant sample size to accurately assign a breed. $\endgroup$
    – user560
    Jun 10 '13 at 22:55
  • 1
    $\begingroup$ @leonardo: The paper managed to separate breeds by genotyping only 5 unrelated dogs of each breed, but yes, I agree that the commercial tests are at a disadvantage because of the lack of a common repository (also they might be using SNPs instead of microsatellites) -- this is probably a significant source of the variation between tests. $\endgroup$
    – blep
    Jun 11 '13 at 17:57
  • 1
    $\begingroup$ Another point that follows from the point made by @leonardo: if in the Parker et al paper, if breeds were separated on the basis of groups of microsatellites rather than microsatellites that were unique to each breed, you'd need a lot more markers to identify the heritage of a MBD with any reasonable certainty, so that you'd have clusters of linked markers (the idea being that each cluster was inherited from a single purebred dog). $\endgroup$
    – blep
    Jun 12 '13 at 18:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.