What a genetic test can't tell you
A DNA result is usually the firmest evidence in a breeding decision. It is also narrower than it looks on the page. Knowing where a result stops is what keeps it from being asked to carry a claim it cannot support.
A test answers one question about one variant
A panel covers a list of variants, and a clear result is clear for that list. It says nothing about a variant that was never on the panel, and nothing about a condition with no test yet. Our guide on how to read a genetic test result covers the notation; the point here is the boundary around it.
Some tests follow a marker rather than the variant
A direct test detects the variant itself. A linked test detects a marker that sits near the variant and usually travels with it. Usually is the operative word: the association was established in particular breeds and lines, and it can be weaker outside them.
The International Partnership for Dogs’ Harmonization of Genetic Testing for Dogs gathers details from test providers on the tests they offer. Whether a test is direct or linked is a fair question to ask a lab before you buy.
A variant proven in one breed may not carry over
The same condition can be caused by different variants in different breeds, and a test validated in one population may have little evidence behind it in another. The harmonization initiative records breed-specific information on tests, such as research and validation, for this reason. Check that a test is established in your breed, not just available for your species.
An at-risk result is not always a sick animal
Degenerative myelopathy in dogs is the clearest example. It is inherited as an autosomal recessive with incomplete penetrance, so a dog with two copies of the SOD1 risk variant is at risk rather than certain to be affected. The Orthopedic Foundation for Animals notes that not every homozygous dog develops the disease, and that other variants may modify the risk. It is also late onset, so a healthy older dog is not proof the result was wrong.
Read a risk variant as a probability attached to that animal, not as a diagnosis.
Many traits that matter are not single-gene at all
Hip dysplasia is polygenic. No single variant explains more than a fraction of the variation, which is why the OFA’s hip databases are built on radiographic evaluation of the animal rather than a DNA result. Most performance, conformation, and temperament traits sit in the same category.
A panel that is silent on these traits is not telling you they are fine. It is not looking.
What testing does do well
- confirming identity and parentage;
- carrier status for well-characterised single-gene conditions;
- ruling out an affected outcome entirely when one parent is clear, which is what makes recessive-risk screening worth doing before every mating.
That is a large amount of certainty. It is worth protecting by not stretching it.
Using results honestly
- Record the exact test, not just the word “clear”.
- Treat a missing result as missing. A gap is not a pass.
- Keep phenotypic screening alongside DNA for the traits DNA does not cover.
- Revisit older results when a new test or new breed guidance appears.
Where AnimalTrace fits
AnimalTrace keeps each report on the animal’s record as the lab issued it, with the lab named and the test it covered, so what was checked stays visible next to what was not. Breeding checks (Client Pro) score a pairing for the conditions both animals have lab carrier results for, and say which those are. An untested condition stays untested, and never reads as clear.
Sources
- Orthopedic Foundation for Animals, Hip dysplasia
- Orthopedic Foundation for Animals, Degenerative myelopathy
- UC Davis Veterinary Genetics Laboratory, Degenerative myelopathy
- Canadian Kennel Club, Harmonization of Genetic Testing for Dogs
- Mäki and others, Companion Animal Health and Genetics (2025), Report from the 5th International Dog Health Workshop