A dog DNA test is now something you can order online, swab at the kitchen table and read on your phone a few weeks later. The convenience hides an important question: how much can you trust what comes back? The honest answer is that it depends on what the test is measuring. Reading a known DNA variant is technically very reliable. Turning that reading into a statement about your dog's health or ancestry is where most of the uncertainty lies.
Key facts
- Genotyping a known variant is usually very accurate in a well-run laboratory; errors more often come from poor samples or mix-ups.
- A disease variant is a risk factor, not a diagnosis. Some variants almost always cause disease; others only rarely do, or only in certain breeds.
- In a 2018 study of more than 100,000 dogs, about two in five carried at least one copy of a tested disease variant (Donner et al., PLOS Genetics).
- Breed ancestry results are estimates that depend on the company's reference panel, and are least reliable for distant ancestors.
Three different kinds of test
"Dog DNA test" covers three quite different things, and their accuracy is not the same.
1. Health and trait variant tests
These look for specific known changes in the DNA, for example the PRCD variant behind the most common form of progressive retinal atrophy, the MDR1 drug-sensitivity mutation or the SOD1 variant associated with degenerative myelopathy. Modern panels test hundreds of variants at once on a genotyping array, while single-gene tests may use PCR or sequencing. For a variant that the laboratory has validated, the technical accuracy of reading "zero, one or two copies" is high.
2. Parentage and DNA profile tests
These compare a set of standard markers (microsatellites, or increasingly SNPs) between a puppy and its possible parents. They follow international standards set by the International Society for Animal Genetics (ISAG), and kennel clubs accept them for registration. Parentage exclusion is very reliable when all candidate parents are tested; the Finnish Kennel Club, for example, requires profiles from the puppy, the dam and every possible sire to verify a litter.
3. Breed ancestry tests
These compare thousands of markers across the genome with a reference database of purebred dogs and estimate which breeds contributed to a mixed-breed dog. Results for a recent purebred parent are usually solid. Beyond the grandparent generation, small percentages become more uncertain, results can differ between companies, and a breed that is not in the reference panel cannot be detected at all.
Why a positive result is not a diagnosis
The biggest source of confusion is the difference between carrying a variant and being ill. Geneticists use the word penetrance for the share of dogs with a given genotype that actually develop the disease. Some variants are close to fully penetrant: a dog with two copies will almost certainly be affected. Others are much weaker. Degenerative myelopathy is the standard example: many dogs with two copies of the SOD1 variant never show signs, and the disease is a diagnosis of exclusion that a veterinarian makes from clinical signs.
Context matters too. Many variants were discovered in one breed and then added to panels sold for all dogs. When Donner and colleagues analyzed more than a million dogs tested for 250 variants (PLOS Genetics, 2023), they found most variants in both mixed-breed and purebred dogs, but could show full penetrance for only 10 variants and plausible clinical significance for 22 others across diverse breed backgrounds. For many variants found outside their original breed, the clinical meaning is simply unknown.
In 2018, three researchers writing in Nature (Moses, Niemi and Karlsson) warned that consumer pet genetic testing was largely unregulated and that results were being used to make clinical decisions without enough evidence. Their case study was a 13-year-old pug whose positive result for a degenerative myelopathy-associated variant was mistaken for a diagnosis. Their recommendation, which veterinary geneticists widely share, is that a positive result should prompt a conversation with a veterinarian, not a decision on its own.
How to judge a test before you buy it
- Is the variant relevant to your dog's breed? Look for evidence that the variant causes disease in that breed, not just that it was found somewhere. The International Partnership for Dogs runs the Harmonization of Genetic Testing for Dogs (HGTD) database at dogwellnet.com, which lists tests, providers and breed relevance.
- Does the laboratory say how it validates tests? Quality standards, external proficiency testing and a named scientific team are good signs.
- Will your kennel club accept the result? For breeding decisions this matters. In Finland, the Kennel Club decides breed by breed which DNA results it records, and in some cases requires the sample to be taken by a veterinarian or approved sampler with identity checks.
- Is genetic counseling available? A laboratory that offers access to a geneticist or veterinarian to explain results is more useful than one that only sends a report.
How to take a cheek swab sample
Most home tests use buccal (cheek) swabs. Poor sampling is one of the main reasons a test fails or gives a confusing result, because the swab can pick up DNA from food, from another dog or from a puppy's mother. The instructions differ between kits, so always follow the ones in the box, but the principles are the same everywhere:
- Keep the mouth clean. Do not let the dog eat, drink, chew toys or lick other dogs for at least 15 minutes, and ideally longer, before sampling. Separate nursing puppies from the dam for a while first.
- Do not touch the brush. Hold the swab by the handle only.
- Swab firmly. Place the brush between the cheek and the gum and rub firmly against the inside of the cheek for about 15 seconds, rotating it. You need cells, not just saliva.
- Air-dry the swab. Let it dry for at least 15 minutes before putting it back in its package. A wet swab invites mold and bacterial growth.
- Label carefully. Mix-ups between dogs in the same household are a classic cause of "impossible" results.
Genoscoper's original sampling instructions describe the same steps. Blood samples, taken by a veterinarian into EDTA tubes, give more DNA and are required for some official tests.
Bottom line
For a validated variant in the right breed, a reputable laboratory's DNA test is a very reliable way to know whether a dog is clear, a carrier or at risk, and it is the best tool breeders have to avoid producing affected puppies. For everything else, read the result as information to discuss with a veterinarian. The 2016 study of canine breed disease heritage, summarized in the news archive, is a good illustration of how much there still is to learn.
Frequently asked questions
Can a dog DNA test be wrong?
Yes, though rarely for the genotyping itself. Most errors come from contaminated or mislabeled samples, from variants that have not been validated in the dog's breed, or from misreading what a result means. Repeating the test with a fresh sample, or confirming at a second laboratory, settles most doubts.
Are breed ancestry results reliable for mixed breeds?
The main breeds of a recent purebred parent are usually identified well. Small percentages and distant ancestors are less certain, and results can differ between companies with different reference databases.
My dog has one copy of a disease variant. Should I worry?
For recessive conditions, a single copy usually means the dog is a carrier and will not be affected. It matters mainly for breeding. For dominant or incompletely dominant variants, ask your veterinarian what the result means for your dog.
Do mixed-breed dogs carry fewer genetic diseases?
Not fewer variants. Donner et al. (2018) found mixed breeds were more likely to carry common recessive variants as carriers, while purebreds were more likely to have two copies. Mixed breeds can still develop these conditions.
Sources
- Donner J, et al. (2018). Frequency and distribution of 152 genetic disease variants in over 100,000 mixed breed and purebred dogs. PLOS Genetics 14(4):e1007361. doi:10.1371/journal.pgen.1007361.
- Donner J, et al. (2023). Genetic prevalence and clinical relevance of canine Mendelian disease variants in over one million dogs. PLOS Genetics 19(2):e1010651. doi:10.1371/journal.pgen.1010651.
- Moses L, Niemi S, Karlsson E (2018). Pet genomics medicine runs wild. Nature 559:470–472.
- Suomen Kennelliitto (Finnish Kennel Club): DNA-ohje (rules for DNA identification and test results).
- International Partnership for Dogs: Harmonization of Genetic Testing for Dogs (HGTD), dogwellnet.com.