How to Read Your Dog's DNA Results

The headline of any result is the breed breakdown, a list of breeds with a percentage beside each, adding up to roughly one hundred. The single most useful thing to know is that not all of those numbers are equally reliable, and reading them well is mostly about weighting them correctly.

The large percentages are the trustworthy part. A dog reported as forty percent one breed and thirty percent another has clear, recent ancestry from those breeds, and you can take that to the bank. The small percentages are a different story. A three or five percent sliver is the least certain part of the whole report, often reflecting a very distant ancestor or simply genetic background noise, and it is better read as a hint than as a fact. If your dog comes back with a long tail of tiny percentages, do not over-interpret them. The story of your dog is in the big slices, not the crumbs. This is also why the same deeply-mixed dog can get slightly different trace breeds from different tests, a point we cover in how accurate dog DNA tests are.

Two labels tend to stop rescue owners in their tracks, because nobody explains them in advance. Both are real, honest answers, not glitches.

Supermutt, as Embark defines it, is the catch-all for distant ancestry that has been mixed so many generations back that the surviving DNA fragments are too small to pin to a single breed. Your dog's ancestors were once identifiable purebreds, but generations of mixing have worn the signal down to fragments, and the test gathers those together under one label so the percentages still total one hundred. A big Supermutt slice is not the test giving up; it is an accurate description of a very deeply mixed dog, which describes a great many rescues.

Village dog is different and often surprises people even more. As Embark explains, it refers to a distinct, free-breeding population of dogs that is not descended from modern breeds at all, with its own ancestral genetic signature. The free-roaming and street dogs of many regions belong to these populations, so a rescue imported from those parts of the world can legitimately read as part or mostly village dog. Again, this is a correct, meaningful result, not an error, and for an internationally-adopted rescue it is often the most honest answer there is.

Below the breed breakdown, most reports build out a few more panels, and some are more useful than others.

The family tree is the test's best reconstruction of your dog's recent ancestors, estimating what the parents and grandparents likely were. Treat it as a well-informed estimate rather than a certified pedigree, but it is a satisfying way to see how the breed mix came together. The trait predictions cover things like coat type, shedding, and colour, drawn from the genes behind those features; they are a fun confirmation of what you can already see, and occasionally explain a quirk of coat or colouring.

The genuinely practical panel is the size and weight prediction. For an adult dog it just confirms the obvious, but for a mystery-breed puppy it is the most valuable number in the whole report, because the genes controlling adult size are well understood and a DNA estimate beats guessing from paw size every time. If you have adopted a puppy of unknown parentage, this is the figure that tells you whether you are raising a lapdog or a couch-filler, and lets you plan for it. It pairs naturally with our guide to finding out what breed your rescue dog is.

The feature that turns a breed report into something people cry over is the relative finder. It matches your dog against every other dog in the test's database and shows the ones that share DNA, along with how much, which lets you identify genetic relatives ranging from close siblings to distant cousins. Embark's is the largest and most established, and it is the feature owners most often credit with genuinely surprising finds, from a rescue's littermate turning up across the country to, occasionally, a parent.

One point of confusion is worth clearing up: shared DNA between full siblings is not always exactly fifty percent. Siblings inherit different halves of each parent's DNA, so full littermates average around fifty percent but can land somewhat above or below, and they are still true siblings. Half-siblings, grandparents, and cousins show lower shared-DNA figures accordingly. If the relative finder is a big draw for you, it is one of the reasons to lean toward Embark, which we lay out in the Embark versus Wisdom Panel comparison.

Some links on this page are Amazon affiliate links — we may earn a small commission at no extra cost to you. It never changes which tests we recommend.

If you bought a breed-plus-health test, your report also includes a genetic health section, and it deserves a different mindset from the breed results. A health result is a screening flag, not a diagnosis: a marked condition means your dog carries a genetic variant associated with it, not that your dog has it or definitely will. The terms clear, carrier, and at risk trip people up constantly, and “at risk” in particular does not mean “affected.”

The single most useful health result to look for is MDR1, a drug-sensitivity variant that genuinely changes which medications are safe for your dog, which is worth flagging to your vet especially before any surgery. Beyond that, take any concerning result to your veterinarian to confirm and interpret rather than acting on the dashboard alone, and remember that a clean panel is not a guarantee of a healthy dog, since no test screens for everything. We will cover reading the health section in depth in a dedicated guide.