What a DNA parentage test actually proves

A pedigree certificate names a sire and a dam. A DNA parentage test is the only part of the process that checks whether those names are true. Everything else in a pedigree — the generations behind the parents, the titles, the health results attached further up the tree — is inherited trust, built on the assumption that every parentage claim below it was accurate at the time it was made. Take away the test and a pedigree is simply what someone wrote down.

What the test actually does

A parentage test does not read a dog's or a foal's whole genome, and it does not look for disease. It compares a small, standard panel of DNA markers — short tandem repeats in dogs, the same basic method in horses — between the offspring and its claimed parents, and checks whether the offspring's markers could plausibly have come from that pairing. Each parent passes down one copy of each marker; if the offspring carries a version that neither claimed parent has, that parent is excluded. It's a process of elimination, not a reconstruction of ancestry, and it works the same way whether the question is a single disputed puppy or an entire multi-sire litter.

The American Kennel Club has run a DNA profiling programme since 1996 and moved to its current panel — thirteen STR markers plus a gender marker — in 2003. Every new profile submitted is checked against the AKC's full database of registered dogs and litters going back to January 2000, which lets the organisation confirm parentage to greater than 99% confidence. The database itself has grown large enough that the comparison is a meaningful check rather than a formality: it now holds close to a million profiles, with more than 40,000 new samples arriving every year. Even against that scale, fewer than 3% of samples need follow-up research before a parentage question is resolved — a small number, but not a zero one, which is the point. The test exists because the small number isn't zero.

The Kennel Club in the UK runs a comparable service, free of charge, built around the same cheek-swab sampling used for routine DNA health panels. It's aimed specifically at the cases where parentage is genuinely in question — multiple-sire matings above all — rather than at verifying every litter as a matter of course.

When it stops being optional

For most litters, a parentage test is something a breeder can choose to run. For a defined set of situations, the major registries don't leave it to choice. The AKC requires every sire to be DNA profiled before his fourth litter in a calendar year, or his seventh in a lifetime — the Frequently Used Sires rule, aimed at the small number of dogs siring disproportionate numbers of litters. Any litter with more than one possible sire requires every candidate sire, the dam, and every puppy to be profiled before registration, a rule that has applied since 1998 and today no longer carries a separate penalty fee for the extra testing. Stud dogs whose semen is collected and stored for later use must be profiled before that semen can be used at all, and any dog imported and registered since March 2006 needs a profile on file before a US-born litter from that dog can be registered.

Horse breeding runs the same logic through a different mechanism. Weatherbys has provided parentage testing for British and Irish thoroughbreds since 1985, moving from blood typing to DNA technology in 1998, and processing a DNA sample is now treated as a routine part of registering a foal rather than an optional extra. New Zealand Thoroughbred Racing sets out the clearest version of the rule: every foal must be branded, microchipped and parentage tested by 31 July of the season it was born in, using a mane hair sample compared against samples already on file for the sire and dam. If the dam's own DNA isn't already in the system — because she changed hands, or because the original sample was never taken — her current owner has to supply one before the foal can be registered at all. No sample, no foal on the stud book, regardless of what the covering certificate says happened in the paddock.

What it doesn't tell you

A parentage result that confirms the two named parents is good evidence that the people managing the mating recorded it accurately — nothing more. It says nothing about the health of either parent, nothing about the quality or soundness of the resulting puppy or foal, and nothing about the generations further back in the pedigree, which the test was never asked to check. A sire with a clean parentage result can still carry a recessive condition his owner has never tested for; see what a DNA health panel actually tests for for the test that actually answers that question, and what a coefficient of inbreeding actually tells you for what the rest of a verified pedigree can and can't tell a buyer once parentage itself isn't in doubt. On the foal side specifically, what a pedigree can't tell you about a foal covers the same gap from the other direction — a confirmed pedigree is a starting point for judging an individual animal, not a substitute for it.

It's also worth being precise about what "confirmed" means in practice. A parentage test checks the claimed parents against the offspring; it cannot tell a buyer whether the breeder chose those two animals for good reasons. Verification and judgement are different questions, and a marketplace that only checks the first one has done a smaller job than it sounds like.

What to actually ask

For a dog, ask whether the sire and dam are both DNA profiled with the relevant registry, and whether the litter itself has been profiled where the breeding situation calls for it — multiple sires, a frequently used sire, or imported stock. For a horse, ask whether the foal's parentage sample has been processed and matched against samples already on file for both parents, not merely collected and sent off. In both cases, a breeder running a legitimate programme will usually have the paperwork already, because the registries increasingly require it before they'll register the animal at all — which is a lower bar to clear than most buyers assume, and one worth checking rather than taking on trust.

---

Sources

• AKC DNA Program: Past, Present and Future
• Parentage Verification — The Kennel Club
• GB Thoroughbred DNA Samples 2020 — Weatherbys
• Foal Identification — New Zealand Thoroughbred Racing

— Rene