What 12 public canine DNA test pages told us
Research note · Public webpage audit · 30 September 2026
When a breeder compares DNA tests, the public test page is often the first place they look. Can they tell exactly which variant the test detects? Does the page explain which breeds are relevant and what a clear result leaves open?
We examined 12 public canine test pages: four named conditions at each of three labs. This is original AnimalTrace research about the wording on those pages. It does not test the labs, their reports, or the clinical value of the tests. The sample is small and deliberately selected; its percentages should not be applied to other tests or labs.
How we did the audit
We selected pages for progressive rod-cone degeneration (PRA-prcd), degenerative myelopathy (DM; SOD1 exon 2 or common-variant page), exercise-induced collapse (EIC; DNM1 or retriever-type page), and von Willebrand disease type 1 (vWD1). On 30 September 2026, we read the standalone pages at UC Davis Veterinary Genetics Laboratory, Laboklin UK, and Orivet. All 12 selected pages were accessible.
Our protocol sets out four page-level questions:
- Does the page give a positioned DNA or protein change, coordinate, or rsID? A gene name or test shorthand alone does not qualify.
- Does it explicitly state an inheritance mode?
- Does it name a breed linked to the test or variant, or say the test is for all breeds?
- Does it explicitly say what a normal result does not rule out?
For the fourth question, we marked unclear when a page explained a normal result and mentioned another form or variant elsewhere, without connecting those statements. We coded the selected page itself, including its visible sections, rather than searching a lab’s entire site. A “no” describes this page and rule; it does not mean the information is unavailable from the lab.
The full data table records the URL, page title, date, each code, and the page section behind it. One person performed the coding. There was no independent veterinary genetics review, so these findings are descriptive and do not make clinical judgments.
What we found
| Page feature | Yes | Unclear | No |
|---|---|---|---|
| Positioned variant | 8/12 | 0/12 | 4/12 |
| Explicit inheritance mode | 12/12 | 0/12 | 0/12 |
| Breed relevance on the page | 11/12 | 1/12 | 0/12 |
| Explicit limit of a normal result | 1/12 | 5/12 | 6/12 |
The variant appeared as a positioned change on all four selected UC Davis pages and all four selected Orivet pages. The four selected Laboklin pages named the condition, gene, exon, or mutation without a positioned change on the page. We therefore could not match their exact molecular targets to another lab’s similarly named page from those page descriptions alone.
Every page stated an inheritance mode. That count measures the presence of a statement, not its correctness. For example, Orivet’s vWD1 page uses a recessive description in its product details and says the disorder has also been described as dominant in its prose. We recorded both descriptions and did not resolve the clinical interpretation.
Only the UC Davis PRA-prcd page directly links a normal result to what it cannot exclude: other PRA mutations. Five more pages discuss another form or variant apart from their normal-result text. Under our narrow rule, those were unclear, not an explicit warning about a normal result. The six remaining pages did not meet either category.
All twelve page-level results
Yes means the specified wording appeared on the selected page. Unclear means the page gave related context without meeting the rule above.
| Lab and test page | Variant | Inheritance | Breed | Clear-result limit |
|---|---|---|---|---|
| UC Davis · PRA-prcd | Yes | Yes | Yes | Yes |
| UC Davis · DM | Yes | Yes | Yes | Unclear |
| UC Davis · EIC | Yes | Yes | Yes | No |
| UC Davis · vWD1 | Yes | Yes | Yes | Unclear |
| Laboklin · PRA-prcd | No | Yes | Yes | Unclear |
| Laboklin · DM | No | Yes | Yes | Unclear |
| Laboklin · EIC | No | Yes | Yes | No |
| Laboklin · vWD1 | No | Yes | Yes | Unclear |
| Orivet · PRA-prcd | Yes | Yes | Yes | No |
| Orivet · DM | Yes | Yes | Yes | No |
| Orivet · EIC | Yes | Yes | Yes | No |
| Orivet · vWD1 | Yes | Yes | Unclear | No |
For Orivet’s vWD1 page, the prose names a breed in a discussion of presurgical screening, but the visible page does not say that breed is appropriate for its specific listed variant. Its “associated breeds” prompt did not show a list in the page text we reviewed, so breed relevance was unclear by our rule.
What this can and cannot tell us
These results show how much information a reader could extract from these particular pages on this date. They cannot tell whether one lab’s assay is more accurate, whether a test belongs in a breed’s health plan, or how to use a result for an individual dog. Linked references, customer reports, and direct lab support may provide information that the selected page does not.
We originally planned to sample Paw Print Genetics as the third lab. During page-availability screening, its home page announced a merger with Orivet and the former test pages we checked were no longer usable as current page evidence. We amended the protocol before formal coding and substituted Orivet. This makes the sample an availability-based pilot, not a representative survey. The page content may also change after the collection date.
For a breeder, the useful follow-up is a short list of questions: Which exact variant does this test cover? Is it established for my dog’s breed? What does a clear result leave untested? If the page does not answer, ask the lab and your veterinarian before using the result in a breeding or health decision. Our DNA panel guide explains how those answers fit with the rest of a breed’s health plan.