The Chihuahua is the good twist in this story. The St Bernard is the correction. A 2025 dog-genetics study found detectable wolf ancestry in many modern dogs, including tiny Chihuahuas. But the researchers’ museum reports no detectable wolf ancestry in the St Bernard examples—not the positive finding our earlier article described. The result is interesting enough without making it bigger than it is.
My takeaway: enjoy the ancestry story, but do not turn a small DNA signal into a personality diagnosis. A wolf-like face, a “wolfiness” score and recent wolf ancestry are three different things. Here is what the research establishes, where the earlier claims went wrong, and what I would ask before paying for a test.
Source review and correction: 19 October 2026. This is a research explainer, not our own genetic experiment, a test of commercial kits or veterinary advice. Photographs illustrate the topic; they do not establish the pictured animals’ DNA results.
The actual study—not an unnamed “surprising” discovery
The paper is A legacy of genetic entanglement with wolves shapes modern dogs, published in PNAS on 24 November 2025. Audrey T. Lin and colleagues investigated gene flow between dogs and wolves after domestication. The American Museum of Natural History’s account identifies the research team and explains the findings.
This distinction matters: dogs’ deep evolutionary descent from wolves is not the same question as later interbreeding. The paper examines evidence of that later genetic exchange. It does not announce that scientists have just discovered dogs are related to wolves.
The indexed paper abstract and results report detectable wolf ancestry in 64.1% of the sampled breed dogs, spanning 264 breeds, or 79.8% of the breeds analysed. All 280 village-dog genomes in the relevant analysis showed it. Those are sample findings, not a claim that every dog everywhere carries the same detectable segments.
The individual-dog percentage and breed percentage have different denominators. A breed can count as represented when a sampled member has a signal; that does not mean every member has it. I would keep that distinction beside any headline saying “most breeds” or “most dogs”.
Which breeds had wolf ancestry—and which did not?
The museum’s summary gives the following examples. These are results and ranges reported for the study, not a substitute for testing an individual pet:
| Breed or group | Reported finding | How I would read it |
|---|---|---|
| Chihuahua | About 0.2% wolf ancestry. | Small size is not a reason to rule out a detectable historical signal. |
| St Bernard, bullmastiff and Neapolitan mastiff | No detectable wolf ancestry in the museum’s account of the study. | A large dog is not automatically a positive example; this corrects our earlier St Bernard claim. |
| Great Anglo-French tricolour hound | About 4.7–5.7%. | The reported range is more useful than an invented breed-wide single number. |
| Shiloh shepherd | About 2.7%. | A study example, not an estimate for every shepherd-looking rescue dog. |
| Czechoslovakian and Saarloos wolfdogs | About 23–40%; these were deliberately developed through wolf–dog crossing. | Keep these intentional crosses separate from the much smaller signals in many ordinary breed dogs. |
I would not guess the answer from ears, coat or size. The Chihuahua and St Bernard comparison is a better illustration of the study than a dramatic photograph of a howling animal. Nor does “not detected” erase dogs’ ancient origins: it describes what this analysis identified.
Why the Chihuahua is interesting without the exaggeration
At 0.2%, the reported Chihuahua finding is small, not evidence of a recent wolf parent. The interesting point is the mismatch with an appearance-based assumption. It gives us a reason to look at the data rather than decide that a lap dog cannot belong in the story.
Why the St Bernard passage needed correcting
The earlier article described wolf markers in St Bernards as an important discovery. That reversed the source’s example. The museum uses the breed to show that some large guardian dogs do not have detectable wolf ancestry in the analysis. We have corrected the text rather than preserve an appealing but wrong contrast.
How researchers look for genetic exchange
A SNP is a variation at one DNA position. The paper’s local-ancestry analysis uses patterns across genomic regions, comparing reference dog and wolf populations to infer ancestry along chromosomes. This is a statistical method—not a search for one uniquely “wild” gene that labels an animal.
Mitochondrial DNA and Y-chromosome lineages provide different views of ancestry from the nuclear genome. Following one maternal or paternal line is not equivalent to estimating ancestry across the genome. I would ask which question a test answers before comparing its result with this study.
The researchers report interbreeding from many generations ago on average. That estimate does not date each pet’s last wolf ancestor or tell you how many generations a percentage on a consumer report represents. An inferred segment also depends on the reference samples and method; a missing signal is not proof that no historical exchange ever occurred.
A commercial “wolfiness” score is not recent wolf ancestry
This is the part I would check first on an owner’s report. Embark explains that its wolfiness score uses markers in candidate domestication regions. It explicitly says the score does not represent recent wolf ancestry; ancestry of that kind is reported separately in breed results.
That is Embark’s explanation of its own product, not our independent validation of its accuracy. It is nevertheless enough to reject a common misreading: a wolfiness score cannot simply be relabelled as a percentage of recent wolf ancestry. Do not compare it directly with the research percentages in the table.
| What you are looking at | Question it addresses | What not to infer |
|---|---|---|
| Research estimate of introgressed ancestry | Which genomic regions the study infers came from later gene flow. | An exact recent family tree or an individual behavioural assessment. |
| Embark wolfiness score | Markers in its selected candidate domestication regions. | That the score is recent wolf ancestry or a danger rating. |
| A provider’s breed-ancestry result | What its reference panel and model infer about the tested animal. | That another company or research method measures the same thing identically. |
| A photograph or breed label | Appearance or reported identity. | A measured DNA percentage. |
There is no universal “trace” percentage rule here
Our earlier version grouped less than 1–5% as distant ancestry and 5–20% as more recent, then treated a “true wolfdog” as needing a higher percentage. The cited research and provider explanation do not establish those universal cut-offs. They have been removed.
I would read the provider’s definition, confidence and limitations instead. A result is not a reason to reverse-engineer a precise family tree from a percentage alone. If a report raises a concern, ask the laboratory to explain exactly what it measured before making decisions about the animal.
Does wolf DNA explain temperament?
I would not use it to diagnose why a dog howls, chases or dislikes strangers. The museum describes associations with kennel-club breed descriptors, but warns that those descriptions are biased and that direct genetic causation is not established. That is not an experiment showing that a particular DNA percentage causes a particular pet’s behaviour.
A separate 2022 Science study, Ancestry-inclusive dog genomics challenges popular breed stereotypes, surveyed owners of 18,385 dogs and genotyped 2,155. It found that breed explained about 9% of individual behavioural variation in its data. That is not “genes explain only 9%”: breed labels and genetics are not interchangeable.
The practical conclusion is narrower and better: assess the dog in front of you. Keep a record of the behaviour, its context and any changes, then discuss concerns with a veterinarian or suitably qualified behaviour professional. An ancestry headline is not an individual assessment, and a warm description such as “gentle giant” is not one either.
What about health and physical adaptations?
The study discusses possible functional consequences of gene flow, including enrichment of smell-related genes in village dogs. That is a research finding about genetic patterns, not a health report for your pet. It does not justify our earlier broad suggestion that trace ancestry has negligible health effects in general, nor an opposing claim that it predicts illness.
A health-screening result asks a different question from an ancestry result. If a commercial kit reports a disease-associated variant, take the actual report to a veterinarian for interpretation. I would not change diet, medication or exercise because a general ancestry article made a “wild genes” connection.
Before buying a dog DNA test in the UAE
I would start with the question I wanted answered. Breed curiosity, a possible inherited condition and a concern about recent ancestry are not one purchase brief. Ask the provider the following before paying:
- Measurement: does the product report breed ancestry, a wolfiness score, health variants or all three? Ask for a sample report and definitions.
- Method: which reference populations and confidence measures support the result? Does the company explain small or unresolved percentages?
- Interpretation: who can explain a surprising result, and does that support cover your particular question?
- UAE logistics: can the company deliver the kit and accept the returned sample from your location? Confirm the instructions and full cost directly; we have not tested a UAE purchase or return route.
- Privacy: read the sample-storage, research-consent, data-sharing and deletion terms before sending DNA.
- Health follow-up: ask your veterinarian whether the chosen health screen addresses the concern. More reported traits do not automatically make it the right test.
This is a buying checklist, not an endorsement of a laboratory or proof that the largest database is always the most accurate. I would want transparent limitations and useful support, not simply the most dramatic ancestry badge.
My verdict: a better ancestry story, not a new label for your pet
The research is worth reading because it complicates a tidy dog-versus-wolf story. It does not need promises that every wagging tail hides a detectable wolf segment, that a small percentage explains temperament, or that a consumer score is a recent family tree.
Enjoy the finding; read the result correctly. That is my recommendation. Keep population research, a laboratory report and your own dog’s needs separate. For practical care rather than ancestry curiosity, our UAE pet-insurance guide and Dubai dog-hydrotherapy guide cover different decisions; neither replaces veterinary assessment.
Sources and review scope
The correction uses the paper’s indexed abstract and results, the research team’s museum account and Embark’s own score definition, with the separate 2022 behavioural study for context. Full paper access was intermittent and later presented a browser challenge; this is not a reproduction of its analysis or a review of every supplementary breed result. We have not sampled dogs, interviewed the authors or independently compared testing laboratories.
