Genotype Imputation, Simply Explained (With Dog DNA)
Genotype imputation fills in the DNA positions a test didn't read directly, by comparing what it did read with a reference panel of dogs. Here's how it works, what a 2022 dog study found about its error rate, and its limits.
By PuppyGenes · How we research and write our guides
Genotype imputation is a way of filling in a dog's genotypes at DNA positions a test didn't read directly. It compares the stretches of DNA that were read with a reference panel of dogs whose genomes are well known, and uses the patterns the dog shares with them to estimate the missing positions. The result is a well-supported estimate rather than a direct reading.
The short version
- It fills gaps: imputation estimates genotypes at positions that weren't read, or were read too thinly to call.
- It works because DNA travels in blocks: neighbouring positions tend to be inherited together.
- It needs a reference panel of well-characterized dogs to compare with.
- It's good, not perfect: a 2022 study of dogs measured an error rate of 1.5% after quality filtering.
- It's weaker where the panel is thin, such as for rare variants or under-represented populations.
How imputation works
Neighbouring DNA positions tend to be inherited together in blocks called haplotypes, so dogs that share ancestry share long matching stretches of DNA. The 2005 dog genome study found long-range haplotypes across the entire dog genome, part of what makes imputation work well in dogs.
Imputation uses those shared stretches. If a dog's DNA matches a set of reference dogs at the positions that were read, the positions in between are likely to match too. It's a bit like reading a sentence with a few smudged letters: if you know the language well, you can fill in the gaps with confidence, though not certainty. Our guide to reference panels explains why the comparison dogs matter so much.
How well it works in dogs
A 2022 study tested imputation in dogs directly. The researchers took 97 high-coverage genomes from 51 breeds, thinned them to about one read per position to simulate low-pass sequencing, imputed the missing genotypes using a reference panel of 676 dogs, and compared the results with the full genomes. A quality filter that kept about 80% of the 14 million imputed positions lowered the error rate from 3.0% to 1.5%.
About seven million positions with common variants remained after filtering, with an average imputation quality score of 0.95 on a scale where 1 is best.
Where imputation is weaker
Imputation can only fill in patterns its reference panel contains. It works best for common variants and for dogs whose ancestry is well represented in the panel, and less well for rare variants or populations the panel barely includes. That's why it matters which dogs a test compares with, and why a result inferred this way isn't the same as reading every position directly.
What this means for a DNA report
The PuppyGenes report combines sequencing evidence with reference-based predictions. Sequencing evidence is combined with canine reference data to estimate ancestry, included traits and genetic diversity, and an inferred result is not a definitive direct reading of every genetic position. Our science page explains the method, and our guide to reading a dog DNA report explains how confidence is shown.
Frequently asked questions
- What is genotype imputation?
- It's a way of estimating a dog's genotypes at DNA positions that weren't read directly, by comparing the DNA that was read with a reference panel of well-characterized dogs.
- Is imputed DNA data accurate?
- It can be very good but isn't perfect. In a 2022 study of dogs, quality filtering lowered the imputation error rate from 3.0% to 1.5%. How well it works depends on the reference panel and on how common a variant is.
- Why do DNA tests use imputation?
- Reading every position directly, many times over, costs more. Imputation lets a test estimate far more positions than it reads, using what's known from its reference panel.
- What is a reference panel in imputation?
- It's the set of well-characterized genomes that imputation compares a sample with. The more relevant the panel is to a dog's ancestry, the better it can fill in that dog's DNA.
From general to personal
See how this applies to your dog specifically.
Our at-home Complete Dog DNA Test adds your dog's own results to these tips across six report sections: breed ancestry, coat colour, physical appearance, temperament and size predictions, genetic diversity and oral microbiome. The report does not include genetic disease or health screening.