Pedigree-tracking improved population monitoring of recessive disease-causing alleles in pedigreed canine populations
This study demonstrates that utilizing probabilistic pedigree-based inference to integrate all available genetic and pedigree data provides a more comprehensive and accurate monitoring of recessive disease-causing allele frequencies in pedigreed dog populations compared to relying solely on direct genetic test results or hereditary status.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
Most dogs that are purebred belong to closed families, where the lineage of every puppy is carefully recorded in a registry. This practice helps preserve specific traits, but it also means that harmful genetic mutations can hide quietly within a breed, passed down from parent to child without anyone knowing. Many of these inherited diseases are recessive, meaning a dog only gets sick if it inherits two copies of the bad gene—one from its mother and one from its father. If a dog has just one copy, it remains healthy but acts as a carrier, silently passing the mutation to its puppies. For decades, breeders have relied on genetic tests to find these carriers and avoid mating them together. However, these tests are not perfect. They only tell the story of the dogs that were actually tested, leaving a vast number of untested dogs in the family tree with unknown genetic histories. This creates a blind spot: breeders might think a disease is disappearing because the tested dogs are healthy, while the bad gene is actually still hiding in the untested parts of the population.
A team of researchers at the University of Edinburgh set out to fix this blind spot by using the existing family trees to guess the genetic status of every dog, even those that were never tested. They focused on five different dog breeds registered with the Royal Kennel Club in the UK, looking at two specific eye diseases: one that causes the lens of the eye to slip out of place, and another that leads to blindness. Instead of just counting the results from the genetic tests that already existed, the team used a computer program to trace the flow of genes through the entire family history. They treated the family tree like a map, using the known results from tested dogs to estimate the likelihood of the bad gene being present in their untested relatives, parents, and ancestors. To make these guesses more accurate, they grouped the oldest dogs in the family trees—those with no recorded parents—into categories based on when they were born and where they came from, acknowledging that dogs from different eras or countries might have carried the gene at different rates.
The researchers found that looking at the whole family tree told a very different story than looking only at the tested dogs. In the years before genetic tests were widely available, the traditional method of counting only tested dogs suggested the bad genes were rare. However, when the team used their new method to estimate the status of the untested dogs, they found the bad genes were actually more common than previously thought. This difference became even more pronounced after genetic tests were introduced. While the old method showed a sharp drop in the bad gene frequency, the new method showed a much slower decline. The reason for this discrepancy is that breeders often mate healthy dogs with known carriers to avoid making sick puppies. Under the old system, the puppies from these matings were often ignored because they didn't have a test result, so their hidden carrier status was missed. The new method, by contrast, accounted for these puppies, revealing that the bad gene was still circulating in the population, just hidden inside healthy carriers.
To ensure their guesses were reliable, the team tested their method by hiding real test results from the computer and seeing if it could guess them correctly. The results were promising, with the method correctly identifying the genetic status of dogs most of the time, though it was slightly better at identifying healthy dogs than carriers or sick dogs. They also ran computer simulations to see how the method would perform under different scenarios, such as if breeders only tested dogs that were already sick or only tested the healthiest ones. These simulations showed that the method was robust, though it tended to slightly underestimate the bad gene frequency in the very old parts of the family trees where information was scarce. Crucially, the simulations confirmed that the real-world data from the kennel clubs was not as biased as the worst-case scenarios, suggesting the new estimates are a trustworthy reflection of reality.
The study concludes that relying solely on genetic test results gives an incomplete picture of a breed's health. By combining those test results with the full family history, breeders and kennel clubs can get a much clearer view of how common a disease-causing gene really is. This approach does not suggest that the disease is out of control, but rather that it is more persistent than previously believed, hiding in plain sight within healthy carriers. The researchers emphasize that this method is best used for understanding the health of the entire breed population over time, rather than for diagnosing individual dogs, as the accuracy for predicting the status of a single untested dog is not yet high enough to replace a real test. Instead, this tool offers a way to monitor the overall genetic health of a breed, helping to guide responsible breeding decisions that can reduce the risk of disease while preserving the genetic diversity that keeps these dogs healthy and strong.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.