How should we interpret the results of racial identification tests?

Before discussing the interpretation, it is essential to understand how the analysis is performed. Much of our work focuses on issues related to genetic variation. Using computational tools, we seek to understand the genetic structure of dog populations, identifying patterns of genetic variation within each breed and between them. Through thousands of specific points in the DNA that vary between individuals (also called SNPs), it is possible to group dogs that share similar genetic characteristics, indicating a closer genetic relationship between them.

This makes it possible to understand the ancestry of dogs, identify specific breeds or lineages, and even investigate issues such as genetic diversity and population structure.

How do we interpret this diversity in canine DNA?

When discussing the diversity of canine DNA, it's interesting to remember how purebred dogs exhibit a homogeneous phenotype. This means that all dogs of a given breed share similar patterns, whether in terms of size, coat, head conformation, ears, and behavior. Just as the characteristics (these phenotypes) are visible to the naked eye, we can observe similarities in the DNA of these dogs. This similarity, in many cases, can reach 100%, but it is common for some variation to exist within a breed.

The issue of population structure refers, for example, to different lineages of the same breed, such as dogs of different origins (Europe or USA) of a given breed. In this case, the Brazilian lineage may present small genetic variations in locally bred dogs, resulting in a detection of DNA compatibility that differs from 100%. This means that, although it is a purebred dog, in this example, it may only have 95% compatibility with the DNA present in the database that was formed by dogs from other populations. Therefore, it is crucial to have purebred dogs born in Brazil to capture this local variability and make the tests more accurate. To this end, we at petgenoma are continuously expanding our DNA database of Brazilian purebred dogs.

When it comes to mixed-breed dogs, they exhibit a naturally diverse genetic mosaic, a result of natural selection. In the results for these dogs, it's possible to observe percentages of admixture that indicate approximately in which generation this admixture occurred. For example, if a tested dog has 50% purebred ancestry, this means that one of the dog's parents was of a breed present in the database. If it has 25% purebred ancestry, it means that one of the grandparents was of that breed. When the percentage is approximately 12% of another breed, it indicates that one of the great-grandparents was of that breed. If the percentage is less than 10%, the admixture occurred more than three generations ago, making it imprecise to determine the exact breed admixture.

An important aspect to consider is the accuracy of the mixture reported in the results. The lower the percentage, the lower the accuracy. When trying to capture small variations in DNA variability within the same species (dogs), it can be challenging when dealing with percentages below 10%. Therefore, in pursuit of more accurate results, it is recommended to focus on mixture rates above 10%.

To conclude the interpretation of breed identification results, it is important to highlight that these results are highly dependent on the laboratory's genetic database and how that database is structured. This is because your dog's DNA will be compared to (similar) DNA present in that database, and the breed from the database that has a similarity will be in the breed identification result. Therefore, comparisons between results from different laboratories, especially when dealing with greater genetic diversity (as in the case of mixed-breed dogs), may vary due to the breeds and origins of the dogs present in each laboratory's database.