Recently we have had questions from dog breeder health advisors about using inbreeding tools, particularly pedigree-based tools such as estimating Coefficients of Inbreeding (COI). This blog discussion describes COIs, some advantages and disadvantages, and provides thoughts on usage. 

• Pedigree-based COI estimations (Coefficient of Inbreeding)
• Genomic COI calculations

Often associated with COI:

• Popular sire tracking. Access to breed-wide data will vary by country, breed/kennel club. Some clubs restrict numbers of litters born to sires/dams to reduce the impact of popular sires. 
• Breed diversity reports/research. Some kennel clubs, breed clubs, researchers and commercial genetic test providers have undertaken breed reports on population differences, including genetic diversity and inbreeding, popular sires/dams, etc.  

There are also other genetic tools used to estimate genetic diversity, such as MHC/DLA testing, but for the purposes of this article, we are focusing on COI-based concepts. We hope to follow up with more information on all types of genetic testing-type diversity tests in another article. 

Pedigree-based COI

This tool uses pedigree information to estimate how closely related a dog's parents are. The more ancestors a dog's parents share, the more likely that duplicate genes will have been passed down, accumulating over generations (called identity by decent, or IBD) and thus reducing genetic diversity. Pedigree COI probability calculations are based on the dog's relatedness, not specific genes or regions of the genome. It does not attempt (nor claim) to be able to tell you which parts of the genome a dog has randomly inherited from the dam or sire, but instead focuses on how many relatives are held in common throughout the dog's known ancestry. All dogs in the same litter will have the same COI probability. Pedigree-based COI probabilities can not account for any influence of recombination or specific genes/sections of genetic code. 

Advantages:

There are some advantages to pedigree-based COI. There are a number of kennel and breed clubs who have invested in providing pedigree COIs to members, allowing free or low-cost access. There are also many free tools available for breeders to use, particularly when they have a breed or type that is not registered with a kennel club. If the COI is based on a large number of generations, ideally as many generations as known, and good-quality pedigrees, it can be informative and helpful in selecting dogs for breeding that are less closely related. It can also be used to look at changes over time in a fairly simple way that doesn't require special technology or access to the dogs. It may also have the advantage of being based on something familiar to breeders - pedigrees - and therefore may overcome some of the education challenges other genetic diversity tools may face. Pedigree-based COI calculations have been used effectively to impact genetic diversity across a wide range of animal breeding practices. 

Disadvantages:

The big challenge is confidence in the accuracy and quantity of the pedigree data used. If based on a small number of generations, or on inaccurate pedigree information, the COI can be wildly inaccurate in measuring relatedness in a meaningful way. There have been numerous studies that have shown that restricting the number of generations included in the calculations can be very misleading. Examples have shown a COI based on 5-generations, vs 10+ generations can give a false impression that the COI is much lower, and not reflect the accumulation of inbreeding/relatives in common over time. There can also be the temptation to have generational calculation cut-offs to either reduce the time required for COI calculations or to compare different breeds to each other. This leaves COI vulnerable to being used to support practices and strategies that may agree with biased agendas, but do not improve dog health. 

To provide good-quality COIs requires large amounts of pedigree data, time to analyze this data, and some expertise. Without this, there is a very high risk that decisions could be made on poor information. 

Can we calculate better? Genomic COI:

Genomic COI works off the same concept of IDB as pedigree COI, but does this by comparing tens of thousands of markers across the dog's genome (or, in theory, could use a dog's whole genome). It aims to capture specific swathes of homozygous markers of the genome inherited over generations, rather than estimate the probability that duplicate parts of the genome have been inherited. In this way, it should be more reflective of what precisely has been inherited in common, or what is IDB, over all of the generations of a breed's development, and not be restricted to a specific number of generations. 

Advantages:

Genomic COI is more precise, and if based on a large volume of genomic data, has a high standard of accuracy. It can also be more precise in reflecting an individual dog's level of inbreeding compared to a defined breed population. There is also the potential for much more precise estimations of inbreeding risk or improving diversity in matings. This could be especially helpful in choosing between two potential dogs who are otherwise equal in breeding qualities. 

Disadvantages:

There are costs for testing (usually around the $100-200 range, but often with other genetic tests included), and you need to have access to the dogs to provide genetic samples. There may be challenges (e.g. costs, compatibility) for integration into kennel club databases or other resources trying to collate health, pedigree, and genetic diversity information. Like all genetic diversity genetic tests, identifying presence or absence of areas of homozygous genetic markers does not mean identifying what those markers may be coding for. The assumption with genetic diversity is that the more diversity that is in place, the less likely that undesirable traits will be passed down. However, there could be rare or less common markers for a reason, such as an association with a disease or other undesirable traits. 

What about other genetic diversity tests?

We hope to follow up this blog with more articles exploring different kinds of genetic diversity genetic testing tools. 

How could COI be useful?

There are many ways a COI % can be used. For the individual breeder, the COI can be one way to add in genetic diversity as a consideration in breeding plans for individual matings, or as a way to track breeding plans over time. For breed clubs, or kennel clubs, it can be particularly useful in observing breed-wide trends and changes. Many clubs have recommendations based on the current breed-average - usually that a breeder should aim to breed at or below the current breed average COI. This is not necessarily to breed the lowest COI possible, and certainly not to prioritize COI over the health or health risks of an individual dog.

For example, the Finnish Breed Club encourages breeders to integrate COI information into breeding plans based on the first 5 generations to avoid high COI% matings in the near-generations, as a way to reduce the likelihood of passing on deleterious genes that could impact the individual dog's health. The Swedish Kennel Club monitors 5-gen COI over time, to look at near-generation trends and changes. And, the Kennel Club (UK) provides the COI for the individual dog, and the breed average, based on all available generations included in the calculation. They also include the number of generations included in the calculation as this can impact how informative the COI may be (e.g. 3 gen is more dynamic and changeable than 22 gen). The recommendation for the Kennel Club is therefore to breed at or below the breed average. These examples illustrate the differences between focusing on the near-generations which could reflect "fast" inbreeding and also intentional inbreeding vs. COIs calculated with 10-20 generations more suitable for describing the situation and trends for the whole breed.

There is also the consideration of COI changes over time. It is usually more concerning if a breed sees a rapid, drastic increase in inbreeding, leading to a rapid loss of genetic diversity, which, in a closed breeding system, cannot be regained without outcrossing to a new breed or "unrelated" breed population. 

Rule of thumb: aim to breed for type, not relatedness...

Even without specific tools, breeders can start by considering their own practices when selecting mates. Options could include: avoiding popular sires and their close relatives who have been used extensively for breeding, exploring lines that have not used for breeding previously, seeking dogs from populations that are “new” to them (e.g. field/working vs show), aiming to seek dogs that have desirable traits but are not as likely to be closely related:

• Use available tools to help make breeding plans - including genomic COI or genetic diversity tests, pedigree COI if that's the best you can access, with as many generations as possible. 
• Use genetic trait/disease tests and clinical screening schemes, as well as knowledge of familial longevity, behaviour, etc.  to help select the healthiest representatives of the breed
• Use MORE dogs of both sexes. Most breeds use less than 10% of the available dogs for breeding... you're losing 90% of the possible genetic diversity! 
• Genetic diversity is an important factor of breeding for the Big Picture. It should be one consideration in breeding. 

There is much more to explore in genetic diversity tools and applications both for individual breeders' plans, and breed-wide strategies than can be discussed in this short blog. Look for more information to come!

More resources and reference publications:

Get a GRIHP! on several breeds offers a variety of international information on a breed, including genetic diversity reports. Find them here. 

Relatedness of Breeds

Genetic Diversity webinar excerpt

Genetic Diversity and the Big Picture

Janes M, Lewis TW, Ilska JJ, Woolliams JA. The usage of Mate Select, a web-based selection tool for pedigree dogs for promoting sustainable breeding. Canine Med Genet. 2020 Oct 19;7(1):14. doi: 10.1186/s40575-020-00094-8. PMID: 33372639; PMCID: PMC7574414.

Photo by Blue Bird from Pexels