Breeders want to produce puppies that are healthy, long-lived, and have the type and traits that define the phenotype and behavior of the breed. The main tool used to guide breeding decisions is the pedigree, which records the history of the dogs in the breed. This can be paired with health and trait information to select dogs for mating that are likely to produce the desired offspring.

This is how we have bred purebred dogs for more than 100 years. In the last decade, the development of DNA tests for specific genes has allowed breeders to "see" what they're doing, allowing selection of dogs that have specific genes for traits as well as those carrying deleterious mutations, taking some of the guesswork out breeding for phenotypic traits like coat characteristics (color, long or short, straight or curly, etc) or body size, as well as an ever-increasing list of mutations associated with diseases or disorders. 

Breeders are familiar with the standard 5 (or more) generation pedigree chart that shows ancestors of a specific dog. However, most breeders are not familiar with the treasure trove of information that can be extracted from a pedigree database. Here are some of the statistics about a breed that can be extracted from a pedigree database and which should be part of every breeder's toolbox for proper genetic management. These terms are "jargon" and likely unfamiliar to you - and they go by somewhat nerdy names - but fear not. Keep the definitions in front of you for reference, and look up definitions as necessary when you review the data for your breeds or others. 

To make the best possible breeding decisions, breeders should have these tools in their tool box. 

Effective number of founders (fe)
Most breeds go back to a small population of "founder dogs", those individuals that contributed to the allelic variation in the original population. The genes in these founders define the size of the original gene pool. However, some of this original diversity is lost over time due to genetic drift (random chance) and selection, so the current population's gene pool will be smaller.

To estimate the size of the current gene pool, we can compute the "effective number of founders" (fe). This is an estimate of the number of founders that would produce the current genetic diversity of the population if all contributed equally to subsequent generations. This is a measure of the fraction of the genes contributed by the founders that still remain in the population. Alternatively, it represents the fraction of the original genetic diversity of the breed that has been lost due to genetic drift or selection.

Founder genome equivalents (fg)
This is the number of hypothetical founder dogs that could produce the same genetic diversity of the current population if each founder contributed equally and there was no loss of founder alleles through genetic drift. 

Effective population size (Ne)
Effective population size is the size of an "ideal" population of animals that would have the same rate of inbreeding, or decrease in genetic diversity due to genetic drift, as the real population of interest. It is not the same as census size, which is the actual number of individuals in a population. Ne reflects the "genetic size" of the population in that it depends on the number of animals that are breeding, which will always be less than the census size of the population. The rate of inbreeding in a population increases as Ne get smaller, so it tells you about the "genetic behavior" of the population in terms of how fast inbreeding will increase in the future. For a sustainably breeding population, the value of Ne should be at least 100, and some recommend at least 500 (the rate of inbreeding will be slower with higher Ne).

Equivalent Complete generations (EqG)
This is an estimate of the completeness of the pedigree database you are using. Missing data will reduce EqG and result in underestimation of inbreeding calculated from a pedigree database. A more complete pedigree database will have a higher EqG. 

These five statistics computed from a pedigree database should be available for every breed as the basic tools that can be used to offer insight into the present genetic status of your breed. The are fundamental to genetic management plans to control inbreeding (and therefore genetic disease) as well as mating strategies to protect genetic diversity. Instead of flipping through piles of printed pedigrees trying to estimate inbreeding and relatedness of individual dogs, these statistics provide quantitative information that can help you make the best possible breeding decisions. 

 Here are some examples of these statistics for a variety of breeds (from Mabunda et al 2022).

A quick look at these data can tell you about the genetic status of each breed. For example, the Finnish Spitz has the most complete pedigree database of these breeds (EqG = 10.88), so the statistics for this breed will be more reliable  than for other breeds with lower EqG. The inbreeding coefficient (F) for the Finnish Spitz calculated from the pedigree data is 6.33%. As explained above, this is much lower than the inbreeding coefficient determined from DNA (F = 28.5; Bannasch et al 2021). The ratio of fe/fa (1.5) indicates a loss of some of the genetic diversity that was present in the founder population. The values of Ne range from 384 (Ca de Bestiar) to only 3.2 (Bouvier des Ardennes), and the value for Finnish Spitz is 75.53. 
​
​Ne can be increased by balancing the number of male and females used in breeding. Since fewer males are used than females in most breeds, the simple solution to low Ne is  increasing the number of males used in the breeding program.

The data for these breeds can give you an idea of the genetic health of each breed. Some have small founder populations (e.g., fe = 3; based on the pedigree data), and only two populations have gene pools greater than 100 (fe = 117 and 205.5, for Border Collies in Hungary and Australia, respectively). Some breeds have dangerously low Ne (Bouvier des Ardennes), some have experienced a significant bottleneck (Border collie, Hungary), high inbreeding (36.45% and 44.7%, for Czech Spotted dog and Bouvier des Ardennes, respectively.

This information about a breed can give you a "heads up" spot check of a breed's genetic health. Having a large population size can be very misleading, but beyond the number of dogs and maybe an underestimate of the inbreeding coefficient, breeders do not have the basic information about the genetics of the population that is summarized here.

These statistics are essential tools that should be in the breeder's toolbox. Of course, every breed should have a good pedigree database, and if yours doesn't already this should be motivation to remedy this. If you do have a good pedigree database, these statistics can be computed for your current breed population, and refreshed to reveal the consequences of breeding decisions on genetic health.

If your breed does not have a good pedigree database, urge your fellow breeders to work on this. The database should be global and should go back as far as possible. 

With a pedigree database in hand, I can generate the statistics I describe here to provide the first window into the current genetic status of your breed. If you are breeding for preservation and health, this information is essential.

REFERENCES
​
Bannasch et al 2021 The effect of inbreeding, body size and morphology on health in dog breeds. Canine Medicine and Genetics  8:12. https://doi.org/10.1186/s40575-021-00111-4.

Mabunda et al 2022 Evaluation of Genetic Diversity in Dog Breeds Using Pedigree and Molecular Analysis: A Review. Diversity 14:1054. https:// doi.org/10.3390/d14121054.

To learn more about the genetics of dogs, check out
ICB's online courses

***************************************

Visit our Facebook Groups

ICB Institute of Canine Biology
...the latest canine news and research

ICB Breeding for the Future
...the science of animal breeding

By Carol Beuchat PhD

I am often contacted by breeders that are concerned about inbreeding and loss of genetic diversity in their breed. They recognize that these things matter over both the short and long term, and that they affect the burden of genetic health issues in the breed. The question from the breeder is "What do we do about this?"

Invariably, they bring up the issue of what dogs they should cross to, either individuals in the breed, or other dogs that are completely unrelated (e.g., another breed, mixed breed, etc.)

Breeders always want to discuss this right from the start. But, for a genetic rehabilitation project, you need to identify the specific problems you need to solve before you make decisions about what to do. The "What do we do" questions will get answered after we first ask "What are the genetic problems that need to be solved?"

Genetic analyses, whether from pedigrees or DNA, will produce information containing a bunch of terms you are probably not familiar with. To use this information, you will need to understand the vocabulary and what the data produced can tell you. Do not think you can learn this stuff from chats in Facebook groups. If you're going to put in the time and make an investment in doing these analyses for your breed, please take the ICB online courses. I am not aware of any other educational resources that are developed specifically for dog breeders and that cover the essential information while requiring no background in science. You need to invest some time in your own education or you won't know how to use the information from data analyses. Remember, the goal here is to figure out what breeders need to do to solve specific genetic problems in the breed. For this, you will need to understand the information provided. Take the courses. 

I have created a table (below) with some information about the types of basic information that you can extract from both pedigree data and genotypes. In general, I would advise breeders that only intend to use data from one of these sources to use pedigrees. It might seem that DNA should be better because it uses "high tech" techniques to provide information about chromosomes and genes. But, in fact, you will find that a pedigree analysis provides all the information you need to do a good job of genetic management of a breed. It requires no effort (or cost) from breeders to keep it up to date, and it will provide critical information for you generation after generation, forever. There is some overlap in what can be extracted from pedigree versus DNA data, but pedigrees might be better for some things and DNA better for others. Again, if you have the option of doing both, that is the most useful thing to do.

If you want to use DNA genotype data for your breed, you should use SNP (single nucleotide polymorphism) analysis, not microsatellites (or single tandem repeat markers, STR). Embark uses the most popular, research grade analysis based on high-density SNPs. Their output can be combined with data produced by any research lab using the same chip (e.g., data from a study). Wisdom also uses SNPs but they have a lower resolution chip of their own design and their data cannot be combined with Embark's. Also (and more importantly) they do not make the genotype data available to the user, so third-party analysis (as I am describing here) is not possible. Microsattelite data cannot be combined with SNPs, and they do not provide the resolution needed for planning a genetic rehabilitation program for a breed. However, STRs can provide data for the genes for the immune system (DLA, dog leucocyte antigen), and this is useful to know so you can breed for high diversity in these genes specifically.

The table below includes some additional information about the convenience and logistics of obtaining and managing data from pedigrees versus DNA genotypes. 

Armed with the information from analysis of your pedigree database or DNA genotyping, you are ready to start summarizing the issues that need to be addressed. For this, you will also need to know some other things - the actual number of living dogs, how many of these can be used in a breeding program (i.e, not spayed or neutered), and how many breeders there are and where they live (e.g., country). This is also the time to pull out all the information you have about phenotypic traits that you care about (e.g., coat color, variety, etc.) as well as known health issues and associated genotypes when known.

All of this is information that should be available about your breed anyway, but usually the motivation for getting things together is the realization by breeders that there are issues about genetic health that need to be addressed. With some information extracted from pedigrees or DNA genotypes, we can take a lot of the guesswork out of making breeding decisions. A little effort invested now in data preparation will yield benefits for the breed far into the future.

I have provided the reference for a nice paper by Mabunda et al (see References) about using pedigree and molecular data to evaluate genetic health of dog breeds. It contains a useful summary of the information provided by pedigree and genotype analyses and how to use it. You will see some jargon that might be new to you, so you can start here to become familiar with meanings and concepts. There is much more to know about how to set up a genetic rehabilitation program for your breed, but this is the place to start.

If you would like to know more about the genetic status of your breed and what can be done to improve it, contact me and we can discuss where to start.