By Carol Beuchat PhD
Facebook is an echo chamber of misinformation, myths, and unfounded opinions about the coefficient of inbreeding, which is unfortunate because there is probably no statistic (other than the kinship coefficient, which is related) that is more useful to breeders who want to minimize the risk of health problems in their puppies.
Here's my list of the 10 really essential things you need to understand about the coefficient of inbreeding.
Here's my list of the 10 really essential things you need to understand about the coefficient of inbreeding.
| 1) The coefficient of inbreeding (COI) is the probability of an individual inheriting two copies of an allele from an ancestor on both sides of the pedigree. Every dog has two alleles at each locus—one inherited from its mother and one from its father. At every locus, there is a 50:50 chance (a probability of 0.5) of passing on either of the two alleles to its offspring. This process is random and happens independently at every locus, in each generation. 2) COI quantifies the chance of homozygosity at any locus; therefore, it is also equal to the risk of producing a genetic disorder caused by the inheritance of two copies of a recessive mutation. The estimated number of recessive deleterious mutations carried by the average dog is thought to be around 50-100. This number represents mutations that are hidden in heterozygous carriers and could result in a genetic disorder if a dog inherits two copies (homozygosity) of the same mutation. |  |
3) The relationship between inbreeding (COI) and genetic disease (risk) due to homozygosity of recessive mutations is linear. So, risk increases with increasing COI.
Remember that most loci are heterozygous in healthy, freely breeding animals. In dogs, homozygosity results from breeding related dogs. Loss of the normal state of heterozygosity results in inbreeding depression, a suite of negative effects referred to as loss of “fitness.” This includes things like litter size, body size, puppy mortality, lifespan, etc. This effect occurs with ANY level of inbreeding. No level of inbreeding is “safe.”
4) The estimated probability of inheriting two copies of the same allele from an ancestor can be calculated from pedigree data that goes back to that ancestor.
Because dogs are in a closed stud book, there is a finite number of “founders,” and all of the alleles the breed will ever have were contributed by those dogs. The COI of a current dog will be the probability of inheriting an allele that was present in one of the founder dogs and passed on from generation to generation. If those dogs are not present in the pedigree used to estimate COI, the true level of inbreeding will be underestimated. If you use a five-generation pedigree, COI will only estimate the probability of inheriting two copies of an allele from an ancestor in the fifth generation. This means that COI based on a 5 generation pedigree tells you about inbreeding only over those five generations. This will not be the true level of inbreeding so it will not reflect actual risk of producing a genetic disorder resulting from homozygosity of a recessive mutation.
Selecting a specific number of generations (e.g., 5 or 10) to use in calculating COI generation after generation will result in a systematic bias in the result. This is because the current generation gets farther and farther away from the inbreeding that has occurred in earlier ancestors, with the result that the calculated COI will start going down. In fact, for animals in a closed gene pool, inbreeding can only increase over generations (although there might be blips up or down in the average). You can not breed your way back to low inbreeding in a closed population of animals beyond making better use of animals with low relatedness (kinship) to the rest of the population. Breeding related animals in a closed gene pool will always result in an increase in inbreeding over time.
5) The fraction of a dog’s DNA that is homozygous due to inbreeding can be estimated from genotype data from the runs of homozygosity (ROH).
Inbreeding calculated from ROH is an estimate based on several assumptions, one of which is the length of homozygous “runs” (i.e., blocks of homozygosity) that reflect actual inbreeding; i.e., there must be a decision about the minimum length of ROH to be included in the estimate, and this must be specified in the analysis. Choosing different block lengths will result in different estimates of inbreeding, so the expertise of the analyst is critical for the quality of the inbreeding estimate. Determining the “exact” amount of inbreeding in an animal would require information about the entire genome and the ability to identify when homozygosity is caused by inheritance of two copies of the same allele inherited from the same ancestor (“inherited by descent”, IBD) versus two copies of the same allele that did not come from a single ancestor (“identical by state”, IBS). So the COI provided with a DNA analysis using SNP data (e.g., the Illumina Canine SNP panel) is an estimate from calculations based on a number of assumptions.
6) The notion that COI is a dusty relic from the good old days a century ago and is obsolete today is false.
The coefficient of inbreeding is just as relevant today as it was when it was first derived by Sewell Wright in the 1920s. This is because it provides a good estimate of homozygosity due to inbreeding, which is proportional to the risk of genetic disorders caused by recessive mutations. Because it remains the best predictor of genetic risk due to inbreeding, it is widely and routinely used today by animal breeders. Those claiming that COI is irrelevant or obsolete have an inadequate understanding of population genetics and especially do not understand this most basic statistic in the science of animal breeding.
COI is the best predictor of the risk of deleterious effects caused by homozygosity of recessive mutations, whether determined from pedigree data or DNA. If your goal is to breed dogs that are as healthy as possible, you definitely want to know this. The risk of adverse effects due to inbreeding is proportional to COI; risk goes up as COI increases.
COI estimated from pedigree data will depend on the depth of the pedigree data. Deep, complete (no missing data) pedigrees provide good estimates of predicted COI that are usually comparable to homozygosity estimated from DNA. For dogs, 20 complete generations of pedigree data will provide a useful estimate of inbreeding. Note, however, that the risk of genetic disease from homozygous recessive mutations accrues from the lowest levels of inbreeding; COI of only 3% is associated with an increased frequency of seizures in humans. Livestock breeders understand that every 1% increase in inbreeding has deleterious effects. Consequently, the time to worry about inbreeding is when it is very low, when every additional percentage of inbreeding reduces fitness. The negative effects of inbreeding usually outweigh the benefits by about COI of 10% (so livestock breeders try to keep COI below about 6%). So,a 10% COI is not “okay” or acceptable; it represents an average of 10% reduction in health and fitness due to loss of heterozygosity. This should be a bright red line for breeders. It is not the case that COI below 10% is "safe." The risk of deleterious effects is lower but still significant at 8%, or 5%, or even 3%.
7) The meaning or relevance of COI is not a matter of personal opinion.
You might have your own level of acceptable risk in your breeding program, but COI is a quantitative estimate of homozygosity for which the deleterious effects are well documented. Accepting a COI of 10% or 15% as “okay” implicitly accepts the same level of risk of negative effects for health. DNA testing can eliminate the 25% risk of producing offspring that are homozygous for a recessive allele from parents who are both carriers. Carriers produce a 25% risk of genetic disorders, which breeders are willing to pay to avoid. A 25% COI reflects the same level of genetic risk from ANY recessive mutation, including the ones we don’t know about. Paying for DNA testing but then producing a litter with COI of 25% (or more!) reflects a failure to understand what DNA testing tells you, as well as an inadequate understanding of the genetics of inbreeding. Your opinion about inbreeding coefficients is not relevant. It is the best statistic we have to quantify the risk of genetic disorders caused by recessive mutations.
8) COI predicts the frequency of homozygosity of alleles that are identical by descent; it is not a measure of genetic (allelic) diversity.
COI is the fraction of loci that are homozygous for an allele inherited from an ancestor on both sides of the pedigree. By itself, it does not tell you about genetic diversity. (Again, it is simply a probability of homozygosity.) However, breeding in a closed gene pool results in the loss of alleles by two means in every generation – from inbreeding and from genetic drift. So, inbreeding results in reduced genetic diversity, but this is not quantified by the inbreeding coefficient. Furthermore, genetic drift can result in reduced genetic diversity with no effect on inbreeding.
There are specific, objective metrics to quantify genetic diversity. It is common to see COI used in the context of discussions of loss of genetic diversity, but understand that this is because inbreeding results in loss of genetic diversity, so they are correlated. But remember that COI is specifically about the risk of homozygosity of alleles, not an estimate of genetic diversity.
9) Linebreeding is inbreeding, with exactly the same risks as ANY breeding of related animals.
Linebreeding is a breeding strategy designed to increase the genetic representation of a specific ancestor in an animal. Done properly, inbreeding from other ancestors should not be affected. That is, homozygosity of genes passed down from that ancestor should increase, without otherwise increasing the overall level of inbreeding. Linebreeding and inbreeding both involve the crossing of related dogs, and the consequences for homozygosity and risk of producing genetic disease follow the same rules.
10) COI is not "just a tool".
The coefficient of inbreeding is a quantitative estimate of the homozygosity of alleles that are identical by descent. This is the best statistic we have for the risk of producing genetic disease or inbreeding depression in the animals we breed. We should be using COI in the planning of every litter. There is simply nothing better, because it tells us the specific thing we want to know. The quality of COI estimates will depend on the quality of the data on which it is based- pedigrees should be deep and complete (no missing data), and DNA genotypes should be based on a very large number of loci (e.g., 100,000+ SNPs) distributed across every chromosome.
If you learned anything useful here, check out ICB's FREE online course, "COI Bootcamp," which is available from the ICB website.
Remember that most loci are heterozygous in healthy, freely breeding animals. In dogs, homozygosity results from breeding related dogs. Loss of the normal state of heterozygosity results in inbreeding depression, a suite of negative effects referred to as loss of “fitness.” This includes things like litter size, body size, puppy mortality, lifespan, etc. This effect occurs with ANY level of inbreeding. No level of inbreeding is “safe.”
4) The estimated probability of inheriting two copies of the same allele from an ancestor can be calculated from pedigree data that goes back to that ancestor.
Because dogs are in a closed stud book, there is a finite number of “founders,” and all of the alleles the breed will ever have were contributed by those dogs. The COI of a current dog will be the probability of inheriting an allele that was present in one of the founder dogs and passed on from generation to generation. If those dogs are not present in the pedigree used to estimate COI, the true level of inbreeding will be underestimated. If you use a five-generation pedigree, COI will only estimate the probability of inheriting two copies of an allele from an ancestor in the fifth generation. This means that COI based on a 5 generation pedigree tells you about inbreeding only over those five generations. This will not be the true level of inbreeding so it will not reflect actual risk of producing a genetic disorder resulting from homozygosity of a recessive mutation.
Selecting a specific number of generations (e.g., 5 or 10) to use in calculating COI generation after generation will result in a systematic bias in the result. This is because the current generation gets farther and farther away from the inbreeding that has occurred in earlier ancestors, with the result that the calculated COI will start going down. In fact, for animals in a closed gene pool, inbreeding can only increase over generations (although there might be blips up or down in the average). You can not breed your way back to low inbreeding in a closed population of animals beyond making better use of animals with low relatedness (kinship) to the rest of the population. Breeding related animals in a closed gene pool will always result in an increase in inbreeding over time.
5) The fraction of a dog’s DNA that is homozygous due to inbreeding can be estimated from genotype data from the runs of homozygosity (ROH).
Inbreeding calculated from ROH is an estimate based on several assumptions, one of which is the length of homozygous “runs” (i.e., blocks of homozygosity) that reflect actual inbreeding; i.e., there must be a decision about the minimum length of ROH to be included in the estimate, and this must be specified in the analysis. Choosing different block lengths will result in different estimates of inbreeding, so the expertise of the analyst is critical for the quality of the inbreeding estimate. Determining the “exact” amount of inbreeding in an animal would require information about the entire genome and the ability to identify when homozygosity is caused by inheritance of two copies of the same allele inherited from the same ancestor (“inherited by descent”, IBD) versus two copies of the same allele that did not come from a single ancestor (“identical by state”, IBS). So the COI provided with a DNA analysis using SNP data (e.g., the Illumina Canine SNP panel) is an estimate from calculations based on a number of assumptions.
6) The notion that COI is a dusty relic from the good old days a century ago and is obsolete today is false.
The coefficient of inbreeding is just as relevant today as it was when it was first derived by Sewell Wright in the 1920s. This is because it provides a good estimate of homozygosity due to inbreeding, which is proportional to the risk of genetic disorders caused by recessive mutations. Because it remains the best predictor of genetic risk due to inbreeding, it is widely and routinely used today by animal breeders. Those claiming that COI is irrelevant or obsolete have an inadequate understanding of population genetics and especially do not understand this most basic statistic in the science of animal breeding.
COI is the best predictor of the risk of deleterious effects caused by homozygosity of recessive mutations, whether determined from pedigree data or DNA. If your goal is to breed dogs that are as healthy as possible, you definitely want to know this. The risk of adverse effects due to inbreeding is proportional to COI; risk goes up as COI increases.
COI estimated from pedigree data will depend on the depth of the pedigree data. Deep, complete (no missing data) pedigrees provide good estimates of predicted COI that are usually comparable to homozygosity estimated from DNA. For dogs, 20 complete generations of pedigree data will provide a useful estimate of inbreeding. Note, however, that the risk of genetic disease from homozygous recessive mutations accrues from the lowest levels of inbreeding; COI of only 3% is associated with an increased frequency of seizures in humans. Livestock breeders understand that every 1% increase in inbreeding has deleterious effects. Consequently, the time to worry about inbreeding is when it is very low, when every additional percentage of inbreeding reduces fitness. The negative effects of inbreeding usually outweigh the benefits by about COI of 10% (so livestock breeders try to keep COI below about 6%). So,a 10% COI is not “okay” or acceptable; it represents an average of 10% reduction in health and fitness due to loss of heterozygosity. This should be a bright red line for breeders. It is not the case that COI below 10% is "safe." The risk of deleterious effects is lower but still significant at 8%, or 5%, or even 3%.
7) The meaning or relevance of COI is not a matter of personal opinion.
You might have your own level of acceptable risk in your breeding program, but COI is a quantitative estimate of homozygosity for which the deleterious effects are well documented. Accepting a COI of 10% or 15% as “okay” implicitly accepts the same level of risk of negative effects for health. DNA testing can eliminate the 25% risk of producing offspring that are homozygous for a recessive allele from parents who are both carriers. Carriers produce a 25% risk of genetic disorders, which breeders are willing to pay to avoid. A 25% COI reflects the same level of genetic risk from ANY recessive mutation, including the ones we don’t know about. Paying for DNA testing but then producing a litter with COI of 25% (or more!) reflects a failure to understand what DNA testing tells you, as well as an inadequate understanding of the genetics of inbreeding. Your opinion about inbreeding coefficients is not relevant. It is the best statistic we have to quantify the risk of genetic disorders caused by recessive mutations.
8) COI predicts the frequency of homozygosity of alleles that are identical by descent; it is not a measure of genetic (allelic) diversity.
COI is the fraction of loci that are homozygous for an allele inherited from an ancestor on both sides of the pedigree. By itself, it does not tell you about genetic diversity. (Again, it is simply a probability of homozygosity.) However, breeding in a closed gene pool results in the loss of alleles by two means in every generation – from inbreeding and from genetic drift. So, inbreeding results in reduced genetic diversity, but this is not quantified by the inbreeding coefficient. Furthermore, genetic drift can result in reduced genetic diversity with no effect on inbreeding.
There are specific, objective metrics to quantify genetic diversity. It is common to see COI used in the context of discussions of loss of genetic diversity, but understand that this is because inbreeding results in loss of genetic diversity, so they are correlated. But remember that COI is specifically about the risk of homozygosity of alleles, not an estimate of genetic diversity.
9) Linebreeding is inbreeding, with exactly the same risks as ANY breeding of related animals.
Linebreeding is a breeding strategy designed to increase the genetic representation of a specific ancestor in an animal. Done properly, inbreeding from other ancestors should not be affected. That is, homozygosity of genes passed down from that ancestor should increase, without otherwise increasing the overall level of inbreeding. Linebreeding and inbreeding both involve the crossing of related dogs, and the consequences for homozygosity and risk of producing genetic disease follow the same rules.
10) COI is not "just a tool".
The coefficient of inbreeding is a quantitative estimate of the homozygosity of alleles that are identical by descent. This is the best statistic we have for the risk of producing genetic disease or inbreeding depression in the animals we breed. We should be using COI in the planning of every litter. There is simply nothing better, because it tells us the specific thing we want to know. The quality of COI estimates will depend on the quality of the data on which it is based- pedigrees should be deep and complete (no missing data), and DNA genotypes should be based on a very large number of loci (e.g., 100,000+ SNPs) distributed across every chromosome.
If you learned anything useful here, check out ICB's FREE online course, "COI Bootcamp," which is available from the ICB website.
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
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
