These are the textbook figures for a clean pedigree. Where two cats are related in more than one way at once — half siblings that also share a grandsire, say — the real coefficient is higher, so use the pedigree mode below.
COEFFICIENT OF INBREEDING
25%
High risk
Stillbirths, defects, reduced kitten survival and immune deficiency all rise sharply at this level. This pairing is not recommended. If the goal is fixing a trait, use more distantly related cats and take two steps instead of one.
Both parents are shared, contributing 12.5% each for a total of 25%.
What the coefficient actually measures
The coefficient of inbreeding is a probability, not a vague sense of how closely two cats are related: pick any locus in the kitten, and COI is the chance that its two alleles are copies of the same gene in the same shared ancestor. A COI of 25% does not mean a quarter of the genome is faulty — it means that on average one locus in four is homozygous by descent. That is precisely where the risk comes from. Harmful recessives sit harmlessly in the heterozygous state; inbreeding raises the chance that sire and dam pass down the same copy, the recessive becomes homozygous, and the defect shows.
Wright's path method: where the number comes from
Find the ancestors the sire and dam have in common, then trace every path that runs up from the sire to that ancestor and back down to the dam. Each path contributes (1/2) raised to (n₁+n₂+1), where n₁ and n₂ are the generations from sire and dam to the ancestor. Take full siblings: their shared sire is one generation from each, contributing (1/2)³ = 12.5%, the shared dam contributes another 12.5%, and the total is 25%. If the common ancestor is itself inbred, multiply by (1+F) to fold in its own homozygosity — which is why figures tend to climb as you trace further back.
Three things to know before you act on the number
First, the figure only holds within the pedigree you entered. Every blank slot is treated as unrelated, so a three-generation COI of 0% can become 6% over five generations — imported cats are the usual case of an underestimate. Second, COI does not replace DNA testing: it is a probability, not a statement about what this cat actually carries, so HCM, PKD and PRA still need their own tests where a commercial panel exists. Third, a low coefficient does not make a good pairing; type, temperament and breed standard all still apply. This tool only gets the genetics right.
Frequently asked questions
What COI is safe for cats?
The common working bands are: under 6.25% is low risk, 6.25–12.5% is the usual line-breeding range and acceptable with a cumulative ceiling, 12.5–25% calls for caution and should follow breed-relevant DNA screening, and 25% or above is not recommended. Full siblings and parent-offspring pairings both land exactly at 25%, where stillbirths, defects and reduced kitten survival all rise sharply.
Is line breeding the same as inbreeding?
Mathematically yes; the difference is degree and intent. Line breeding means deliberately repeating one proven ancestor at a greater genetic distance, usually holding the figure between 6.25% and 12.5% to fix a trait that has already been verified. Inbreeding usually refers to high-coefficient pairings at 25% or more. What matters is not the label but two numbers: this litter’s COI, and how many generations the line has been stacking on the same ancestor.
How is the coefficient of inbreeding calculated?
By Wright's path method. Find every ancestor common to the sire and dam, then for each path running sire → common ancestor → dam, the contribution is (1/2) raised to (n1+n2+1), where n1 and n2 are the generations from sire and dam up to that ancestor, multiplied by (1+F) to account for the ancestor's own inbreeding. Sum every path. This calculator's pedigree mode implements exactly that, and treats identically spelled ancestors as the same cat.
Can half siblings be bred together?
A half-sibling pairing gives a COI of 12.5%, right on the line between acceptable and caution. It is technically workable if both cats have cleared the breed’s main DNA tests, the shared parent itself has a low coefficient, and you accept that the kittens will need outcrossing to bring the figure back down. If the reason is simply that these are the two cats you happen to own, rather than a specific trait you are fixing, it is not worth doing.
Why does this differ from my registry's COI?
Almost always because of generation depth. A registry database computes over every generation it holds — often ten or more — while this tool works from the three generations you enter, and any ancestor left blank is treated as unrelated. A three-generation COI of 0% can easily become 6% once five generations are counted. Use this calculator for a quick decision on a specific pairing, and your registry database when the full ancestry is available.