is there a solid formula for determining the percentage of a breed in a horses pedigree?
Well, the most obvious answer is:
Ancestors of a certain breed in one generation divided by total horses in that generation?
In the fifth generation, you have 32 horses. If 12 of them are TB, then the horse is 37.5% TB
2 Likes
Are you speaking of XX, XO, O? Or just how much of a specific breed (or WB registry) a horse has in its pedigree? If it’s the later, I generally I do a 5-generation (62 horses) work-back. So it would just be number of whatever breed/registry you’re looking for divided by 62. If the horse was half GOV, it would be 31/62 = .5 or 50%
Not sure where you get 62 from.
2 to the 5th (3g-grandparents) is 32
2 to the 6th (4g-grandparents) is 64.
There is no generation that has 62 horses.
Aside from that, the basic calculation is as described in posts 2 and 3.
If some of the horses in the chosen generations are part-breds (of whatever “bred” you are looking at), you have to go back further generations, until all the horses are either “full” the breed you are interested in, or “none of” the breed you are interested in.
It’s ALL the horses in a five generation pedigree: 2 + 4 + 8 + 16 + 32 = 62.
I guess that’s a more “thorough” way?
No, that doesn’t work.
Suppose that both parents are halfbreds.(and assume we are talking about TB percent). The horse is 50% TB.
If you look at the grandparents generation, there will be 2 TB and 2 Non-TB, giving 2/4 = 50% which is the correct result
But with your approach you would have
(2 TB + 2 Non-TB) (grandparents) + 2 Non-TB (parents) giving 2/6 =33.33…% which is NOT correct.
1 Like
If you look at what I wrote, I’m only counting all 62 relatives when dealing with WBs. I"m thinking of pedigrees that have a registered Oldenburg that’s actually a Hanoverian x TB who are both approved GOV. Looking at the 5th generation only will not tell you what the offspring went on to be registered as. But perhaps I misunderstood the question…
So you are saying that if you bred an Oldenburg to a Hanoverian, and the offspring was registered ISR. and then you bred the ISR to a TB you would count the offspring as 3TB, 1 Oldenburg, 1 Hanoverian and 1 ISR and say the horse was 1/6th Hanoverian?
That is weird.
In the cross you describe (ISR/Hanno x TB), there are in theory 16 Hannos in a 5-generation pedigree, or 25%. So the G1 offspring is 25% Hanno (assuming all behind are Hanno), 25% ISR (assuming all behind are ISR) and 50% TB.
If you want to break out the ISR from GOV, it’s 1/14/15 or 1.6% ISR, 23% GOV, 24% Hanno, and then 50% TB.
Are you looking for % of blood (TB, Arabian), or % of registry? The math is all done the same way, the difference is just the variables you’re looking at.
You can’t get % blood just by looking at whether a horse in the pedigree is registered TB or Hano/Old/DWB/etc, since that Hano might be 50% TB
1 Like
Yes… and also, this was the essential nugget at the heart of an old (and misunderstood/under-appreciated) point of divergence between Francis Galton and Karl Pearson, two circa-1900 biometricians who set up systems for quantitatively studying breeding patterns. Other folks since have continued in that effort, similarly trying to figure out degrees of genetic relationship between relatives (or even breeds).
Awesome question, OP, and one that was has been considered in mathematical terms since the late 1890s. It gets a tad technical. Do you want to go there? Lemme know, and I’ll see if I can give you the semi-layman’s** run-down of the way to consider the degrees of relatedness between breeds and individual horses belonging to one or more breeds.
**I’m not a theoretical population geneticist, but I used to be a historian of science who worked on that field so I kinda/sorta get it. I’d like to think that I’m good at explaining it to “the rest of us.”
Yes… but the horses back in that 5th generation each contribute more or less influence to the one at the top of the pedigree depending on the breeding patterns in the generation between them. For example, the horse who appears once in the pedigree, way back in that 5th generation, AND who has no relatives in any of the subsequent generations has less influence that would that same horse who was, say, line-bred all the way up to one of the parents of the horse in questions. In this second case, more of his inheritance remains in the parental generation because of the line-breeding behind them, back to that one horse in the 5th generation.
See what I mean? You can’t just divide the total influence by the number of animals in the pedigree.
