In another topic, roanart said
I totally agree with Atjan’s comments in that topic. Geneticists and animal scientists use a computation called the inbreeding coefficient to estimate how closely related two parents are and how much inbreeding would be present in their offspring. The calculated number estimates what percentage of the offspring’s genes are doubled up due to inbreeding. So, the smaller the number, the less inbreeding exists.
All horse breeds are inbred to some extent. For the Thoroughbred, where breeders intentionally avoid inbreeding as if it were the plague, the inbreeding coefficient is still about 8%. For breeds like the Lipizzaner which have a small number of stallion and mare lines and a history of close breeding, the average would be somewhat higher, perhaps somewhere around15%.
The inbreeding coefficient is based on common names appearing in both the stallion and mare's pedigrees. It decreases by 1/2 for each generation back in the pedigree that the common name appears. So, having C. Soja 4 generations back in the pedigrees of a stallion and mare would increase the inbreeding coefficient of their foal by about 0.2%.
Dr. Oulehla, the former Director of the SRS and Bundesgestut Piber, lectured in the U.S.in 1995 and 2001. In those lectures, he said that he used a three generation rule of thumb: if the stallion and mare had no common parents or grandparents, the foal resulting from the cross would be acceptable in terms of inbreeding. This rule of thumb means that the stallion and mare may have common great grandparents. In the cases where they have exactly one common great grandparent, it would increase the inbreeding coefficient by about 1.5%.
About 15 years ago, I wrote an article on inbreeding and for the USLR magazine. Here’s a summary of what I tried to say in the article:
- The amount of inbreeding in a proposed mating can be evaluated by looking at the pedigrees and statistically estimating the percentage of genes that would be duplicated by inbreeding.
- The farther back in the pedigrees of a stallion and mare you see duplicated names, the less important they are:
- Names duplicated in the first two or three generations have a big influence.
- Duplicated names that appear only in the fourth generation or farther back are pretty meaningless as they contribute at most 0.2% to the inbreeding coefficient.
- Names that appear in the first two or three generations of one parent's pedigree and are duplicated in the fourth generation (or farther back) in the other parent's pedigree have a modest influence (1.5% or less). For example, if the stallion's sire appears in the 4th generation of the mare's pedigree, that would produce 1.5% inbreeding. If the stallion's sire appears in the mare's 5th generation, that's about 0.75%. Similarly, if the stallion's granddaddy appears in the mare's 4th generation, that's also 0.75%.
- To avoid doing arithmetic for each mating you consider, evaluate the match as follows:
- You can ignore repeated names in either pedigree if you do not see the same name in the other pedigree. For example, if the stallion has two crosses to Maestoso Santuzza but the mare has none, ignore it. In such a case, you would have an inbred parent but the foal would not be inbred. It's what breeders call an outcross.
- Any name that appears in the first two generations of the stallion's pedigree should not appear anywhere in the first two generations of the mare's pedigree
- If the pedigrees pass that test, the inbreeding coefficient for the resulting foal will be no more than 3% from the first 3 generations and perhaps less. To get as high as 3%, one of the stallion's parents would have to a great grandparent of the mare or the other way around (one of the mare’s parents is a great grand parent of the stallion).
- If you extend this approach to 3 generations (i.e. No name that appears in the first 3 generations of the stallion's pedigree appears in the first 3 generations of the mare's pedigree), the foal's 4 generation inbreeding coefficient would be 1.5% or less.
- The foal's inbreeding coefficient for 5 or 6 generations, might be a little higher if there are more duplications farther back but you can usually ignore them.
AntwortenLöschenJohn,
This is a great post. You made a subject that causes brain meltdown in many of us understandable.
Romy
Thank you John!
AntwortenLöschenSee: http://www.highflyer.supanet.com/coefficient.htm
In the Lipizzaner literature and recent research projects, Wright's coefficient is used.
Thank you, John! I'm going to read and reread your post. It's a topic that comes up much in our Lipizzaner list.
AntwortenLöschenOkay, so Conversano Toscana II is, on his topline -- >340 Conversano Sagana --> 203 - Conversano IV Soja. Three generations. And my mare is 5 generations. Should be absolutely fine.
AntwortenLöschenSomeone else told me that the best match is breeding "like to like", but I had thought I wanted to breed to a stallion that could improve my mare's "faults". How does the SRS/Piber do this?
The "like to like" method is a good choice if you are trying to set a particular trait or type in your stock. It usually takes several generations to achieve that goal so it's not the best method for someone breeding a single mare. It's better suited to breeders who are using several mares each year.
AntwortenLöschenThe "improve on the mare" approach is probably the best choice for any individual mare. In using this method to select a mating, some people imagine the foal getting the worst traits of both the sire and dam. If they don't like the imaginary result, they don't make the mating. Others take the imagination step back a generation and look at combinations of the grandparents' worst traits.
Whatever method you use to select a mating, you're rolling the dice in terms of what the foal will look like. That's true even if you have made the cross before. You can produce 4 full brothers from a cross and get 4 very different looking stallions!
In his U.S. lectures, Dr O said that he selected 5 potential mates for a mare based
on conformation and type. He then evaluated the pedigrees. If the pedigree shows
that the stallion and mare have any of the parents or same grandparents, hedidn't consider the mating any further as the stallion and mare are too closely related. He said he also used a computer program to calculate a coefficient of inbreeding for each mating which gives an indication of how closely the stallion's ancestors were related to the mare's ancestors. If the inbreeding coefficient is higher than the limit recommended by geneticists, he discarded that stallion as a possible mate for the mare because this indicates that their ancestors were closely related. When he was done with the selection, he had a match in conformation and type, a pedigree he likes and a relatively low rate of inbreeding. With these techniques, Dr.
Oulehla estimated that it will take 100 years for inbreeding in the Piber herd to build up enough that they will need to outcross again.
So, that's what they used to do at Piber. I don't know how Dr Max is making the mating decisions as I've never heard him speak on the subject. I would imagine that his approach is similar to Dr O's as he must consider the comformation, type, and pedigrees in making his decision.
Hrm, hrm! Boy, this is fascinating :)
AntwortenLöschenDr O gave my girl 9s for her neck and shoulder and, I might add, she DOES have a lovely neck and shoulder. Not so good on the rest -- 6s and 7s. Her legs are very long. Her back is long. Her croup is flat and she's camped out a bit in the hind end. And she's 15.3hh. What she can do is extend like a warmblood, but without all that toe-flicking silliness. She looks like a carriage horse.
I do have a program that will calculate all this for me -- it's called Hawkeye, a horse version of Breedmate (dogs) -- so I'm going to put the pedigree in there and see what it comes up with. I'll do that tonight.
Another question -- it's said that the foal will usually take on the personality of the dam. Is that true as well?
roanart wrote:
AntwortenLöschen"I do have a program that will calculate all this for me -- it's called Hawkeye, a horse version of Breedmate (dogs) -- so I'm going to put the pedigree in there and see what it comes up with. I'll do that tonight."
If you only have a 5 generation pedigree, you won't see much of an inbreeding coefficient unless the pedigrees contain some close relatives. When I mentioned breed averages earlier, I was thinking of in depth pedigrees going back to the foundation stock. Over hundreds of years, inbreeding builds up gradually even if breeders are conscientious about following the 3 or 4 generation rule of thumb. That's how Thoroughbreds got up to an inbreeding coefficient of 8%.
Hawkeye does use Wright's formula which is still the standard method among animal scientists. Haridman's method (described in the link Atjan posted) is a fairly recent development based on similar principles but it is so complicated that I've never seen anyone use it other than that link.
O yes, Dr. Max (Dobretsberger) is following the same principal.
AntwortenLöschenEvery year the Obergestütsmeister Leopold Weiss proposes a list of those broodmares which will be used in that coming season. (In average about 50 mares)
First the computer program counts the possible combination with the selected breeding stallions for that season, which indicates a ranking of preference, counting by the Wright-system, purely based on the genealogical data.
The next step is considering the individual mares and stallions. And their parents with their heritable features. This contains all their positive and negative aspects: character ("interior") including workability, breed characteristics, exterior, movements, height, body mass, leg and foot quality, ability for collection, color.
Followed by an estimation of these aspects within a possible combination (completely based on the practical breeding experience of Dr.Dobretsberger and Mr. Weiss)
The input of the the SRS-riders and stable personnel about character, workability, physical potential etcetera is crucial foregoing information!
Then every mare is seen individually in the stable, and discussed by Dr.Dobretsberger and Mr. Weiss, and (sometimes) an appropriate extra adviser.
For the majority of the mares it is rather easy to select the ideal partner for that year. But for some, a discussion and a defined consideration is required. Sometimes the best genealogical combination is not the best in practice, considering interior or exterior. Sometimes long term strategic decisions must be made, f.e. considering the existence of a mare family or male dynasty. Really all kinds of considerations may occur.
This process is really so interesting! If you see the foals next year... it is like Christmas presents ;-) You know what you could expect considering your foregoing wish-list and the form of the package.... but you can always be surprised (of perhaps disappointed...) about the final present ;-))
Santa Claus (read: Nature) always follows his own way, at the end, you know...
this is a great thread---thanks john and all!
AntwortenLöschenif been breeding dog fore over 17 years, and inbreeding and line breeding are the moste inportant way of breeding.
AntwortenLöschenwe have a saying: inbreeding and line breeding is not the cause of hereditary lacks but the outcome of a polluted genepole