female families

Understanding pedigrees, inbreeding, dosage, etc.

Moderators: Roguelet, hpkingjr, WaveMaster, Lucy

stancaris
Restricted Stakes Winner
Posts: 763
Joined: Thu Sep 16, 2004 11:24 am

female families

Postby stancaris » Sat Jan 21, 2012 9:03 am

I am interested in researching an angle on the Derby and BC Classic with regard to female family numbers but would like to know which families are uncertain due to DNA research (haplotype research done in England I believe a few years ago). Does anyone know which families are the ones that are in error? So, when I find a horse from say family 1, 2 etc can I be reasonably sure that horse actually came from that family?

xfactor fan
Breeder's Cup Winner
Posts: 2212
Joined: Thu Sep 16, 2004 8:46 pm

Postby xfactor fan » Sat Jan 21, 2012 9:27 am

You might want to start wlith the Hill study that is posted on the TB Heritage site. It also shows which families have the same type of mtDNA.

There's another study that Pan Zareta posted a while ago, perhaps it could be posted again? I've managed to lose the link before having the time to read it.

User avatar
Pan Zareta
Breeder's Cup Winner
Posts: 2074
Joined: Wed Dec 22, 2004 10:55 am
Location: west TX boonies

Postby Pan Zareta » Sat Jan 21, 2012 10:59 am

xfactor fan wrote:There's another study that Pan Zareta posted a while ago, perhaps it could be posted again? I've managed to lose the link before having the time to read it.


You may mean the 2006 paper by Harrison & Turrion-Gomez
http://thoroughbredgenetics.com/Mitochondrion%206(2006)%2053-66.pdf
They reported 28 incorrect sub-branches in 19 of 33 unspecified TB female lines that were sampled.

pedigreeann
Weanling
Posts: 33
Joined: Thu Jul 12, 2012 7:55 am
Location: Lexington

Postby pedigreeann » Mon Jul 16, 2012 9:15 am

On pedix (a Yahoo group for pedigree geeks), we discussed some of this stuff a while back.

The highly influential family stemming from Americus Girl (Nasrullah, Royal Charger, etc.) does not match up with other branches of 9-c and hence had been labeled 9-c* in many sources. Interestingly, the Bobinsky-Zamoysky Tables had a notation for the mare who supposedly connects this lot with the rest of 9-c that stated that she had half-siblings who raced as half-breds. So this line has always had question-marks about its source.

Looking at the sources-mares in two misplaced sub-branches, it became pretty clear, at least to me, that two filly foals who were at the same farm at the same time were somehow switched before they became broodmares.

Numerical studies are fine, but one must look at the whole picture in context. Some inconsistencies are to be expected from a Stud Book that was established 100 years after many of the foundation stock (especially the mares) were producing offspring. Most of the mistakes of assignments are from long ago, some of it from Lowe's own research when he was assigning family numbers in the 1880s and 90s.
There is no such thing as too much Teddy.

parlo
Grade I Winner
Posts: 1515
Joined: Fri Jun 03, 2005 9:36 am
Location: Germany

Postby parlo » Mon Jul 16, 2012 10:52 am

My question: why does no one change the obviously wrong "official" historic dam-lines according to the expertise we got from those mulecular-genetic research of Harrison / Turion-Gomez and me be later research of other scientists?

User avatar
Pan Zareta
Breeder's Cup Winner
Posts: 2074
Joined: Wed Dec 22, 2004 10:55 am
Location: west TX boonies

Postby Pan Zareta » Mon Jul 16, 2012 12:06 pm

pedigreeann wrote:The highly influential family stemming from Americus Girl (Nasrullah, Royal Charger, etc.) does not match up with other branches of 9-c and hence had been labeled 9-c* in many sources. Interestingly, the Bobinsky-Zamoysky Tables had a notation for the mare who supposedly connects this lot with the rest of 9-c that stated that she had half-siblings who raced as half-breds. So this line has always had question-marks about its source


I realize some concluded the findings re. fam. 9 reported by Hill et al. 2002 confirmed a suspected error at the Young Sir Peter Mare (9th dam by stud book record of Americus Girl). However, the linear timeline for family 9 (Fig.1, p.291, Hill et al. 2002) indicates a much earlier common dam for the six family 9 samples exhibiting their haplotype "G" than the Young Sir Peter mare whose pedigree was characterized as fraudulent ed.5, GSB1, (1891, the source for the note in B-Z).

Additionally, S.P. Harrison, whose study (finally published in 2006) was based on a sample size 10x the size of Hill's, made the following remarks to TBpedix 9/11/2002 "In respect of Family 9 and 12. The only sub-families of these that we have found to carry the same mtDNA variant are 9c and 12c (12b is different). Most of the rest of Family 9 are unique...".

Most plausibly, all six of Hill's haplotype "G" samples from family 9 had the mare x Crab, tap of 9-c, in common (again, by stud book record). This doesn't necessarily validate the traditional pedigree for the Y. Sir Peter Mare, contradicting the remarks 1891. It could be a simple coincidence representing two founders with the same mtDNA haplotype, or she could indeed have been a true descendant in tf of the 9-c tap, but with an incorrect or falsified pedigree.

parlo wrote:My question: why does no one change the obviously wrong "official" historic dam-lines according to the expertise we got from those mulecular-genetic research of Harrison / Turion-Gomez and me be later research of other scientists?


In their report as finally published 2006 Harrison & Turrion-Gomez did not correlate mtDNA haplotypes with any specific family, pedigree, or lineage. That report alone is relatively useless for making corrections to the historic record. Other researchers before and since have offered such information, but it's usually not specific enough to pinpoint at precisely which mare the error occurred.

xfactor fan
Breeder's Cup Winner
Posts: 2212
Joined: Thu Sep 16, 2004 8:46 pm

Postby xfactor fan » Tue Jul 17, 2012 11:33 am

parlo, the powers in charge of American racing don't want to change things.

Rather than replacing female families, it might be better to have mtDNA typeing done when the horse is registered. And post the results. It would only take one generation to sort out most of the mistakes.

zinn21
3rd Year Sire
Posts: 3307
Joined: Mon Jan 31, 2005 8:23 pm

Postby zinn21 » Tue Jul 17, 2012 8:09 pm

Rather than replacing female families, it might be better to have mtDNA typeing done when the horse is registered. And post the results. It would only take one generation to sort out most of the mistakes.


Now that's a good idea..

User avatar
Pan Zareta
Breeder's Cup Winner
Posts: 2074
Joined: Wed Dec 22, 2004 10:55 am
Location: west TX boonies

Postby Pan Zareta » Wed Jul 18, 2012 7:13 am

xfactor fan wrote:the powers in charge of American racing don't want to change things.


It's the powers in charge of the Thoroughbred registries that matter in this regard. The JC here, as well as Weatherby's, made it clear a decade ago that they are not interested in using genetic testing to correct historic errors, following precedents set when parentage verification by (autosomal) DNA was first implemented.

Rather than replacing female families, it might be better to have mtDNA typeing done when the horse is registered. And post the results.


Don't we wish! ;) Unfortunately, mtDNA haplotyping is pretty expensive. Neither the registry nor the avg. TB owner is going to be happy about assuming that expense. And any attempt to publish the results could infringe on a patent held by a UK company.

pedigreeann
Weanling
Posts: 33
Joined: Thu Jul 12, 2012 7:55 am
Location: Lexington

Postby pedigreeann » Wed Jul 18, 2012 8:45 am

The family numbers are not official sanctioned by the authorities that register TBs. They were assigned by Bruce Lowe in the 1890s as part of his proposed method for breeding TBs. They have turned out to be useful for keeping track of female lines, so everybody used them.

Bobinsky and Zamoysky found that they couldn't fit all the branches they wanted on single pages, so broke up the families up in a, b, c, etc. These lettered branches have no official sanction, either, and they are not always the most logical divisions, being based on space considerations.

If we all decided that the offspring of mare x belongs in family 8, not family 11 (say), we could rewrite our databases and put her where she belongs. We have done so before, with the great racer Selima, imported to the US (the Colonies) in 1750. Lowe put her in family 15, based on an assumed ancestry of her dam, but later a private studbook came to light that put her solidly in family 21. No stud books have been corrected that I know of, but we who care know.
There is no such thing as too much Teddy.

parlo
Grade I Winner
Posts: 1515
Joined: Fri Jun 03, 2005 9:36 am
Location: Germany

Postby parlo » Wed Jul 18, 2012 10:46 am

pedigreeann wrote:The family numbers ... were assigned by Bruce Lowe in the 1890s as part of his proposed method for breeding TBs. ....


Well, don't mix up cause and consequence. :wink:

Lowe intended to trace and record dam lines - as the German Goos has done before / parallel to him.

For classification he used numbers that he ordered in the totals of wins of 3 GB-classics. And then he made a mental error in assuming that that total of classic wins is an expression of higher quality, neglecting the quantities of his families. Then he started to group his families (classic, running, sire, outside families) to further push that idea of a certain quality which he thought to be expressed by his family-numbers. After that he thought, he had discovered / created a breeding system.

xfactor fan
Breeder's Cup Winner
Posts: 2212
Joined: Thu Sep 16, 2004 8:46 pm

Postby xfactor fan » Wed Jul 18, 2012 11:42 am

I'm not sure that people are using the Lowe system for their breeding programs. It is the only tool out there that specifically tracks the female line.

In our age of genetic testing, we now know that mtDNA is important and it is passed on via female lines. And with the research out there connecting mtDNA with distance preferences, it becomes another tool in the breeder's kit.

Anyone know if the UK guy got a patent on the use of mtDNA in horses? If so, patents have set time limits. And would just publishing the mtDNA results cause problems?

User avatar
Pan Zareta
Breeder's Cup Winner
Posts: 2074
Joined: Wed Dec 22, 2004 10:55 am
Location: west TX boonies

Postby Pan Zareta » Wed Jul 18, 2012 2:11 pm

xfactor fan wrote:Anyone know if the UK guy got a patent on the use of mtDNA in horses?


http://patentscope.wipo.int/search/en/WO2006003436

And would just publishing the mtDNA results cause problems


Yes, it could be construed as patent infringement.

xfactor fan
Breeder's Cup Winner
Posts: 2212
Joined: Thu Sep 16, 2004 8:46 pm

Postby xfactor fan » Thu Jul 19, 2012 10:00 am

Well, patents do expire--the term used to be 17 years. Don't know how long this one will run.

But if you test your own horse, get a mtDNA type, then put that information up on this data base, it would be a paten infringment? That seems a bit crazy.