Home

0564817
Ratings Generated in 2012
Featured Stallion

The Genetics of Nicking

I have just finished reading Thoroughbred Breeding: Pedigree Theories and the Science of Genetics by Dr. Matthew Binns and Mr. Tony Morris, and it is a thoroughly interesting and worthwhile read. I can recommend it for anyone involved in breeding racehorses (see it at Amazon.com). The book covers a large range of topics within Thoroughbred breeding as it relates to pedigree theory and does this well, with a healthy dose of skepticism along with the right balance between covering the topics that should be of interest to any Thoroughbred breeder, and covering them in depth enough for the breeder with a little more understanding of the machinations of genetics.

As I said earlier, Morris and Binns cover a lot of topics within Thoroughbred breeding and unsurprisingly see fit to devote one of its 24 chapters to the concept of Thoroughbred nicking, a subject that has attracted its fair set of supporters and detractors alike. In just the second paragraph of this chapter, Morris and Binns cut straight to the heart of what Thoroughbred nicking is truly about, and why to date a lot of popular Thoroughbred nicking services haven’t truly addressed the best way to handle nicking at all. Morris and Binns write:

“When looking at nicks, the questions to be considered include: what proportion of poor horses were produced by the nick, what proportion of top-quality winners were produced by the stallion with mares by other stallions, and what proportion of top-quality races were run by the produce of those mares when bred to other stallions? That last angle is frequently omitted from discussions of nicks, even those that take a statistical approach. It is obviously important, as superior dams will often produce top-class stakes-winners by a number of different stallions, allowing the supposed nick to be seen in its proper perspective.”

While their book was published in July of 2010 Morris and Binns, whether by design or otherwise, have just described the exact computation that the TrueNicks rating is built on. The TrueNicks rating, first established and published in 2007, is calculated by factoring two separate computations:

  • 1. The Sire Improvement Index (SII), a comparison of the percentage of stakes winners that a sire (or sire line) has achieved with mares by the specific broodmare sire (or broodmare sire line), with the percentage of stakes winners that the specific sire (or those specific members of the sire line) has achieved with all other mares; and
  • 2. The Broodmare Sire Improvement Index (BSII), a comparison of the percentage of stakes winners produced by daughters of a particular broodmare sire (or broodmare sire line) with the percentage of stakes winners from the same mares when bred to all other sires.

The resultant figure shows the stakes winner-to-starter production rate of the nick, compared to the stakes winner-to-starter production rate of the sire/sire line and broodmare sire/broodmare sire line when bred to representatives of all other lines. We here at TrueNicks are obviously glad that Tony Morris and Matthew Binns, both leaders in their respective fields, have pointed out where the severe limitations of other nicking services lie, and have verified that the way that TrueNicks designed its calculation so many years ago is the best way to create a nick rating service.

The chapter goes on to talk about how, from a genetic standpoint, Thoroughbred nicking could be explained in terms of genetics. Binns describes what is possibly occurring when one sire line shows a distinct affinity for daughters of another sire.

“An alternative genetic explanation for nicking is that the daughters of some stallions are able to provide combinations of ‘good’ variants of genes that complement those of the stallion with whom the nick takes place. Not all of the broodmare sire’s daughters would receive the correct combinations of complimentary genes; the shuffling of genetic material that takes place during sperm and egg production would ensure variations. However, those that did would have an increased chance of producing better offspring when mated to that stallion. In this case we might expect that the nick would often be reversible, with the daughters of the stallion also proving complementary with the broodmare sire.”

This is a very interesting point to think about. Let’s put this in a real life example. Sadler’s Wells did very well when mated to Darshaan mares (see High Chaparral). There were 169 foals bred on this cross, from 75 different mares, with 28 stakes winners (17%) coming from 21 different mares. These specific Darshaan mares did not perform as well when bred to all other stallions, nor did Sadler's Wells with mares by all other broodmare sires. So if we can work our way back, lets for a minute presume that of these 75 mares, only 50% of these, or 37 of them, got the “good variants of genes” from their sire Darshaan and presume that these 37 mares had a 50% chance of passing these on to their foals when mated to Sadler’s Wells which would bring us back to 19 mares – close enough to the 21 in reality to show you how nicks really do work from a genetic sense.

Of course all this begs the question, if only 50% of Darshaan mares mated to Sadler’s Wells had the “good variants of genes,” would it not have been better to have known if the Darshaan mare had these genes in the first place, before she was mated to Sadler’s Wells? After all, if the good variants of genes were indeed required to produce a superior runner on the cross, half of the Darshaan mares bred to him may well have had no chance of producing a superior runner by Sadler’s Wells in the first place. Equally, if only 50% of the foals bred on the Sadler’s Wells/Darshaan nick expressed the good variant of genes, would it not have been better to know this if the resultant foal was on their way through the sales ring?

Finally, Binns mentions that “the nick would often be reversible.” In the case of Sadler’s Wells as a broodmare sire, when his daughters have been mated to Darshaan and his sons it has also proved itself a superior nick with a TrueNicks rating of A+ (see Conduit and Hard Top). It seems that in the case of Sadler’s Wells and Darshaan, not only are we talking about a solid nick rating, but indeed a complete genetic affinity between these two stallions.

Filed under: , ,

13 Comments:

Byron,

As to your "real life example"-"work our way back" (Darshaan mares bred to Sadler's Wells): your stated schematic would require Sadler's Wells to possess total homozygosity to the complementary "good variants of genes" (from Darshaan) to support the percentages offered in your conclusion. While not discussed, the spectre of variable homozygosity from one stallion to another could factor into potential genetic explanations for nicking. Lastly, if one is presumming that an all or nothing scenario exists relative to the achievements of these end product nicks (which seems to follow from your example), then wastage (due to all environmental factors) should further erode your figures.  

sceptre 30 Nov 2010 3:08 PM

Sceptre,

I would indeed require Sadler's Wells to possess total or near total homozygosity to the complementary genes. While one is pensioned and the other dead, I would suggest that with enough horses bred on the cross and enough sons of Sadler's Wells out there and enough daughters of Darshaan that a geneticist could indeed rebuild the genotype for both to see if this held true, but that is something that I will leave to the geneticists.

The intention was not to make the numbers absolute, rather show how in a practical sense it all may work. No doubt environment plays a part, even within those horses that were stakes winners that were bred on the nick. i.e not all of the horses bred on the cross that presumably got the good variants of genes (and this is a presumption) ended up as Gr 1 stakes winners - there are some that are Gr. 1 winners and some that are listed winners. Environment (the way the horse was raised, the trainer, jockey, race conditions, etc) may well have eroded the latter's ability to get to the top where others made it.  

Byron Rogers 30 Nov 2010 4:13 PM

Is it true that the sire will pass on his X chromosome unmodified to his daughters?  While the X chromosome from the dam is always subject to recombination?  

If that is the case, then the genetic lottery probability would be considerably reduced in the first generation from the daughter.

vineyridge 01 Dec 2010 10:22 AM

I am new to the horse industry personally, although my father used to train.  I am analytical which tends to move me toward nicking in general.  That said, it appears to me that the majority of the focus is on sires and their impact on the breed.  If the dam provides potentially 50%+ of the genetic material, it would seem logical to me to focus equally on the girls.  Yet, when one reads a racing form, they give 4 ancestral names, 3 of which are male.

Therefore, I'm attempting to learn, not only the tail-male side of breeding relationships, but also, tail-female.  There are obviously some sires that improve their mares, but there are many who performed well because of the mares they had access to. I'm compiling a list of mares who've produced multiple stakes winners and I find trends in female families regardless of which sires they are bred to.  That has led me to think the girls are just as important as the boys.  Am I incorrect, or is all of that logic included in the mathematics of your proprietary formulas?

Mark 01 Dec 2010 3:58 PM

Mark,

Excellent comment. Yes, the mare is just as important as the sire. In effect, their genetic contribution to the foal is equal.

The difference is a matter of statistics. A sire can have 1000+ foals in his lifetime, while a mare can have only 10-15. While the mare's genetic contribution to her foals is no less significant than the sire's, statistical analysis of her influence is more difficult because of the small sample size.

This is why our nicking calculation references the correlation between sire and broodmare sire. If both sire and dam had large numbers of offspring to study, then it would be simpler to compare their respective influences.

By grouping mares of the same broodmare sire line together, we are able to identify certain trends. Mares sired by the same broodmare sire have a degree of genetic commonality, and therefore we can compare the average to the behavior of specific crosses.

The nick rating goes hand-in-hand with pedigree analysis. The strike rate of a nick helps qualify study of the pedigree and female family. It's not an either-or scenario, but a tool that gives breeders more information to make a decision.

Ian

Ian Tapp 01 Dec 2010 9:41 PM

Vineyridge:

Your question(s) seem to be directed to this topic's author, so I didn't wish to intrude. But unlike me, Byron is probably a busy guy, and I now noticed that Ian, his colleague, has responded to one. For what it's worth, I'll try to answer your questions. Your first paragraph of statements (albeit formed as questions) is, to my knowledge, accurate, while your second paragraph remarks are incorrect. As you no doubt realize the horse has 64 chromosomes-an individual x chromosome being but one of the 64. To modify somewhat your wording, I wouldn't think that 1 out of 64 could be called a considerable proportion even had your basic premise (2nd paragraph) been correct, which it isn't. At the embryo stage, if there are two x chromosomes one becomes a Barr Body which is inactive. There is a 50/50 chance as to which of the two x chromosomes will be rendered inactive (loses it ability to be involved in polypeptide synthesis). So, merely from the standpoint of Byron's example (a hypothesis I don't share), his figures retain their validity.    

sceptre 02 Dec 2010 12:53 AM

Maybe I just don't understand enough.  

The x chromosome is by far the largest one there is.  It seems to have a lot of genes that function in other ways than just reproduction.  Otherwise there wouldn't be sex linked characteristics.  The X and Y chromosomes could function as some sort of controllers for the rest of the horse building process.  At least that's a possible hypothesis.

Even though one of the two Xs in fillies is rendered inactive in her at the outset, both are used in egg formation through recombination and then meiosis, aren't they?  Recombination always seems to involve chance.  A colt, on the other hand, never gets anything from the sire's X chromosome, because he doesn't get it at all.  So whatever happens in fillies that have that chromosome activated cannot happen in colts from the same sire.  The colt only gets his dam's recombined X, which he then passes on in "pure" form to his daughters.

This is all very deep and difficult to comprehend.

vineyridge 02 Dec 2010 2:21 PM

Vineyridge,

I now understand better your point-I had interpreted differently your 2nd paragraph's phrase "...reduced in the first generation from the daughter."-Since, unlike autosomes that might be inherited from Darshaan to a Sadler's Wells-Darshaan daughter offspring, a Darshaan "x" if inherited by same offspring could be "deactivated", thus in some cells (of that offsping) a loss of that potential Darshaan "positivity". So, let's forget about the Barr Body issue and, instead, focus upon the rest:

Yes, a Darshaan daughter will receive an intact ("unmodified"/non-recombined) Darshaan "x" chromosome. Whereas this same Darshaan daughter will also receive Darshaan autosomes that have undergone recombination (resulting in some being "better" than "average", and some worse than "average")--("average" relating to Darshaan's potentiality to cause a + nick with Sadler's Wells)... So, going back to Byron's schematic, if one assumes that only 50% of these Darshaan daughters inherited the "good variants of (autosomal) genes" ("good" here meaning that they served a + nick with Sadler's Wells), I assume you (Vineyridge) believe that an intact Darshaan "x" chromosome is a given positive as it relates to the Sadler's Wells-Darshaan nick. Said another way-that it is more likely than a 50/50 chance that a Darshaan "x" would yield a relative "positive" (regarding this nick)? I wouldn't presume the Darshaan "x" to be a given +, but in light of crosses' data I'd be inclined that it does have a better than 50/50 chance...But, I see this all as very slight, certainly not "considerable". And while the "x" chromosome may be slightly larger (than the autosomes) I'm not so sure that the genetic material it contains relative to overall performance is any greater than that found in the autosomes. Lastly, why is it more likely that "...some sort of controllers..." are located on the "x" chromosome, rather than on the autosomes?      

sceptre 02 Dec 2010 6:19 PM

Just speculation, mind you;  I've do no studies of the literature so am talking out of a very large hat.  

But think how many fundamental physical differences there are between males and females.  I was just reading something today about scientific studies that are suggesting a connection between testosterone which is much reduced in the female and anxiety disorders which are far more common in women.  And there are many such differences which I would suggest have to come from the sex chromosome, the number of which *might* be 21, but then I might be remembering wrong.  Both sexes have many things in common, like the skeleton, most of the organs, etc. But they don't always end up the same or functioning the same in the male and female.  Something has to control the differentiation, and the logical place is in the XX or XY chromosome.

vineyridge 04 Dec 2010 6:44 PM

Vineyridge,

Sorry for the delay in replying. Sceptre has indeed answered some of the questions that you have stated but there one statement that you made that don't quite hold true.

"The x chromosome is by far the largest one there is." In terms of the # of genes per chromosome, that isn't quite right. Chromosome #1 of the Equine genome assembly (EquCab2.0) has almost 2200 genes on it while Chromosome #7 has over 1400. The x Chromosome has a little under 1400, so in terms of being "the largest one there is", I would suggest that chromosome #1 gets the prize.

This of course doesn't mean that Chromosome #1 is where all the action is, as it relates to thoroughbred performance. Indeed there was a study released last week (www.ncbi.nlm.nih.gov/.../21115378) whereby they looked at the candidate genes for elite performance in the human field and applied these genes to the equine genome. They found that if, and that is a big if, the same genes that separate out elite performance in humans, apply to horses (specifically thoroughbreds) then chromosomes #4 and #12 in the equine genome are over-represented with candidate genes for performance. Of course they are presuming that the genes for performance in humans is the same in horses which it will probably not (completely) be.

Truth be known, elite performance in thoroughbreds will most likely be polygenic with the combined effect of literally hundreds of genetic variants. These variants could form part of a total genotypic score in thoroughbreds where a horse with elite potential may get a score of say 75 out of 100, but the construction of that 75 could be made in any number of ways. Importantly, in humans at least there have already been studies to show that a mean 'total genotype score' is a significantly better indicator of athletic performance than the reflection of one polymorphism from a single candidate gene.

Byron Rogers 05 Dec 2010 7:45 PM

I read the entire give and take twice. WOW!!! I admit I comprehend so little of the science and specifics of genetics that this question may be foolish but here goes. Is there much that can be agreed upon or is there little than can be agreed upon with respect to predicting outcomes of mating thoroughbreds?  As a retired Marine fighter pilot, top gun graduate, and as an owner of a business that is somewhat successful I believe that I have some skills in making decisions. I have not been successful in my thoroughbred mating endeavor as yet but I have one decent looking yearling and my two weanlings have the best conformation of any I have produced in my 11 years in this game [that does not mean they have the intelligence or aptitude to endure the training regimen]. My favorite line is good at business- horrible at horses. I have read the TrueNicks blogs with great interest for a good while. I enjoy the challenge of trying to breed, raise, train and race a stakes caliber thoroughbred. The variables after birth are so many that it is frightening to consider how the odds are stacked against getting to the point of having a stakes caliber thoroughbred even if you have bred a physically and mentally talented thoroughbred. Never the less, I will continue this endeavor until my wife makes me stop. I would like to believe that there are likely outcomes when mating thoroughbreds.  Is there any common ground for those of you who have spent a life time studying this or is there so little that can be agreed upon that I should consult the mystics for my breeding advice?

dave york 06 Dec 2010 3:42 PM

Dave York,

If you're looking for consensus- I, for one, agree with the basic science and conclusions offered in Byron's closing paragraph. It's really stated quite well, and sums up the likely near-complete "mechanism" for end result (phenotypic) variation. Too often, mating decisions are derived from false mechanistic assumptions. If nothing else, one should attempt to understand what was offered here, and adjust their thinking acordingly. There are many additional nuances to all this (both quantitatively and qualitatively), some known, some yet unknown, but as an overview it's likely very near to reality.  

sceptre 06 Dec 2010 11:50 PM

Dave York-

The closing paragraph to which I referred is the one contained in Byron's (brogers) 05 Dec 2010 reply to Vineyridge. Sorry for my lack of clarity.

sceptre 07 Dec 2010 11:44 AM

Leave a Comment

All comments are moderated and must be approved before they are posted. The blog author reserves the right to edit or omit any comment.

  (Appears with your comment) (required)
  (Will not be published) (required)
  (required)