There has certainly been great debate since the introduction of genomics about how accurate the information is. While some breeders have gone full throttle on the use of genomic test sires, others are still very hesitant in the use of these yet to be daughter proven sires. For many the question remains,” how much can we trust these sires?”
Recently I had a conversation with a breeder and he said “They are only 70% reliable and you can’t really trust that.” To which I argued, “Actually you can trust that a fair bit.”
Some time ago CDN published that for genomically evaluated bulls with 65% reliable gLPIs, breeders can expect 95% of the time that their official proof will be within 670 LPI points (within about 18-20%). This means that we can be 95% sure that the current top gLPI sire, Suntor Joyride, will be higher than +2813 LPI, once he has his official progeny proven index that is over 90% reliable. That boils down to say that at least 95% of the time Joyride would end up with an official proof that would rank him in the top 10 in Canada. That is the worst case scenario.
When you apply this to your breeding program when you’re using a genomic young sire, you can take 670 LPI points or approximately 455 TPI points off their predicted index and they will achieve that number or higher 95% of the time. For example, take the #1 gPA TPI sire, Seagull-Bay Supersire, who has a current gPA TPI of +2527 and you can be 95% certain that his daughter proof that is over 90% reliable will be at least +2072. That would place him in the top 77 sires in the US (260 points behind current proven leader Observer). Remember that is 95% of the time he would be there at least. Not a bad worst-case scenario. (Read more: The Truth About Genomic Indexes – “Show Me” That They Work)
Pattern vs. Rank
The question that really comes to mind for me is not necessarily how do they rank, but rather how good is genomics at predicting the sire’s breeding pattern? Rankings will change all the time as new sires are added and breeders continue to push the envelope on genetic advancement. I am more concerned about how good genomics is at predicting the strengths and weaknesses of a sire.
To look closer at this, I decided to compare Maple-Downs-I G W Atwood’s genomic proof pattern vs. his now daughter proven pattern. Since Atwood is now over 95% reliable, it is safe to assess his current strengths and weaknesses, remembering that he was heavily used based on his genomic proof. In looking at Atwood’s genomic indexes, you would have said that he was a strong components sire with low production. His type pattern was that he would leave you outstanding daughters with great mammary systems, feet & legs and loads of dairy strength, but needs to be protected on rumps. Looking at his actual daughter performance you would see the same exact pattern. While yes his rump score is lower than his genomic index would have indicated, it was an area that genomics did say needed to be protected.
The interesting pattern that we have started to see is that the greatest variance from genomic prediction to actual proof is in the areas of health and fertility. Logically that makes sense, since most of these traits have a lower reliability. What we are noticing here is that genomic sires due to tend to follow the pattern of their sires for health and fertility traits more so than those of their dams. This makes sense too, since there is a larger data set in the sire’s health and fertility index than in the index of a dam. So next time you are looking at a genomic test sires health and fertility traits be sure to also check out those traits for his sire, as that may be as much a predictor of his potential as are his own indexes.
Prior to the introduction of Genomics in 2008, there was great attention paid to how young sires were sampled. AI companies worked very hard at getting a young sire sampled in as many different herds and different environments as possible, in order to get an accurate proof. Since the introduction of Genomics this has actually changed drastically. It is now to the point where the top genomic sires are actually used very selectively.
Young sires are no longer randomly sampled. In today’s genomic age, a lot of the systems and controls are gone. Yes, many of the sires are still offered to all breeders, but these high-ranking young sires are sold at a much higher price, and marketed much heavier. In addition, often the first release semen is only used on contract mattings on extremely high index, carefully selected mates. This too results in anything but random sampling and in reality is almost the perfect method for receiving an inflated proof. It isn’t just because of the actual mates they are being used on, but also because of the care the resulting calves will receive.
Genetic evaluation systems assume that all animals in the herd are treated equally. Yet while there is nothing wrong with a breeder wanting to ensure their return on their investment in these top genetic animals, it certainly causes many problems when accounting for it in the genetic evaluations of these animals. Most “animal-model” genetic evaluations in the world account for the genetic merit of a sire’s mates. However, when the US first added females to their genomic reference set, they actually got lower reliabilities as a result of inaccuracies in female’s proofs, due to preferential treatment. That is why some countries actually leave female genomic data out of their reference sets, as a large portion of the females are these high index animals that, in many cases, have received preferential treatment. In the US they actually implemented a scaling-effect adjustment to bring those top females down. The US has also implemented a new single-step model that includes genomic and traditional data together designed to account for this in bull proofs. Other countries are also looking for potential solutions. This includes possibly withholding early data from evaluations, as well as other options. The challenge is that no one has found a real solution to the actual problem and steps so far just mask the issue with scale downs and other band-aids. This preferential treatment problem is going to get greater attention, as more high profile genomic sires, priced high and heavily marketed will start to receive proofs in 2013.
The Bullvine Bottom Line
More and more Genomic young sires are now receiving their daughter proven proofs and many, such as Observer, have come through with flying colors. While rankings may change, the important thing to remember is that the genomic indexes did accurately predict breeding patterns. In that case, if you took the effort to make sure you used the sire because he was the correct mate for the animal, then the majority of the time the resulting progeny should be fine. If instead you used the sire just because of how he ranked and then his ranking changed, well then yes, you are going to find that you may not be as happy. The key thing to remember in any mating you are doing is know your goals. Make sure you breed towards them by selecting the sire that best accelerates those traits that you are breeding for and fixes the challenges of the cow you are breeding to. When you do that, you can be very confident in using genomic young sires to deliver the results you are looking for.
Not sure what all this hype about genomics is all about?
Want to learn what it is and what it means to your breeding program?