Smoke and mirrors – Genomics and TOP traits
I was looking at the RAS (Ranking of Active Sires) List late last week, and I just about cried. The average Udder Overall score of the top 20 Friesian bulls ranked by BW, on the 20th February, was 0.38. This is in comparison to the overall Friesian cow breed average of 0.29! No wonder I hear so many complaints in our industry about cows with poor udders, and no wonder why we see such slow improvement in udders in our national herd. A Friesian cow breed average of 0.29 PLUS the average of the top 20 bulls of 0.38 DIVIDED BY TWO (50% genes from dam and 50% genes from sire) = 0.34. Udder improvement at a snail’s pace!
I decided to dig a bit deeper into this conundrum and discovered that 17 of those top 20 Friesian bulls were sired by genomic sires. That is to say, sires that haven’t had any daughters herd tested yet, or been ‘proven’ in industry parlance. Sires that were selected to be used for contract matings on the basis of what each breeding company’s genomic model had indicated their likely future performance to be. Of those 17 genomically-sired bulls, only three are greater than 0.5 for udder overall – Denovo Atlantis at 1.4, Jones Revelation at 0.92 and Mattajude Manu at 0.85.
The fact that those 3 bulls are all sired by genomic sires in not question here. The bigger question is – are we prepared to accept (from our breeding companies) a success rate of 3 bulls out of 17 with an Udder Overall of greater than 0.5? That’s less than 20% success at breeding a bull with an Udder Overall capable of effecting some ACTUAL improvement at a herd level.
Something else to bear in mind is that a significant portion of those top 20 bulls will have been heavily promoted as a part of a ‘teams’ approach, and used extensively in herds before any of their daughters were born and had been TOP (Traits Other than Production) scored. If you were expecting your breeding company’s bull team to deliver you an improvement in udders, you’ve unfortunately dipped out and possibly gone backwards in some cases. That’s a bitter pill to swallow when you’re paying upwards of $30/straw!
If udders this poor are the outcome of using genomic bulls to create the next generation of ‘rockstars’, then I have grave concerns as to the reliability of the genomic modelling used to predict the TOP traits of the subsequent generation; there is “a lack of confidence [from] farmers in the use of genomically-evaluated young sires and sons of sires. A range of factors stemming from lower reliability genomic estimates are causing this…” (IWG Report, p.2). Is it any wonder why farmers’ are reluctant to use genomic sires and their sons when poor udders and other substandard TOP traits get dished up to them and they pay a premium for that privilege?!
In my mind, either one or both of the following two things are happening to cause such a woeful expression of TOP traits and lack of improvement between generations.
1) Breeding companies are losing access to the very best dams each breed has to offer because the dam’s owner has decided that they no longer want to give the company access to those animals. This means the quality of the dam pool available to the breeding companies has become smaller and weaker; whether it’s mating dams with below-par TOP traits or having fewer options available is a moot point – the effect is still the same.
2) The genomic modelling used to predict the likelihood of any given TOP trait needs SIGNIFICANT work. This has as much to do with the QUALITY of the phenotypic data imputed into those models as it does with sheer data quantity. The IWG report (p.2) notes “the significant decline in the quality and accuracy of New Zealand’s phenotypic data. In the genomics era, high quality phenotypic data is of even greater importance to be able to make reliable, unbiased genomic estimates… one estimate is that quality [phenotypic] data is available for only 40,000 cows or 10% of the target reference population size [of 400,000 cows]. Our industry doesn’t suffer from having enough genomic data; it has plenty of that courtesy of LIC and CRV. It’s not all in one central place though, which is a story for another day! What our industry does suffer from, however, is a lack of quality phenotypic data used in support of genomic evaluations.
So where does all this leave Mr and Mrs Farmer then? Well, I think it leaves them with more questions than answers, and a decision to make. Do you trade off some BW for better TOP traits now to create a cow capable of lasting in your herd so you can milk fewer of them sooner? Or, do you continue your all-out assault on BW and hope that the udders will get better at some stage and only then milk less cows? The ‘efficiency’ and GHG conversation isn’t going away any time soon, and only one of those two pathways will help you. Choose wisely.
