Genomics is creating a new buzz in the genetics industry and is getting quite a bit of press, as you may have noticed. I am convinced it is not just “techno-babble,” either. It’s a technology that promises to drive productivity and genetic improvement on progressive dairies.

So I wanted to use this space to answer some of the main questions I’ve been asked when meeting with dairy producers.

What do we want to create?
A number of times in past Progressive Dairyman columns, I have referenced the objective to make the “perfect cow,” or the “4-event cow.” This is a cow that has an easy and uneventful calving, a single A.I. breeding, a single pregnancy check with positive results and a dry-off.

These are the kind of cows that are easy to manage, seamlessly fit into production systems, and are generally unnoticed by workers and managers alike. They are the no-nonsense, problem-free cows.

Creating these cows just got a whole lot easier … and quicker …thanks to genomics.

What is genomics?
Researchers at the USDA have produced a commercially available method that tests 50,000 locations on the bovine genome. Based on the sequence of genes the animal inherited at each location from parents, a prediction can be made as to what that animal’s future performance will be on the range of production, health and type traits. Thanks to genomics, with better predictive power we can know the genetic merit of a sire when the bull is born.

This might not sound like much, but genomic selection has improved our predictive power by the equivalent of about 10 daughters for every trait we evaluate. More accurate and reliable predictions mean better information and less risk for producers when implementing breeding strategies for their herd. That is a substantial impact.

How does it work on my farm?
I hate to go back to school and put an equation in this article, but here it goes. Genetic progress is determined by the following: Progress = (Selection intensity x Accuracy x Variation)/ Generation interval

It’s complicated math, but the important thing is this: Genetic progress is maximized when selection intensity, accuracy and genetic variation are high, while generation interval is minimized. With genomic selection, both the accuracy and generation interval can be substantially affected.

First, we can reduce the generation interval by using younger bulls as sires because we can now get the information through a genome test instead of waiting for a progeny test. Second, we will increase the accuracy, because A.I. companies won’t sample any poor bulls in the future. In other words, the genetic bar will be raised, and producers will be the winners.

Will genomic testing end progeny testing?
While it is hard to predict the future, it is feasible to perceive a future where progeny testing is substantially different than it is today. However, in the short-term, I doubt it will have a noticeable impact as the results of the test will have to be verified with information over the next three years. The most immediate effect it will have on progeny test herds is that they will receive significantly better genetics to sample, as the bottom-end young sires will likely not ever be sampled.

What does it take to test an animal?
A DNA sample of blood, tissue or hair will do it. Participating labs (currently Geneseek and GIVF) then extract the DNA, use the SNP (single nucleotide polymorphisms) chip developed by Illumina to determine the genotype of the animal at each of the 50,000 SNP locations. The USDA then uses the SNP data in their prediction model to estimate the effects on the PTA for each of 26 traits for the animal.

What is the result of the test?
A new predicted transmitting ability (PTA) with significantly higher accuracy for each of 26 traits is the result. The new PTA will include parent average information plus genomic information. The added information increases the reliability of the PTA, actually reflecting the true transmitting ability of the specific sire. The reliability effectively goes from around 35 percent to 70 percent for most traits, which is the equivalent of having 10 daughters in a sire’s proof and reducing the range of genetic outcomes.

Have A.I. companies tested any bulls with current high reliability to see how the test compares to actual data?
Yes, the USDA used older bulls (sires of sons born prior to 1998) to predict the PTA’s of bulls born between 1998 and 2002. All the older bulls currently have second-crop info and were the basis to predict the PTA’s from the genotypes. The reliability of a genome-tested animal is about 70 percent (varies by trait), which is equivalent to 10 milking daughters.

Who can test animals?
Only the A.I. cooperators in the research project are allowed to test males for the next five years, but anyone can test a female immediately. If you do want to test a female, it is a good idea to contact your A.I. stud provider to ensure the best pricing.

What would the advantages be of testing a female?
Testing females will tell you which of your animals will transmit the best offspring. Females get PTA’s too, and just like bulls, some cows transmit better genes than others. If you have 10 animals that you deem “genetically superior,” you could test them and find out which of those are the best transmitters and how well they do in fact transmit. Finding an elite female in your herd may encourage you to flush the animal to get multiple offspring from this top transmitter.

Will I be able to sort through genome-tested sires and use the ‘best of the best’?
It will be possible to do this starting in January 2009, as the information on tested sires will be released on proof day. However, each A.I. company will determine which sires to market, and if they will market any “genome-tested” sires. Also, be careful in assuming the highest-ranking bull is the best. Oftentimes bulls with more than 100 daughters have their PTA’s significantly change, so these genome-tested sires (with the equivalent of only 10 daughters) are certain to change fairly substantially. It will still be important to use a group of sires to increase the overall accuracy of the group.

What impact will this have on other genetic tests available today?
This is a completely new process, of which the accuracy and predictability is exponentially better than anything that has been previously available. This new technology will make any other genetic tests currently or previously available completely obsolete.

Is the USDA the only global entity doing research in this area?
No, there are other countries that have done research that is similar in scope, including Holland and New Zealand. The USDA research is the most comprehensive using the 50,000 SNP chip. The National Association of Animal Breeders, which includes many of the major North American A.I. cooperators, will have exclusive rights to the USDA research for five years.

Will genomics affect how A.I. companies source their genetics?
It is likely to change, and each company will have its own strategy. Everyone will now be on a more level playing field because genomic information will make the search and selection process more transparent (i.e., we can see the genes, not just cows and records). There may also be shrink in the perceived gap of genetic quality between “breeder herds” and “commercial herds.” Again, this is because the true genotypes can now be tested, so it will be less of a guessing game as to which family, cow or young bull is the best.

Will this make it possible to select for more traits?
It is possible; however, in order to predict which genes have an effect on a given trait, we must have some data that shows performance for that trait. In other words, if we desire an evaluation for something like resistance to ketosis, we still need to collect data on that trait in the field before we can predict which genes have the desired effect. So it will take time to develop research we can have confidence in.

Are some companies already collecting this type of data?
Yes, data is being collected on additional traits by a few companies.

What is the expected change for a genome-tested animal?
The expected change would be zero; however, there is a normally distributed curve around the average change of zero. Basic statistics tells us that 95 percent of animals should change less than two standard deviations, plus or minus, for any given trait. So we can be confident in the results of the test, but variation within the expected range is still substantial - 2 STD of Net Merit is about ±$110! It is still very important to remember that genomic information reduces the expected range by about 50 percent over a parent average alone.

How much can full siblings vary?
There can be a substantial difference between siblings. Full siblings share 50 percent of their genes on average, so with only 50 percent of their genes in common the actual outcome can vary substantially. That’s caused by the 50 percent of genes they don’t have in common. One “real-world” example has two full brothers that differ by more than $200 Net Merit, 200 TPI points, nearly 2 points PTAT, 2 months of Productive Life, 2 on Daughter Pregnancy rate and even a difference of 2 on calving ease score.

Will inbreeding be a concern?
Inbreeding on the farm level should always be accounted for, and this does have the opportunity to increase inbreeding to some extent. However, in my opinion it is going to have the opposite effect because with genomic testing A.I. companies can be more aggressive in developing outcross sires by investing only about $250 to see if the genome test results merit a given “outcross” bull should enter an A.I. program.

Therefore, the potential to find more genetic outliers that are outcrosses from the population will become much less costly and therefore will be done more often. This will, in effect, increase the effective population size and decrease inbreeding.

Will genomics have an effect on crossbreeding?
Most producers that are currently crossbreeding are doing so for a few simple reasons: longer-lasting cows, more fertile cows, easier calving and healthier animals in general. Genomic selection will positively impact these traits quicker and more substantially than other traits and has the potential to do so to a larger magnitude that what is gained through long-term crossbreeding. The most important thing in any breeding program is setting your goals and then putting a plan in place to achieve them.

Does this mean I will have to pay more for semen?
Genomic selection certainly makes semen more valuable – especially young sire semen, and as always the extreme outliers in the breed will command a good price. The good news is that for the same dollars as you spend now, it is very likely you will be able to buy even better genetics in the future. Fortunately for you the additional value that the elite bulls offer (through performance of the genetically superior offspring) will always be much greater than additional dollars you have to pay for their semen, and this will not change. PD

Nate Zwald for Progressive Dairyman