BIF 2018: Experiences with Implementation of Single-Step at American Angus, One Year In

Steve Miller


Angus Genetics Inc has four research scientists. They do single-step genomic evaluations for Angus, Canadian Angus, Charolais, Canadian Charolais, and Maine-Anjou.

Angus switched to single-step on July 7th, 2017. Miller feels like the switch to single-step was a monumental switch, should have made the Time Magazine year in review!

When breeders see bulls re-rank, they then understand was a correlation of 0.9 really means! On July 7th, Steve Miller talked to two breeders. On July 8th he was feel cool and casual like Ferris Buller. The next week, then the calls came.
"I've seen this before, in a couple of months it will all blow over" a friend told Miller. And that proved true.

The switch to single-step was kind of like the stages of grief with getting a flat tire.
Quiet phase: Listening, evaluating, asking yourself if you have a flat tire.
Expressive phase: Cussing, kicking dirt, frustrated.
Have to Move Forward phase: Fixing the tire and getting back on the road.

Communication was crucial, both before the changed occurred and phone calls after the change.

There is no way AGI would go back to the methods used before the switch to single-step. The new evaluation is that much better.

On July 7th, Angus changed to single-step. But they also made a lot of other changes to the evaluation. The changes were frequently due to the changes in the carcass model, by accounting for the selection bias in which animals had carcass and ultrasound data.

Single-step is now a brand.

Any genetic evaluaiton is only as good as the data behind it.

Genotyped database has doubled in 96 weeks. On June 16, 2018 they reached over 500,000 samples. They receive over 3,000 samples per week since November 2017.

So, after the storm of single-step passed, what questions now cause the phone to ring.
How can two flush mates have this different of EPDs?

The power in the pedigree. Creating the famous A matrix, the relationships between all of the animals in the evaluation. When we use pedigree data, we use the expected relationships between animals. But, with genomic data, we actually use the real relationships between the animals.

Recombination between paternal and maternal chromosomes cause unique chromosomes being inherited by the progeny of a bull.

We can look at 4 bulls.

Bull Rank Whole Chromosome 6 Top Half of Chromosome Bottom Half of Chromosome
Bull A 1 6 9
Bull B 51 7 167
Bull C 58 187 1
Bull D 183 171 184
Rank out of top 193 Sires for Birth Weight genomic value.

Look at variation in calf crop. One quarter of the variation comes from the dams. On quarter of the variation comes from the sires. One half of the variation comes from Mendelian sampling, the random shuffle of genes and chromosomes between generations.

As a breeder, we should love variation. Variation is what makes genetic progress possible.

Theory and what AGI is putting out the door matches. up.

Consider two full sib ET bulls. Pedigree says they should share 1/32 of their DNA with Shearbrooke Shoshone. But, what does the DNA say? Calf with higher birth weight EPD is 15% more related with ancestors is the pedigree with higher birth weight EPDs.

Post a Comment

Popular posts from this blog

ARSBC 2018: EPDs and Reasonable Expectations in Commercial Crossbred Operations

Reverend Bayes and Cattle Breeding

Do You Own Your Job or a Business? Are You Working for a Lunatic?