Yesterday my cows got turned out on grass. Despite it being a cloudy day, I think they were pretty happy!
Let me tell you the story of these three cows. Last August my mother brought three of her two-year-old cows to Missouri. In New Mexico they have been struggling with drought, so the cows would benefit from some extra feed. And they did. In the first three months their body condition score increased by one (almost two) units. But, then my pastures ran out of grass, so I started feeding hay, and I started to observe big differences between the cows. I started to notice their feet were sore, and I soon realized I was dealing with fescue toxicosis. Two of the three responded rapidly to 20% protein cube supplementation, but the third one never did. So, I have made a selection decision, and this fall after she weans her calf, the fescue sensitive cow will be culled from my herd.
In a previous post, one commenter mentioned gene-by-environment interactions. In my case, my cow likely has a genetic predisposition to fescue toxicosis sensitivity and then she was placed in an environment containing fescue with a high level of endophytes. Thus, an example where genes and environment interact. This highlights that EPDs are one of many tools that we use when making selection decisions. EPDs only account for additive genetics, and do not consider gene-by-gene or gene-by-environment interactions (additive genetic variation is inherited in a predictable manner, thus the reason we use additive genetic variation to predict EPDs). In the beef industry we still lack genetic evaluations for many economically important traits such as structure and locomotion, feed efficiency, disease susceptibility, and fertility. Currently, we still have to rely on phenotypic selection (using our eyes) to make selection decisions for these traits.
Graybull quoted Johann Zietsman who stated EPDs for traits like fertility with strong gene-by-environment could not be calculated. Zietsman broke fertility into hormonal balance and body condition. He left out an important player, embryonic lethal variants. These are variants which "break" genes, causing the protein coded by the gene not to function properly. When an embryo inherits two copies of an embryonic lethal variant, the fetus is aborted sometime during the pregnancy, perhaps as early as the first 45 days. In the next four years we will produce a fertility EPD, which will be incorporated into economic selection indexes, that will account for the embryonic lethal load an animal carries. We will deliver a genomic EPD which directly tests the number of variants (alleles) which decrease fertility, and this test will be directly comparable across environments.
Furthermore, 10% of the variation in fertility is due to additive genetics, which excludes gene-by-environment effects. Keep in mind that in EPD calculations, animals are compared to their contemporary groups which are all managed in the same environment. By selecting for fertility EPDs, such as heifer pregnancy, we can improve the 10% of fertility that is due to genetic merit (in other words additive genetics estimated by EPDs). To further improve fertility, we need to take a systematic approach and use best practices to properly manage the fertility in our herds and use phenotypic selection to identify any gene-by-environment interactions.