What are they?
Expected Progeny Differences, better known as EPDs, have been used in the beef industry since the 1970s. However, confusion still surrounds their application. EPDs describe exactly what their name implies: The differences which we would expect to see between progeny of two different animals. This statement assumes that we are mating the animals to animals with the same genetic merit. For example, two potential sires mated to the same group of cows.
What do all these numbers mean?
If I am considering using two potential herd sires shown in the table below, I am going to be interested in the difference between their EPDs (the last line in the table), rather than the actual EPD value itself. Based on this information, I would expect that sire 2’s progeny will be, on average, 1.3 lbs. heavier at birth, 5 lbs. heavier at weaning, and 2 lbs. heavier at yearling than the progeny of sire 1. Keep in mind that comparisons between EPDs of two different sires are ONLY valid between bulls of the same breed. For example, you cannot compare a Simmental with an Angus bull and get a valid comparison.
Table 1: Comparison of EPDs on two prospective herd sires.
If the ranking of a bull’s EPD in comparison to all other bulls in that breed is of interest, the rank can be located in the sire summary published for that bull’s breed. Simply mating to animals with the largest or highest ranking EPDs is not always the best strategy for genetic improvement. It is important to consider the resources (labor, forage, feed, etc.) available to you and to select bulls and females that will generate optimum performance (not maximum!) given the resources available in your production environment.
How good are EPD predictions?
All EPDs have an associated accuracy value that is typically listed below the EPD (for an example, see the table below). The accuracy value reflects the confidence in the EPD prediction and will be reflected as a number between 0 and 1. Zero means that there is no confidence in the EPD prediction while a 1 would mean complete confidence in the EPD prediction. Generally, EPDs on yearling bulls will be low (between 0.05 and 0.35) and will increase over time if performance records for the bull’s progeny are reported to the corresponding breed association. Commonly used AI bulls will often have EPD accuracies that exceed 0.8 and are considered proven sires because the recorded evidence gives us a high degree of confidence in the EPD prediction.
Even low accuracy EPDs provide a more direct route to select for the animal’s genetic merit than just evaluating these traits by sight (phenotype). Visual evaluation of an animal’s phenotype should only be used in cases where there are no EPD predictions for the trait (soundness and movement, for example). In addition, use of high accuracy bulls in an AI program can help to manage risk at breeding time by allowing the breeder to choose bulls that possess the characteristics they desire with a greater confidence than simply using a yearling herd sire.
How do I use them?
Anyone currently aware of EPDs and genetic improvement trends knows that the number of EPDs provided by breed associations is increasing. There are more EPDs published than any single producer can, or should select for at one time.
click image to zoom
It is important to identify those EPDs that are important to your production system and to select using only those metrics. For example, identify important output traits (such as weaning weight) and traits limiting in the environment (like milk production) and then choose the EPDs that best fit those parameters and select only on those EPDs. It is important to monitor your genetic progress by recording performance data for the offspring you produce. For commercial producers, it is not important to record as much data as should be expected from a seedstock enterprise, but it is still a good idea to collect calving ease data (assisted vs unassisted), relevant output data (weaning weights, if selling at weaning), and cowherd data (mature weight, etc.), as well as any additional data on traits important to your production enterprise. Remember, once you have accumulated a group of animals that fit your environment and available resources well, sometimes the best genetic change is no genetic change!
Source: Megan Rolf, Oklahoma State University