Carcass expected progeny differences (EPDs) are an important tool for producers to use when kept in perspective with other Economically Relevant Traits (ERTs).
First let’s define what ERTs are. This concept came about when Red Angus Association of America (RAAA) passed a strategic plan in 1994 calling for the adoption of 20 EPDs by the year 2000. After talking to influential breeders like Gary Rolland of Fort Hays University and Butch Schuler, it became evident that they should do the opposite and produce the fewest EPDs possible to cover growth, reproduction, carcass and maintenance. But the question was how to accomplish this goal? This is where Dr. Bruce Golden of Colorado State University (CSU) and his colleagues Drs. Rick Bourdon, Mark Enns and Dorian Garrick came in. They devised a system where traits were classified as indicator traits or ERTs. Indicator traits would never be published as EPDs but would only be used as a correlated trait to add accuracy to an ERT EPD.
The classic example of an ERT, which I have used before, is scrotal circumference and heifer pregnancy.
Yearling scrotal circumference in bulls is a useful indicator of puberty; thus bulls with larger scrotal circumference should have daughters that reach puberty at an earlier age, thus having a better chance of becoming pregnant to calve as two year olds. However, the ERT is a heifer pregnancy EPD, not the indicator trait which is a scrotal circumference EPD. In the ory, one would use the scrotal data in the heifer pregnancy model to add accuracy, but remember the trait of interest is heifer pregnancy. (As a side note, in most breeds, scrotal circumference is confounded between size and puberty and should not be used in the calculation of a heifer pregnancy EPD.)
Although American Angus Association (AAA) always led the industry in the collection of carcass and then ultrasound data, in 2000, RAAA and CSU was the first to implement ERTs on carcass EPDs by combining carcass (ERT) and ultrasound data (correlated or indicator trait) into a single carcass EPD. Simmentals quickly followed, and other breeds followed suit over the years.
AAA has a reputation for publishing lots of EPDs for indicator traits. However, recently, AAA has shown leadership in following the concept of ERTs with their carcass EPDs by combining genomics, ultrasound and carcass data into one analysis. The new EPDs are on a carcass basis with ultrasound and genomic data adding accuracy.
In 2007, the Beef Improvement Federation (BIF) recommended correctly that carcass was the ERT, not ultrasound, which is an indicator trait, and that carcass EPDs should be expressed on a carcass basis.
In the spring of 2008, AAA combined their ultrasound and carcass EPDs, which had been printed separately before, into a single set of EPDs. In the summer of 2009, they treated the genomic data the same, using it as a correlated trait to add accuracy to the carcass EPDs.
One problem does exist biologically with selecting against ultrasound backfat (BF). Ultrasound BF is positively correlated yearling growth and ribeye area (REA). Meaning the betterdoing yearling bulls do not tend to be the leanest. That just makes sense. However, the opposite correlation exists between carcass BF and growth and carcass REA. Meaning cattle with fatter carcasses tend to be slower growing and smaller REA. When combining carcass and ultrasound EPDs would have been an opportunity to drop ultrasound BF from the EPD calculation, but that path was not chosen by most all breeds.
Genomics have been a tougher nut to crack. At first we just had markers, which did not explain enough variation to be overly useful.
However, in 2006, the American Simmental Association and Dr. Dick Quaas of Cornell University pioneered the use of genomics by successfully utilizing specific markers into their tenderness EPD. The next evolution in genomics was high density panels, which instead of testing a few significant markers, identified a whole host of SNPs (very basic places on the genome) that explain variation for a particular trait. The problem for the various breeds is that it is becoming more evident that these panels will need to “trained” for a specific breed for them to be useful in the production of EPDs across the full host of traits (there is a possibility of this changing when we move to the next generation of even higher density panels).
As it stands right now, AAA implemented genomic data into their carcass EPDs in the fall of 2009 and into the rest of their traits this year. Meanwhile, Hereford collected enough samples and has run the parameters that they are now in a position to implement. Simmental has the largest training population of animals that have gone through the 50K high density panel and hope to have their EPD model parameters done for implementation as soon as next spring. Developing a breed specific panel is very expensive, but Limousin got lucky when Dr. Taylor at University of Missouri needed samples to run 50K panels on and happened to have a large bank of Limousin samples on hand. The preliminary parameters were all run to include them into EPDs, but Limousin has not yet taken advantage of their opportunity. Some other breeds are working hard to catch up, but because of time and expense involved, several are waiting on the sidelines.
Through the wise leadership of Dr. Doyle Wilson of Iowa State and John Crouch of AAA, the centralized ultrasound processing system was designed in which full contemporary groups were ultrasounded by certified technicians, interpreted in a certified centralized laboratory and the data sent directly to the association to avoid data sorting. This system has since been adopted by practically all the other breed associations. When it came to genomics, AAA copied this model for collecting genomic data.
In the end, EPDs accuracies will be driven by the phenotypes of the trait of economic importance. This means collection of carcass data will continue to be important. However, ultrasound and genomic data allow a producer to add a lot of accuracy to an animal’s EPDs much younger in its life, thus should be collected where breed appropriate. They should only be shown in a single ERT EPD since calculations by geneticists who came up with the ERT concept demonstrated that when all are shown separately as indicator traits, accuracy of selection can actually be decreased. I think gathering as much accuracy early for bulls will be especially important this year. With record feeder calf prices and lower cow inventory, we may be looking at all-time average prices paid for bulls next spring for all breeds; however, it will probably be for fewer bulls. — Dr. Bob Hough [Dr. Bob Hough has served as the executive vice president of the Red Angus Association of America and more recently as executive vice president of the North American Limousin Foundation from 2009 to early 2011. He is now a consultant, a freelance writer and semi-retired.]