Genetic evaluations: Refining the toolbox

The evolution of performance measurements continues to progress, thanks in part to the influence of genomic data.

March 20, 2019

5 Min Read
Genetic beef toolbox
Guzaliia Filimonova / Getty Images

By B. Lynn Gordon

All around us, technology is dramatically changing the way we live. Cell phones have become mini-computers and having a camera in the barn to monitor calving is no longer a specialty item but now is standard. Measuring and calculating beef cattle performance data is not exempt from the influx of new technology as well.

In the 1980s, performance data was generated with pedigree and phenotype data. In the 90s, the industry gained a greater understanding of genetic markers and utilized them to help predict genomic-enhanced EPDs (GE-EPD).

Fast forward to 2017 where single-step predictions incorporate genomic data (generated from DNA) along with pedigree and phenotype information to result in genetic predicted EPDs.

“One of the primary goals of genetic evaluation has been and always will be to increase prediction accuracy—between leveraging DNA and evolving statistical methodology and computational strategies, we have recently made a quantum leap in prediction accuracy,” says Wade Shafer, Ph.D, executive secretary of the American Simmental Association.

The progression

The progression to a single-step evaluation has not come without questions, as the industry tries to wrap its arms around the best way to measure data and the best methods to educate cattlemen about genetic evaluations. As genetic suppliers, seedstock producers have been challenged to keep up with the pace of new technology to measure their cattle’s performance and help their customers interpret the changing evaluation systems.

The shift to single-step EPDs, which most breed associations have adopted within the past two years, has producers trying to embrace shifts in the numerical values, whether positive or negative, in their herd sires, cowherds or bloodlines associated with their program.

“The 1980s ushered in a revolution in modern genetic evaluation. Though the revolution was underpinned by statistical methodology created by Charles Henderson in the late ‘60s and ‘70s, it wasn’t until the mid ‘80s that computers were powerful enough to utilize it on breed-sized databases, Shafer says.

“From that point to this day, the animal breeding community has advanced Henderson’s methodology and leveraged the ever-increasing power of computing to improve the accuracy of genetic prediction.”

Shafer further summarizes, “Since the late ‘90s, our industry has struggled with leveraging the information made available by advancements in DNA technology. For the first decade, DNA results were analyzed by DNA labs and the end product was used independently of the National Cattle Evaluation (NCE). This era produced no progress in NCE from DNA information.”

The following decade brought about period in which information was harvested from DNA tests via stand-alone procedures and subsequently blended with NCE results from the traditional evaluation. Though the blending approach had definite shortcomings, it did improve the predictive accuracy of the NCE.

“That said, we have long known the single-step approach to be the superior means of utilizing DNA in genetic evaluation. To implement single-step, however, advances in methodology along with computational translation were needed. Those things have come to pass, which has allowed single-step platforms to be utilized in the NCE – which represents a monumental step in the evolution of genetic evaluation,” Shafer says.

Approach of single-step analysis

Lee Leachman, president of the Beef Improvement Federation (BIF) and owner of Leachman Cattle of Colorado, Fort Collins, Colo., agrees. “The single-step analysis is the best way to incorporate DNA information into the EPDs on animals. Additionally, single-step requires that the entire breed’s database is analyzed each week to incorporate DNA and the latest performance information,” he says.  

“By transition to a single-step platform and leveraging associated advancement in methodology, we are able to influence substantially more information from a DNA test than previously possible,” says Shafer. “We are seeing accuracy increases equivalent to adding 25-progeny for some traits.”

The knowledge and accessibility to this amount and quality of data, especially on young sires, promoted much of the interest of breed associations to adopt single-step evaluations.

Comparing EPDs

This shift to single-step genetic evaluations has led to questions, such as, “Can I still compare the previous EPDs on my herd sire to his new numbers?” “My cow herd numbers went up (or down), why?” “Did this calculation make that much difference across the breed?” 

Leachman explains the differences in the two systems. “First, the accuracy is now higher on young animals than it was before. Secondly, the animal’s own performance record now contributes less to their EPDs, given all of the information we get from one-step DNA incorporation. 

“For example, you will find bulls that have birth weights that would normally preclude them from being used on heifers, but DNA might say that they indeed are heifer bulls.”

As a genetic supplier and leader of the industry’s primary performance-based organization, BIF, Leachman understands the importance of educating bull buyers and customers about the industry’s changes in the EPD calculations.

“I think the bottom line is that we need to convey that EPDs on genomic-enhanced animals are significantly more accurate than are those on animals whose EPDs are not genomic-enhanced. For example, at Leachman Cattle of Colorado, we will not use an unproven purebred bull unless that animal has genomic-enhanced EPDs.” 

“One thing I like to remind producers of is that there is nothing magic about DNA," says Shafer. "It is a source of information that can add a significant amount of accuracy to genetic prediction in one fell swoop.

“That said, at this point we cannot achieve high levels of accuracy via DNA tests. Substantial numbers of phenotypes are still required,” he adds.

“Further, for DNA tests to be effective, large numbers of phenotypes are a necessity — so keep collecting them. Do not let DNA take your eye off the ball. Other than the fact that we can now achieve substantially higher accuracy on young animals, nothing has changed. The sound animal breeding principles adopted for decades are as valid now as they have ever been — so stick to the game plan.”

Gordon is a freelance agricultural writer and regular contributor to BEEF from Sioux Falls, S.D.

Editor's Note: There are two primary single-step evaluations currently being used. One utilizes software from the University of Georgia and calculates data for Angus Genetics Inc. primarily. The second is the BOLT software calculating data for the American Hereford Association and International Genetic Solutions.

Subscribe to Our Newsletters
BEEF Magazine is the source for beef production, management and market news.

You May Also Like