The number of genotyped animals within the American Angus Association database has increased more during the last four years than it previously ever had. Such growth means the “single-step
machine” is ready for maintenance as computing times get slower, according to Andre Garcia, Angus Genetics Inc. (AGI). Regular maintenance of models, genetic parameters and algorithms helps ensure the data systems are accurate, efficient and up to date. On May 28, AGI will run an update to the core group of animals to allow for a more accurate representation of the current genotyped population.
Since 2017, AGI and the American Angus Association have worked together to implement and run the single-step genomic best linear unbiased prediction (ssGBLUP) machine. This machine uses genomic information to build a genomic relationship matrix (GRM). GRM’s are then combined with traditional pedigrees to increase the accuracy of the evaluations. However, a challenge in running these evaluations is presented when the number of genotyped animals is more than 150,000.
The University of Georgia created a solution to this challenge as they developed a procedure called “Algorithm for Proven and Young” (APY). APY sets a group of reference animals to represent the entire genotyped population. This group is used to calculate the inverse of the GRM for all genotyped animals. APY has been used since the “single-step machine” was implemented at the Association in 2017 to run the weekly evaluations of millions of animals and more than 950,000 genotyped animals.
Read more about the “single-step machine," why these updates are taking place and how this affects Angus breeders in Garcia’s “By The Numbers” column in the May 2021 issue of the Angus Journal.
Editor’s note: If you have questions regarding these updates, contact an AGI customer service representative at 816.383.5100 or at angus@angus.org.