Monday, December 9, 2013

Angus Association Announces Third Recalibration of Zoetis HD 50K Prediction

The American Angus Association announced the 3rd recalibration of the Zoetis HD 50K product. The previous recalibration was based on about 40,000 animals and this recalibration is based on about 51,000 animals.I want to highlight a few points from Crystal Albers' interview with Dr. Kent Andersen of Zoetis and Tonya Amen of Angus Genetics Inc.
First, genomic predictions and genomic-enhanced EPDs are self-improving. Every animal tested can be used for the next round of improvements and recalibration.
Second, genomic predictions reduce the risk and improve the accuracy of purchase decisions for commercial producers.
Third, for the first time the HD 50K product produces genomic-enhanced EPDs for heifer pregnancy.
Fourth, Zoetis recognizes the need for more aggressive marketing of animals with genomic-enhanced EPDs to see a greater return on the investment. They use a website called GenomeXchange where their customers have the opportunity to list information about their operation and advertise cattle that have been tested.

Don't hesitate to contact me if you would like to discuss how genomic technologies could fit into your operation.

*Note: this is not an endorsement of a particular product, simply a discussion of changes in genomic prediction.


Jared Decker said...

From a post on
how and why did any re-sortings or rankings occur?
Genomic-enhanced EPDs (and all EPDs in general) are predictions based on random effects. Random effects always have a certain degree of uncertainty, and with EPDs we express this uncertainty as an EPD accuracy. When we add additional data to random effects the predictions change. When genomic predictions are re-estimated the random effects change. This is especially significant for families within the breed have not been previously sampled. When these families are sampled the predictions for animals within these lineages change. This is the major source of re-ranking in recalibration.

Jared Decker said...

A more technical explanation. When new haplotypes are sampled from the population the genomic predictions become more robust. Animals which previously had inaccurate predictions because they carry unsampled haplotypes now have more accurate estimates.