Submitted to: Journal of Dairy Science
Publication Type: Abstract Only
Publication Acceptance Date: 2/24/2011
Publication Date: 6/30/2011
Citation: Wiggans, G.R., Van Raden, P.M., Cooper, T.A. 2011. Adjustment of deregressed values from cow evaluations to have the similar mean and variance as bull deregressed values. Journal of Animal Science 89(E-Suppl. 1)/Journal of Dairy Science 94(E-Suppl. 1):165(abstr. 37).
Technical Abstract: Traditional evaluations of cows with genotypes have been adjusted since April 2010 to be compatible with evaluations of bulls to improve their accuracy in estimation SNP effects. This adjustment made them incomparable with traditional evaluations of cows. Recent work has improved the adjustment by using a bull only direct genomic value to group cow according to the amount of adjustment required. This adjustment is not applied to all cows because the grouping requires a genotype and adjustments based on genotyped cows probably would over adjust other cows. To create an adjustment for all cows, Mendelian sampling (MS), the difference between PTA and parent average (PA), was calculated for milk, fat, and protein. These were deregressed by dividing by a function of reliability with the reliability from parent average removed. Standard deviations (SD) of these values were grouped by reliability. A linear function of reliability was estimated as a multiplicative adjustment to reduce the SD from cows to that of bulls with similar reliability. Averages of PA by birth year were subtracted from PA to create within year PA deviation groups and mean deregressed MS was calculated for bulls and cows by group. These means varied across PA deviation groups, those for bulls fell and those for cows rose. The differences were fit by linear regression on PA deviation and used to adjust cow deregressed MS. Processing cows and bulls in birth date order enabled propagation of the adjusted PTA to the PA of progeny. This adjustment reduced the PTA of cows with high PA and increased those of cows with low PA, but did not change the estimates of genetic trend because the adjustment was within birth year. It also reduced the within birth year variance of cow evaluations. This adjustment does not replace the adjustment of evaluations of genotyped cows because the use of bull only direct genomic values more precisely tailors the adjustment to the individual cow. This improved the comparability of evaluations of cows with and without genotypes.