Detection of genomic pre-selection with Mendelian sampling variance test

Abstract

Genomic pre-selection of bull calves is potentially one additional source of bias in the international evaluations, given it has not been adequately accounted for in the national evaluations. The aim of this simulation study was to investigate whether it is possible to detect the effect of genomic pre-selection in the MS means or variances obtained by the MS-trend validation test. In total, 50 replicates were generated under control and genomic pre-selection schemes utilizing structures of the real data and pedigree from a medium size cow population during a 20-year time interval. All bulls in the last 10 birth year classes were assumed to be genomically pre-selected. Therefore, their Mendelian sampling terms were inflated with a value corresponding to the selection of best 10% of the genomically tested bull calves. After start of genomic pre-selection, both the true and estimated means of Mendelian sampling terms and breeding values rose sharply in bulls, although the effects of estimated means were more moderate. A clear decrease was found in true and estimated genetic variances of bulls, but the effect was temporary and thus hard to exploit. Daughters of genomically pre-selected bulls had higher true and estimated breeding values compared to the control scheme, only slightly elevated MS means and no effect in the genetic variances.