DEVELOPMENT OF AN APPROXIMATE MULTIVARIATE TWO-STEP APPROACH FOR THE JOINT GENETIC EVALUATION OF AUSTRIAN AND GERMAN DAIRY CATTLE

Abstract

Multivariate genetic evaluation in modern dairy cattle breeding programs became important in the last decades. The simultaneous estimation of all production and functional traits is still demanding. Different meta-models are used to overcome several constraints. The aim of this study was to conduct an approximate multivariate two-step procedure applied to de-regressed breeding values and yield deviations of five fertility traits of Austrian Pinzgau cattle and to compare results with routinely estimated breeding values. The approximate two-step procedure applied to de-regressed breeding values performed better than the procedure applied to yield deviations. Spearman rank correlations for all animals, sires and cows were between 0.996 and 0.999 for the procedure applied to de-regressed breeding values and between 0.866 and 0.995 for the procedure applied to yield deviations. The results are encouraging to move from the currently used selection index in routine genetic evaluation towards an approximate two-step procedure applied to de-regressed breeding values.

Keywords

• approximate multiple trait
• de-regressed breeding values
• yield deviation
• fertility
• cattle