Effect of selection for maternal and direct genetic effects on genetic improvement of litter size in swine.

Optimum weighting ratios of maternal:direct EBV for litter size using an animal model were examined to achieve maximum genetic improvement (direct plus maternal response). Stochastic simulation of a 120-sow herd over a 10-yr period of selection was used (20 replicates). Directional selection was based on a merit function of maternal and direct EBV for first-parity litter size. Optimum weighting ratios for maternal to direct EBV in pure- and crossbreeding schemes with different genetic correlations between maternal and direct effects were obtained. Genetic gain in maternal and direct effects was more sensitive to change in weighting ratios of maternal to direct estimates of breeding values under an animal model than earlier theoretical studies showed for selection index. In the purebreeding scheme, the weighting ratios of maternal:direct effects of 1:1, 1.25:1, and -.5:1 resulted in the highest overall response of 3.11, 1.73, and .69 pigs after 10 yr of selection with genetic correlations between maternal and direct effects of 0, -.5, and -.9, respectively. In the crossbreeding scheme with a male dam line selected for direct effects only and a female dam line selected for an optimum weighting ratio of maternal and direct effects the overall response was always higher with 3.19, 1.89, and 1.31 pigs/10 yr for the genetic correlations 0, -.5, and -.9, respectively, than in the purebreeding scheme. With a large negative correlation between maternal and direct effects a meaningful overall response of litter size was achieved only in the crossbreeding scheme. The effect of negative weighting of maternal effects to increase direct response with a large negative correlation between maternal and direct effects was also examined as well as the influence of weighting ratios on accuracy of evaluation, additive genetic variance, prediction error variance, and correlation between maternal and direct EBV.