Genomics for phenotype prediction and management purposes

Pre-assuming accuracies of genomic breeding values larger than 0.7 for a moderate heritability production trait, and larger than 0.5 for a low heritability functional trait, additional profit from genotyping female calves or heifers compensates costs for genotyping in commercial herds. Herd management will be improved by including SNP information into electronically mating software, e.g., through the exploitation of non-additive genetic effects and via controlling of inbreeding and genetic relationships. Random forest methodology can infer binary disease phenotypes in validation sets with moderate accuracy, also for a small number of diseased genotyped animals in training sets. Genomic random regression models can be used to predict genomic breeding values for animals without phenotypes in, e.g., harsh environments.