Increasing Accuracy of Genomic Predictions for Economically Important Traits in Dairy Cattle

Milk production and reproductive performance are two economically important traits contributing to the profitability and overall efficiency in the dairy cattle industry. Major improvements in selection for these traits have been achieved with the advent of high-throughput genotyping and the availability of SNP information. This progress allows for genomic selection (GS) to be applied, which relies on linkage disequilibrium (LD) between genotyped SNPs and casual mutations. However, GS based on functional SNP genotypes should be less reliant on LD patterns shared by training and validation populations and might lead to more accurate genomic predictions. Developing gene networks and biological pathways from genome wide association studies (GWAS) of multiple traits may help define sets of interrelated genes and their underlying pathways to further define these functional SNPs. In this study, GWAS results for milk production and fertility traits as well as conformation, auxiliary, and health traits were used to identify positional candidate genes and then functional candidate genes were selected, using a variety of bioinformatics tools. Indels and SNPs of potential importance in those genes were identified through annotation of variants obtained from whole genome re-sequencing, RNA-seq and public databases. From the set of annotated variants more than 135,000 markers suitable for inclusion in a custom Affymetrix genotyping panel were selected. The performance of this panel will be evaluated by genotyping approximately 1300 animals for which phenotypes are already available.