Identifying Polymorphisms Underlying QTLs in Bos taurus using GWAS and Whole Genome Sequence Data

Genomic selection in beef cattle has the potential for improving response to selection. However, the accuracy of molecular estimates of breeding value (MEBV) diminishes in advanced generations of selection if phenotypes and genotypes are not routinely captured and the prediction model retrained. A partial solution is to identify and include causal variants in the MEBV prediction model. In this study, we investigate the potential for identifying the large effect causal polymorphisms which underlie quantitative trait loci in Angus cattle. We performed a genome-wide association study (GWAS) for sixteen traits by genotyping 3,570 animals with the BovineSNP50 assay and estimated SNP allele substitution effects (ASEs) by genomic best linear unbiased prediction. Prospective large effect QTLs were next identified using graphical analysis of the ASEs for each of the 16 traits. QTL genotypes were then estimated for each individual in the GWAS using their SNP genotypes within a 5 SNP window centered on the most likely position of each QTL. Eleven of these individuals were whole genome resequenced to an average depth of ~24X coverage and aligned to the UMD3.1 reference assembly. Within each QTL region we screened the sequenced bulls for polymorphisms that were concordant with predicted QTL genotypes and labeled those that matched in 10 or more as potential causal polymorphisms. Our next step will focus on annotation information about the set of QTL candidates and the development of an assay to facilitate their genotyping in several breeds.