Genome Wide Association using Imputed Sequence Data in Dairy Cattle with the 1000 Bull Genomes Project Data Set

Genome wide association (GWAS) with whole genome sequence data on phenotyped individuals would have the advantage that the causative mutations are actually in the data set, as opposed to GWAS that use SNP on standard arrays, which relies on (usually imperfect) linkage disequilibrium between these SNP and the causative mutations.  However, the cost of sequencing is such that the very large numbers of animals required for GWAS to identify mutations affecting complex traits are unlikely to be sequenced.  An alternative strategy is to sequence key ancestors of the population, then impute the genotypes for the sequence variants into much larger reference sets with phenotypes and SNP panel genotypes.  The 1000 Bull Genomes Project is building this database of sequenced key ancestor bulls for the bovine research community.  Run3.0 of the project included 429 full genome sequences of key ancestors  of 15 breeds, sequenced at an average of 10.5 fold coverage.  There were 31.8 million filtered sequence variants detected in the sequences, including 29.1 million SNP and 1.7 million insertion-deletions.  Agreement of sequence genotypes to genotypes from a 800K SNP array in the sequenced bulls was excellent at 98.8%.  To demonstrate the approach, we  imputed sequence data into dairy cattle populations, genotyped with 800K SNP arrays, in regions where QTL affecting milk production and other traits had previously been identified.  Association studies with this imputed sequence data identified putative causative mutations, some of which were supported by gene expression data (RNASeq).  These early results are encouraging and suggest the 1000 Bull Genomes project will be a critical resource for GWAS, and genomic prediction, in dairy and beef cattle in the future.