Functional annotation and Bayesian fine-mapping reveals candidate genes for important agronomic traits in Holstein bulls

Authors: JIANG, JICAI - University Of Maryland; Cole, John; FREEBERN, ELLEN - University Of Maryland; DA, YANG - University Of Minnesota; Vanraden, Paul; MA, LI - University Of Maryland

Submitted to: Communications Biology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/29/2019
Publication Date: 6/18/2019
Citation: Jiang, J., Cole, J.B., Freebern, E., Da, Y., Van Raden, P.M., Ma, L. 2019. Functional annotation and Bayesian fine-mapping reveals candidate genes for important agronomic traits in Holstein bulls. Communications Biology. 2:212. https://doi.org/10.1038/s42003-019-0454-y.
DOI: https://doi.org/10.1038/s42003-019-0454-y

Interpretive Summary: Genome-wide association studies are used to identify regions of the genome that are associated with the control of traits of economic importance. Detailed study of those regions leads to improved understanding of biology, as well as better models for predicting animal genetic merit. In this study, a new approach -- Bayesian Fine-Mapping – was used to determine how 3 million DNA variants in more than 27,000 Holstein bulls were associated with 35 production, reproduction, and body conformation traits. This new methods method also makes use of the predicted functional changes associated with each variant. For example, some DNA variants produce substantial changes in the protein sequence coded for in a gene, while other changes have little or no apparent effect on proteins. Including that information helped identify genes associated with these traits that had not been identified in previous studies. The use of 3 million DNA variants allows us to pinpoint with a high degree of accuracy the regions of the genome that are associated with regulation of many important traits in dairy cattle.

Technical Abstract: mputation has been routinely applied to ascertain sequence variants in large genotyped populations based on reference populations of sequenced individuals. With increasing numbers of animals sequenced and the implementation of the 1000 Bull Genomes Project, fine-mapping of causal variants for complex traits is becoming possible in cattle. After quality control and LD pruning, we imputed 3 million selected sequence variants to over 27,000 Holstein bulls. These bulls were selected to have highly reliable phenotype (breeding values) for 35 production, reproduction, and body conformation traits. We first performed whole-genome single-marker scan for the 35 traits using the mixed-model based association tests. The single-trait association statistics were then merged into multi-trait analyses of 3 groups of traits, production, reproduction, and body conformation, respectively. Both single- and multi-trait GWAS results were used to pick 2-Mb candidate genomic regions for fine-mapping studies. We developed a fast Bayesian Fine-MAPping approach (BFMAP) to fine-map the candidate genomic regions to single-gene resolution and to integrate fine-mapping with functional enrichment analysis. Our fine-mapping identified many promising candidate genes for dairy traits, including ABCC9 VPS13B, MGST1, SCD, MKL1, CSN1S1 for production traits, CHEK2, GC, KALRN for reproduction traits, and TMTC2, ARRDC3, ZNF613, CCND2, FGF6 for body conformation traits. Based on existing functional annotation data available for the cattle genome, we revealed biologically meaningful enrichment in our fine-mapped variants that can be readily tested for future functional validation. Collectively, we developed a fast Bayesian approach for fine-mapping and enrichment analysis, generated a list of candidate genes and variants of complex traits for functional studies, and expanded our understanding of the genetic basis of complex traits in dairy cattle.