Abstract #384

# 384
Genome-wide association studies for fertility traits in Brown Swiss cattle using single SNP regression and Bayesian approaches applied to high-density SNP array information.
Beat Bapst*1, Christine F. Baes1,2, Franz R. Seefried1, Marlies A. Dolezal5, Christine Flury2, Heidi Signer-Hasler2, Dorian Garrick3, Christian Stricker4, Johann Sölkner7, Alessandro Bagnato6, Ingolf Russ8, Klemen Potocnik9, Birgit Gredler1, 1Qualitas AG, Zug, Switzerland, 2Bern University of Applied Sciences, Zollikofen, Bern, Switzerland, 3Iowa State University, Ames, IA, 4agn Genetics, Davos, Graubünden, Switzerland, 5University of Veterinary Medicine Vienna, Vienna, Austria, 6Dept. Vespa University of Milan, Milan, Italy, 7University of Natural Resources and Life Sciences, Vienna, Austria, 8Tierzuchtforschung e.V. (TZF), Poing, Bavaria, Germany, 9University of Ljubljana, Ljubljana, Slovenia.

Large-scale, in-depth, genome-wide analyses of various economically important traits are now possible due to the availability of medium (50K; 54,609 SNP) and high-density (HD; 777,962 SNP) SNP arrays. Genome-wide association studies (GWAS) provide a powerful tool for identifying associations between phenotypes and variants in the underlying genome. Here we present GWAS results for 5 fertility traits in Brown Swiss cattle. Traits analyzed included non-return-rates in heifers (NRR_H) and cows (NRR_C), days to first service (DFS), and intervals between first and last insemination in heifers (IFL_H) and in cows (IFL_C). Imputation from 50K to HD was carried out with FImpute. After filtering out SNPs with minor allele frequency <0.05, a total of 654,847 imputed SNPs on 29 bovine autosomes remained. Deregressed breeding values of 1,502 (IFL_H) to 3,379 (NRR_C) progeny-tested bulls were used as phenotypes. Single SNP GWAS (M1) were conducted with the SNP & Variation Suite package GoldenHelix. The Efficient Mixed-Model Association expedited (EMMAX) algorithm was applied to account for population stratification. Additionally, bulls with more than 75% Original Braunvieh pedigree background were excluded. Significance levels were determined by false discovery rates. Multi-SNP GWAS (M2) using Bayesian methodology implemented in GenSel were also conducted. In those analyses, all SNPs were fitted simultaneously. A 1 Mb non-overlapping window approach that accumulated contributions of adjacent SNPs was used to identify associated genomic regions. The M1 analyses using genome-wide significance levels showed that regions on BTA3, 17, and 25 were associated with NRR_C, a region on BTA29 was associated with DFS, and a region on BTA8 was associated with IFL_C. The significant SNPs in these regions were also represented in the 1 Mb windows accounting for the largest proportion of genetic variance in the corresponding traits from M2. The promising results presented here provide evidence that specific regions of the genome are associated with fertility traits in Brown Swiss cattle.

Key Words: GWAS, fertility, dairy cattle