Genomic prediction of male fertility in Brown Swiss cattle

Abstract

Bull fertility has been recognized as an important factor affecting dairy herd fertility. The objective of this study was to assess the feasibility of predicting male fertility in Brown Swiss cattle using genomic data. The dataset consisted of 1,102 Italian Brown Swiss bulls with sire conception rate (SCR) records and genotype data for roughly 480k SNP. The analyses included the use of linear kernel-based regression models fitting all SNPs or incorporating markers with large effect. Predictive performance was evaluated in 5-fold cross-validation using the correlation between observed and predicted SCR values and mean squared error of prediction. The entire SNP set exhibited predictive correlations around 0.19. Interestingly, the inclusion of 2 markers with large effect yielded predictive correlations around 0.32. Overall, using linear kernel-based models fitting markers with large effect is a promising approach. Our findings could help Brown Swiss breeders make enhanced genome-guided management and selection decisions on male fertility.