Investigation of reliability of genomic predictions in the admixed Nordic Red dairy cattle

Authors

• Mahlako L. Makgahlela Department of Agricultural Sciences, P. O. Box 27 FIN-00014 University of Helsinki, Finland
• E. A. Mäntysaari MTT Agrifood Research Finland, Biotechnology and Food Research, Biometrical Genetics, 31600 Jokioinen, Finland
• I. Strandén MTT Agrifood Research Finland, Biotechnology and Food Research, Biometrical Genetics, 31600 Jokioinen, Finland
• M. Koivula MTT Agrifood Research Finland, Biotechnology and Food Research, Biometrical Genetics, 31600 Jokioinen, Finland
• U.S. Nielsen 3Danish Agricultural Advisory Service, Udkaersvej 15, 8200 Aarhus, Denmark
• M. J. Sillanpää Department of Mathematics and Statistics, P.O. Box 68 FIN-00014 University of Helsinki, Finland; Department of Mathematical Science and Department of Biology, P.O. Box 3000 FIN-90014 University of Oulu, Finland
• J. Juga Department of Agricultural Sciences, P. O. Box 27 FIN-00014 University of Helsinki, Finland

DOI: https://doi.org/10.33354/smst.75539

Keywords:

reliability, genomic breeding values, admixed breeds, breed proportions

Abstract

The success of genomic selection (GS) in small breeds which are likely to have admixed structures has been minimal. This is because accuracy of GS depends on the extent of linkage disequilibrium (LD) between markers and quantitative trait loci (QTL) and LD depends on the genetic structure of the population and marker density. In the current study, we evaluate reliability of genomic predictions in young unproven bulls, when interactions between marker effects and breed of origin are accounted for in the Nordic Red dairy cattle (RDC). The population structure of the RDC is admixed. Data consisted of animal breed proportions calculated from the full pedigree, deregressed proofs (DRP) of published estimated breeding values (EBV) for yield traits and genotypic data for 37,595 SNP markers. Direct genomic breeding values (DGV) were estimated using 2 models, one accounting for breed-specific effects and other assuming uniform population. Validation reliabilities were calculated as the squared correlation between DRP and DGV (r2DRP, DGV), corrected by the mean reliability ofDRP. Using the breed-specific model increased the reliability of DGV by 2% and 3% for milk and protein, respectively, when compared to homogeneous population GBLUP model. The exception was for fat, where there was no gain in reliability. Estimated validation reliabilities were low for milk (0.32) and protein (0.32) and slightly higher (0.42) for fat.