The accuracy of genomic breeding value for production trait in Iranian Holstein Dairy Cattle using parametric and non-parametric methods

Genomic Selection (GS) is a tool for prediction of breeding values for quantitative traits. For a successful application of GS, accuracy of predicted genomic breeding value (GEBV) is a key issue to consider. Here we investigated the accuracy of GEBV in 345 genotyped Iranian Holstein cattle. The study was performed on milk, fat, protein yield and somatic cell count. Four methods G-BLUP, Bayes B, Reproducing kernel Hilbert Spaces (RKHS) and Neural Networks (NN) were used to predict genomic breeding values and their accuracies. The GEBV accuracies varied between 0.39 for somatic cell count to 0.73 for fat yield. Bayes B gave the highest accuracies among methods. Bayes B and non- parametric methods tended to produce inflated predictions ( slope of the regression of GBV on EBV greater than 1). However, in all traits, lower estimates of MSE were obtained using G- BLUP. Bayes B regression model are of interest for future applications of genomic selection in this population, but further improvements are needed to reduce deflation of their predictions.