Genetic parameters and accuracy of traditional and genomic breeding values for resistance to infectious bovine keratoconjunctivitis in Hereford

Comin, H. B.; Campos, G. S. [UNESP]; Domingues, R.; Gaspar, E. B.; Sollero, B. P.; Cardoso, F. F.

Genetic parameters and accuracy of traditional and genomic breeding values for resistance to infectious bovine keratoconjunctivitis in Hereford

Data

2022-10-01

Autores

Tipo

Artigo

Resumo

The objectives of this research were to estimate genetic parameters using a threshold animal model by traditional (pedigree and phenotype), single-step genomic BLUP (ssGBLUP) and Weighted ssGBLUP (WssGBLUP) methods, and assess the relative increase in accuracy with the addition of genomic information on the estimation of breeding values for the incidence of infectious bovine keratoconjunctivitis (IBK) in Hereford cattle. The data used in this study were obtained from animals belonging to four farms of Delta G breeders association and genetically evaluated by the Delta G breeding program, born between 2015 and 2017, totaling 293,716 individuals in the pedigree data and 1,579 phenotypes for IBK (1 = absence of disease or 2 = presence of disease). The genotypic data, after quality control, retained 37,505 SNP markers and 1,401 samples. A Bayesian framework and a threshold animal model were used to estimate variance components and to predict breeding values. Animals with genotype and phenotype for the traits studied were divided into 3 groups by random and k-means clustering cross-validation strategies. The estimate of direct heritability for resistance to IBK was 0.12 ± 0.01 and 0.10 ± 0.01 through PBLUP and ssGBLUP methods. The mean realized prediction accuracies based on linear regression statistics were 0.36, 0.65 and 0.70, respectively for PBLUP, ssGBLUP and WssGBLUP; therefore, the use of the genomic relationship matrices was superior to the pedigree-based method. When using the WssGBLUP method, there was an additional increase in prediction accuracy when compared to the ssGBLUP method, which was already substantially better than PBLUP. The prediction of breeding values using genomic information by the single-step methods should result in more accurate selection decisions compared to traditional pedigree evaluations and, consequently, genomic selection can be readily applied with moderately high reliability to help reduce the incidence of IBK in Hereford cattle.