Predição genômica para o caráter mocho em bovinos da raça brahman utilizando uma metapopulação

Abstract

The absence of horns, or analogously the polled trait, is a characteristic that not only ensures better animal welfare conditions by reducing the dehorning process but also contributes to economic indices in the beef industry due to the reduction of injuries and bruising in leather and muscle tissue. Therefore, this study aims to perform genomic prediction for the polled trait in Brahman cattle, as well as identify the effects that influence prediction ability and estimates of heritability, accuracy, bias, and dispersion of genomic values for this trait through the use of a metapopulation. This study was conducted with phenotypic records related to the polled trait, along with pedigree and genomic information for Brahman and Nellore cattle, which were provided by Regalito, Guaporé (OB Brand), and Estâncias Espiritú farms participating in the genetic improvement program of the National Association of Breeders and Researchers (ANCP). Different phenotypic classifications, obtained through visual assessment, were analyzed to evaluate the variance components and genomic prediction ability of horn development in Brahman cattle. For this purpose, analysis models were adopted that included different strategies for including genetic effects. Two scenarios were considered: one involving only Brahman cattle and another simultaneously comprising Nellore and Brahman breeds, as well as their metafounders. For variance component estimates, a Bayesian approach model was used with the aid of Gibbsf90+ software. In genomic prediction for the polled trait, the threshold model was used through the single-step genomic best linear unbiased predictor method (ssGBLUP). Heritability estimates obtained for the polled trait in Brahman cattle, using single-breed models, showed moderate to high magnitude, ranging from 0.49 to 0.72, with the highest estimate observed with the binary model and the lowest estimate corresponding to the model where observed phenotypes were adopted. The genomic prediction model consisting of sex effects and SNPs present on sex chromosomes, associated with the use of alternative phenotypes and the adoption of the multiracial population and their metafounders, showed the best predictive capacity for the polled trait in Brahman cattle. The use of single-breed models was not appropriate for predicting the genomic value of this trait for the breed under study. The results obtained indicate the high potential for increasing the allelic frequency of genes associated with the polled phenotype in Brahman cattle populations.