Genomic prediction for reproductive, carcass and feed efficiency traits in Nelore cattle using biological information and different criteria in the incorporation of candidate genomic regions

Abstract

The project aims to evaluate the inclusion of inserted SNPs in possible causative QTNs for genomic evaluation of reproductive, carcass and efficiency traits using the ssGBLUP method; to use genome-wide association studies (GWAS) in a population of Nelore cattle using phenotypes related to the probability of early delivery (PP30), stayability (STAY), rib eye area (REA), subcutaneous fat thickness (SFT) and residual feed intake (RFI); to explore the genomic regions identified by the GWAS using functional analysis methods to detect the main candidate genomic regions that explain the additive genetic variance of the studied phenotypes; insert in the analysis of prediction SNPs present in possible causative regions of the phenotypes studied, based on the literature and determine the impact on the accuracy of the genomic predictions by the inclusion of SNPs inserted in causative genomic regions. The data set used in this study will be composed of Nelore animals from a breeding program, the National Association of Breeders and Researchers (ANCP, Ribeirão Preto, Brazil). The database includes records of 66,496 females and their relatives, totaling approximately 176,000 phenotypic records for reproductive, carcass and food efficiency traits. A total of approximately 30,000 animals were genotyped with the low density panel. For imputation, a reference population will be used with 963 founder animals genotyped with the Illumina BovineHD BeadChip. Markers for genotyped and reference populations were excluded for call rate (<90%), lower allele frequency (MAF <0.05%), Mendelian conflicts (> 1.0%) and for markers located on non-autosomal chromosomes. Two methods for genetic evaluation: PBLUP and ssGBLUP, will be used. The covariance components and genetic parameters will be estimated considering a linear animal model (carcass and feed efficiency traits) and a threshold animal model (reproductive traits). The G matrix will be constructed using different combinations of SNPs and weighting. The Pearson correlation between the EBV (obtained using the complete data set) and the DGV (estimated in the validation subset, corresponding to the predictive capacity of the genetic evaluation) will be estimated. The results of this study provide an opportunity to analyze the effects of insertion of SNPs near or within candidate genomic regions or causative QTNs in the accuracy of genetic assessments, thus allowing a more accurate selection of animals with higher genetic merit, resulting in increased genetic progress. (AU)