Impact of Multiple Sire Mating System on the Accuracy of Genomic Breeding Value Prediction in a Beef Cattle Population Under Selection

The objective of this study was to investigate the application of BLUP and ssGBLUP in different scenarios of uncertain paternity using data from a Nellore cattle population. The analyzed data set was provided by the National Association of Farmers and Researchers (ANCP). The data set contained information from 18 Nellore herds located in the southeast and mid-west regions of Brazil, which participate in the ANCP breeding program. A total of 60,325 records for weight adjusted at 450 days (W450) were used. The mean values ± standard deviations were as follows: 290.20 ± 50.26 kg and the contemporary groups (CG) were and defined as farm, year of birth, season of birth, sex and management group. Records with values above or below the range of 3.5 standard deviations from the CG mean were excluded, as well as CGs with less than five animals. The variance components were estimated using BLUP and ssGBLUP methods. The relationship matrix (A) was created with different proportions of animals with unknown sires (0, 25, 50, 75, and 100% of multiple sires). All models included contemporary groups as fixed effects. The breeding value (EBV/GEBV) accuracy was calculated according to BIF and evaluated in each scenario with eight groups of animals: ALL = all animals in the population, BULL = only bulls with ten or more progenies; GEN = genotyped animals, GENwithPHEN = genotyped animals with phenotypes, GENwithoutPHEN = genotyped animals without phenotypes, YOUNG = male and female young animals without phenotypes, YwithoutGEN = young animals without phenotypes and genotypes, and YwithGEN = young animals without phenotypes and with genotypes. The additive genetic variance decreases as the proportion of multiple sire increased in the population for both methods. Prediction accuracies ranged from 0.02 to 0.46 and from 0.12 to 0.48 for BLUP and ssGBLUP, respectively. In general, for all scenarios, the EBV/GEBV prediction accuracy decreased as the proportion of MS in the population increased, however, the decrease of EBV accuracy was more intense compared to GEBV accuracy which was of 8.8%, 4.3%, 46.2%, 27.3%, 82.4%, 25.0%, 18.8% and 87.5% for ALL, BULL, GEN, GENwithPHE, GENwithoutPHE, YOUNG, YwithoutGEN and YwithGEN groups, respectively. The breeding values for W450 were influenced by presence of the paternity uncertainly in the pedigree. The presence of paternity uncertainly affects more intensively the breeding value of young animals. The genotyped young animals were benefited from the application of ssGBLUP, particularly in situations with missing pedigree.