Evaluation of genotyping strategies in situations of uncertainty paternity and impact on genomic evaluation in beef cattle

Multiple service sire (MS) is the most common mating system in extensive beef production systems mainly due the facility and low management cost. However, MS does not allow the paternity identification, which makes the incompleteness of pedigree one of the main obstacles for accurate genetic evaluations. There is a grown interest to investigate the use of genomic data in uncertain paternity models aiming to increase the accuracy and to decrease bias in genetic evaluations. Therefore, the objective of this study was to investigate the application of BLUP and single step genomic BLUP (ssGBLUP) under different scenarios of uncertain paternity, using simulated and real data in beef cattle population. Genotypes, pedigree, and phenotypes for age at first calving (AFC) and weight at 550 days (W550) were simulated using heritabilities based on real data (0.12 for AFC and 0.34 for W550). The simulated genome had a total length of 2,333 cM, with 735,293 biallelic markers and 7,000 QTLs randomly distributed over 29 BTA. It was assumed that QTLs explained 100% of the genetic variance. For QTL, the amount of alleles per loci ranged randomly from two to four. Uncertain paternity scenarios using 0%, 25%, 50%, 75%, and 100% were studied. Four ways of scaling the mean of the genomic matrix (G) to match the mean of the pedigree relationship matrix among genotyped animals (A22) were tested. Accuracy, bias and inflation were investigated for five groups of animals: ALL = all animals; BULL = only bulls; GEN = genotyped animals; FEM = females; YOUNG = young males. The use of genomic information in the model (ssGBLUP) provided more accurated prediction (ranging from 0.31 to 0.97) than traditional BLUP (ranging from 0.05 to 0.97), especially in the YOUNG group. In real data, all models included contemporary groups and age at calving in classes as fixed effects. The accuracy of the estimated breeding value (EBV/GEBV) prediction was calculated 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. Accuracies of EBV (BLUP method) ranged from 0.02 to 0.46 for W450 and 0.04 to 0.18 for AFC, while the accuracies of GEBV (ssGBLUP) ranged from 0.13 to 0.48 for W450 and 0.16 to 0.33 for AFC. The results obtained in simulated and real data showed that EBV and GEBV accuracy decreased as the proportion of MS increased. Additionally, the use of genomic information in the genetic evaluation by ssGBLUP increases the accuracy of evaluation, especially for animals with few number of information, such as young animals. (AU)