The objectives of this study were to 1) evaluate marker associations for genomic regions of interest and significant ontology terms, 2) evaluate and compare 4 models for their efficacy in predicting genetic merit, 3) evaluate and compare the impact of using breed-of-origin genotypes in a Bayesian prediction model, and 4) evaluate the effects of data partitioning using family structure on predictions. Nellore-Angus F2, F3 and half-sibling calves were used with records for overall temperament at weaning (OTW; a subjective scoring system; n = 769) and Warner-Bratzler shear force (WBSF; a measure of tenderness; n = 389). After filtering, 34,913 markers were available for use. Bayesian methods employed were BayesB (using ̂) and BayesC (using π = 0 and ̂) in GenSel software, where, after estimation, π ̂ = 0.995 or 0.997 for WBSF or OTW, respectively. No regions associated with either trait were found using π ̂, but when π = 0 associated regions were identified (37 and 147 regions for OTW and WBSF, respectively). Comparison of genomic estimated breeding values from these 3 Bayesian models to an animal model showed that BayesC procedures (using ̂) had the highest accuracy for both traits, but that BayesB had the lowest indication of bias in either case. Using a subset of the population (n = 440), genotypes based on the breed in which the alleles originated from (i.e., breed-of-origin genotypes) were assigned to markers mapped to autosomes (n = 34,449), and incorporated into prediction analyses using BayesB (π ̂ = 0.997) with or without nucleotide-based genotypes. In either case, there was an increase in accuracy when breed-of-origin genotypes were incorporated into prediction analyses. Data partitions based on family structure resulted in 13 distinct training and validations groups. Relationship of individuals in the training with validation individuals did have an impact in some cases, but not all. There was poor prediction of genomic estimated breeding values for individuals in the validation population using BayesB methods, but performed better in all cases than breeding values generated using an animal model. Future studies incorporating breed-of-origin genotypes are of interest to determine if accuracy is improved in these groups.

Thesis (M.Sc. Agriculture (Animal Production)) -- University of Limpopo, 2020 / Stillbirth is a trait of high economic importance in dairy cattle and is increasingly being included in dairy cattle breeding objectives worldwide. In South Africa, however, there is limited information on stillbirth that can be used to improve this trait genetically. Currently, there are no estimated breeding values (EBVs) for any measures of calving performance produced under the national genetic evaluation programme. The current study was, therefore, conducted to assess the incidence of stillbirth and estimate the genetic and environmental influences on maternal effects for stillbirth in South African Holstein cattle, to enable estimation of breeding values for the trait. Data used in the study comprised 13 143 calving records of 7 723 Holstein cows, from 41 herds, participating in the National Dairy Animal Recording and Improvement Scheme during the period 2014 to 2018. Incidence of stillbirth was determined using the PROC FREQ procedure and environmental effects were tested by the General Linear Models (GLM) procedure of Statistical Analysis System (SAS 9.4, 2016). Maternal heritability of stillbirth was estimated by the Restricted Maximum Likelihood (REML) procedure, using the ASReml software (Gilmour et al., 2018). The analyses were carried out using a threshold animal model and a repeatability animal model, where the latter considered stillbirth in different parities as repeated measures of the same trait. Environmental effects significantly influencing stillbirth (p<0.05) were herd-year season of calving, dam parity and calf sex, and these were included in the model for variance component estimation. Estimates of maternal heritability effects from the threshold animal model were 0.12±0.04, 0.15±0.08 and 0.14±0.06 for parities 1 to 3, respectively. The repeatability animal model gave a heritability estimate of 0.09±0.03 and a repeatability of 0.18±0.03. The moderate estimates of maternal heritability indicate scope for reducing incidence of stillbirth by selectively breeding cows that are less genetically predisposed to calving dead calves. Stillbirth in different parities should not be considered as the same trait, as indicated by the low repeatability estimate. Results of the current study estimate genetic parameters that are required to compute accurate estimated breeding values (EBVs) for stillbirth, which will enable South African Holstein farmers to select for reduced stillbirths, thus improving calving performance / National Research Foundation (NRF), University of Limpopo (UL) and Department of Agriculture Forestry and Fisheries (DAFF)