Genomic selection has been widely used in terrestrial animals but has had limited application in aquaculture due to relatively high genotyping costs. Genomic information has an important role in improving the prediction accuracy of breeding values, especially for traits that are difficult or expensive to measure. The purposes of this study were to (i) further evaluate the use of genomic information to improve prediction accuracies of breeding values from, (ii) compare different prediction methods (BayesA, BayesCπ and GBLUP) on prediction accuracies in our field data, and (iii) investigate the effects of different SNP marker densities on prediction accuracies of traits in the Portuguese oyster (Crassostrea angulata). The traits studied are all of economic importance and included morphometric traits (shell length, shell width, shell depth, shell weight), edibility traits (tenderness, taste, moisture content), and disease traits (Polydora sp. and Marteilioides chungmuensis). A total of 18,849 single nucleotide polymorphisms were obtained from genotyping by sequencing and used to estimate genetic parameters (heritability and genetic correlation) and the prediction accuracy of genomic selection for these traits. Multi-locus mixed model analysis indicated high estimates of heritability for edibility traits; 0.44 for moisture content, 0.59 for taste, and 0.72 for tenderness. The morphometric traits, shell length, shell width, shell depth and shell weight had estimated genomic heritabilities ranging from 0.28 to 0.55. The genomic heritabilities were relatively low for the disease related traits: Polydora sp. prevalence (0.11) and M. chungmuensis (0.10). Genomic correlations between whole weight and other morphometric traits were from moderate to high and positive (0.58–0.90). However, unfavourably positive genomic correlations were observed between whole weight and the disease traits (0.35–0.37). The genomic best linear unbiased prediction method (GBLUP) showed slightly higher accuracy for the traits studied (0.240–0.794) compared with both BayesA and BayesCπ methods but these differences were not significant. In addition, there is a large potential for using low-density SNP markers for genomic selection in this population at a number of 3000 SNPs. Therefore, there is the prospect to improve morphometric, edibility and disease related traits using genomic information in this species.
27 June 2019 in Journal of Animal Breeding and Genetics
Gilmour, Thompson, and Cullis (Biometrics, 1995, 51, 1440) presented the average information residual maximum likelihood (REML) algorithm for efficient variance parameter estimation in the linear mixed model. That paper dealt specifically with traditional variance component models, but the algorithm was quickly applied to more general models and implemented in several REML packages including ASReml (Gilmour et al., Biometrics, 2015, 51, 1440). This paper outlines the theory with respect to these more general models, describes the main issues encountered in fitting these models and how they have been addressed in the ASReml software. The issues covered are the basics steps in the implementation of the algorithm, keeping parameters within the parameter space, maximizing sparsity, avoiding issues associated with unstructured variance matrices by using the factor‐analytic structure and handling singularities in marker‐based relationship matrices and current work.
Genetic parameters for test-day milk yield, lactation persistency, and age at first calving (as a fertility trait) were estimated for the first 4 lactations in multiple-breed dairy cows in low-, medium-, and high-production systems in Kenya. Data included 223,285 test-day milk yield records from 11,450 cows calving from 1990 to 2015 in 148 herds. A multivariate random regression model was used to estimate variance and covariance components. The fixed effects in the model included herd, year, and test month, and age as a covariate. The lactation profile over days in milk (DIM) was fitted as a cubic smoothing spline. Random effects included herd, year, and test month interaction effects, genetic group effects, and additive genetic and permanent environmental effects modeled with a cubic Legendre polynomial function. The residual variance was heterogeneous with 11 classes. Consequently, the variance components were varied over the lactation and with the production system. The estimated heritability for milk yield was lower in the low-production system (0.04-0.48) than in the medium- (0.22-0.59) and high-production (0.21-0 60) systems. The genetic correlations estimated between different DIM within lactations decreased as the time interval increased, becoming negative between the ends of the lactations in the low- and medium-production systems. Low (0.05) to medium (0.60) genetic correlations were estimated among first lactation test-day milk yields across the 3 production systems. Genetic correlations between the first lactation test-day milk yield and age at first calving ranged from 0.27 to 0.49, 0 to 0.81, and -0.08 to 0.27 in the low-, medium-, and high-production systems, respectively. Medium to high heritabilities (0.17-0.44) were estimated for persistency, with moderate to high (0.30-0.87) genetic correlations between 305-d milk yield and persistency. This indicates that genetic improvement in persistency would lead to increased milk yield. The low to medium genetic correlations between test-day milk yield between production systems indicate that sires may be re-ranked between production systems. Therefore, we conclude that sires should be selected based on a genetic evaluation within the target production system.
1. The heritability (h2) of liveweight (LW) in ostriches can be highly variable, depending on age at recording. The objective of this study was to consider random regression (RR) as an alternative to the multi-trait (MT) structure for the analysis of repeated measures of LW.2. The data included 74 683 LW phenotypes recorded from 10 052 birds aged between 20 and 410 days (d) of age. Statistical analysis included single trait (ST), MT and RR analysis in a linear mixed model framework using the ASREML V4.2 software.3. For ST and MT, six traits were defined to represent LW at 28, 77, 150, 230, 300 and 365 d of age. Random variance components included direct genetic and maternal permanent environment (PE) effects. A MT analysis including all six traits converged.4. For RR, the data was transformed (LW + 10)-0.5 due to difficulty in dealing with large scale effects. The final RR model fitted direct genetic and animal PE components as third degree Legendre polynomials and heterogeneous residuals.5. The h2 estimates was in agreement across analysis, ranging from moderate (0.16-0.20) for W28 to high (0.41-0.51) for W230 to W365. Importantly, the genetic relationship between LW recorded as a chick and juvenile was only moderate (~0.35 to 0.55). The correlations between RR and MT EBVs for the six traits were 0.85, 0.54, 0.65, 0.75, 0.83 and 0.91, showing a considerable level of re-ranking.6. This study reaffirmed age dependent genetic variation when determining LW in ostriches. The RR structure was useful for overcoming the dimension problem of MT analysis, but was susceptible to scale effects present in the data, despite transformation. It remains unknown whether the need for cubic terms reflected scale or animal effects.