Genomic prediction for additive and dominance effects of censored traits in pigs

Abstract

ABSTRACT. Age at the time of slaughter is a commonly used trait in animal breeding programs. Since studying this trait involves incomplete observations (censoring), analysis can be performed using survival models or modified linear models, for example, by sampling censored data from truncated normal distributions. For genomic selection, the greatest genetic gains can be achieved by including non-additive genetic effects like dominance. Thus, censored traits with effects on both survival models have not yet been studied under a genomic selection approach. We aimed to predict genomic values using the Cox model with dominance effects and compare these results with the linear model with and without censoring. Linear models were fitted via the maximum likelihood method. For censored data, sampling through the truncated normal distribution was used, and the model was called the truncated normal linear via Gibbs sampling (TNL). We used an F 2 pig population; the response variable was time (days) from birth to slaughter. Data were previously adjusted for fixed effects of sex and contemporary group. The model predictive ability was calculated based on correlation of predicted genomic values with adjusted phenotypic values. The results showed that both with and without censoring, there was high agreement between Cox and linear models in selection of individuals and markers. Despite including the dominance effect, there was no increase in predictive ability. This study showed, for the first time, the possibility of performing genomic prediction of traits with censored records while using the Cox survival model with additive and dominance effects.

Genome-Enabled Prediction of Genetic Values for Using Radial Basis Function Neural Networks

“…In general, the results showed a decrease in accuracy when dominance increased in both methods. This result differs from those obtained by some authors (Denis et al, 2011;Almeida-Filho et al, 2016;Santos et al, 2016;Toro and Varona, 2018), who reported that the incorporation of the dominance component in the linear model used lead to an improvement in the prediction process of complex traits. Concerning the RMSE values, for G-BLUP, the higher the influence of dominance, the higher the RMSE value obtained, and the adjustment of the model to the effects of dominance greatly diminishes the estimates of the error.…”

Subset selection of markers for the genome-enabled prediction of genetic values using radial basis function neural networks

“…These results suggest that the degree of simulated epistasis, in which only dual interactions between subsequent markers are considered, was not a determining factor in differentiating the fit of regression models and neural networks. In terms of dominance, as already reported in the literature (Long et al, 2007;Denis & Bouvet, 2011;Almeida Filho et al, 2016;Santos et al, 2016;Xu et al, 2018), that factor is not regarded as a problem in genomic prediction studies. Therefore, even if nonparametric models (such as artificial neural networks) do not need to impose strong assumptions upon the phenotype-genotype relationship, presenting the potential to capture interactions between loci by the interactions between neurons of different layers (Gianola, Fernando, & Stella, 2006;2010), a substantial improvement in the prediction process depends on the level of epistasis present.…”

Subset selection of markers for genome-enabled prediction of genetic val-ues using radial basis function neural networks

“…These results suggest that the degree of simulated epistasis, in which only dual interactions between subsequent markers are considered, was not a determining factor in differentiating the fit of regression models and neural networks. In terms of dominance, as already reported in the literature [22,23,24,25,26], that is not regarded as a problem in genomic prediction studies. Therefore, even if non-parametric models such as artificial neural networks do not need to impose strong assumptions upon the phenotype-genotype relationship presenting the potential to capture interactions between loci by the interactions between neurons of different layers [27,2], a substantial improvement in the prediction process depends on the level of epistasis present.…”