A Variational Bayes Genomic-Enabled Prediction Model with Genotype × Environment Interaction

Abstract

There are Bayesian and non-Bayesian genomic models that take into account G×E interactions. However, the computational cost of implementing Bayesian models is high, and becomes almost impossible when the number of genotypes, environments, and traits is very large, while, in non-Bayesian models, there are often important and unsolved convergence problems. The variational Bayes method is popular in machine learning, and, by approximating the probability distributions through optimization, it tends to be faster than Markov Chain Monte Carlo methods. For this reason, in this paper, we propose a new genomic variational Bayes version of the Bayesian genomic model with G×E using half-t priors on each standard deviation (SD) term to guarantee highly noninformative and posterior inferences that are not sensitive to the choice of hyper-parameters. We show the complete theoretical derivation of the full conditional and the variational posterior distributions, and their implementations. We used eight experimental genomic maize and wheat data sets to illustrate the new proposed variational Bayes approximation, and compared its predictions and implementation time with a standard Bayesian genomic model with G×E. Results indicated that prediction accuracies are slightly higher in the standard Bayesian model with G×E than in its variational counterpart, but, in terms of computation time, the variational Bayes genomic model with G×E is, in general, 10 times faster than the conventional Bayesian genomic model with G×E. For this reason, the proposed model may be a useful tool for researchers who need to predict and select genotypes in several environments.

A large-scale genome-wide enrichment analysis identifies new trait-associated genes, pathways and tissues across 31 human phenotypes*

“…Although these methods are computationally convenient in large datasets, and often produce reliable results (e.g. [8,10,[107][108][109][110][111][112][113][114][115]), they also have features to be aware of. One feature is that when multiple SNPs in strong LD are associated with a trait, the variational approximation tends to select one of them and ignore the others.…”

Multi-environment Genomic Prediction of Plant Traits Using Deep Learners With Dense Architecture

“…These three data sets are made up of a total of 309 maize lines which were used by Crossa et al (2013) and Montesinos-López et al (2016, 2017). Traits evaluated were grain yield (GY; data set 1), anthesis-silking interval (ASI; data set 2), and plant height (PH; data set 3); each of these traits was measured in three environments (Env1, Env2, and Env3).…”

Benchmarking between item based collaborative filtering algorithm and genomic best linear unbiased prediction (GBLUP) model in terms of prediction accuracy for wheat and maize//Estudio comparativo en términos de capacidad predictiva para datos de trigo y maíz entre el algoritmo de filtrado colaborativo y el modelo genómico mejor predictor lineal insesgado (GBLUP)

“…This data set is based on that of Montesinos-López et al (2017). It is composed of a sample of 309 maize lines evaluated for three traits: anthesis-silking interval (ASI), plant height (PH), grain yield (GY), each of them evaluated in three optimal storm environments (Env1, Env2 and Env3).…”

“…We have identified this data set as Maize. For more details, see the study of Montesinos-López et al (2017).…”