Bayesian Additive Regression Trees for Genotype by Environment Interaction Models

We propose a new class of models for the estimation of Genotype by Environment (GxE) interactions in plantbased genetics. Our approach, named AMBARTI, uses semiparametric Bayesian Additive Regression Trees to accurately capture marginal genotypic and environment effects along with their interaction in a fully Bayesian model. We demonstrate that our approach is competitive or superior to the traditional AMMI models widely used in the literature via both simulation and a real world data set. Furthermore, we introduce new types of visualisation to properly assess both the marginal and interactive predictions from the model. An R package that implements our approach is available at https://github.com/ebprado/ambarti .