Genotype environment interaction effects in multi environment models for Fusarium head blight resistance in wheat

Abstract Fusarium head blight (FHB) is a prevalent wheat disease causing yield losses and mycotoxin contamination threatening food security and safety. As such, disease resistant cultivars must be developed through multi-environment replicated field experiments. In the case of FHB, disease resistance is quantitative and greatly influenced by genotype x environment (G × E) interactions (GEI). In this sense, it could be beneficial to include GEI in genomic prediction models, which are commonly used to select the candidate lines using genome-wide markers and pedigree information. In the present study, we used six different mixed models, three each for single trait and multi-trait scenario; Environment and Line (EL), Environment and Genomic information (EG) and Environment, Genomic information and Genomic information x Environment (EG&GxE) for performing genomic predictions in two spring wheat breeding panels. These models were used to predict sets of lines within an environment and across environments. The EG&GxE model, which includes GEI, enhanced prediction abilities in many cases for both single trait and multi-trait scenarios. Borrowing information from the genotypes evaluated in correlated environments could help the models to predict the performance of genotypes in new environments. Overall, GEI contributed to higher predictive abilities, in addition to main genetic and environmental effects.