Predicting Hybrid Performance in Complex Scenarios: The G2F Gxe Maize Project.

The differential sensitive of certain plant genotypes to different environmental influences, also known as genotype by environment (GXE) interaction, creates significant challenges for the deployment of whole genome prediction tools in a plant breeding context. Using the framework of the Genomes to Fields (G2F) G X E Maize project, the goal of this work is to assess the potential of predicting performance between and within environments through the incorporation of environmental component information together with the GXE component. Between 2014 and 2015, the G2F G X E Maize project evaluated a collection of hybrids involving close to 1,000 diverse maize inbreds across more than 40 unique sites in North America. Hybrids were evaluated for relevant phenological and agronomic characteristics. Climatic information such as temperature, precipitation, solar radiation and wind speed, among others was also collected in all locations. Overall, the inclusion of robust climatic information increases prediction ability, but the prediction of extreme scenarios (i.e., genotypes in unobserved environments) remains challenging.