Simplification of complex environmental variations on maize-phenotype predictability

Crop yield is driven by the crop's genotypes, how the crop is managed, and the environment in which the crop is grown. Advances in sensor technology and data acquisition have unleashed our ability to measure in fine detail a crop's growing environment. However, harnessing such massive amounts of data can be a limiting factor in discovering, advancing, and transforming phenotype predictability. This project is the first step in a larger effort to understand how environmental conditions and genomes affect crop growth and phenotype predictions. We hypothesized that an environmental index in phenotypic assessments can integrate the environmental complexity into a single holistic metric in each location enabling unbiased phenotypic comparisons of hybrid's stability across environments. The data being used for this inquiry extracted from the Genomes to Fields (G2F) Initiative, which includes the genomic, phenotypic, and environmental data for a set of maize hybrids grown in multiple locations across the US since 2014. This study's objectives are two-folded (1) to use a yield stability index to simplify the complexity of environmental variation on maize yields; and (2) to develop a python-based code that illustrates how such an approach can be used to compare hybrid performance across locations and time. For this, we explore yield stability as a measure of constant yield output across trials and environments. The simplification of environmental variations on the yield -the stability index-was applied on 21 maize hybrids across 109 environments. Using this approach, we found that 11 of 21 of tested hybrids display above-average stability, with one also displaying above-average yield. Future research will investigate the specific contributions to these environmental indices, such as weather factors near-critical crop development stages.