Understanding Genotype × Environment × Management Interactions and Their Implementation in Predictive Breeding

Genotype × environment × management (G×E×M) interactions are of major importance because they shape the stability and adaptability of crop varieties and have a substantial impact on agricultural productivity and food security. In the past, genotypes exhibiting strong G×E×M interactions were commonly removed from consideration. Given the pressures imposed by climate change and restrictions on fertilizer inputs, leveraging these interactions can contribute to greater yield stability. However, modeling G×E×M interactions is challenging due to high collinearity among environmental covariates and the high dimensionality of the data. Factor analytic models have proven to be a flexible and robust approach for modeling genetic variance-covariance matrices, providing a more parsimonious alternative to unstructured models in multi-environment trial analyses.
The project utilizes integrated factor analytic mixed models that incorporate both observed and latent environmental covariates to disentangle genotype × environment interactions and estimate overall performance and stability parameters. These models also enable predictions for previously untested genotypes and future growing seasons. Moreover, they allow the decomposition of marker effects into components that are either stable across environments or specific to particular environmental and management conditions. In summary, a thorough understanding of genotype × environment × management interactions is crucial for breeding crop varieties that are resilient, adaptable, and high yielding across diverse environmental and management scenarios.