Comparing a Mixed Model Approach to Traditional Stability Estimators for Mapping Genotype by Environment Interactions...

Area of Science:

• Agronomy
• Genetics
• Plant Breeding

Background:

• Genotype by environment (GxE) interactions significantly influence crop productivity, particularly in soybeans (Glycine max).
• Numerous methodologies exist to quantify yield stability and model GxE interactions, but their correspondence in genome-wide association studies (GWAS) is unclear.
• Understanding these interactions is crucial for developing stable and high-yielding crop varieties.

Purpose of the Study:

• To compare GWAS results for soybean yield and yield stability using different analytical approaches.
• To determine the utility and intersection of traditional stability measures versus direct GxE interaction modeling in GWAS.
• To identify genetic loci associated with yield and yield stability in soybean.

Main Methods:

• Conducted GWAS on 213 soybean lines across 11 environments.
• Employed univariate and multivariate conventional stability estimates as phenotypes.
• Utilized a mixed model incorporating marker by environment interactions as a random effect for yield.

Main Results:

• Discovered 106 total quantitative trait loci (QTL) across all analyses.
• Genetic loci significant in the mixed model for yield (including GxE) were distinct from those identified using traditional stability measures.
• QTL identified through direct GxE interaction mapping explained more yield variance and often caused genotype rank changes between environments.

Conclusions:

• Explicitly mapping GxE interactions in GWAS is more effective for identifying loci with significant impact on soybean yield variance and stability.
• Traditional stability measures may not capture the full picture of GxE interactions relevant for breeding.
• Investigating GxE interactions in multiple contexts is essential for effectively manipulating them in soybean breeding programs.