Grouping genotypes by a cluster method directly related to genotype-environment interaction mean square

Area of Science:

• Agricultural Science
• Genetics
• Biostatistics

Background:

• Genotype by environment (GE) interactions complicate the interpretation of experimental data.
• Large GE interactions obscure the identification of superior genotypes across diverse environments.

Purpose of the Study:

• To propose a novel cluster method for grouping genotypes based on their response patterns across environments.
• To facilitate the interpretation of GE experimental data, particularly when GE interactions are substantial.

Main Methods:

• A dissimilarity index is defined using distance adjusted for average genotype effects.
• Sokal and Michener's unweighted pair-group method is employed within the clustering algorithm.
• The index's equivalence to within-group GE interaction mean square under 2-way ANOVA is demonstrated.

Main Results:

• The proposed index effectively quantifies genotype dissimilarity in response to environments.
• Using an F-value as a stopping criterion results in clusters with no significant GE interaction.
• Genotypes within identified clusters can be reliably compared based on their average effects.

Conclusions:

• The developed cluster method simplifies the analysis of GE interactions.
• This approach enables more straightforward identification and comparison of genotypes across different environmental conditions.
• The method provides a statistically sound framework for interpreting complex GE data.