Interpreting genotype-by-environment interaction using redundancy analysis.

Area of Science:

• Quantitative Genetics
• Statistical Modeling

Background:

• Interpreting genotype-by-environment (GxE) interaction with measured environmental variables is complex.
• Existing methods either extract environmental characteristics or directly incorporate measured variables into models.

Purpose of the Study:

• To present and theoretically describe redundancy analysis (RA) for interpreting GxE interaction.
• To demonstrate RA's derivation from existing methods like singular-value decomposition and factorial regression.
• To highlight RA's utility, especially with concomitant environmental information.

Main Methods:

• Redundancy analysis is presented as a method that directly incorporates measured environmental variables.
• RA is shown to be derivable from singular-value decomposition of residuals from additivity.
• RA is also shown to be derivable from factorial regression through axis rotation and dimensionality reduction.

Main Results:

• Redundancy analysis is positioned as a member of the second group of GxE interpretation methods.
• The paper provides a theoretical treatment of redundancy analysis.
• A practical example compares RA results with other mentioned methods.

Conclusions:

• Redundancy analysis is a valuable tool for interpreting genotype-by-environment interaction.
• RA is particularly useful when considering concomitant environmental information.
• The method offers a direct way to integrate measured environmental variables into GxE models.