Fully efficient, two-stage analysis of multi-environment trials with directional dominance and multi-trait genomic...

Area of Science:

• Plant Breeding
• Quantitative Genetics
• Bioinformatics

Background:

• Genomic selection (GS) is crucial for crop improvement.
• Plant breeders often struggle to leverage complex multi-environment and multi-trait data effectively.
• Existing software may require advanced programming skills or lack specific functionalities.

Purpose of the Study:

• To introduce R/StageWise, a novel package for efficient two-stage genomic selection analysis.
• To enable breeders to utilize multi-environment, multi-trait datasets, including dominance and polyploidy.
• To provide a user-friendly tool that bypasses the need for advanced programming.

Main Methods:

• Implemented a fully efficient two-stage analysis, using the full variance-covariance matrix from Stage 1 in Stage 2.
• Incorporated directional dominance analysis, including for polyploid species, to address inbreeding depression.
• Developed multi-trait selection indices, with options for restricted indices to constrain trait responses.

Main Results:

• StageWise analysis significantly improved model fit (reduced AIC) for potato traits (maturity, yield, fry color) by incorporating Stage 1 errors into Stage 2.
• Dominance heterosis explained a small proportion of variation (0.03 for yield), likely due to limited genomic inbreeding coefficient variation.
• Restricted selection indices successfully identified coefficients to maximize genetic merit without negatively impacting maturity, despite a high genetic correlation between yield and maturity.

Conclusions:

• R/StageWise provides a powerful and accessible solution for advanced genomic selection in plant breeding.
• The package facilitates the effective use of complex datasets, including those with dominance effects and polyploidy.
• StageWise enables sophisticated breeding strategies, such as restricted selection indices, for optimized crop genetic improvement.