E-GP

E-GP integrates enviromic assembly with genomic data to refine whole-genome predictions (GP) by modelling environmental relatedness and genotype-by-environment interactions for plant breeding.

Key Features:

• Enviromic Assembly Approach: Uses the envirotype concept to define environment quality through core environmental typologies and their frequencies and derives markers that reflect environmental similarity.
• Integration with Genomic Data: Combines additive and dominance genomic effects and genomic kinships with enviromic data to represent phenotypic variation and capture phenotypic plasticity.
• Optimized Multi-Environment Trials (MET): Implements a genetic algorithm scheme to design optimized METs that integrate enviromic assembly with genomic kinships and enable in-silico simulation of genotype-environment combinations for phenotyping.
• Applications in Phenotypic Data Management: Enhances GP model training under scarce phenotypic information across diverse environments and aids early screening for yield plasticity via optimized phenotyping strategies.
• Validation and Performance: Demonstrated in tropical maize where E-GP outperformed benchmark GP models across scenarios, emphasizing genotype-environment representativeness over MET size for training accuracy.
• Theoretical Insights: Characterizes how envirotype-phenotype covariances within phenotypic records influence GP accuracy and affect predictive breeding approaches.

Scientific Applications:

• Plant breeding: Improves genomic prediction accuracy and selection decisions by integrating environmental relatedness into GP models.
• Phenotyping strategy optimization: Guides optimized MET design and in-silico selection of genotype-environment combinations to prioritize phenotyping for yield plasticity.
• Climate-adaptive predictive breeding: Leverages environmental databases and enviromic assembly to anticipate future scenarios and support development of genotype-environment representative models for resilient crop varieties.

Methodology:

Applies enviromic assembly using envirotype typologies and frequencies, integrates additive and dominance genomic effects and genomic kinships with enviromic markers, employs a genetic algorithm to optimize MET design, and performs in-silico simulations of genotype-environment combinations.