GbyE

GbyE integrates genetic and environmental data to detect gene-by-environment interactions and improve genome-wide association studies (GWAS) and genomic selection (GS).

Key Features:

• Detection of Significant Genetic Loci: Uses a Power-FDR curve to identify significant loci across varying levels of genetic correlation and heritability, including low heritability scenarios.
• Separate Estimation of Additive and Interactive Effects: Employs a genotype design model constructed via the Kronecker product to distinctly estimate additive and interactive effects.
• Enhanced Prediction Accuracy: Improves genomic selection prediction accuracy and increases performance of BRR (Bayesian Ridge Regression), BayesA, and BayesLASSO by 9.4%, 9.1%, and 11% respectively.
• Robustness Across Environments: Maintains superior prediction performance even when data from a single environment are absent, particularly under high genetic correlation and heritability.
• Versatility in Phenotypic Prediction: Predicts phenotypes across single and multiple environments for inference populations with incomplete or variable environmental data.

Scientific Applications:

• GWAS Enhancement: Reveals gene-by-environment interactive effects to increase statistical power and identify additional significant markers in GWAS.
• Genomic Selection Improvement: Increases prediction accuracy for breeding programs and other applications requiring precise genomic predictions.
• Understanding Complex Traits: Elucidates interactive relationships between genes and environments to aid study of traits influenced by both genetics and environmental factors.

Methodology:

Applies a genotype design model based on the Kronecker product to separate additive and interactive effects, uses a Power-FDR curve for locus detection, and evaluates prediction performance using BRR, BayesA, and BayesLASSO.