GEInfo

GEInfo implements quasi-likelihood and penalization methods to integrate prior literature-based information for enhanced gene-environment (G-E) interaction analysis within linear, logistic, and Poisson regression frameworks.

Key Features:

• Incorporation of Prior Information: Integrates prior information from published literature into G-E interaction models using a quasi-likelihood plus penalization approach to improve power in high-dimensional settings with weaker interaction signals.
• Model Extensions: Extends the quasi-likelihood and penalization framework to linear, logistic, and Poisson regression models for analysis of continuous, binary, and count outcomes.
• Alternative Method Implementations: Provides implementations of alternative analytical methods to enable direct comparison of different G-E interaction approaches.
• Visualization Tools: Includes visualization routines for exploring estimated interaction effects and comparing results across methods.

Scientific Applications:

• Genetic Epidemiology: Identification and characterization of gene-environment interactions underlying complex disease etiology.
• Risk Stratification and Precision Medicine: Informing risk assessment based on combined genetic and environmental profiles.
• Translational Research: Supporting application of G-E interaction analyses in both basic research and clinical study contexts.

Methodology:

Integration of prior information via a quasi-likelihood framework combined with penalization techniques, applied within linear, logistic, and Poisson regression models to address high-dimensionality and weak interaction signals.