Detecting Gene-Environment Interactions for a Quantitative Trait in a Genome-Wide Association Study

Area of Science:

• Genetics
• Biostatistics
• Epidemiology

Background:

• Genome-wide association studies (GWAS) primarily detect marginal genetic effects.
• Complex traits often result from gene-environment (G × E) interactions, which may involve SNPs with weak marginal effects.
• Standard genome-wide interaction scans (GWIS) have low power, especially after correcting for multiple testing.

Purpose of the Study:

• To address the low power and potential invalidity of existing two-step GWIS methods.
• To propose a novel, valid, and more powerful two-step GWIS approach.
• To identify novel genetic associations with complex traits influenced by G × E interactions.

Main Methods:

• Evaluation of existing two-step GWIS methods, including the Paré et al. approach screening on variance heterogeneity.
• Development and simulation of a novel two-step GWIS method combining marginal effect and variance heterogeneity screening.
• Application of the proposed method to a G × ethnicity scan for childhood lung function.

Main Results:

• The Paré et al. approach demonstrates an inflated false-positive rate when an environmental marginal effect is present.
• The proposed novel two-step GWIS method shows improved power and a valid false-positive rate in simulations.
• A single nucleotide polymorphism (SNP) near the MARCO locus associated with childhood lung function was identified, missed by previous marginal-effect scans.

Conclusions:

• The novel two-step GWIS method offers a more powerful and valid approach for detecting G × E interactions compared to existing methods.
• This method can identify important genetic associations missed by traditional GWAS.
• The findings highlight the importance of considering G × E interactions in genetic studies of complex traits.