Identifying gene-environment interactions across genome-wide, twin, and polygenic risk score approaches

Area of Science:

• Behavioral Genetics
• Quantitative Genetics
• Genomics

Background:

• Genome-wide gene-environment interaction (GxE) research has faced challenges due to low statistical power and skepticism.
• Twin and polygenic risk score (PRS) studies indicate that GxE is widespread and significantly impacts complex genetic traits.

Purpose of the Study:

• To demonstrate consistent findings across twin, PRS, and genome-wide approaches for identifying GxE.
• To evaluate the strengths and limitations of each method in detecting GxE.

Main Methods:

• Conducted simulation studies to generate datasets applicable to twin, PRS, and genome-wide association studies (GWAS).
• Compared the consistency and power of different methodologies in identifying gene-environment interactions.

Main Results:

• All three approaches (twin, PRS, and genome-wide) successfully detected GxE, highlighting methodological consistency.
• Genome-wide approaches identify specific interacting variants but can lack power for polygenic traits.
• PRS methods aggregate effects but may underestimate genetic signals if discovery sample power is low; twin studies are robust to polygenicity and effect distribution.

Conclusions:

• Multiple valid methods exist for detecting GxE, stemming from shared assumptions about complex trait genetic architecture.
• The robustness of these diverse methods in identifying genomic moderation underscores the pervasiveness of GxE.