Detecting gene-environment interactions from multiple continuous traits

Area of Science:

• Genetics
• Biostatistics
• Epidemiology

Background:

• Gene-environment interactions (GxE) influence disease risk, varying with environmental exposures.
• Complex diseases often manifest with multiple continuous traits (e.g., obesity, diabetes).

Purpose of the Study:

• To develop and evaluate a multivariate scale test (MST) for detecting GxE in diseases with multiple traits.
• To identify specific traits and environmental factors contributing to significant GxE signals.

Main Methods:

• Developed a multivariate scale test (MST) to analyze GxE across multiple continuous traits.
• Compared the statistical power of MST against the univariate scale test (UST) via simulations.
• Applied MST to large-scale genetic data from the Taiwan Biobank.

Main Results:

• MST demonstrated potential for increased power compared to UST by integrating multiple traits and employing a less stringent multiple testing penalty.
• Application to Taiwan Biobank data identified 18 independent variance quantitative trait loci and 41 significant GxE signals across eight trait domains.
• The study analyzed over 2.5 million SNPs in nearly 119,000 individuals.

Conclusions:

• MST is a powerful tool for uncovering complex gene-environment interactions in diseases characterized by multiple traits.
• The identified GxE signals provide insights into the genetic architecture of complex diseases.
• The developed methodology and findings are publicly available for further research.