Robust association tests under different genetic models, allowing for binary or quantitative traits and covariates.

Area of Science:

• Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Genetic association studies commonly use additive models, which can reduce power if the true genetic model is misspecified.
• Existing robust association tests primarily focus on binary traits and lack covariate handling.
• Accurate genetic model specification is crucial for reliable outcome association analysis.

Purpose of the Study:

• To develop an analytic approach for robust genetic association tests using the MAX3 statistic.
• To extend robust association testing to accommodate quantitative and binary traits with covariates.
• To implement the methodology in an accessible R package for broader application.

Main Methods:

• Utilized the maximum of three test statistics (MAX3) under additive, dominant, and recessive models.
• Developed an analytic method for p-value adjustment to maintain Type I error rates.
• Validated theoretical calculations against bootstrap resampling procedures.

Main Results:

• The developed analytic approach for MAX3 robust association tests accurately calculates p-values.
• The methodology effectively handles both quantitative and binary traits in the presence of covariates.
• The p-values derived from theoretical calculations closely align with bootstrap resampling results.

Conclusions:

• The MAX3 approach offers a more robust method for genetic association testing compared to single-model tests.
• The RobustSNP R package provides a versatile tool for robust association analysis in various study designs, including Genome-Wide Association Studies (GWAS).
• This methodology enhances the power and reliability of genetic association studies.