Robust joint analysis with data fusion in two-stage quantitative trait genome-wide association studies

Area of Science:

• Genetics
• Statistical genetics
• Bioinformatics

Background:

• Genome-wide association studies (GWASs) are crucial for identifying disease-associated genetic variants.
• Two-stage designs are common in GWASs, but robust methods for quantitative traits are limited.
• Existing robust methods primarily focus on binary traits, leaving a gap for quantitative trait analysis.

Purpose of the Study:

• To develop a powerful, robust joint analysis method for quantitative traits in two-stage GWASs.
• To address the challenge of calculating statistical significance for robust tests in quantitative trait analysis.
• To improve the detection of genetic variants associated with quantitative traits under model uncertainty.

Main Methods:

• Developed a novel F-statistic-based robust joint analysis method.
• Utilized combined raw data from both stages of two-staged GWASs.
• Derived explicit expressions for calculating statistical significance and power.

Main Results:

• The proposed method demonstrates substantial robustness compared to the F-test based on the additive model when the genetic model is unknown.
• Simulations confirm the superior performance of the new method.
• The method was illustrated using an example for rheumatoid arthritis (RA).

Conclusions:

• The developed F-statistic-based method offers a robust and powerful approach for quantitative trait analysis in two-stage GWASs.
• This method enhances the ability to identify genetic variants associated with quantitative traits, even with unknown genetic models.
• The findings have implications for genetic research in complex diseases like rheumatoid arthritis.