Novel score test to increase power in association test by integrating external controls

Area of Science:

• Genomics
• Statistical Genetics
• Bioinformatics

Background:

• Genetic association studies identify disease-risk variants using high-quality genotyped data.
• External control genomes offer a cost-effective way to increase the power of association testing.
• Existing methods like iECAT integrate external controls but lack covariate adjustment.

Purpose of the Study:

• To develop a novel score-based test for genetic association studies that allows for covariate adjustment.
• To improve the power of association testing by integrating both internal and external control samples.
• To address limitations of the original iECAT method regarding covariate adjustment.

Main Methods:

• Proposed a novel score-based test incorporating a shrinkage score statistic.
• Weighted sum of score statistics from internal and external control samples.
• Assessed batch effects by comparing internal and external control samples at the variant level.

Main Results:

• The proposed method demonstrated increased power compared to the original iECAT.
• The novel test maintained type I error rates in simulation studies.
• Successfully applied the method to age-related macular degeneration (AMD) association studies.

Conclusions:

• The developed score test extends iECAT's utility by enabling covariate adjustment and enhancing statistical power.
• This method refines the identification of disease-causing genetic variants.
• Improved statistical approaches are crucial for advancing human genome research.