The eigen higher criticism and eigen Berk-Jones tests for multiple trait association studies based on GWAS summary...

Area of Science:

• Genetics
• Statistical genetics
• Bioinformatics

Background:

• Genome-wide association studies (GWAS) typically analyze one trait at a time.
• Identifying genetic variants associated with multiple correlated traits is challenging.
• Existing methods may miss variants with complex association patterns across traits.

Purpose of the Study:

• To develop novel statistical methods for testing genetic variant association with multiple correlated traits using GWAS summary statistics.
• To enhance the power and robustness of genetic association analyses in large-scale studies.
• To identify genetic variants influencing multiple lipid traits missed by single-trait analyses.

Main Methods:

• Proposed eigen higher criticism and eigen Berk-Jones testing procedures.
• Developed an omnibus (OMNI) test using the aggregated Cauchy association test for robust performance.
• Methods compute p-values analytically, suitable for large-scale studies.
• Validated methods through extensive simulation studies.

Main Results:

• Proposed tests maintain correct type I error rates.
• Tests demonstrate greater power in specific settings.
• The OMNI test consistently provides robust power across diverse scenarios.
• Application to Global Lipids Genetics Consortium data identified additional significant genetic variants.

Conclusions:

• The proposed eigen higher criticism, eigen Berk-Jones, and OMNI tests are effective for multi-trait genetic association studies.
• These methods improve the identification of genetic variants influencing multiple traits.
• The EBMMT R package facilitates the application of these novel statistical approaches.