Divided-and-combined association test for pleiotropic effects with GWAS summary statistics.

Area of Science:

• Genetics
• Statistical genetics
• Bioinformatics

Background:

• Genome-wide association studies (GWAS) traditionally test one-to-one genetic variant-phenotype associations.
• Existing methods are underpowered due to ignoring pleiotropic effects (one gene influencing multiple traits).
• Public GWAS summary statistics offer potential for multi-trait analyses.

Purpose of the Study:

• To develop a novel statistical method for detecting pleiotropic effects using GWAS summary statistics.
• To enhance the power of association testing by integrating multi-trait information.
• To improve the identification of genes associated with complex diseases.

Main Methods:

• Proposes the divided-and-combined association test (DCAT) for pleiotropic effect detection.
• Constructs a Z-score covariance matrix using phenotype and genotype correlations via Kronecker product.
• Employs a family of quadratic statistics with weighted covariance matrices and Cauchy combination test for p-values.

Main Results:

• DCAT controls type I error rates effectively.
• Achieves superior statistical power compared to existing methods in most simulation scenarios.
• Identifies more significant genes in a real-data analysis of cardiovascular diseases.

Conclusions:

• DCAT offers a powerful and robust approach for detecting pleiotropic effects in GWAS.
• The method effectively leverages multi-trait GWAS summary statistics for enhanced genetic discovery.
• DCAT outperforms existing methods in identifying disease-associated genes, particularly for complex traits.