False discovery rate control in genome-wide association studies with population structure

Area of Science:

• Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Genome-wide association studies (GWAS) are crucial for understanding polygenic traits.
• Standard GWAS methods often struggle with complex genetic architectures and population structures.
• Controlling the false discovery rate (FDR) is essential for reliable genetic discoveries.

Purpose of the Study:

• To develop a comprehensive statistical framework for analyzing GWAS data of polygenic traits.
• To produce interpretable findings while effectively controlling the false discovery rate.
• To offer a powerful alternative to standard GWAS approaches without making parametric assumptions.

Main Methods:

• A novel statistical framework utilizing multivariate algorithms.
• Generation of 'knockoffs' (imperfect copies) of genetic variables as negative controls.
• Correction for linkage disequilibrium and unknown population structure (ancestry, relatedness).

Main Results:

• The method demonstrates validity and effectiveness through extensive simulations.
• Application to UK Biobank data shows high power compared to state-of-the-art methods.
• Most discoveries made by the method are validated by comparisons with existing studies.

Conclusions:

• The proposed framework provides a powerful and flexible approach for GWAS of polygenic traits.
• It effectively controls the false discovery rate and accounts for complex genetic factors.
• Fast, publicly available software enables analysis of Biobank-scale genetic datasets.