Exploiting pleiotropy to enhance variant discovery with functional false discovery rates.

Area of Science:

• Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Recruitment costs limit sample sizes in genome-wide association studies (GWASs), hindering genetic variant discovery.
• Existing methods may not fully leverage information from related traits to maximize statistical power.

Purpose of the Study:

• To introduce the surrogate functional false discovery rate (sfFDR) framework for integrating summary statistics of related traits.
• To enhance power and improve the discovery of genetic variants in GWASs.

Main Methods:

• The sfFDR framework integrates summary statistics from related traits.
• It provides estimates of functional local false discovery rate (fFDR) and q values.
• It derives functional P values for type I error control and functional local Bayes' factors for post-GWAS analyses.

Main Results:

• sfFDR substantially increased power in a UK Biobank study of obesity-related traits, equivalent to a 52% sample size increase.
• Eight additional significant SNPs were discovered in a GWAS for eosinophilic granulomatosis with polyangiitis.
• The method demonstrated utility in both small and large-scale genetic studies.

Conclusions:

• Exploiting related traits through the sfFDR framework significantly enhances power in GWASs.
• This approach facilitates the discovery of genetic variants, particularly in rare diseases.
• sfFDR offers a valuable tool for optimizing genetic association studies.