Prioritized subset analysis: improving power in genome-wide association studies

Area of Science:

• Genetics
• Statistical genomics
• Bioinformatics

Background:

• Genome-wide association studies (GWAS) are crucial for understanding complex disease genetics.
• Existing candidate gene information can enhance GWAS power but is often underutilized.
• Traditional GWAS methods treat all genetic markers equally, potentially reducing power to detect variants.

Purpose of the Study:

• To introduce a novel analytical approach, Prioritized Subset Analysis (PSA), for GWAS.
• To leverage prior biological information to improve the detection of disease-associated variants.
• To compare the statistical power of PSA against standard whole-genome analysis.

Main Methods:

• PSA involves pre-selecting a subset of genetic markers from known candidate regions.
• The false discovery rate (FDR) procedure is applied separately to the prioritized subset and the remaining markers.
• This approach allows for differential analysis based on prior biological knowledge.

Main Results:

• PSA demonstrates superior statistical power compared to whole-genome single-step FDR adjustment.
• The extent of power improvement is contingent on the proportion and significance of associated single nucleotide polymorphisms (SNPs) within the prioritized subset.
• Negligible power loss is observed for disease loci outside the prioritized subset.

Conclusions:

• PSA offers a flexible framework for integrating diverse sources of prior biological information into GWAS.
• This method represents a valuable advancement for genetic analysis in complex diseases.
• PSA can enhance the efficiency and effectiveness of variant detection in genetic association studies.