Two-stage joint selection method to identify candidate markers from genome-wide association studies

Area of Science:

• Genetics
• Epidemiology
• Computational Biology

Background:

• Gene interactions and environmental factors significantly influence disease susceptibility.
• Single-marker analyses in genome-wide association studies (GWAS) may miss genes with effects mediated through interactions.
• Rheumatoid arthritis (RA) risk is influenced by complex genetic architectures.

Purpose of the Study:

• To develop and validate a novel two-stage joint single-nucleotide polymorphism (SNP) analysis method for detecting gene-gene interactions in GWAS.
• To identify novel genetic factors and epistatic effects associated with rheumatoid arthritis (RA).
• To improve the detection of disease susceptibility genes that exhibit marginal effects in traditional analyses.

Main Methods:

• A two-stage model selection procedure was employed using Genetic Analysis Workshop 16 (GAW16) GWAS data for RA.
• Stage 1 involved an exhaustive two-dimensional search to identify promising SNP and SNP pairs.
• Stage 2 utilized LASSO regression for SNP selection, followed by validation using Wellcome Trust Case Control Consortium RA data.

Main Results:

• The proposed method successfully replicated known RA risk genes.
• Novel genes and their epistatic (gene-gene interaction) effects on RA were identified.
• The approach demonstrated the ability to detect joint genetic effects missed by single-marker analyses.

Conclusions:

• The developed two-dimensional scan-based analysis is effective for real-world GWAS data, particularly for identifying complex genetic interactions.
• This method enhances the discovery of RA-associated genes by accounting for epistasis.
• The findings highlight the importance of considering joint genetic effects in understanding disease etiology.