KGG

KGG performs knowledge-based genome-wide analysis to prioritize single nucleotide polymorphisms (SNPs), conduct gene-based association tests using SNP p-values, and perform pathway and protein-protein interaction (PPI) module-level analyses for genome-wide association studies (GWAS).

Key Features:

• Knowledge-Based Weighting for SNP Prioritization: Employs a knowledge-based weighting method to prioritize SNPs using existing biological information.
• Gene-Based Association Tests Using SNP P-Values: Conducts gene-based association tests that aggregate SNP p-values to address multiple testing in GWAS.
• Biological Module-Level Analysis: Performs pathway and protein-protein interaction (PPI) network enrichment analyses to interpret significant genes at the module level.
• Multivariate Gene-Based Association Test (MGAS): Implements the Multivariate Gene-Based Association Test by Extended Simes Procedure (MGAS) for multivariate phenotype analysis in unrelated individuals.
• Demonstrated Superior Statistical Power: Simulation studies show MGAS has superior power compared to GATES, multiple regression, and MANOVA.
• Re-analysis Outcome: Re-analysis of a metabolic dataset using MGAS identified 32 FDR-controlled genome-wide significant genes and 12 regions harboring multiple genes, 30 of which were not reported in the original analyses.

Scientific Applications:

• Multivariate Phenotype Analysis: Enables gene-based testing of complex traits by considering multiple phenotype dimensions simultaneously with MGAS.
• SNP Prioritization in GWAS: Prioritizes SNPs for follow-up based on biological knowledge to focus on variants with higher relevance to phenotypes.
• Pathway and PPI Network Interpretation: Identifies enriched pathways and PPI modules to provide biological context for associated genes.
• Omics Re-analysis to Discover Novel Associations: Facilitates re-analysis of datasets (e.g., metabolomics) to detect additional genome-wide significant genes and regions missed by original analyses.

Methodology:

Methods explicitly include a knowledge-based weighting method for SNP prioritization, gene-based association tests that aggregate SNP p-values, pathway and PPI network enrichment analyses, and the Multivariate Gene-Based Association Test by Extended Simes Procedure (MGAS); MGAS was evaluated via simulation studies against GATES, multiple regression, and MANOVA.