Consensus Pathway Analysis (CPA)

Consensus Pathway Analysis (CPA) integrates multiple pathway analysis methods to identify and prioritize biological pathways from lists of differentially expressed genes, gene names with fold changes, and expression matrices for genomic and transcriptomic interpretation.

Key Features:

• Diverse Analytical Methods: Integrates Gene Set Enrichment Analysis (GSEA), Global Test for GSA, Fast GSEA (FGSEA), Pathway Activity Difference of Group (PADOG), Impact Analysis, Over-Representation Analysis (ORA)/Webgestalt, Kolmogorov-Smirnov test (KS-test), and Wilcoxon rank-sum test.
• Meta-Analysis Capability: Performs meta-analysis across multiple datasets to synthesize findings and increase statistical power.
• Method and Dataset Integration: Supports combination of different analytical methods and datasets for consensus pathway identification.
• Pathway–Gene Relationship Analysis: Examines relationships between pathways and constituent genes to aid interpretation of impacted pathways.
• Flexible Input Options: Accepts lists of differentially expressed genes, gene names with associated fold changes, and expression matrices, and supports import from NCBI GEO.
• Multi-Organism Support: Uses KEGG (Kyoto Encyclopedia of Genes and Genomes) and Gene Ontology databases for analyses across multiple organisms.

Scientific Applications:

• Genomics: Identifies pathway-level alterations from differential gene lists and fold-change data derived from genomic studies.
• Transcriptomics: Detects impacted pathways using expression matrices and gene-level statistics from transcriptomic experiments.
• Systems Biology: Integrates methods and datasets to infer pathway interactions and mechanistic hypotheses.
• Meta-Analyses and Large-Scale Studies: Synthesizes results across studies to prioritize conserved or consistently impacted pathways.

Methodology:

Applies GSEA, Global Test for GSA, FGSEA, PADOG, Impact Analysis, ORA/Webgestalt, Kolmogorov-Smirnov test (KS-test), and Wilcoxon rank-sum test, and supports dataset-level meta-analysis and combination of method outputs.