CTpathway

CTpathway performs crosstalk-based pathway enrichment analysis by integrating pathway crosstalk, transcription factor-gene regulations, and protein-protein interactions to identify disease-risk pathways in cancer.

Key Features:

• Global Pathway Crosstalk Map (GPCM): A GPCM comprising over 440,000 edges constructed by integrating pathways from eight resources, transcription factor-gene regulations, and protein-protein interactions.
• Risk Score Assignment: Assigns gene risk scores by combining gene differential expression data with crosstalk effects within the GPCM to identify pathways enriched for high-risk genes.
• Enhanced Accuracy and Reproducibility: Benchmarking on over 8,300 expression profiles across ten cancer tissues and blood samples indicated higher accuracy, reproducibility, and speed compared with state-of-the-art methods.
• Identification of Novel Pathways: Recapitulates known risk pathways and identifies previously unreported critical pathways specific to individual cancer types.
• Stage-Specific Analysis: Detects risk pathways across cancer stages, including early-stage cancers characterized by a limited number of differentially expressed genes.
• Versatility with Data Types: Analyzes both bulk and single-cell RNA-seq profiles to predict tissue- and cell type-specific risk pathways.

Scientific Applications:

• Disease-risk pathway identification: Identification and prioritization of disease-risk pathways in cancer using integrated network crosstalk and differential expression information.
• Tissue- and cell-type-specific pathway prediction: Prediction of cancer tissue- and cell type-specific risk pathways from bulk and single-cell RNA-seq data.
• Support for translational research: Revealing candidate pathways to inform personalized medicine and targeted therapy development.

Methodology:

Constructs a Global Pathway Crosstalk Map by integrating pathways from eight resources, transcription factor-gene regulations, and protein-protein interactions (GPCM >440,000 edges); computes gene risk scores by combining differential expression with crosstalk effects within the GPCM; and benchmarks performance using analysis of over 8,300 expression profiles across ten cancer tissues and blood samples, supporting both bulk and single-cell RNA-seq inputs.