GeneNetFusion

GeneNetFusion constructs fused gene co-expression networks and identifies gene biomarkers to characterize cancer-type-specific gene associations using TCGA datasets for kidney renal clear cell carcinoma, liver hepatocellular carcinoma, and prostate adenocarcinoma.

Key Features:

• Fused co-expression networks: Constructs integrated co-expression networks by combining multiple gene networks to capture cancer-type-specific associations.
• Integration of multiple similarity measures: Builds gene networks derived from different similarity measures and integrates them for fusion.
• Normal and cancer-specific fusion: Combines normal and cancer-specific networks for each cancer type to highlight differential network structure.
• Community extraction: Extracts communities of genes from the fused networks to define coherent gene modules.
• Functional analysis of communities: Subjects extracted gene communities to functional analyses to interpret biological roles.
• Literature cross-referencing: Cross-references identified gene communities with existing literature to assess biological relevance.
• Computational validation: Assesses relevance of identified gene sets by comparing their performance in normal/cancer classification against other gene signatures, including differentially expressed genes.
• TCGA dataset focus: Applies the pipeline to The Cancer Genome Atlas (TCGA) datasets for kidney renal clear cell carcinoma, liver hepatocellular carcinoma, and prostate adenocarcinoma.
• Data-driven pipeline: Implements a data-driven computational workflow to extract and evaluate gene associations.

Scientific Applications:

• Gene biomarker discovery: Identifies biologically and computationally significant gene biomarkers relevant to cancer pathogenesis and diagnostics.
• Gene-to-gene association analysis: Explores gene-to-gene associations and network structure specific to each cancer type.
• Normal versus cancer classification: Enables comparative assessment of gene signatures for discriminating normal and cancer samples.
• Functional interpretation of gene modules: Facilitates functional annotation of gene communities and validation against literature findings.
• TCGA cancer-specific studies: Supports focused analyses on kidney renal clear cell carcinoma, liver hepatocellular carcinoma, and prostate adenocarcinoma TCGA datasets.

Methodology:

Integrates multiple gene networks derived from different similarity measures, fuses normal and cancer-specific networks per cancer type, extracts gene communities, performs functional analyses and literature cross-referencing, and validates gene sets by comparing normal/cancer classification performance against other gene signatures including differentially expressed genes.