BetweenNet

BetweenNet identifies and prioritizes cancer driver genes by integrating patient-specific genomic data with protein-protein interaction networks and network-based analyses.

Key Features:

• Integration of Genomic Data: Combines patient-specific genomic data with protein-protein interaction networks to contextualize genes within molecular interaction networks.
• Betweenness Centrality Measure: Uses betweenness centrality to identify outlier genes characterized as dysregulated genes specific to each patient.
• Bipartite Graph Construction: Constructs a bipartite graph linking mutated genes and identified outlier genes to represent their relationships.
• Random-Walk Process: Applies a random-walk algorithm on the bipartite graph to prioritize mutated genes and help distinguish driver from passenger mutations.

Scientific Applications:

• Cancer Driver Prioritization: Ranks mutated genes to prioritize candidate cancer drivers in individual patients.
• Comparative Evaluation: Evaluated against state-of-the-art methods across lung, breast, and pan-cancer datasets with improved recovery of known cancer genes.
• Functional Enrichment Analysis: BetweenNet-ranked genes show overlaps in Gene Ontology (GO) terms and reference pathways with established cancer genes.
• Precision Oncology: Uses patient-specific prioritization to support applications in precision oncology and individualized analysis.

Methodology:

Integrates patient-specific genomic data with protein-protein interaction networks; computes betweenness centrality to identify patient-specific outlier/dysregulated genes; constructs a bipartite graph between mutated and outlier genes; performs a random-walk process on the bipartite graph to prioritize mutated genes.