DriverNet

DriverNet identifies driver genes in cancer by integrating mutation, copy number variation (CNV), and gene expression data to detect mutations that alter mRNA expression networks.

Key Features:

• Integration of Genomic and Transcriptomic Data: DriverNet combines mutation data, copy number variation (CNV) data, and gene expression data to construct an influence graph linking genetic alterations to transcriptional outputs.
• Identification of Driver Mutations: DriverNet uses an influence graph and a greedy algorithm to identify genes that are frequently mutated across patients and whose mutations significantly change mRNA expression of connected genes.
• Discovery of Rare Drivers and Pathway Co‑modification: Applied to four cancer datasets, DriverNet uncovers rare candidate driver mutations and highlights modulation of oncogenic and metabolic pathways via copy number co‑modification of adjacent drivers.

Scientific Applications:

• Detection of rare driver mutations: Reveals the prevalence of rare driver mutations that disrupt transcriptional networks.
• Analysis of pathway co‑modification: Provides insights into simultaneous modulation of oncogenic and metabolic pathways through copy number co‑modification of adjacent drivers.
• Assessment of mutation impact on expression networks: Offers a computational method to assess the impact of individual mutations on gene expression networks.

Methodology:

DriverNet constructs an influence graph from mutation, CNV, and gene expression data and applies a greedy algorithm to identify genes whose mutations significantly alter mRNA expression of connected genes.