Consensus Cancer driver gene Caller (C3)

Consensus Cancer Driver Gene Caller (C3) identifies consensus driver genes from somatic mutations identified by next-generation sequencing by integrating six complementary analytical strategies to distinguish driver mutations from passenger mutations.

Key Features:

• Integrative strategy: Combines frequency-based, machine learning-based, functional bias-based, clustering-based, statistical model-based, and network-based methods to generate consensus driver calls.
• Frequency-based method: Includes a frequency-based approach to assess recurrence of mutations across samples.
• Machine learning-based method: Incorporates machine learning-based analyses as one component of the consensus framework.
• Functional bias-based method: Applies functional bias-based analyses to evaluate mutation impact on gene function.
• Clustering-based method: Uses clustering-based analyses to detect spatial or pattern-based mutation signals.
• Statistical model-based method: Employs statistical model-based analyses to assess significance of candidate drivers.
• Network-based method: Integrates network-based analyses to contextualize genes within interaction networks.
• Statistical evaluation: Provides statistical evaluations of integration results.
• Interpretable visualizations: Produces interpretable visualizations of integrated results.
• Implementation: Implemented in Python.
• Input data: Operates on somatic mutation data generated by next-generation sequencing.

Scientific Applications:

• Driver gene discovery: Identification of consensus cancer driver genes from large somatic mutation datasets.
• Mutation prioritization: Prioritization of candidate driver mutations versus passenger mutations.
• Cancer genomics research: Support for studies aiming to elucidate cancer biology and inform therapeutic target selection.

Methodology:

Implements an integrative computational framework in Python that combines frequency-based, machine learning-based, functional bias-based, clustering-based, statistical model-based, and network-based methods to analyze somatic mutations from next-generation sequencing and produce consensus driver gene calls with statistical evaluation and visualizations.