e-Driver

e-Driver identifies protein regions with biased distributions of somatic missense mutations to detect cancer driver genes and infer positive selection.

Key Features:

• Region-focused analysis: Analyzes distributions of somatic missense mutations across protein functional regions rather than treating genes as single entities.
• Functional region types: Evaluates domains and intrinsically disordered regions to localize mutation enrichment.
• Mutation-rate bias detection: Identifies regions with mutation rate bias compared to other parts of the same protein as evidence of positive selection.
• 3D structural integration: Leverages three-dimensional (3D) protein structures to pinpoint structural features enriched in cancer somatic mutations.
• Intermediate resolution: Operates between single-residue and whole-gene analyses to improve statistical reliability and resolution.
• Benchmarking on cancer datasets: Applied to The Cancer Genome Atlas (TCGA) and compared against four other methods to identify novel candidate drivers.

Scientific Applications:

• Cancer driver gene discovery: Detects candidate driver genes by locating protein regions under positive selection from somatic missense mutations.
• Localization of driver regions: Pinpoints specific protein domains and intrinsically disordered regions likely contributing to oncogenesis.
• Structural hotspot identification: Maps mutation-enriched structural features using 3D protein structures to inform mechanistic interpretation.
• Method benchmarking: Enables comparative evaluation of driver-detection approaches on large cancer genome datasets such as TCGA.

Methodology:

Examines the internal distribution of somatic missense mutations across protein functional regions (domains and intrinsically disordered regions), identifies regions with mutation-rate bias relative to other parts of the same protein as evidence of positive selection, and integrates three-dimensional (3D) protein structures to detect structural features enriched in somatic mutations.