Cancermuts

Cancermuts automates retrieval and annotation of cancer missense mutations from cBioPortal and COSMIC and annotates them with REVEL pathogenicity scores plus protein-context features such as post-translational modification sites, structured and unstructured regions, and short linear motifs to prioritize variants.

Key Features:

• Database integration: Retrieves known missense cancer mutations from cBioPortal and COSMIC and records mutation sources.
• Pathogenicity scoring: Annotates variants with REVEL pathogenicity scores.
• Protein-context annotation: Maps post-translational modification sites, structured and unstructured regions, and short linear motifs onto protein sequences.
• Contextual classification: Classifies mutations by REVEL score, sequence position, and local protein context.
• Analysis of intrinsically disordered proteins: Supports detailed annotation of intrinsically disordered proteins, exemplified by AMBRA1.
• Variant prioritization linked to experiments: Prioritized mutations that were experimentally validated, including two AMBRA1 variants with enhanced tumorigenic potential.
• Target-agnostic input: Operates starting from minimal information for any protein target.

Scientific Applications:

• Prioritization of pathogenic variants: Ranks missense mutations for downstream functional assessment in cancer genomics.
• Contextual functional interpretation: Assesses potential functional impact via proximity to PTMs, structured regions, disordered regions, and short linear motifs.
• Study of intrinsically disordered proteins: Enables analysis of mutation effects in proteins like AMBRA1 that contain disordered regions.
• Experimental candidate selection: Identifies mutations for cellular and molecular validation, including studies in melanoma.

Methodology:

Implemented in Python; retrieves mutations from cBioPortal and COSMIC; annotates variants with REVEL scores and mutation source; maps post-translational modification sites, structured/unstructured regions, and short linear motifs; and classifies mutations by REVEL score, sequence position, and local context.