What problem does it solve?
When you have a set of protein residues — from DMS hotspot screens, ClinVar recurrent variants, literature reports, or conservation analysis — you need to explain why they matter biologically. A residue in a catalytic triad means something different from one in a ligand pocket, a protein-protein interface, or a PTM site, and no single data source can distinguish these cases.
Core Features & Use Cases
- Multi-evidence mechanism calling: Combines structural annotation (interface, ligand pocket, core, secondary structure), UniProt features (active sites, binding sites, PTMs, disulfides), optional SAE feature evidence, and optional DMS effect data to assign each residue cluster a mechanism: catalytic, ligand-binding, interface, structural-core, PTM, regulatory, or unknown.
- Flexible input paths: Accepts user-provided residue positions from any source (ClinVar, literature, conservation) or auto-detects top-K hotspots from a DMS effect matrix, with a mandatory premise check that verifies user-named positions actually rank as hotspots in the given assay.
- Publication-ready visualization: Generates an annotated DMS heatmap with sequence strip, structural annotation track, and per-hotspot mechanism callouts, with explicit alignment verification to avoid off-by-N registration errors.
- Use Case: A researcher studying KRAS asks why positions 12 and 13 are critical. The skill checks their DMS rank, finds them in the GTP-binding P-loop via structural and UniProt evidence, and reports a ligand-binding mechanism with supporting SAE feature labels.
Quick Start
Explain why residues 12, 13, and 116 of KRAS (UniProt P01116, PDB 6VJJ chain A) are functionally critical using structural, UniProt, and SAE evidence.
