molclaw-residue-mapper

What problem does it solve?

Accurately reconcile residue numbering across UniProt canonical positions, PDB author numbering (including insertion codes and offsets), and tool-internal sequential numbering so that residue-specific analyses (ProLIF, PLIP, per-residue energy decomposition) are interpreted correctly and silently-misleading conclusions are avoided.

Core Features & Use Cases

• Three-tier mapping strategies: arithmetic offset for predicted structures, DBREF header fast path for RCSB PDBs, and Needleman-Wunsch sequence alignment fallback for robust mapping.
• Queryable forward and reverse lookup: accept UniProt identifiers, PDB author numbers, or tool:internal indices and return CSV/JSON mapping tables with match quality indicators.
• Real-world scenarios: translate ProLIF/PLIP outputs from ESMFold/Boltz-2/Chai-1 predictions back to UniProt numbering, reconcile PDB files with DBREF offsets, and map residues across multi-chain complexes for downstream analysis.

Quick Start

Map the PDB file step05_boltz2_complex.pdb to UniProt P00533 for chain A with predicted=True and input_seq_start=718 and return the mapping as a CSV file.