diffdock

SUMMARY

What it does

DiffDock is a diffusion-based deep learning tool for molecular docking that predicts 3D binding poses of small molecule ligands to protein targets. It supports protein structures (PDB files) or sequences (via ESMFold), processes single complexes or batch virtual screening campaigns, generates confidence scores to assess prediction reliability, and handles diverse ligand inputs (SMILES, SDF, MOL2). It is not designed for binding affinity prediction. The skill provides guidance for installation, environment setup, core workflows (single docking, batch processing, result analysis), confidence score interpretation, parameter customization, advanced techniques (ensemble docking, integration with scoring functions), limitations, troubleshooting, and references to helper scripts and documentation.