Does DiffDock predict binding affinity?

No, DiffDock predicts the 3D binding pose and provides a confidence score regarding the structural accuracy. For binding affinity (ΔG or Kd), you should combine DiffDock results with scoring functions like GNINA or MM/GBSA.

Can I use DiffDock for protein-protein docking?

No, DiffDock is specifically designed for small molecule ligands. For protein-protein docking, tools like DiffDock-PP or AlphaFold-Multimer are recommended.

How do I interpret the confidence scores?

Scores above 0 indicate high confidence, -1.5 to 0 indicate moderate confidence, and scores below -1.5 suggest low prediction certainty.

Is a GPU required to run DiffDock?

While DiffDock can run on a CPU, a GPU is strongly recommended as it typically provides a 10-100x speedup, making practical docking and screening feasible.

What inputs are needed to use this Claude skill?

You need a protein source (either a PDB file or an amino acid sequence) and a ligand description (a SMILES string or a structure file like SDF or MOL2).

DiffDock Molecular Docking

Name: DiffDock Molecular Docking
Author: Microck

byMicrock

•

データサイエンスとML

Predicts 3D protein-ligand binding poses using diffusion-based deep learning for structure-based drug design.

概要

DiffDock is a state-of-the-art computational chemistry skill that enables Claude to perform diffusion-based molecular docking for drug discovery. It allows researchers to predict high-accuracy 3D binding poses for small molecule ligands against protein targets using either PDB files or amino acid sequences. With features for batch virtual screening, confidence scoring, and support for multiple ligand formats like SMILES and SDF, this skill bridges the gap between AI-driven structural biology and practical lead optimization workflows.

主な機能

Supports protein structures via PDB files or sequences via ESMFold
Predicts 3D ligand binding poses with high structural accuracy
81 GitHub stars
Generates confidence scores to rank and validate prediction reliability
Processes diverse ligand inputs including SMILES, SDF, and MOL2
Performs batch virtual screening for large compound libraries

ユースケース

Predicting the binding orientation of a drug candidate to a target protein
Identifying potential binding sites for proteins with unknown experimental structures
Running virtual screening campaigns on libraries of SMILES strings

概要

主な機能

Supports protein structures via PDB files or sequences via ESMFold
Predicts 3D ligand binding poses with high structural accuracy
81 GitHub stars
Generates confidence scores to rank and validate prediction reliability
Processes diverse ligand inputs including SMILES, SDF, and MOL2
Performs batch virtual screening for large compound libraries

ユースケース

Predicting the binding orientation of a drug candidate to a target protein
Identifying potential binding sites for proteins with unknown experimental structures
Running virtual screening campaigns on libraries of SMILES strings