How do I interpret the confidence scores?

Scores above 0 indicate high confidence, scores between -1.5 and 0 are moderate, and scores below -1.5 indicate low confidence in the predicted binding pose.

Does DiffDock predict binding affinity (delta G)?

No, DiffDock specializes in predicting 3D binding poses and providing confidence scores for those poses. For affinity prediction, you should combine DiffDock results with scoring functions like GNINA or MM/GBSA.

Is a GPU required to use the DiffDock skill?

While DiffDock can run on a CPU, a GPU is strongly recommended as it offers a 10-100x speedup, which is critical for practical docking tasks and virtual screening.

Can I run docking if I only have a protein sequence?

Yes, DiffDock can process amino acid sequences by utilizing ESMFold to generate the protein structure before performing the docking procedure.

What ligand file formats are supported?

The skill supports SMILES strings, SDF files, and MOL2 files as inputs for ligand structures.

DiffDock Molecular Docking

Name: DiffDock Molecular Docking
Author: jcafazzo

byjcafazzo

0•

Data Science & ML

Performs diffusion-based molecular docking to predict high-accuracy 3D binding poses between proteins and ligands for drug discovery.

DiffDock is a specialized computational biology tool that leverages diffusion-based deep learning to solve the molecular docking problem with state-of-the-art accuracy. It enables researchers to predict how small molecule ligands interact with protein targets, supporting inputs ranging from PDB files to raw amino acid sequences via ESMFold. This skill streamlines the entire docking pipeline within Claude, from environment setup and batch virtual screening to the analysis of confidence scores, making it an essential asset for structure-based drug design and lead optimization.

Key Features

01Predicts 3D ligand binding poses using advanced diffusion models

02Supports protein structures (PDB) and sequences (ESMFold)

03Generates confidence scores to validate prediction reliability

04Automates batch virtual screening for large chemical libraries

05Analyzes and ranks docking results for downstream affinity scoring

060 GitHub stars

Use Cases

01Generating high-quality starting poses for molecular dynamics or MM/GBSA calculations

02Running large-scale virtual screenings of SMILES libraries against protein targets

03Predicting the binding site and structural pose of novel drug candidates

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jcafazzo/thewolfondchair.ca diffdock

For use in Claude.ai and ChatGPT

Key Features

01Predicts 3D ligand binding poses using advanced diffusion models

02Supports protein structures (PDB) and sequences (ESMFold)

03Generates confidence scores to validate prediction reliability

04Automates batch virtual screening for large chemical libraries

05Analyzes and ranks docking results for downstream affinity scoring

060 GitHub stars

Use Cases

01Generating high-quality starting poses for molecular dynamics or MM/GBSA calculations

02Running large-scale virtual screenings of SMILES libraries against protein targets

03Predicting the binding site and structural pose of novel drug candidates

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jcafazzo/thewolfondchair.ca diffdock

For use in Claude.ai and ChatGPT