diffdock

Name: diffdock
Availability: InStock
Author: robinbarvaag

Predict 3D ligand binding poses to protein targets using diffusion models.

Updated Jan 10, 2026

Search Tags:#virtual screening #drug discovery #molecular docking #cheminformatics #protein-ligand binding #pose prediction

Ready to run

npx skills add https://github.com/robinbarvaag/poynt --skill diffdock-robinbarvaag

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: diffdock
Source: https://github.com/robinbarvaag/poynt/tree/main/.github/skills/diffdock
Command: npx skills add https://github.com/robinbarvaag/poynt --skill diffdock-robinbarvaag

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill automates the prediction of how small molecules (ligands) bind to proteins, a critical step in drug discovery and understanding biological interactions.

Core Features & Use Cases

Molecular Docking: Predicts the 3D binding pose of a ligand to a protein target.
Virtual Screening: Screens large libraries of compounds to identify potential drug candidates.
Structure-Based Drug Design: Aids in designing new molecules with improved binding characteristics.
Use Case: You have a protein target and a list of potential drug compounds (as SMILES strings). Use this Skill to predict which compounds are most likely to bind and where they will bind on the protein.

Quick Start

Use the diffdock skill to predict the binding pose of the ligand with SMILES 'CC(=O)Oc1ccccc1C(=O)O' to the protein structure 'protein.pdb'.

Frequently Asked Questions about diffdock

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I predict protein-ligand binding poses for virtual screening?▼

Predict protein-ligand binding poses by providing a protein structure (PDB) or sequence and a ligand input (SMILES, SDF, or MOL2). The Skill uses diffusion models to output 3D binding poses for structure-based drug design.

What input formats are required for molecular docking pose prediction?▼

Molecular docking pose prediction requires protein structures in PDB format or as sequences, and ligands inputted as SMILES strings, SDF, or MOL2 files to accurately predict 3D binding poses.

Can I screen large compound libraries using SMILES strings for drug discovery?▼

Yes, you can screen large libraries of compounds for drug discovery by inputting a list of potential drug compounds as SMILES strings alongside your protein target to predict binding likelihood and sites.

How do diffusion models work for small molecule docking in cheminformatics?▼

Diffusion models work for small molecule docking by mathematically generating 3D binding poses of ligands to protein targets, facilitating virtual screening and chemical biology research without exhaustive physical simulations.

Does pose prediction work with protein sequences or do I need a PDB structure?▼

Pose prediction works with either a 3D protein structure in PDB format or a protein sequence. Providing the PDB structure or sequence alongside your SMILES ligand ensures accurate binding pose prediction.

What is the best way to identify potential drug candidates from a protein target?▼

The best way to identify potential drug candidates is structure-based drug design: supply a protein target and compound library to predict which molecules most likely bind and where they attach on the protein.