diffdock

Name: diffdock
Availability: InStock
Author: HolobiomicsLab

Generate docking poses and confidence scores for protein-ligand complexes.

4|1|Updated Jun 18, 2025

Search Tags:#molecular docking #protein-ligand #pose prediction #confidence scores #diffdock #drug design

Ready to run

npx skills add https://github.com/HolobiomicsLab/Toolomics --skill diffdock-holobiomicslab

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: diffdock
Source: https://github.com/HolobiomicsLab/Toolomics/tree/main/mcp_host/skills/scientific-skills/scientific-skills/diffdock
Command: npx skills add https://github.com/HolobiomicsLab/Toolomics --skill diffdock-holobiomicslab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, rdkit, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

Automating the generation of plausible protein-ligand binding poses and associated confidence scores to accelerate structure-based drug design workflows.

Core Features & Use Cases

Diffusion-model based docking of small molecules to protein targets
Generates 3D binding poses and per-pose confidence scores
Supports single and batch docking, virtual screening, and ensemble docking with multiple protein conformations
Works with PDB, SMILES, and RDKit-supported inputs; integrates with downstream rescoring tools

Quick Start

Run DiffDock with a protein structure and a ligand description to generate pose predictions and confidence rankings.

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 and generate confidence scores for drug design?▼

Protein-ligand binding pose prediction and confidence scoring are automated using a diffusion-model approach. It generates 3D binding poses and per-pose confidence scores to accelerate structure-based drug design workflows.

Can I run batch docking and virtual screening using SMILES and PDB inputs?▼

Batch docking and virtual screening are supported using PDB structures and SMILES inputs. The workflow processes multiple ligands across diverse protein targets, generating 3D binding poses and confidence rankings.

Do I need RDKit and pandas to configure molecular docking workflows via YAML or CLI?▼

RDKit and pandas are required dependencies for the molecular docking workflow. Configuration is handled through YAML or CLI, producing poses and confidence outputs from PDB structures and ligand files.

What is the best way to perform ensemble docking across multiple protein conformations?▼

Ensemble docking across multiple protein conformations is supported natively. It automates pose generation and confidence scoring for protein-ligand complexes, integrating with downstream rescoring tools.

Does diffusion-based molecular docking work with downstream rescoring tools?▼

Diffusion-based molecular docking integrates with downstream rescoring tools. It outputs 3D binding poses and per-pose confidence scores that feed directly into structure-based drug design workflows.

Why use confidence scores for pose prediction instead of standard molecular docking methods?▼

Confidence scores rank generated binding poses by reliability, unlike standard molecular docking methods. This diffusion-based approach produces 3D binding poses with associated confidence metrics for virtual screening workflows.