Is a GPU required to use this skill?

While DiffDock can run on a CPU, a GPU is strongly recommended as it offers a 10-100x speedup, making it much more practical for research and screening tasks.

What ligand formats does DiffDock support?

DiffDock is highly flexible, supporting SMILES strings directly as text, as well as standard structural files like SDF and MOL2.

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 suggest low prediction certainty. High confidence refers to the model's certainty of the pose, not necessarily the strength of the binding.

Does DiffDock predict binding affinity?

No, DiffDock is specifically designed to predict 3D binding poses and provide a confidence score regarding the structural accuracy. It should be combined with scoring functions like GNINA or MM/GBSA for affinity (ΔG) assessment.

Can I use DiffDock for protein-protein docking?

No, this skill is optimized for small molecule ligands (typically 100-1000 Da). For protein-protein interactions, specialized tools like DiffDock-PP or AlphaFold-Multimer are more appropriate.

DiffDock Molecular Docking

Name: DiffDock Molecular Docking
Author: K-Dense-AI

byK-Dense-AI

•

2,188

データサイエンスとML

Predicts high-accuracy 3D protein-ligand binding poses using state-of-the-art diffusion-based deep learning models.

概要

DiffDock is a specialized scientific skill for Claude Code designed to accelerate structure-based drug discovery and computational biology. It leverages diffusion models to predict the most probable 3D orientations (poses) of small molecule ligands when bound to target proteins, providing both structural coordinates and reliability confidence scores. This skill enables researchers to automate complex bioinformatics workflows, ranging from single protein-ligand docking to large-scale virtual screening and ensemble docking, significantly reducing the manual effort required to model molecular interactions.

主な機能

Batch processing capabilities for virtual screening campaigns
Support for PDB files, amino acid sequences, and SMILES strings
2,188 GitHub stars
Automated confidence scoring to assess prediction reliability
High-accuracy 3D ligand binding pose prediction
Seamless integration with ESMFold for sequence-to-structure workflows

ユースケース

Predicting the binding orientation of novel drug candidates to target proteins
Conducting virtual screenings of compound libraries to identify lead molecules
Analyzing molecular interactions for structure-based drug design and optimization

概要

主な機能

Batch processing capabilities for virtual screening campaigns
Support for PDB files, amino acid sequences, and SMILES strings
2,188 GitHub stars
Automated confidence scoring to assess prediction reliability
High-accuracy 3D ligand binding pose prediction
Seamless integration with ESMFold for sequence-to-structure workflows

ユースケース

Predicting the binding orientation of novel drug candidates to target proteins
Conducting virtual screenings of compound libraries to identify lead molecules
Analyzing molecular interactions for structure-based drug design and optimization