skills.rest
HolobiomicsLab
HolobiomicsLabOFFICIAL
Explore all skills by this author →
WebsiteGitHub

torchdrug

Build and evaluate PyTorch graph neural networks on biomedical datasets.

4|1|Updated Jun 18, 2025
Search Tags:#drug-discovery#retrosynthesis#protein-modeling#graph-neural-networks#knowledge-graphs#torchdrug
One-click install
npx skills add https://github.com/HolobiomicsLab/Toolomics --skill torchdrug-holobiomicslab
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: torchdrug
Source: https://github.com/HolobiomicsLab/Toolomics/tree/main/mcp_host/skills/scientific-skills/scientific-skills/torchdrug
Command: npx skills add https://github.com/HolobiomicsLab/Toolomics --skill torchdrug-holobiomicslab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

TorchDrug provides a PyTorch-native toolkit for building, training, and deploying graph neural networks on molecules, proteins, and biomedical knowledge graphs, enabling researchers to develop custom models for drug discovery, protein modeling, and knowledge-graph reasoning.

Core Features & Use Cases

  • Extensive model library: 20+ architectures (GIN, GCN, GAT, SchNet, GearNet, RotatE, etc.) and 40+ datasets spanning molecular, protein, and KG tasks.
  • End-to-end workflows: property prediction, protein modeling, retrosynthesis planning, molecular generation, and knowledge graph reasoning.
  • Integrations: RDKit for chemistry, AlphaFold/ESM for structure, PyTorch Lightning for training, and pre-training utilities, plus cross-tool workflows.
  • Reference content: detailed references for datasets, architectures, and workflows to guide implementation.

Quick Start

Install TorchDrug, load BBBP, instantiate a GIN model, and run a simple molecular property-prediction task.

Frequently Asked Questions about torchdrug

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

FAQPage Schema
How do I build graph neural networks for molecular property prediction?▼

To build graph neural networks for molecular property prediction, use this PyTorch-based toolkit to load datasets like BBBP or HIV, instantiate architectures such as GIN or GAT, and run end-to-end training workflows.

Can I use PyTorch Lightning for training models on biomedical knowledge graphs?▼

Yes, you can use PyTorch Lightning for training models on biomedical knowledge graphs. The toolkit integrates PyTorch Lightning to handle training loops for knowledge graph reasoning tasks across datasets like Hetionet.

What graph neural network architectures are available for protein modeling?▼

Available graph neural network architectures for protein modeling include GearNet and SchNet. These models support structural protein modeling tasks with integrations for AlphaFold and ESM embeddings.

How do I perform retrosynthesis planning with a graph neural network?▼

Perform retrosynthesis planning with a graph neural network by utilizing the toolkit's end-to-end molecular workflows, which apply reasoning models to break down target molecules into available reactants.

Does this drug discovery toolkit integrate with RDKit for chemistry operations?▼

Yes, this drug discovery toolkit integrates with RDKit for chemistry operations. It leverages RDKit to handle molecular structures and feeds the processed graph data into PyTorch model architectures.

What datasets are supported for knowledge graph reasoning in drug discovery?▼

Supported datasets for knowledge graph reasoning in drug discovery include Hetionet and ZINC. The toolkit provides 40+ datasets spanning molecular, protein, and biomedical knowledge graph tasks.

Related Skills

Explore More Education & Research Skills→
SologaSologa
1

torchdrug

Create and apply PyTorch-native Graph Neural Networks for molecular and protein data analysis.

Community
Advanced
jimmc414jimmc414
557

torchdrug

Predict molecular properties from SMILES strings using a PyTorch-based toolkit.

Community
Advanced
founderphantomfounderphantom

pytorch-lightning

Transform raw PyTorch code into structured training pipelines with PyTorch Lightning.

Community
Advanced
Lord1EgyptLord1Egypt
2

deepchem

Train graph neural networks for molecular property prediction on chemical datasets.

Community
Advanced