skills.rest
Lord1Egypt
Lord1EgyptCOMMUNITY
Explore all skills by this author →
GitHub

torchdrug

Implement PyTorch-native graph neural networks for molecular and protein property prediction.

2|Updated Jun 4, 2026
Search Tags:#bioinformatics#drug discovery#pytorch#cheminformatics#graph neural networks#protein modeling
One-click install
npx skills add https://github.com/Lord1Egypt/scientific-agent-toolkit --skill torchdrug-lord1egypt
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: torchdrug
Source: https://github.com/Lord1Egypt/scientific-agent-toolkit/tree/main/scientific-skills/torchdrug
Command: npx skills add https://github.com/Lord1Egypt/scientific-agent-toolkit --skill torchdrug-lord1egypt

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires torch, torchdrug, rdkit, pytorch-lightning, and includes references (resource) components.

What problem does it solve?

This skill addresses the complexity of implementing graph-based machine learning for biological and chemical data, providing a unified framework to bridge the gap between raw molecular/protein structures and predictive modeling.

Core Features & Use Cases

  • Molecular Property Prediction: Predict ADMET properties, toxicity, and quantum chemical features using state-of-the-art GNNs like GIN and SchNet.
  • Protein Modeling: Analyze protein sequences and 3D structures for function, stability, and interaction prediction using specialized architectures like GearNet and ESM.
  • Knowledge Graph Reasoning: Perform link prediction and drug repurposing on biomedical knowledge graphs like Hetionet.
  • Use Case: A researcher can use this skill to train a graph neural network to predict the blood-brain barrier penetration of a new drug candidate by loading the BBBP dataset and applying a GIN model.

Quick Start

Use the torchdrug skill to load the BBBP dataset and train a GIN model for molecular property prediction.

Frequently Asked Questions about torchdrug

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

FAQPage Schema
How do I train a graph neural network for molecular property prediction?▼

To train a graph neural network for molecular property prediction, you load datasets like BBBP and apply PyTorch-native GNN architectures such as GIN or SchNet to predict ADMET properties and toxicity.

Can I use PyTorch for protein modeling and function prediction?▼

Yes, you can use PyTorch for protein modeling by analyzing protein sequences and 3D structures with specialized architectures like GearNet and ESM to predict protein function, stability, and interactions.

What is the best way to perform drug repurposing on biomedical knowledge graphs?▼

The best way to perform drug repurposing on biomedical knowledge graphs is by executing link prediction and knowledge graph reasoning on datasets like Hetionet using PyTorch-native graph neural networks.

Do I need RDKit to build graph neural networks for drug discovery?▼

Yes, you need RDKit along with PyTorch and PyTorch Lightning for graph construction and model training, providing the necessary cheminformatics framework to bridge raw molecular structures with predictive modeling.

Can I generate novel molecular structures using graph neural networks?▼

Yes, you can perform de novo molecular generation using this unified framework, leveraging PyTorch-native graph neural networks to design novel molecular structures for drug discovery workflows.

Related Skills

Explore More Education & Research Skills→
LeonChaoXLeonChaoX
783

torchdrug

Train PyTorch-native graph neural networks on TorchDrug datasets for molecular and protein tasks.

Community
Intermediate
jaechang-hitsjaechang-hits
298

torchdrug

Train GNN models for molecular property prediction using PyTorch and TorchDrug.

Community
Advanced
SologaSologa
1

torchdrug

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

Community
Advanced
biomaps-infrabiomaps-infra

lib-torchdrug

Develop drug discovery AI models with PyTorch-native graph neural networks.

Official
Advanced