skills.rest
Zeyuyang-0420
Zeyuyang-0420COMMUNITY
Explore all skills by this author →
GitHub

torchdrug

Predict molecular properties and model proteins using graph neural networks.

Updated May 8, 2026
Search Tags:#knowledge graph#graph neural networks#retrosynthesis#protein modeling#molecular property prediction
One-click install
npx skills add https://github.com/Zeyuyang-0420/bio-ai-research-skills --skill torchdrug-zeyuyang-0420
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: torchdrug
Source: https://github.com/Zeyuyang-0420/bio-ai-research-skills/tree/main/categories/ml-training-ai/torchdrug
Command: npx skills add https://github.com/Zeyuyang-0420/bio-ai-research-skills --skill torchdrug-zeyuyang-0420

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires torchdrug, rdkit, torch, torchvision, torchtext, numpy, scipy, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

TorchDrug tackles complex molecular science problems such as property prediction, protein modeling, knowledge graph reasoning, and retrosynthesis by applying advanced graph neural networks.

Core Features & Use Cases

  • Molecular Property Prediction: Predict molecular properties like solubility, toxicity, and activity using state-of-the-art GNNs.
  • Protein Modeling: Model protein functions, structures, and interactions with sequence or 3D structure data.
  • Knowledge Graph Reasoning: Predict missing links and relationships in biological knowledge graphs for drug repurposing and disease understanding.
  • Retrosynthesis: Plan synthetic routes from target molecules to starting materials, aiding in synthesis planning and route optimization.
  • Use Case: Imagine you need to design a novel drug candidate. Use TorchDrug to predict its properties, model its structure, and generate potential starting materials for synthesis.

Quick Start

Use the torchdrug skill to train a GNN model to predict molecular properties on the BBBP dataset.

Frequently Asked Questions about torchdrug

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

FAQPage Schema
How do I predict molecular properties like toxicity using graph neural networks?▼

You can predict molecular properties using graph neural networks by training models on datasets like BBBP. This Skill applies GNNs to evaluate solubility, toxicity, and activity for drug discovery.

What is the best way to perform retrosynthesis planning for target molecules?▼

The best way to perform retrosynthesis planning involves using GNNs to map synthetic routes from target molecules to starting materials. This Skill supports synthesis planning and route optimization for drug discovery.

Does this Skill support knowledge graph reasoning for drug repurposing?▼

Yes, it supports knowledge graph reasoning for drug repurposing. It uses GNNs to predict missing links and relationships in biological knowledge graphs to aid disease understanding and drug repurposing.

Can I model protein structures and interactions with PyTorch?▼

Yes, you can model protein structures and interactions with PyTorch. This Skill uses PyTorch and GNNs to model protein functions, structures, and interactions using sequence or D structure data.

Do I need RDKit and PyTorch to run molecular property prediction?▼

Yes, you need RDKit and PyTorch to run molecular property prediction. This Skill requires torchdrug, RDKit, PyTorch, and other dependencies like NumPy and SciPy to perform GNN training and molecular modeling.

Related Skills

galeepgaleep

molfeat

Convert SMILES strings and RDKit molecules into numerical features for machine learning.

Community
Advanced
jaechang-hitsjaechang-hits
298

torchdrug

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

Community
Advanced
jasrajtulsijasrajtulsi

phylogenetics

Build rooted phylogenetic trees from biological sequences using MAFFT, IQ-TREE 2, and ETE3.

Community
Advanced
Hung-3008Hung-3008
1

torchdrug

Build, train, and deploy graph neural network workflows for molecules and proteins.

Community
Advanced