Apply graph neural networks to drug discovery tasks with PyTorch.
npx skills add https://github.com/jackspace/ClaudeSkillz --skill torchdrug
Please help me install this Agent Skill. Skill: torchdrug Source: https://github.com/jackspace/ClaudeSkillz/tree/main/skills/scientific-pkg-torchdrug Command: npx skills add https://github.com/jackspace/ClaudeSkillz --skill torchdrug
TorchDrug provides a comprehensive PyTorch-based toolbox for drug discovery and molecular science, applying graph neural networks to molecules, proteins, and biological knowledge graphs.
Install via pip, load datasets, define models, and train end-to-end using standard PyTorch workflows.
High-intent search queries and answers about installing and using this skill.
Graph neural networks model molecules as graphs where atoms are nodes and bonds are edges. TorchDrug provides GNN architectures like GIN, GAT, and SchNet integrated with PyTorch, enabling end-to-end molecular property prediction on datasets with RDKit-compatible SMILES strings and protein sequences.
Yes. TorchDrug supports protein modeling with compatibility for AlphaFold and ESM protein representations. You can integrate protein sequences and structures into graph-based workflows for knowledge graph reasoning and biomedical applications.
TorchDrug provides graph-based molecular generation and retrosynthesis planning across SMILES strings and reaction graphs. Its 20+ model architectures and 40+ datasets enable flexible experimentation with PyTorch workflows for drug discovery.
TorchDrug integrates molecules, proteins, and biomedical knowledge graphs as unified graph representations. Apply GNN-based reasoning to knowledge graphs for drug discovery tasks like entity linking, relation prediction, and molecular reasoning.
Yes. TorchDrug accepts SMILES strings for molecules, protein sequences and structures for proteins, and reaction graphs. RDKit integration enables seamless conversion and standardization across formats within PyTorch-based workflows.
TorchDrug supports 40+ datasets, but graph construction and GNN forward passes scale with molecular size and graph density. Consider memory constraints on GPU when processing large proteins or dense biomedical knowledge graphs in batch training.
LeonChaoXTrain PyTorch-native graph neural networks on TorchDrug datasets for molecular and protein tasks.
rubenslivTrain graph neural networks on molecular and biomedical datasets for drug discovery.
RamanEbrahimiProcess molecular and protein data with graph neural networks for drug discovery tasks.
swaruplabBuild and train graph neural network models on molecular and protein data.
