DeepChem (Claude Skill)

Claude skill that drives DeepChem, a deep-learning framework for molecular machine learning built on TensorFlow and PyTorch — graph neural networks for property prediction, featurization, pre-trained models, and the MoleculeNet benchmark suite.


Type	Claude Skill
Supplier	K-Dense Inc. (community OSS); DeepChem maintained by the `deepchem/deepchem` project
Availability	GA — actively maintained 2025–2026
Pricing	Free / OSS skill (MIT collection); DeepChem itself is MIT-licensed
Capabilities	Read/Write — Claude executes DeepChem via Python/Bash to featurize molecules, train models, and run MoleculeNet benchmarks

How to install

Also packaged in the SciAgent-Skills collection (jaechang-hits (community OSS, CC BY 4.0)): clone jaechang-hits/SciAgent-Skills and run /plugin install sciagent-skills in Claude Code (or copy skills/structural-biology-drug-discovery/deepchem into ~/.claude/skills/).
Claude Code / Claude.ai — Skills CLI (recommended):
```
npx skills add K-Dense-AI/scientific-agent-skills
```
Installs the K-Dense collection; enable the deepchem skill when prompted (also works in Cursor/Codex via the Agent Skills spec; requires Node ≥ 18).

Claude Code / Claude Desktop — manual clone:

git clone https://github.com/K-Dense-AI/scientific-agent-skills
cp -r scientific-agent-skills/skills/deepchem ~/.claude/skills/
pip install deepchem

Project-scoped alternative: copy into .claude/skills/ instead of ~/.claude/skills/.

What it does

SKILL.md with recipes for:

Featurization — molecular graphs, circular / Morgan fingerprints, 2D/3D descriptors, ConvMol, Weave, MolGraphConv
Graph neural networks — GCN, GAT, MPNN, AttentiveFP, ChemBERTa, and pre-trained molecular foundation models for property prediction
MoleculeNet benchmarks — 50+ curated datasets covering quantum-chemistry (QM7/QM8/QM9), physical-chemistry (ESOL, FreeSolv, Lipophilicity), biophysics (PCBA, MUV, HIV), and physiology (Tox21, ToxCast, SIDER, ClinTox, BBBP)
Workflows — dataset loading, train/valid/test splits (random, scaffold, stratified), model training and hyperparameter optimization, model interpretation
Integration with RDKit for chemical sanitization and feature extraction

Primary use cases: Toxicity prediction (Tox21, ClinTox, SIDER), ADMET property modeling, virtual-screening rank-ordering, binding-affinity prediction, quantum-property regression, blood-brain-barrier permeability classification.

Notes

Pairs with the pytdc, medchem, datamol, molfeat, and rdkit-skill entries — DeepChem supplies models and MoleculeNet datasets while those skills supply orthogonal featurizers, filters, and benchmark splits. DeepChem auto-downloads benchmark datasets on first use (typically tens of MB to a few GB depending on dataset); allow disk and network access. TensorFlow / PyTorch backends are selectable per model class; GPU is optional but recommended for graph-neural-network training. Skill is documentation plus Python recipes — Claude calls DeepChem locally via Bash/Python.

Sources

Installed this tool?

Share feedback — install path, OS, errors, workarounds. The form opens with this tool pre-selected and a link back to this page.