deepchem
Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first PyTorch workflows use torchdrug; for benchmark datasets use pytdc.
npx skills add majiayu000/claude-skill-registry --skill deepchem-k-dense-ai-scientific-agent-ski-2 --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Guides the agent to perform molecular machine learning tasks using DeepChem, including loading and featurizing molecular data, training models (traditional ML and GNNs), using MoleculeNet benchmarks, transfer learning, and making predictions. Focuses on experimentation with various featurizers (fingerprints, descriptors, graph-based) and pretrained models, plus workflow scripts for solubility prediction, graph neural networks, and transfer learning.
How it works
Describes loading data via Loader classes (CSVLoader, SDFLoader, FASTALoader, etc.), selecting featurizers (CircularFingerprint, RDKitDescriptors, MolGraphConvFeaturizer, GroverFeaturizer), and choosing splitters (ScaffoldSplitter, RandomSplitter, etc.). Provides example code for traditional ML, deep learning, and graph neural networks, plus workflows for quick benchmarks, custom data prediction, and transfer learning. Includes scripts for solubility prediction, graph neural networks, and transfer learning, with command-line usage. Recommends patterns like Scaffold splitting, feature normalization, and starting simple before scaling, plus handling imbalanced data and memory considerations.
When to use it
Use when you need molecular data loading/processing, property prediction (solubility, toxicity, ADMET), model training on chemical/biological datasets, MoleculeNet benchmarks, feature conversion to fingerprints/descriptors/graph representations, graph neural networks, and transfer learning with pretrained models.
What it can touch
Uses tools and libraries via Python code examples:
deepchemcore APIs for data loading, featurization, modeling, and evaluation- Models:
SklearnModel,MultitaskRegressor,GCNModel,GATModel,AttentiveFPModel,HuggingFaceModel,GroverModel - Datasets and utilities:
dc.molnet,dc.splits,dc.feat,dc.data,dc.trans,dc.metrics - Scripts:
scripts/predict_solubility.py,scripts/graph_neural_network.py,scripts/transfer_learning.pyNote: Frontmatter requires Python 3.7–3.11 and backends PyTorch, TensorFlow, or JAX; RDKit is a core dependency.
Caveats
- Requires specific Python version range and backend frameworks; ensure environment matches compatibility notes.
- GNNs and transformers require appropriate backends and potentially large datasets; model choice should reflect dataset size and compute resources.
- Some workflows assume ScaffoldSplitter usage for molecular data to avoid leakage.
- Content emphasizes patterns and best practices but does not guarantee predictive success or performance improvements.
# DeepChem ## Overview DeepChem is a comprehensive Python library for applying machine learning to chemistry, materials science, and biology. Enable molecular property prediction, drug discovery, materials design, and biomolecule analysis through specialized neural networks, molecular featurization methods, and pretrained models. **Version note:** Examples target **deepchem 2.8.0** (PyPI stable, Apr 2024). Requires **Python 3.7–3.11** (`<3.12` on PyPI). Core utilities (loaders, featurizers, MoleculeNet) work without a DL backend; GNN and transformer models need the matching extra (`torch`, `tensorflow`, or `jax`). Install the backend framework first when using GPU builds. ## When to Use This Skill This skill should be used when: - Loading and processing molecular data (SMILES strings, SDF files, protein sequences) - Predicting molecular properties (solubility, toxicity, binding affinity, ADMET properties) - Training models on chemical/biological datasets - Using MoleculeNet benchmark datasets (Tox21, BBBP, Delaney, etc.) - Converting molecules to ML-ready features (fingerprints, graph representations, descriptors) - Implementing graph neural networks for molecules (GCN, GAT, MPNN,
- Overview
- When to Use This Skill
- Core Capabilities
- 1. Molecular Data Loading and Processing
- 2. Molecular Featurization
- 3. Data Splitting
- 4. Model Selection and Training
- 5. MoleculeNet Benchmarks
- 6. Transfer Learning
- 7. Model Evaluation
- 8. Making Predictions
- Typical Workflows
- Workflow A: Quick Benchmark Evaluation
- Workflow B: Custom Data Prediction
Use Delaney benchmark python scripts/predict_solubility.py Use custom data python scripts/predict_solubility.py \ Train GCN on Tox21 python scripts/graph_neural_network.py --model gcn --dataset tox21 Train AttentiveFP on custom data python scripts/graph_neural_network.py \ Fine-tune ChemBERTa on BBBP python scripts/transfer_learning.py --model chemberta --dataset bbbp
What does the deepchem skill do?
Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first PyTorch workflows use torchdrug; for benchmark datasets use pytdc.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill deepchem-k-dense-ai-scientific-agent-ski-2 --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.