deepchem
Category: chemistry-and-drug-discovery · License: MIT · Version: 1.5 · Includes scripts
Tags: molecular machine learning, ADMET, property prediction, drug discovery, cheminformatics
When to use it
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.Example requests
Ask your agent in plain language, for example:- “Train a quick solubility model on the Delaney dataset and tell me the test RMSE.”
- “I have compounds.csv with columns ‘smiles’ and ‘herg_ic50_class’ (about 4,000 rows, heavily imbalanced, ~8% positives). I tried a random 80/20 split and got 0.97 AUC, which seems too good. Can you rebuild this in DeepChem with a proper split and compare ECFP+RF against a GCN? I’m on Python 3.11 and getting ‘No module named torch’ warnings when I import deepchem.”
- “I want to predict whether small molecules cross the blood-brain barrier, and I’d like to try lots of different ways of turning structures into numeric inputs (fingerprints, descriptors, graph representations) and compare them on a standard benchmark set. Where do I start?”
Install
/plugin marketplace add KalarisLabs/research-agent-skills and install the whole chemistry-and-drug-discovery category with /plugin install chemistry-and-drug-discovery@research-agent-skills.