Production-ready reinforcement learning algorithms (PPO, SAC, DQN, TD3, DDPG, A2C) with scikit-learn-like API. Use for standard RL experiments, quick prototyping, and well-documented algorithm implementations. Best for single-agent RL with Gymnasium environments. For high-performance parallel training, multi-agent systems, or custom vectorized environments, use pufferlib instead.
ToolUniverse workflow — Chemical Compound Retrieval
Access ZINC (230M+ purchasable compounds). Search by ZINC ID/SMILES, similarity searches, 3D-ready structures for docking, analog discovery, for virtual screening and drug discovery.
Query ChEMBL bioactive molecules and drug discovery data. Search compounds by structure/properties, retrieve bioactivity data (IC50, Ki), find inhibitors, perform SAR studies, for medicinal chemistry.
Graph Neural Networks (PyG). Node/graph classification, link prediction, GCN, GAT, GraphSAGE, heterogeneous graphs, molecular property prediction, for geometric deep learning.
QM/MM hybrid simulations with adaptive sampling for enzyme mechanisms and reaction dynamics. Combines quantum mechanics (reactive center) with molecular mechanics (protein/solvent) for accurate transition state and reaction pathway calculations. Supports metadynamics, umbrella sampling, and accelerated MD for enhanced conformational sampling.
Generate candidate crystal structures by element substitution in prototype structures
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.
Google quantum computing framework. Use when targeting Google Quantum AI hardware, designing noise-aware circuits, or running quantum characterization experiments. Best for Google hardware, noise modeling, and low-level circuit design. For IBM hardware use qiskit; for quantum ML with autodiff use pennylane; for physics simulations use qutip.
Query OECD export restriction policies on critical raw materials with corpus-search enrichment
Query openFDA API for drugs, devices, adverse events, recalls, regulatory submissions (510k, PMA), substance identification (UNII), for FDA regulatory data analysis and safety research.
ToolUniverse workflow — Drug Drug Interaction
Generates comprehensive drug research reports with compound disambiguation, evidence grading, and mandatory completeness sections. Covers identity, chemistry, pharmacology, targets, clinical trials, safety, pharmacogenomics, and ADMET properties. Use when users ask about drugs, medications, therapeutics, or need drug profiling, safety assessment, or clinical development research.
Access and analyze comprehensive drug information from the DrugBank database including drug properties, interactions, targets, pathways, chemical structures, and pharmacology data. This skill should be used when working with pharmaceutical data, drug discovery research, pharmacology studies, drug-drug interaction analysis, target identification, chemical similarity searches, ADMET predictions, or any task requiring detailed drug and drug target information from DrugBank.
ToolUniverse workflow — Gwas Drug Discovery
Lightweight WSI tile extraction and preprocessing. Use for basic slide processing tissue detection, tile extraction, stain normalization for H&E images. Best for simple pipelines, dataset preparation, quick tile-based analysis. For advanced spatial proteomics, multiplexed imaging, or deep learning pipelines use pathml.
Query PubChem via PUG-REST API/PubChemPy (110M+ compounds). Search by name/CID/SMILES, retrieve properties, similarity/substructure searches, bioactivity, for cheminformatics.
3D protein structure search via RCSB PDB. Input MUST be a protein/gene name (e.g. 'KRAS', 'EGFR', 'BTK') or a 4-character PDB ID (e.g. '6OIM'). Returns zero results for drug/chemistry phrases such as 'covalent inhibitors' or 'warhead selectivity'. Strip all drug qualifiers and pass only the target protein name or PDB ID.
