Documentation

Introduction

MolHub is a drug-discovery data platform — quantum chemistry parsing, automated docking, and ML-ready datasets in a single API.

What is MolHub?#

MolHub closes the gap between raw computational chemistry outputs and machine-learning workflows. Upload a Gaussian or ORCA log, drop a SMILES string, or post a PDB ID — and get back normalized, semantically validated data ready for AutoDock Vina, PyTorch, or HuggingFace datasets.

Looking for the fastest path? Jump to Quick start to run your first docking job in under five minutes.

Core modules#

LigandHub

SMILES → 3D → PDBQT → Vina → SDF (reassembled)

Onto-CompChem

Gaussian/ORCA log parsing + semantic triples

MolProphet

AI-driven binding pocket detection & repair

PyMolHub SDK

Python client, mirrors the REST API 1:1

Quick tour#

A complete end-to-end example using the Python SDK — install, authenticate, submit a docking job, and parse the results.

quickstart.py

from pymolhub import MolHub

mh = MolHub(api_key="mh_live_***")

# 1. Upload a molecule
mol = mh.molecules.create(smiles="CC(=O)Oc1ccccc1C(=O)O", generate_3d=True)

# 2. Submit a docking job
job = mh.docking.submit(receptor="1M17", ligands=[mol.id])
job.wait()

# 3. Read the top pose
print(job.results()[0].affinity_kcal_mol)

All endpoints are also available as curl and have TypeScript bindings — see the REST API reference.

Next steps#

Quick start — your first docking job in 5 minutes
Installation — install the PyMolHub SDK
Authentication — generate and rotate API keys
Your first docking job — guided walkthrough

Quick start