BioNeMo Drug Discovery
A 3-step recipe with NVIDIA NIM. We can help you get access.
Recipe · 3 steps
BioNeMo Drug Discovery
with NVIDIA NIM
Fold a protein, dock a ligand, then generate analog molecules.
- Predict the 3D structure of the target protein from its amino acid sequence with the AlphaFold2 NIM, returning PDB-format structures.
- Dock the seed ligand against the predicted protein structure with DiffDock to predict candidate binding poses.
- Generate optimized analog molecules around the seed ligand SMILES with MolMIM, optimizing for drug-likeness (QED).
What you'll need to have ready
- apiKey: NVIDIA developer API key (nvapi-...) sent as a Bearer token.
- sequence: Single-letter amino acid sequence of the target protein.
- ligandSmiles: Seed ligand as a SMILES string, used for docking and analog generation.
- numPoses: Number of docking poses to predict.
- numMolecules: Number of analog molecules to generate.
For your developer
Arazzo workflow nvidia-nim:bionemo-drug-discovery, 3 operations: predictProteinStructure → dockLigand → generateMolecules. Specs on APIs.io.