LigandMPNN
Protein sequence design that explicitly models small molecules, nucleotides and metals, and also outputs side-chain conformations.
| Category | Protein structure & design |
|---|---|
| GitHub | dauparas/LigandMPNN |
| First public | 2023-12-23 (source) |
| Paper | https://doi.org/10.1038/s41592-025-02626-1 · Nature Methods (2025) |
| Code license | MIT |
| Weights license | code and model parameters MIT per README |
| What this allows | Permissive (commercial OK) Closed-source commercial use allowed; keep the license and copyright notices. What do these mean? |
| License source | github.com/dauparas/LigandMPNN#readme · checked 2026-10-18 |
| Pricing | Open source |
| Tags | #inverse-folding#design#binder-design |
| Last checked | · links verified |
Compare with… LigandMPNN vs ProteinMPNN
Can I use LigandMPNN commercially?
Yes. The recorded license is permissive, so closed-source commercial use is allowed if you keep the notices.
- Code: MIT
- Weights: code and model parameters MIT per README
- Source: github.com/dauparas/LigandMPNN#readme · license checked 2026-10-18
What the license types mean · not legal advice.
Specs
| Task | Sequence design (inverse folding) with ligand, nucleotide and metal context; side-chain packing |
|---|---|
| Modalities | Protein with small molecules, nucleotides, metals |
| MSA required | — |
| Open weights | Yes (MIT) |
| GPU need | — |
| GitHub stars | 645 |
| Interface | Package |
Compare with…
Pick 1–3 other Protein structure & design tools to compare side by side with LigandMPNN.
Related
DeepMind's open-source protein structure prediction model (monomer and multimer).
Structure prediction of proteins, nucleic acids, ligands and their complexes; hosted web server plus open code, weights on request for non-commercial use.
Open biomolecular interaction models (Boltz-1, Boltz-2) for complex structure and binding-affinity prediction.