Pinal
Pinal is a protein structure & design tool whose code is published by westlake-repl on GitHub, first public in 2025, described in Nature (2026). Westlake University model that designs protein sequences from a natural-language description: a text-to-structure model (T2struc) proposes structure tokens, then SaProt-T writes the sequence.
| Category | Protein structure & design |
|---|---|
| Developer (GitHub) | westlake-repl |
| Task | Protein sequence design from natural-language instructions |
| Interface | Python API, web server |
| Website | www.denovo-pinal.com |
| GitHub | westlake-repl/Denovo-Pinal |
| First public | 2025-01-08 (source) |
| Paper | https://doi.org/10.1038/s41586-026-11205-z · Nature (2026) |
| Code license | Apache-2.0 |
| License note | code Apache-2.0 (GitHub LICENSE) |
| What this allows | Permissive (commercial OK) Use, modify and ship it inside closed-source commercial products; keep the license text and copyright notices. What do these mean? |
| License source | github.com/westlake-repl/Denovo-Pinal · checked 2026-10-23 |
| Pricing | Open source |
| Tags | #protein-design#text-to-protein#protein-language-model |
| Last checked | · links verified |
Can I use Pinal commercially?
Yes. The recorded license is permissive, so closed-source commercial use is allowed if you keep the notices.
- Code: Apache-2.0
- Source: github.com/westlake-repl/Denovo-Pinal · license checked 2026-10-23
What the license types mean · not legal advice.
Specs
| Task | Protein sequence design from natural-language instructions |
|---|---|
| Modalities | — |
| MSA required | — |
| Open weights | Yes (Hugging Face westlake-repl/Pinal) |
| GPU need | — |
| GitHub stars | 138 |
| Interface | Python API, web server |
Compare with…
Pick 1–3 other Protein structure & design tools to compare side by side with Pinal.
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.