---
license: apache-2.0
---
*De novo* protein sequence design model based on pretrained protein language model [Progen2](https://arxiv.org/abs/2206.13517), multiple structure encoders: [ESMif](https://www.biorxiv.org/lookup/doi/10.1101/2022.04.10.487779), [chroma](https://www.nature.com/articles/s41586-023-06728-8), and [ProRefiner](https://github.com/veghen/ProRefiner).

Trained on [CATH 4.2](https://www.cathdb.info/) and [AFDB](https://alphafold.ebi.ac.uk/).