Consistent Synthetic Sequences Unlock Structural Diversity in Fully Atomistic De Novo Protein Design
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917137790009344 |
|---|---|
| author | Reidenbach, Danny Cao, Zhonglin Zhang, Zuobai Didi, Kieran Geffner, Tomas Zhou, Guoqing Tang, Jian Dallago, Christian Vahdat, Arash Kucukbenli, Emine Kreis, Karsten |
| author_facet | Reidenbach, Danny Cao, Zhonglin Zhang, Zuobai Didi, Kieran Geffner, Tomas Zhou, Guoqing Tang, Jian Dallago, Christian Vahdat, Arash Kucukbenli, Emine Kreis, Karsten |
| contents | High-quality training datasets are crucial for the development of effective protein design models, but existing synthetic datasets often include unfavorable sequence-structure pairs, impairing generative model performance. We leverage ProteinMPNN, whose sequences are experimentally favorable as well as amenable to folding, together with structure prediction models to align high-quality synthetic structures with recoverable synthetic sequences. In that way, we create a new dataset designed specifically for training expressive, fully atomistic protein generators. By retraining La-Proteina, which models discrete residue type and side chain structure in a continuous latent space, on this dataset, we achieve new state-of-the-art results, with improvements of +54% in structural diversity and +27% in co-designability. To validate the broad utility of our approach, we further introduce Proteina Atomistica, a unified flow-based framework that jointly learns the distribution of protein backbone structure, discrete sequences, and atomistic side chains without latent variables. We again find that training on our new sequence-structure data dramatically boosts benchmark performance, improving \method's structural diversity by +73% and co-designability by +5%. Our work highlights the critical importance of aligned sequence-structure data for training high-performance de novo protein design models. Our new dataset https://catalog.ngc.nvidia.com/orgs/nvidia/teams/clara/resources/proteina-atomistica_data/files?version=release , the Consistency Distilled Synthetic Protein Database, is made available as an open-source resource. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_01976 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Consistent Synthetic Sequences Unlock Structural Diversity in Fully Atomistic De Novo Protein Design Reidenbach, Danny Cao, Zhonglin Zhang, Zuobai Didi, Kieran Geffner, Tomas Zhou, Guoqing Tang, Jian Dallago, Christian Vahdat, Arash Kucukbenli, Emine Kreis, Karsten Biomolecules High-quality training datasets are crucial for the development of effective protein design models, but existing synthetic datasets often include unfavorable sequence-structure pairs, impairing generative model performance. We leverage ProteinMPNN, whose sequences are experimentally favorable as well as amenable to folding, together with structure prediction models to align high-quality synthetic structures with recoverable synthetic sequences. In that way, we create a new dataset designed specifically for training expressive, fully atomistic protein generators. By retraining La-Proteina, which models discrete residue type and side chain structure in a continuous latent space, on this dataset, we achieve new state-of-the-art results, with improvements of +54% in structural diversity and +27% in co-designability. To validate the broad utility of our approach, we further introduce Proteina Atomistica, a unified flow-based framework that jointly learns the distribution of protein backbone structure, discrete sequences, and atomistic side chains without latent variables. We again find that training on our new sequence-structure data dramatically boosts benchmark performance, improving \method's structural diversity by +73% and co-designability by +5%. Our work highlights the critical importance of aligned sequence-structure data for training high-performance de novo protein design models. Our new dataset https://catalog.ngc.nvidia.com/orgs/nvidia/teams/clara/resources/proteina-atomistica_data/files?version=release , the Consistency Distilled Synthetic Protein Database, is made available as an open-source resource. |
| title | Consistent Synthetic Sequences Unlock Structural Diversity in Fully Atomistic De Novo Protein Design |
| topic | Biomolecules |
| url | https://arxiv.org/abs/2512.01976 |