Improving Protein Sequence Design through Designability Preference Optimization
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916770426650624 |
|---|---|
| author | Xue, Fanglei Kubaney, Andrew Guo, Zhichun Min, Joseph K. Liu, Ge Yang, Yi Baker, David |
| author_facet | Xue, Fanglei Kubaney, Andrew Guo, Zhichun Min, Joseph K. Liu, Ge Yang, Yi Baker, David |
| contents | Protein sequence design methods have demonstrated strong performance in sequence generation for de novo protein design. However, as the training objective was sequence recovery, it does not guarantee designability--the likelihood that a designed sequence folds into the desired structure. To bridge this gap, we redefine the training objective by steering sequence generation toward high designability. To do this, we integrate Direct Preference Optimization (DPO), using AlphaFold pLDDT scores as the preference signal, which significantly improves the in silico design success rate. To further refine sequence generation at a finer, residue-level granularity, we introduce Residue-level Designability Preference Optimization (ResiDPO), which applies residue-level structural rewards and decouples optimization across residues. This enables direct improvement in designability while preserving regions that already perform well. Using a curated dataset with residue-level annotations, we fine-tune LigandMPNN with ResiDPO to obtain EnhancedMPNN, which achieves a nearly 3-fold increase in in silico design success rate (from 6.56% to 17.57%) on a challenging enzyme design benchmark. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_00297 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Improving Protein Sequence Design through Designability Preference Optimization Xue, Fanglei Kubaney, Andrew Guo, Zhichun Min, Joseph K. Liu, Ge Yang, Yi Baker, David Machine Learning Artificial Intelligence Biomolecules Protein sequence design methods have demonstrated strong performance in sequence generation for de novo protein design. However, as the training objective was sequence recovery, it does not guarantee designability--the likelihood that a designed sequence folds into the desired structure. To bridge this gap, we redefine the training objective by steering sequence generation toward high designability. To do this, we integrate Direct Preference Optimization (DPO), using AlphaFold pLDDT scores as the preference signal, which significantly improves the in silico design success rate. To further refine sequence generation at a finer, residue-level granularity, we introduce Residue-level Designability Preference Optimization (ResiDPO), which applies residue-level structural rewards and decouples optimization across residues. This enables direct improvement in designability while preserving regions that already perform well. Using a curated dataset with residue-level annotations, we fine-tune LigandMPNN with ResiDPO to obtain EnhancedMPNN, which achieves a nearly 3-fold increase in in silico design success rate (from 6.56% to 17.57%) on a challenging enzyme design benchmark. |
| title | Improving Protein Sequence Design through Designability Preference Optimization |
| topic | Machine Learning Artificial Intelligence Biomolecules |
| url | https://arxiv.org/abs/2506.00297 |