Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2410.19471 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914549782806528 |
|---|---|
| author | Park, Ryan Hsu, Darren J. Roland, C. Brian Korshunova, Maria Tessler, Chen Mannor, Shie Viessmann, Olivia Trentini, Bruno |
| author_facet | Park, Ryan Hsu, Darren J. Roland, C. Brian Korshunova, Maria Tessler, Chen Mannor, Shie Viessmann, Olivia Trentini, Bruno |
| contents | Inverse folding models play an important role in structure-based design by predicting amino acid sequences that fold into desired reference structures. Models like ProteinMPNN, a message-passing encoder-decoder model, are trained to reliably produce new sequences from a reference structure. However, when applied to peptides, these models are prone to generating repetitive sequences that do not fold into the reference structure. To address this, we fine-tune ProteinMPNN to produce diverse and structurally consistent peptide sequences via Direct Preference Optimization (DPO). We derive two enhancements to DPO: online diversity regularization and domain-specific priors. Additionally, we develop a new understanding on improving diversity in decoder models. When conditioned on OpenFold generated structures, our fine-tuned models achieve state-of-the-art structural similarity scores, improving base ProteinMPNN by at least 8%. Compared to standard DPO, our regularized method achieves up to 20% higher sequence diversity with no loss in structural similarity score. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_19471 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Improving Inverse Folding for Peptide Design with Diversity-regularized Direct Preference Optimization Park, Ryan Hsu, Darren J. Roland, C. Brian Korshunova, Maria Tessler, Chen Mannor, Shie Viessmann, Olivia Trentini, Bruno Machine Learning Artificial Intelligence Inverse folding models play an important role in structure-based design by predicting amino acid sequences that fold into desired reference structures. Models like ProteinMPNN, a message-passing encoder-decoder model, are trained to reliably produce new sequences from a reference structure. However, when applied to peptides, these models are prone to generating repetitive sequences that do not fold into the reference structure. To address this, we fine-tune ProteinMPNN to produce diverse and structurally consistent peptide sequences via Direct Preference Optimization (DPO). We derive two enhancements to DPO: online diversity regularization and domain-specific priors. Additionally, we develop a new understanding on improving diversity in decoder models. When conditioned on OpenFold generated structures, our fine-tuned models achieve state-of-the-art structural similarity scores, improving base ProteinMPNN by at least 8%. Compared to standard DPO, our regularized method achieves up to 20% higher sequence diversity with no loss in structural similarity score. |
| title | Improving Inverse Folding for Peptide Design with Diversity-regularized Direct Preference Optimization |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2410.19471 |