Improving Inverse Folding for Peptide Design with Diversity-Regularized Direct Preference Optimization

Ryan Park, Darren J. Hsu, Chen Tessler, Maria Korshunova, C. Brian Roland, Olivia Viessmann, Bruno Trentini

Keywords: Inverse Folding, Structure-based design, Peptide Design, Direct Preference Optimization

TL;DR: Finetuned ProteinMPNN with up to 20% diversity gains with no loss in structural similarity score through DPO

Abstract: 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 finetune 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 finetuned 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.

Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 11912