Cross-Chirality Generalization by Axial Vectors for Hetero-Chiral Protein-Peptide Interaction Design
Ziyi Yang, Zitong Tian, Yinjun Jia, Tianyi Zhang, Jiqing Zheng, Hao Wang, Yubu Su, Juncai He, Lei Liu, Yanyan Lan · 2026
Read Cross-Chirality Generalization by Axial Vectors for Hetero-Chiral Protein-Peptide Interaction Design by Ziyi Yang, Zitong Tian, Yinjun Jia, Tianyi Zhang, Jiqing Zheng, Hao Wang, Yubu Su, Juncai He, Lei Liu, Yanyan Lan (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-20176.txt.
Description
D-peptide binders targeting L-proteins hold therapeutic promise. Generating them with machine learning remains largely unexplored, even as target-conditioned peptide design advances rapidly. A 2026 arXiv paper by Ziyi Yang and nine co-authors shows cross-chirality generalization is feasible. Injecting axial features into equivariant polar vector features lets models trained on homo-chiral data design hetero-chiral binders.
Implemented within a latent diffusion model, the method designs D-peptide binders that beat existing tools in computational benchmarks and show efficacy in wet-lab validation. To the authors' knowledge, theirs is the first generative AI for de novo D-peptide binder design validated in the wet lab. It offers fresh views on handling chirality in protein design, showing training data from one chirality can generalize to the mirror-image task.
Catalog record
| CREATOR | Ziyi Yang, Zitong Tian, Yinjun Jia, Tianyi Zhang, Jiqing Zheng, Hao Wang, Yubu Su, Juncai He, Lei Liu, Yanyan Lan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 84 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-20176