Tuning ProteinMPNN to reduce protein visibility via MHC Class I through direct preference optimization
Protein Engineering, Design and Selection · 10.01.2025 · 38
DOI: 10.1093/protein/gzaf003 ↗Kurz gefasst
Fachartikel in der Zeitschrift Protein Engineering, Design and Selection, online erschienen am 18.03.2025. Verfasst von Hans-Christof Gasser und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Informatics, University of Edinburgh, Edinburgh, EH8 9AB, United Kingdom. Hauptthemen laut MEDLINE: Histocompatibility Antigens Class I, Protein Engineering, Proteins. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: United Kingdom Research and Innovation, UKRI Centre for Doctoral Training in Biomedical AI at the University of Edinburgh, School of Informatics.
Abstract (Originalsprache)
ProteinMPNN is widely used in protein design workflows due to its ability to identify amino acid sequences that fold into specific 3D protein structures. In our work, we adjust ProteinMPNN to design proteins for a given 3D protein structure with reduced immune-visibility to cytotoxic T lymphocytes that recognize proteins via the MHC-I pathway. To achieve this, we developed a novel framework that integrates direct preference optimization (DPO)—a tuning method originally designed for large language models—with MHC-I peptide presentation predictions. This approach fosters the generation of designs with fewer MHC-I epitopes while preserving the protein’s original structure. Our results demonstrate that DPO effectively reduces MHC-I visibility without compromising the structural integrity of the proteins.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- School of Informatics, University of Edinburgh, Edinburgh, EH8 9AB, United Kingdom.
- School of Biological Sciences, University of Edinburgh, Edinburgh, EH9 3JH, United Kingdom.
- International Centre for Cancer Vaccine Science, University of Gdańsk, Gdańsk, 80-822, Poland.
- Department of Biochemistry and Microbiology, University of Victoria, Victoria, V8W 2Y2, Canada.
- The Canadian Association for Responsible AI in Medicine, Victoria, V8N 4W7, Canada.
ORCID-Profile: Hans-Christof Gasser, Diego A Oyarzún, Javier Antonio Alfaro, Ajitha Rajan
Förderung
- United Kingdom Research and Innovation
- UKRI Centre for Doctoral Training in Biomedical AI at the University of Edinburgh, School of Informatics
Zitieren
APA
Gasser, H. C., Oyarzún, D. A., Alfaro, J. A., & Rajan, A. (2025). Tuning ProteinMPNN to reduce protein visibility via MHC Class I through direct preference optimization. Protein Engineering, Design and Selection, 38. https://doi.org/10.1093/protein/gzaf003
Vancouver
Gasser HC, Oyarzún DA, Alfaro JA, Rajan A. Tuning ProteinMPNN to reduce protein visibility via MHC Class I through direct preference optimization. Protein Engineering, Design and Selection. 2025;38. doi:10.1093/protein/gzaf003
BibTeX
@article{Gasser2025,
author = {Gasser, Hans-Christof and Oyarzún, Diego A and Alfaro, Javier Antonio and Rajan, Ajitha},
title = {Tuning ProteinMPNN to reduce protein visibility via MHC Class I through direct preference optimization},
journal = {Protein Engineering, Design and Selection},
year = {2025},
volume = {38},
doi = {10.1093/protein/gzaf003}
}
EUVIMED