For antibody sequence generative modeling, mixture models may be all you need
Bioinformatics · 02.05.2024 · 40 (5)
DOI: 10.1093/bioinformatics/btae278 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 23.04.2024. Verfasst von Jonathan Parkinson und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemistry and Biochemistry, University of California, San Diego, La Jolla, CA 92093-0359, United States. Hauptthemen laut MEDLINE: Antibodies. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIAID NIH HHS, NIH HHS, National Institutes of Health.
Abstract (Originalsprache)
Motivation Antibody therapeutic candidates must exhibit not only tight binding to their target but also good developability properties, especially low risk of immunogenicity. Results In this work, we fit a simple generative model, SAM, to sixty million human heavy and seventy million human light chains. We show that the probability of a sequence calculated by the model distinguishes human sequences from other species with the same or better accuracy on a variety of benchmark datasets containing >400 million sequences than any other model in the literature, outperforming large language models (LLMs) by large margins. SAM can humanize sequences, generate new sequences, and score sequences for humanness. It is both fast and fully interpretable. Our results highlight the importance of using simple models as baselines for protein engineering tasks. We additionally introduce a new tool for numbering antibody sequences which is orders of magnitude faster than existing tools in the literature. Availability and implementation All tools developed in this study are available at https://github.com/Wang-lab-UCSD/AntPack.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Chemistry and Biochemistry, University of California, San Diego, La Jolla, CA 92093-0359, United States.
- MAP Bioscience, La Jolla, CA 92093, United States.
- Department of Cellular and Molecular Medicine, University of California, San Diego, La Jolla, CA 92093-0359, United States.
ORCID-Profile: Jonathan Parkinson
Förderung
- NIAID NIH HHS
- NIH HHS
- National Institutes of Health
Zitieren
APA
Parkinson, J., & Wang, W. (2024). For antibody sequence generative modeling, mixture models may be all you need. Bioinformatics, 40(5). https://doi.org/10.1093/bioinformatics/btae278
Vancouver
Parkinson J, Wang W. For antibody sequence generative modeling, mixture models may be all you need. Bioinformatics. 2024;40(5). doi:10.1093/bioinformatics/btae278
BibTeX
@article{Parkinson2024,
author = {Parkinson, Jonathan and Wang, Wei},
title = {For antibody sequence generative modeling, mixture models may be all you need},
journal = {Bioinformatics},
year = {2024},
volume = {40},
number = {5},
doi = {10.1093/bioinformatics/btae278}
}
EUVIMED