Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning
PLOS Computational Biology · 20.11.2024 · 20 (11) · e1012618
DOI: 10.1371/journal.pcbi.1012618 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 20.11.2024. Verfasst von Aime Bienfait Igiraneza und 6 weiteren Autor:innen. Beteiligte Einrichtung: Pandemic Sciences Institute, Nuffield Department of Medicine, University of Oxford, Oxford, United Kingdom. Hauptthemen laut MEDLINE: HIV-1, HIV Antibodies, HIV Infections, Machine Learning. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Wellcome Trust, Bill and Melinda Gates Foundation.
Abstract (Originalsprache)
The ability to predict HIV-1 resistance to broadly neutralizing antibodies (bnAbs) will increase bnAb therapeutic benefits. Machine learning is a powerful approach for such prediction. One challenge is that some HIV-1 subtypes in currently available training datasets are underrepresented, which likely affects models’ generalizability across subtypes. A second challenge is that combinations of bnAbs are required to avoid the inevitable resistance to a single bnAb, and computationally determining optimal combinations of bnAbs is an unsolved problem. Recently, machine learning models trained using resistance outcomes for multiple antibodies at once, a strategy called multi-task learning (MTL), have been shown to improve predictions. We develop a new model and show that, beyond the boost in performance, MTL also helps address the previous two challenges. Specifically, we demonstrate empirically that MTL can mitigate bias from underrepresented subtypes, and that MTL allows the model to learn patterns of co-resistance to combinations of antibodies, thus providing tools to predict antibodies’ epitopes and to potentially select optimal bnAb combinations. Our analyses, publicly available at https://github.com/iaime/LBUM , can be adapted to other infectious diseases that are treated with antibody therapy.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Pandemic Sciences Institute, Nuffield Department of Medicine, University of Oxford, Oxford, United Kingdom.
- Big Data Institute, Li Ka Shing Centre for Health Information and Discovery, Nuffield Department of Medicine, University of Oxford, Oxford, United Kingdom.
- Peter Medawar Building for Pathogen Research, Nuffield Department of Medicine, University of Oxford, Oxford, United Kingdom.
- NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, John Radcliffe Hospital, Oxford, United Kingdom.
ORCID-Profile: Aime Bienfait Igiraneza, Panagiota Zacharopoulou, Robert Hinch, Chris Wymant, Lucie Abeler-Dörner, John Frater, Christophe Fraser
Förderung
- Wellcome Trust
- Bill and Melinda Gates Foundation
Zitieren
APA
Igiraneza, A. B., Zacharopoulou, P., Hinch, R., Wymant, C., Abeler-Dörner, L., Frater, J., & Fraser, C. (2024). Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning. PLOS Computational Biology, 20(11), e1012618. https://doi.org/10.1371/journal.pcbi.1012618
Vancouver
Igiraneza AB, Zacharopoulou P, Hinch R, Wymant C, Abeler-Dörner L, Frater J, et al. Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning. PLOS Computational Biology. 2024;20(11):e1012618. doi:10.1371/journal.pcbi.1012618
BibTeX
@article{Igiraneza2024,
author = {Igiraneza, Aime Bienfait and Zacharopoulou, Panagiota and Hinch, Robert and Wymant, Chris and Abeler-Dörner, Lucie and Frater, John and Fraser, Christophe},
title = {Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning},
journal = {PLOS Computational Biology},
year = {2024},
volume = {20},
number = {11},
pages = {e1012618},
doi = {10.1371/journal.pcbi.1012618}
}
EUVIMED