Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning.

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 like...

Full description

Saved in:
Bibliographic Details
Main Authors: Aime Bienfait Igiraneza, Panagiota Zacharopoulou, Robert Hinch, Chris Wymant, Lucie Abeler-Dörner, John Frater, Christophe Fraser
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2024-11-01
Series:PLoS Computational Biology
Online Access:https://doi.org/10.1371/journal.pcbi.1012618
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846128262249447424
author Aime Bienfait Igiraneza
Panagiota Zacharopoulou
Robert Hinch
Chris Wymant
Lucie Abeler-Dörner
John Frater
Christophe Fraser
author_facet Aime Bienfait Igiraneza
Panagiota Zacharopoulou
Robert Hinch
Chris Wymant
Lucie Abeler-Dörner
John Frater
Christophe Fraser
author_sort Aime Bienfait Igiraneza
collection DOAJ
description 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.
format Article
id doaj-art-8f9546c98e2e4f1f89f01ef63ba4a262
institution Kabale University
issn 1553-734X
1553-7358
language English
publishDate 2024-11-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS Computational Biology
spelling doaj-art-8f9546c98e2e4f1f89f01ef63ba4a2622024-12-11T05:31:03ZengPublic Library of Science (PLoS)PLoS Computational Biology1553-734X1553-73582024-11-012011e101261810.1371/journal.pcbi.1012618Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning.Aime Bienfait IgiranezaPanagiota ZacharopoulouRobert HinchChris WymantLucie Abeler-DörnerJohn FraterChristophe FraserThe 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.https://doi.org/10.1371/journal.pcbi.1012618
spellingShingle Aime Bienfait Igiraneza
Panagiota Zacharopoulou
Robert Hinch
Chris Wymant
Lucie Abeler-Dörner
John Frater
Christophe Fraser
Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning.
PLoS Computational Biology
title Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning.
title_full Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning.
title_fullStr Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning.
title_full_unstemmed Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning.
title_short Learning patterns of HIV-1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi-task learning.
title_sort learning patterns of hiv 1 resistance to broadly neutralizing antibodies with reduced subtype bias using multi task learning
url https://doi.org/10.1371/journal.pcbi.1012618
work_keys_str_mv AT aimebienfaitigiraneza learningpatternsofhiv1resistancetobroadlyneutralizingantibodieswithreducedsubtypebiasusingmultitasklearning
AT panagiotazacharopoulou learningpatternsofhiv1resistancetobroadlyneutralizingantibodieswithreducedsubtypebiasusingmultitasklearning
AT roberthinch learningpatternsofhiv1resistancetobroadlyneutralizingantibodieswithreducedsubtypebiasusingmultitasklearning
AT chriswymant learningpatternsofhiv1resistancetobroadlyneutralizingantibodieswithreducedsubtypebiasusingmultitasklearning
AT lucieabelerdorner learningpatternsofhiv1resistancetobroadlyneutralizingantibodieswithreducedsubtypebiasusingmultitasklearning
AT johnfrater learningpatternsofhiv1resistancetobroadlyneutralizingantibodieswithreducedsubtypebiasusingmultitasklearning
AT christophefraser learningpatternsofhiv1resistancetobroadlyneutralizingantibodieswithreducedsubtypebiasusingmultitasklearning