Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report

Many different solutions to predicting the cognate epitope target of a T-cell receptor (TCR) have been proposed. However several questions on the advantages and disadvantages of these different approaches remain unresolved, as most methods have only been evaluated within the context of their initial...

Full description

Saved in:
Bibliographic Details
Main Authors: Pieter Meysman, Justin Barton, Barbara Bravi, Liel Cohen-Lavi, Vadim Karnaukhov, Elias Lilleskov, Alessandro Montemurro, Morten Nielsen, Thierry Mora, Paul Pereira, Anna Postovskaya, María Rodríguez Martínez, Jorge Fernandez-de-Cossio-Diaz, Alexandra Vujkovic, Aleksandra M. Walczak, Anna Weber, Rose Yin, Anne Eugster, Virag Sharma
Format: Article
Language:English
Published: Elsevier 2023-03-01
Series:ImmunoInformatics
Online Access:http://www.sciencedirect.com/science/article/pii/S2667119023000046
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850073566871027712
author Pieter Meysman
Justin Barton
Barbara Bravi
Liel Cohen-Lavi
Vadim Karnaukhov
Elias Lilleskov
Alessandro Montemurro
Morten Nielsen
Thierry Mora
Paul Pereira
Anna Postovskaya
María Rodríguez Martínez
Jorge Fernandez-de-Cossio-Diaz
Alexandra Vujkovic
Aleksandra M. Walczak
Anna Weber
Rose Yin
Anne Eugster
Virag Sharma
author_facet Pieter Meysman
Justin Barton
Barbara Bravi
Liel Cohen-Lavi
Vadim Karnaukhov
Elias Lilleskov
Alessandro Montemurro
Morten Nielsen
Thierry Mora
Paul Pereira
Anna Postovskaya
María Rodríguez Martínez
Jorge Fernandez-de-Cossio-Diaz
Alexandra Vujkovic
Aleksandra M. Walczak
Anna Weber
Rose Yin
Anne Eugster
Virag Sharma
author_sort Pieter Meysman
collection DOAJ
description Many different solutions to predicting the cognate epitope target of a T-cell receptor (TCR) have been proposed. However several questions on the advantages and disadvantages of these different approaches remain unresolved, as most methods have only been evaluated within the context of their initial publications and data sets. Here, we report the findings of the first public TCR-epitope prediction benchmark performed on 23 prediction models in the context of the ImmRep 2022 TCR-epitope specificity workshop. This benchmark revealed that the use of paired-chain alpha-beta, as well as CDR1/2 or V/J information, when available, improves classification obtained with CDR3 data, independent of the underlying approach. In addition, we found that straight-forward distance-based approaches can achieve a respectable performance when compared to more complex machine-learning models. Finally, we highlight the need for a truly independent follow-up benchmark and provide recommendations for the design of such a next benchmark.
format Article
id doaj-art-c337aa296bc14f0e8040baaeb0bebb9e
institution DOAJ
issn 2667-1190
language English
publishDate 2023-03-01
publisher Elsevier
record_format Article
series ImmunoInformatics
spelling doaj-art-c337aa296bc14f0e8040baaeb0bebb9e2025-08-20T02:46:47ZengElsevierImmunoInformatics2667-11902023-03-01910002410.1016/j.immuno.2023.100024Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop reportPieter Meysman0Justin Barton1Barbara Bravi2Liel Cohen-Lavi3Vadim Karnaukhov4Elias Lilleskov5Alessandro Montemurro6Morten Nielsen7Thierry Mora8Paul Pereira9Anna Postovskaya10María Rodríguez Martínez11Jorge Fernandez-de-Cossio-Diaz12Alexandra Vujkovic13Aleksandra M. Walczak14Anna Weber15Rose Yin16Anne Eugster17Virag Sharma18AUDACIS, University of Antwerp, Antwerp, Belgium; ADREM data lab, Department of Computer Science, University of Antwerp, Antwerp, Belgium; Corresponding author at: AUDACIS, University of Antwerp, Antwerp, Belgium.Institute of Structural and Molecular Biology, University of London, London, United KingdomDepartment of Mathematics, Imperial College London, London, United KingdomNational Institute for Biotechnology in the Negev, Ben-Gurion University of the Negev, Beer-Sheva, Israel; Department of Industrial Engineering and Management, Ben-Gurion University of the Negev, Beer-Sheva, IsraelINSERM U932, PSL University, Institut Curie, Paris 75005, France; Laboratoire de physique de l'Ecole normale superieure, CNRS, PSL University, Sorbonne Universite, Universite Paris-Cité, Paris 75005, FranceDepartment of Physics, University of Washington, Seattle, WA, USADepartment of Health Technology, Technical University of Denmark, Lyngby DK-2800, DenmarkDepartment of Health Technology, Technical University of Denmark, Lyngby DK-2800, DenmarkLaboratoire de physique de l'Ecole normale superieure, CNRS, PSL University, Sorbonne Universite, Universite Paris-Cité, Paris 75005, FranceLaboratoire de physique de l'Ecole normale superieure, CNRS, PSL University, Sorbonne Universite, Universite Paris-Cité, Paris 75005, France; Sanofi R&D, Chilly-Mazarin 91380, FranceAUDACIS, University of Antwerp, Antwerp, Belgium; ADREM data lab, Department of Computer Science, University of Antwerp, Antwerp, Belgium; Clinical Virology Unit, Institute of Tropical Medicine, Antwerp, BelgiumIBM Research Europe, Säumerstrasse 4, Rüschlikon 8803, SwitzerlandLaboratoire de physique de l'Ecole normale superieure, CNRS, PSL University, Sorbonne Universite, Universite Paris-Cité, Paris 75005, FranceAUDACIS, University of Antwerp, Antwerp, Belgium; ADREM data lab, Department of Computer Science, University of Antwerp, Antwerp, Belgium; Clinical Virology Unit, Institute of Tropical Medicine, Antwerp, BelgiumLaboratoire de physique de l'Ecole normale superieure, CNRS, PSL University, Sorbonne Universite, Universite Paris-Cité, Paris 75005, FranceIBM Research Europe, Säumerstrasse 4, Rüschlikon 8803, SwitzerlandDepartment of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, USACenter for Regenerative Therapies Dresden, Faculty of Medicine, TU Dresden, Dresden, Germany; Corresponding authors.Department of Chemical Sciences, School of Natural Sciences, University of Limerick, Limerick V94 T9PX, Ireland; Corresponding authors.Many different solutions to predicting the cognate epitope target of a T-cell receptor (TCR) have been proposed. However several questions on the advantages and disadvantages of these different approaches remain unresolved, as most methods have only been evaluated within the context of their initial publications and data sets. Here, we report the findings of the first public TCR-epitope prediction benchmark performed on 23 prediction models in the context of the ImmRep 2022 TCR-epitope specificity workshop. This benchmark revealed that the use of paired-chain alpha-beta, as well as CDR1/2 or V/J information, when available, improves classification obtained with CDR3 data, independent of the underlying approach. In addition, we found that straight-forward distance-based approaches can achieve a respectable performance when compared to more complex machine-learning models. Finally, we highlight the need for a truly independent follow-up benchmark and provide recommendations for the design of such a next benchmark.http://www.sciencedirect.com/science/article/pii/S2667119023000046
spellingShingle Pieter Meysman
Justin Barton
Barbara Bravi
Liel Cohen-Lavi
Vadim Karnaukhov
Elias Lilleskov
Alessandro Montemurro
Morten Nielsen
Thierry Mora
Paul Pereira
Anna Postovskaya
María Rodríguez Martínez
Jorge Fernandez-de-Cossio-Diaz
Alexandra Vujkovic
Aleksandra M. Walczak
Anna Weber
Rose Yin
Anne Eugster
Virag Sharma
Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report
ImmunoInformatics
title Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report
title_full Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report
title_fullStr Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report
title_full_unstemmed Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report
title_short Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report
title_sort benchmarking solutions to the t cell receptor epitope prediction problem immrep22 workshop report
url http://www.sciencedirect.com/science/article/pii/S2667119023000046
work_keys_str_mv AT pietermeysman benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT justinbarton benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT barbarabravi benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT lielcohenlavi benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT vadimkarnaukhov benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT eliaslilleskov benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT alessandromontemurro benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT mortennielsen benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT thierrymora benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT paulpereira benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT annapostovskaya benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT mariarodriguezmartinez benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT jorgefernandezdecossiodiaz benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT alexandravujkovic benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT aleksandramwalczak benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT annaweber benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT roseyin benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT anneeugster benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport
AT viragsharma benchmarkingsolutionstothetcellreceptorepitopepredictionproblemimmrep22workshopreport