Cargando…
BERTrand—peptide:TCR binding prediction using Bidirectional Encoder Representations from Transformers augmented with random TCR pairing
MOTIVATION: The advent of T-cell receptor (TCR) sequencing experiments allowed for a significant increase in the amount of peptide:TCR binding data available and a number of machine-learning models appeared in recent years. High-quality prediction models for a fixed epitope sequence are feasible, pr...
Autores principales: | , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Oxford University Press
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10444968/ https://www.ncbi.nlm.nih.gov/pubmed/37535685 http://dx.doi.org/10.1093/bioinformatics/btad468 |
_version_ | 1785094070913204224 |
---|---|
author | Myronov, Alexander Mazzocco, Giovanni Król, Paulina Plewczynski, Dariusz |
author_facet | Myronov, Alexander Mazzocco, Giovanni Król, Paulina Plewczynski, Dariusz |
author_sort | Myronov, Alexander |
collection | PubMed |
description | MOTIVATION: The advent of T-cell receptor (TCR) sequencing experiments allowed for a significant increase in the amount of peptide:TCR binding data available and a number of machine-learning models appeared in recent years. High-quality prediction models for a fixed epitope sequence are feasible, provided enough known binding TCR sequences are available. However, their performance drops significantly for previously unseen peptides. RESULTS: We prepare the dataset of known peptide:TCR binders and augment it with negative decoys created using healthy donors’ T-cell repertoires. We employ deep learning methods commonly applied in Natural Language Processing to train part a peptide:TCR binding model with a degree of cross-peptide generalization (0.69 AUROC). We demonstrate that BERTrand outperforms the published methods when evaluated on peptide sequences not used during model training. AVAILABILITY AND IMPLEMENTATION: The datasets and the code for model training are available at https://github.com/SFGLab/bertrand. |
format | Online Article Text |
id | pubmed-10444968 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | Oxford University Press |
record_format | MEDLINE/PubMed |
spelling | pubmed-104449682023-08-24 BERTrand—peptide:TCR binding prediction using Bidirectional Encoder Representations from Transformers augmented with random TCR pairing Myronov, Alexander Mazzocco, Giovanni Król, Paulina Plewczynski, Dariusz Bioinformatics Original Paper MOTIVATION: The advent of T-cell receptor (TCR) sequencing experiments allowed for a significant increase in the amount of peptide:TCR binding data available and a number of machine-learning models appeared in recent years. High-quality prediction models for a fixed epitope sequence are feasible, provided enough known binding TCR sequences are available. However, their performance drops significantly for previously unseen peptides. RESULTS: We prepare the dataset of known peptide:TCR binders and augment it with negative decoys created using healthy donors’ T-cell repertoires. We employ deep learning methods commonly applied in Natural Language Processing to train part a peptide:TCR binding model with a degree of cross-peptide generalization (0.69 AUROC). We demonstrate that BERTrand outperforms the published methods when evaluated on peptide sequences not used during model training. AVAILABILITY AND IMPLEMENTATION: The datasets and the code for model training are available at https://github.com/SFGLab/bertrand. Oxford University Press 2023-08-03 /pmc/articles/PMC10444968/ /pubmed/37535685 http://dx.doi.org/10.1093/bioinformatics/btad468 Text en © The Author(s) 2023. Published by Oxford University Press. https://creativecommons.org/licenses/by/4.0/This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. |
spellingShingle | Original Paper Myronov, Alexander Mazzocco, Giovanni Król, Paulina Plewczynski, Dariusz BERTrand—peptide:TCR binding prediction using Bidirectional Encoder Representations from Transformers augmented with random TCR pairing |
title | BERTrand—peptide:TCR binding prediction using Bidirectional Encoder Representations from Transformers augmented with random TCR pairing |
title_full | BERTrand—peptide:TCR binding prediction using Bidirectional Encoder Representations from Transformers augmented with random TCR pairing |
title_fullStr | BERTrand—peptide:TCR binding prediction using Bidirectional Encoder Representations from Transformers augmented with random TCR pairing |
title_full_unstemmed | BERTrand—peptide:TCR binding prediction using Bidirectional Encoder Representations from Transformers augmented with random TCR pairing |
title_short | BERTrand—peptide:TCR binding prediction using Bidirectional Encoder Representations from Transformers augmented with random TCR pairing |
title_sort | bertrand—peptide:tcr binding prediction using bidirectional encoder representations from transformers augmented with random tcr pairing |
topic | Original Paper |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10444968/ https://www.ncbi.nlm.nih.gov/pubmed/37535685 http://dx.doi.org/10.1093/bioinformatics/btad468 |
work_keys_str_mv | AT myronovalexander bertrandpeptidetcrbindingpredictionusingbidirectionalencoderrepresentationsfromtransformersaugmentedwithrandomtcrpairing AT mazzoccogiovanni bertrandpeptidetcrbindingpredictionusingbidirectionalencoderrepresentationsfromtransformersaugmentedwithrandomtcrpairing AT krolpaulina bertrandpeptidetcrbindingpredictionusingbidirectionalencoderrepresentationsfromtransformersaugmentedwithrandomtcrpairing AT plewczynskidariusz bertrandpeptidetcrbindingpredictionusingbidirectionalencoderrepresentationsfromtransformersaugmentedwithrandomtcrpairing |