Predicting TCR-Epitope Binding Specificity Using Deep Metric Learning and Multimodal Learning
Understanding the recognition of specific epitopes by cytotoxic T cells is a central problem in immunology. Although predicting binding between peptides and the class I Major Histocompatibility Complex (MHC) has had success, predicting interactions between T cell receptors (TCRs) and MHC class I-pep...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Genes |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4425/12/4/572 |
| _version_ | 1797412188201156608 |
|---|---|
| author | Alan M. Luu Jacob R. Leistico Tim Miller Somang Kim Jun S. Song |
| author_facet | Alan M. Luu Jacob R. Leistico Tim Miller Somang Kim Jun S. Song |
| author_sort | Alan M. Luu |
| collection | DOAJ |
| description | Understanding the recognition of specific epitopes by cytotoxic T cells is a central problem in immunology. Although predicting binding between peptides and the class I Major Histocompatibility Complex (MHC) has had success, predicting interactions between T cell receptors (TCRs) and MHC class I-peptide complexes (pMHC) remains elusive. This paper utilizes a convolutional neural network model employing deep metric learning and multimodal learning to perform two critical tasks in TCR-epitope binding prediction: identifying the TCRs that bind a given epitope from a TCR repertoire, and identifying the binding epitope of a given TCR from a list of candidate epitopes. Our model can perform both tasks simultaneously and reveals that inconsistent preprocessing of TCR sequences can confound binding prediction. Applying a neural network interpretation method identifies key amino acid sequence patterns and positions within the TCR, important for binding specificity. Contrary to common assumption, known crystal structures of TCR-pMHC complexes show that the predicted salient amino acid positions are not necessarily the closest to the epitopes, implying that physical proximity may not be a good proxy for importance in determining TCR-epitope specificity. Our work thus provides an insight into the learned predictive features of TCR-epitope binding specificity and advances the associated classification tasks. |
| first_indexed | 2024-03-09T04:58:24Z |
| format | Article |
| id | doaj.art-b0c6be31ce7f48e1828441f7dfce0486 |
| institution | Directory Open Access Journal |
| issn | 2073-4425 |
| language | English |
| last_indexed | 2024-03-09T04:58:24Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Genes |
| spelling | doaj.art-b0c6be31ce7f48e1828441f7dfce04862023-12-03T13:02:54ZengMDPI AGGenes2073-44252021-04-0112457210.3390/genes12040572Predicting TCR-Epitope Binding Specificity Using Deep Metric Learning and Multimodal LearningAlan M. Luu0Jacob R. Leistico1Tim Miller2Somang Kim3Jun S. Song4Department of Physics, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USADepartment of Physics, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USADepartment of Physics, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USADepartment of Physics, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USADepartment of Physics, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USAUnderstanding the recognition of specific epitopes by cytotoxic T cells is a central problem in immunology. Although predicting binding between peptides and the class I Major Histocompatibility Complex (MHC) has had success, predicting interactions between T cell receptors (TCRs) and MHC class I-peptide complexes (pMHC) remains elusive. This paper utilizes a convolutional neural network model employing deep metric learning and multimodal learning to perform two critical tasks in TCR-epitope binding prediction: identifying the TCRs that bind a given epitope from a TCR repertoire, and identifying the binding epitope of a given TCR from a list of candidate epitopes. Our model can perform both tasks simultaneously and reveals that inconsistent preprocessing of TCR sequences can confound binding prediction. Applying a neural network interpretation method identifies key amino acid sequence patterns and positions within the TCR, important for binding specificity. Contrary to common assumption, known crystal structures of TCR-pMHC complexes show that the predicted salient amino acid positions are not necessarily the closest to the epitopes, implying that physical proximity may not be a good proxy for importance in determining TCR-epitope specificity. Our work thus provides an insight into the learned predictive features of TCR-epitope binding specificity and advances the associated classification tasks.https://www.mdpi.com/2073-4425/12/4/572T cell receptorsepitope binding specificitydeep learningmetric learningmultimodal learning |
| spellingShingle | Alan M. Luu Jacob R. Leistico Tim Miller Somang Kim Jun S. Song Predicting TCR-Epitope Binding Specificity Using Deep Metric Learning and Multimodal Learning Genes T cell receptors epitope binding specificity deep learning metric learning multimodal learning |
| title | Predicting TCR-Epitope Binding Specificity Using Deep Metric Learning and Multimodal Learning |
| title_full | Predicting TCR-Epitope Binding Specificity Using Deep Metric Learning and Multimodal Learning |
| title_fullStr | Predicting TCR-Epitope Binding Specificity Using Deep Metric Learning and Multimodal Learning |
| title_full_unstemmed | Predicting TCR-Epitope Binding Specificity Using Deep Metric Learning and Multimodal Learning |
| title_short | Predicting TCR-Epitope Binding Specificity Using Deep Metric Learning and Multimodal Learning |
| title_sort | predicting tcr epitope binding specificity using deep metric learning and multimodal learning |
| topic | T cell receptors epitope binding specificity deep learning metric learning multimodal learning |
| url | https://www.mdpi.com/2073-4425/12/4/572 |
| work_keys_str_mv | AT alanmluu predictingtcrepitopebindingspecificityusingdeepmetriclearningandmultimodallearning AT jacobrleistico predictingtcrepitopebindingspecificityusingdeepmetriclearningandmultimodallearning AT timmiller predictingtcrepitopebindingspecificityusingdeepmetriclearningandmultimodallearning AT somangkim predictingtcrepitopebindingspecificityusingdeepmetriclearningandmultimodallearning AT junssong predictingtcrepitopebindingspecificityusingdeepmetriclearningandmultimodallearning |