TEINet: a deep learning framework for prediction of TCR–epitope binding specificity
Abstract The adaptive immune response to foreign antigens is initiated by T-cell receptor (TCR) recognition on the antigens. Recent experimental advances have enabled the generation of a large amount of TCR data and their cognate antigenic targets, allowing machine learning models to predict the bin...
Gespeichert in:
Veröffentlicht in: | Briefings in bioinformatics 2023-03, Vol.24 (2) |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | 2 |
container_start_page | |
container_title | Briefings in bioinformatics |
container_volume | 24 |
creator | Jiang, Yuepeng Huo, Miaozhe Cheng Li, Shuai |
description | Abstract
The adaptive immune response to foreign antigens is initiated by T-cell receptor (TCR) recognition on the antigens. Recent experimental advances have enabled the generation of a large amount of TCR data and their cognate antigenic targets, allowing machine learning models to predict the binding specificity of TCRs. In this work, we present TEINet, a deep learning framework that utilizes transfer learning to address this prediction problem. TEINet employs two separately pretrained encoders to transform TCR and epitope sequences into numerical vectors, which are subsequently fed into a fully connected neural network to predict their binding specificities. A major challenge for binding specificity prediction is the lack of a unified approach to sampling negative data. Here, we first assess the current negative sampling approaches comprehensively and suggest that the Unified Epitope is the most suitable one. Subsequently, we compare TEINet with three baseline methods and observe that TEINet achieves an average AUROC of 0.760, which outperforms baseline methods by 6.4–26%. Furthermore, we investigate the impacts of the pretraining step and notice that excessive pretraining may lower its transferability to the final prediction task. Our results and analysis show that TEINet can make an accurate prediction using only the TCR sequence (CDR3$\beta $) and the epitope sequence, providing novel insights to understand the interactions between TCRs and epitopes. |
doi_str_mv | 10.1093/bib/bbad086 |
format | Article |
fullrecord | <record><control><sourceid>proquest_TOX</sourceid><recordid>TN_cdi_proquest_miscellaneous_2786511483</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><oup_id>10.1093/bib/bbad086</oup_id><sourcerecordid>3049109824</sourcerecordid><originalsourceid>FETCH-LOGICAL-c385t-da4a51c84b1947baaf5322e3a18c5c350d9f4e1714c62c2eb18c5e6b76f4d7523</originalsourceid><addsrcrecordid>eNp9kM1qGzEURkVJqR23q-6LIBACZWJp9DvZBZOkAZNCsdeDpLkqSu3RVJohZJd3yBvmSTLGbhZdZHUvl_N9XA5CXyk5p6Ricxvs3FrTEC0_oCnlShWcCH6026UqBJdsgo5zviekJErTT2jCZEWUFHqK1qur2zvoL7DBDUCHN2BSG9rf2CezhYeY_mAfE-4SNMH1IbY4erxa_Hp5eoYu9LEDbEPb7BK5Axd8cKF__Iw-erPJ8OUwZ2h9fbVa_CiWP29uF5fLwjEt-qIx3AjqNLe04soa4wUrS2CGaiccE6SpPAeqKHeydCXY3R2kVdLzRomSzdDZvrdL8e8Aua-3ITvYbEwLcch1qbQUlHLNRvTkP_Q-Dqkdv6sZ4dVoUpd8pL7vKZdizgl83aWwNemxpqTe2a5H2_XB9kh_O3QOdgvNG_tP7wic7oE4dO82vQLqZ4iI</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3049109824</pqid></control><display><type>article</type><title>TEINet: a deep learning framework for prediction of TCR–epitope binding specificity</title><source>Access via Oxford University Press (Open Access Collection)</source><creator>Jiang, Yuepeng ; Huo, Miaozhe ; Cheng Li, Shuai</creator><creatorcontrib>Jiang, Yuepeng ; Huo, Miaozhe ; Cheng Li, Shuai</creatorcontrib><description>Abstract
The adaptive immune response to foreign antigens is initiated by T-cell receptor (TCR) recognition on the antigens. Recent experimental advances have enabled the generation of a large amount of TCR data and their cognate antigenic targets, allowing machine learning models to predict the binding specificity of TCRs. In this work, we present TEINet, a deep learning framework that utilizes transfer learning to address this prediction problem. TEINet employs two separately pretrained encoders to transform TCR and epitope sequences into numerical vectors, which are subsequently fed into a fully connected neural network to predict their binding specificities. A major challenge for binding specificity prediction is the lack of a unified approach to sampling negative data. Here, we first assess the current negative sampling approaches comprehensively and suggest that the Unified Epitope is the most suitable one. Subsequently, we compare TEINet with three baseline methods and observe that TEINet achieves an average AUROC of 0.760, which outperforms baseline methods by 6.4–26%. Furthermore, we investigate the impacts of the pretraining step and notice that excessive pretraining may lower its transferability to the final prediction task. Our results and analysis show that TEINet can make an accurate prediction using only the TCR sequence (CDR3$\beta $) and the epitope sequence, providing novel insights to understand the interactions between TCRs and epitopes.</description><identifier>ISSN: 1467-5463</identifier><identifier>EISSN: 1477-4054</identifier><identifier>DOI: 10.1093/bib/bbad086</identifier><identifier>PMID: 36907658</identifier><language>eng</language><publisher>England: Oxford University Press</publisher><subject>Adaptive immunity ; Antigens ; Binding ; Complementarity-determining region 3 ; Deep Learning ; Epitopes ; Epitopes, T-Lymphocyte ; Immune response ; Lymphocytes T ; Machine learning ; Neural networks ; Predictions ; Protein Binding ; Receptors, Antigen, T-Cell ; Sampling ; T cell receptors ; Transfer learning</subject><ispartof>Briefings in bioinformatics, 2023-03, Vol.24 (2)</ispartof><rights>The Author(s) 2023. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com 2023</rights><rights>The Author(s) 2023. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.</rights><rights>The Author(s) 2023. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c385t-da4a51c84b1947baaf5322e3a18c5c350d9f4e1714c62c2eb18c5e6b76f4d7523</citedby><cites>FETCH-LOGICAL-c385t-da4a51c84b1947baaf5322e3a18c5c350d9f4e1714c62c2eb18c5e6b76f4d7523</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,1604,27924,27925</link.rule.ids><linktorsrc>$$Uhttps://dx.doi.org/10.1093/bib/bbad086$$EView_record_in_Oxford_University_Press$$FView_record_in_$$GOxford_University_Press</linktorsrc><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/36907658$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Jiang, Yuepeng</creatorcontrib><creatorcontrib>Huo, Miaozhe</creatorcontrib><creatorcontrib>Cheng Li, Shuai</creatorcontrib><title>TEINet: a deep learning framework for prediction of TCR–epitope binding specificity</title><title>Briefings in bioinformatics</title><addtitle>Brief Bioinform</addtitle><description>Abstract
The adaptive immune response to foreign antigens is initiated by T-cell receptor (TCR) recognition on the antigens. Recent experimental advances have enabled the generation of a large amount of TCR data and their cognate antigenic targets, allowing machine learning models to predict the binding specificity of TCRs. In this work, we present TEINet, a deep learning framework that utilizes transfer learning to address this prediction problem. TEINet employs two separately pretrained encoders to transform TCR and epitope sequences into numerical vectors, which are subsequently fed into a fully connected neural network to predict their binding specificities. A major challenge for binding specificity prediction is the lack of a unified approach to sampling negative data. Here, we first assess the current negative sampling approaches comprehensively and suggest that the Unified Epitope is the most suitable one. Subsequently, we compare TEINet with three baseline methods and observe that TEINet achieves an average AUROC of 0.760, which outperforms baseline methods by 6.4–26%. Furthermore, we investigate the impacts of the pretraining step and notice that excessive pretraining may lower its transferability to the final prediction task. Our results and analysis show that TEINet can make an accurate prediction using only the TCR sequence (CDR3$\beta $) and the epitope sequence, providing novel insights to understand the interactions between TCRs and epitopes.</description><subject>Adaptive immunity</subject><subject>Antigens</subject><subject>Binding</subject><subject>Complementarity-determining region 3</subject><subject>Deep Learning</subject><subject>Epitopes</subject><subject>Epitopes, T-Lymphocyte</subject><subject>Immune response</subject><subject>Lymphocytes T</subject><subject>Machine learning</subject><subject>Neural networks</subject><subject>Predictions</subject><subject>Protein Binding</subject><subject>Receptors, Antigen, T-Cell</subject><subject>Sampling</subject><subject>T cell receptors</subject><subject>Transfer learning</subject><issn>1467-5463</issn><issn>1477-4054</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNp9kM1qGzEURkVJqR23q-6LIBACZWJp9DvZBZOkAZNCsdeDpLkqSu3RVJohZJd3yBvmSTLGbhZdZHUvl_N9XA5CXyk5p6Ricxvs3FrTEC0_oCnlShWcCH6026UqBJdsgo5zviekJErTT2jCZEWUFHqK1qur2zvoL7DBDUCHN2BSG9rf2CezhYeY_mAfE-4SNMH1IbY4erxa_Hp5eoYu9LEDbEPb7BK5Axd8cKF__Iw-erPJ8OUwZ2h9fbVa_CiWP29uF5fLwjEt-qIx3AjqNLe04soa4wUrS2CGaiccE6SpPAeqKHeydCXY3R2kVdLzRomSzdDZvrdL8e8Aua-3ITvYbEwLcch1qbQUlHLNRvTkP_Q-Dqkdv6sZ4dVoUpd8pL7vKZdizgl83aWwNemxpqTe2a5H2_XB9kh_O3QOdgvNG_tP7wic7oE4dO82vQLqZ4iI</recordid><startdate>20230319</startdate><enddate>20230319</enddate><creator>Jiang, Yuepeng</creator><creator>Huo, Miaozhe</creator><creator>Cheng Li, Shuai</creator><general>Oxford University Press</general><general>Oxford Publishing Limited (England)</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7QO</scope><scope>7SC</scope><scope>8FD</scope><scope>FR3</scope><scope>JQ2</scope><scope>K9.</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>P64</scope><scope>RC3</scope><scope>7X8</scope></search><sort><creationdate>20230319</creationdate><title>TEINet: a deep learning framework for prediction of TCR–epitope binding specificity</title><author>Jiang, Yuepeng ; Huo, Miaozhe ; Cheng Li, Shuai</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c385t-da4a51c84b1947baaf5322e3a18c5c350d9f4e1714c62c2eb18c5e6b76f4d7523</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Adaptive immunity</topic><topic>Antigens</topic><topic>Binding</topic><topic>Complementarity-determining region 3</topic><topic>Deep Learning</topic><topic>Epitopes</topic><topic>Epitopes, T-Lymphocyte</topic><topic>Immune response</topic><topic>Lymphocytes T</topic><topic>Machine learning</topic><topic>Neural networks</topic><topic>Predictions</topic><topic>Protein Binding</topic><topic>Receptors, Antigen, T-Cell</topic><topic>Sampling</topic><topic>T cell receptors</topic><topic>Transfer learning</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Jiang, Yuepeng</creatorcontrib><creatorcontrib>Huo, Miaozhe</creatorcontrib><creatorcontrib>Cheng Li, Shuai</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>Biotechnology Research Abstracts</collection><collection>Computer and Information Systems Abstracts</collection><collection>Technology Research Database</collection><collection>Engineering Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>ProQuest Health & Medical Complete (Alumni)</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>Biotechnology and BioEngineering Abstracts</collection><collection>Genetics Abstracts</collection><collection>MEDLINE - Academic</collection><jtitle>Briefings in bioinformatics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Jiang, Yuepeng</au><au>Huo, Miaozhe</au><au>Cheng Li, Shuai</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>TEINet: a deep learning framework for prediction of TCR–epitope binding specificity</atitle><jtitle>Briefings in bioinformatics</jtitle><addtitle>Brief Bioinform</addtitle><date>2023-03-19</date><risdate>2023</risdate><volume>24</volume><issue>2</issue><issn>1467-5463</issn><eissn>1477-4054</eissn><abstract>Abstract
The adaptive immune response to foreign antigens is initiated by T-cell receptor (TCR) recognition on the antigens. Recent experimental advances have enabled the generation of a large amount of TCR data and their cognate antigenic targets, allowing machine learning models to predict the binding specificity of TCRs. In this work, we present TEINet, a deep learning framework that utilizes transfer learning to address this prediction problem. TEINet employs two separately pretrained encoders to transform TCR and epitope sequences into numerical vectors, which are subsequently fed into a fully connected neural network to predict their binding specificities. A major challenge for binding specificity prediction is the lack of a unified approach to sampling negative data. Here, we first assess the current negative sampling approaches comprehensively and suggest that the Unified Epitope is the most suitable one. Subsequently, we compare TEINet with three baseline methods and observe that TEINet achieves an average AUROC of 0.760, which outperforms baseline methods by 6.4–26%. Furthermore, we investigate the impacts of the pretraining step and notice that excessive pretraining may lower its transferability to the final prediction task. Our results and analysis show that TEINet can make an accurate prediction using only the TCR sequence (CDR3$\beta $) and the epitope sequence, providing novel insights to understand the interactions between TCRs and epitopes.</abstract><cop>England</cop><pub>Oxford University Press</pub><pmid>36907658</pmid><doi>10.1093/bib/bbad086</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | ISSN: 1467-5463 |
ispartof | Briefings in bioinformatics, 2023-03, Vol.24 (2) |
issn | 1467-5463 1477-4054 |
language | eng |
recordid | cdi_proquest_miscellaneous_2786511483 |
source | Access via Oxford University Press (Open Access Collection) |
subjects | Adaptive immunity Antigens Binding Complementarity-determining region 3 Deep Learning Epitopes Epitopes, T-Lymphocyte Immune response Lymphocytes T Machine learning Neural networks Predictions Protein Binding Receptors, Antigen, T-Cell Sampling T cell receptors Transfer learning |
title | TEINet: a deep learning framework for prediction of TCR–epitope binding specificity |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T00%3A04%3A42IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_TOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=TEINet:%20a%20deep%20learning%20framework%20for%20prediction%20of%20TCR%E2%80%93epitope%20binding%20specificity&rft.jtitle=Briefings%20in%20bioinformatics&rft.au=Jiang,%20Yuepeng&rft.date=2023-03-19&rft.volume=24&rft.issue=2&rft.issn=1467-5463&rft.eissn=1477-4054&rft_id=info:doi/10.1093/bib/bbad086&rft_dat=%3Cproquest_TOX%3E3049109824%3C/proquest_TOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3049109824&rft_id=info:pmid/36907658&rft_oup_id=10.1093/bib/bbad086&rfr_iscdi=true |