Counting is almost all you need
The immune memory repertoire encodes the history of present and past infections and immunological attributes of the individual. As such, multiple methods were proposed to use T-cell receptor (TCR) repertoires to detect disease history. We here show that the counting method outperforms two leading al...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-01-01
|
| Series: | Frontiers in Immunology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fimmu.2022.1031011/full |
| _version_ | 1797946670625849344 |
|---|---|
| author | Ofek Akerman Ofek Akerman Haim Isakov Reut Levi Vladimir Psevkin Yoram Louzoun |
| author_facet | Ofek Akerman Ofek Akerman Haim Isakov Reut Levi Vladimir Psevkin Yoram Louzoun |
| author_sort | Ofek Akerman |
| collection | DOAJ |
| description | The immune memory repertoire encodes the history of present and past infections and immunological attributes of the individual. As such, multiple methods were proposed to use T-cell receptor (TCR) repertoires to detect disease history. We here show that the counting method outperforms two leading algorithms. We then show that the counting can be further improved using a novel attention model to weigh the different TCRs. The attention model is based on the projection of TCRs using a Variational AutoEncoder (VAE). Both counting and attention algorithms predict better than current leading algorithms whether the host had CMV and its HLA alleles. As an intermediate solution between the complex attention model and the very simple counting model, we propose a new Graph Convolutional Network approach that obtains the accuracy of the attention model and the simplicity of the counting model. The code for the models used in the paper is provided at: https://github.com/louzounlab/CountingIsAlmostAllYouNeed. |
| first_indexed | 2024-04-10T21:14:41Z |
| format | Article |
| id | doaj.art-e0d7d7ca99f64f8c8ccfd092235b73b3 |
| institution | Directory Open Access Journal |
| issn | 1664-3224 |
| language | English |
| last_indexed | 2024-04-10T21:14:41Z |
| publishDate | 2023-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Immunology |
| spelling | doaj.art-e0d7d7ca99f64f8c8ccfd092235b73b32023-01-20T12:54:27ZengFrontiers Media S.A.Frontiers in Immunology1664-32242023-01-011310.3389/fimmu.2022.10310111031011Counting is almost all you needOfek Akerman0Ofek Akerman1Haim Isakov2Reut Levi3Vladimir Psevkin4Yoram Louzoun5Department of Mathematics, Bar-Ilan University, Ramat Gan, IsraelDepartment of Computer Science, Bar-Ilan University, Ramat Gan, IsraelDepartment of Mathematics, Bar-Ilan University, Ramat Gan, IsraelDepartment of Mathematics, Bar-Ilan University, Ramat Gan, IsraelDepartment of Mathematics, Bar-Ilan University, Ramat Gan, IsraelDepartment of Mathematics, Bar-Ilan University, Ramat Gan, IsraelThe immune memory repertoire encodes the history of present and past infections and immunological attributes of the individual. As such, multiple methods were proposed to use T-cell receptor (TCR) repertoires to detect disease history. We here show that the counting method outperforms two leading algorithms. We then show that the counting can be further improved using a novel attention model to weigh the different TCRs. The attention model is based on the projection of TCRs using a Variational AutoEncoder (VAE). Both counting and attention algorithms predict better than current leading algorithms whether the host had CMV and its HLA alleles. As an intermediate solution between the complex attention model and the very simple counting model, we propose a new Graph Convolutional Network approach that obtains the accuracy of the attention model and the simplicity of the counting model. The code for the models used in the paper is provided at: https://github.com/louzounlab/CountingIsAlmostAllYouNeed.https://www.frontiersin.org/articles/10.3389/fimmu.2022.1031011/fullrepertoire classificationimmune repertoiremachine learningattentiongraphsT cells |
| spellingShingle | Ofek Akerman Ofek Akerman Haim Isakov Reut Levi Vladimir Psevkin Yoram Louzoun Counting is almost all you need Frontiers in Immunology repertoire classification immune repertoire machine learning attention graphs T cells |
| title | Counting is almost all you need |
| title_full | Counting is almost all you need |
| title_fullStr | Counting is almost all you need |
| title_full_unstemmed | Counting is almost all you need |
| title_short | Counting is almost all you need |
| title_sort | counting is almost all you need |
| topic | repertoire classification immune repertoire machine learning attention graphs T cells |
| url | https://www.frontiersin.org/articles/10.3389/fimmu.2022.1031011/full |
| work_keys_str_mv | AT ofekakerman countingisalmostallyouneed AT ofekakerman countingisalmostallyouneed AT haimisakov countingisalmostallyouneed AT reutlevi countingisalmostallyouneed AT vladimirpsevkin countingisalmostallyouneed AT yoramlouzoun countingisalmostallyouneed |