STAPLER: Efficient learning of TCR-peptide specificity prediction from full-length TCR-peptide data

bioRxiv (Cold Spring Harbor Laboratory)

Abstract

Abstract The prediction of peptide-MHC (pMHC) recognition by αβ T-cell receptors (TCRs) remains a major biomedical challenge. Here, we develop STAPLER (Shared TCR And Peptide Language bidirectional Encoder Representations from transformers), a transformer language model that uses a joint TCRαβ- peptide input to allow the learning of patterns within and between TCRαβ and peptide sequences that encode recognition. First, we demonstrate how data leakage during negative data generation can confound performance estimates of neural network-based models in predicting TCR – pMHC specificity. We then demonstrate that, because of its pre-training and fine-tuning masked language modeling tasks, STAPLER outperforms both neural network-based and distance-based ML models in predicting the recognition of known antigens in an independent dataset, in particular for antigens for which little related data is available. Based on this ability to efficiently learn from limited labeled TCR- peptide data, STAPLER is well-suited to utilize growing TCR – pMHC datasets to achieve accurate prediction of TCR – pMHC specificity.

Authors 8

1. Bjørn P. Y. Kwee Aachen Department of Hematology, Leiden University Medical Center

   Leiden University Medical Center · The Netherlands Cancer Institute · Oncode Institute

   Affiliation as printed

   Department of Hematology, Leiden University Medical Center, Leiden, The Netherlands

   Division of Molecular Oncology & Immunology, Oncode Institute, The Netherlands Cancer Institute, Amsterdam, The Netherlands

2. Marius Messemaker Aachen Department of Hematology, Leiden University Medical Center

   Leiden University Medical Center · The Netherlands Cancer Institute · Oncode Institute

   Affiliation as printed

   Department of Hematology, Leiden University Medical Center, Leiden, The Netherlands

   Division of Molecular Oncology & Immunology, Oncode Institute, The Netherlands Cancer Institute, Amsterdam, The Netherlands

3. Eric Marcus

   The Netherlands Cancer Institute · University of Amsterdam

   Affiliation as printed

   Division of Radiation Oncology, The Netherlands Cancer Institute, Amsterdam, The Netherlands

   Informatics Institute, University of Amsterdam, Amsterdam, The Netherlands

4. Giacomo Oliveira

   Harvard University · Dana-Farber Cancer Institute

   Affiliation as printed

   Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA

   Harvard Medical School, Boston, MA, USA

5. Wouter Scheper

   The Netherlands Cancer Institute · Oncode Institute

   Affiliation as printed

   Division of Molecular Oncology & Immunology, Oncode Institute, The Netherlands Cancer Institute, Amsterdam, The Netherlands

6. Catherine J. Wu

   Broad Institute · Brigham and Women's Hospital · Harvard University · Dana-Farber Cancer Institute

   Affiliation as printed

   Broad Institute of MIT and Harvard, Cambridge, MA, USA

   Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA

   Department of Medicine, Brigham and Women’s Hospital, Boston, MA, USA

   Harvard Medical School, Boston, MA, USA

   Department of Medicine, Brigham and Women's Hospital, Boston, MA, USA

7. Jonas Teuwen

   Memorial Sloan Kettering Cancer Center · Radboud University Medical Center · The Netherlands Cancer Institute · University of Amsterdam

   Affiliation as printed

   Department of Medical Imaging, Radboud University Medical Center, Nijmegen, the Netherlands

   Department of Radiology, Memorial Sloan Kettering Cancer Center, New York City, NY, USA

   Division of Radiation Oncology, The Netherlands Cancer Institute, Amsterdam, The Netherlands

   Informatics Institute, University of Amsterdam, Amsterdam, The Netherlands

8. Ton N. Schumacher corresponding Aachen Department of Hematology, Leiden University Medical Center

   Leiden University Medical Center · The Netherlands Cancer Institute · Oncode Institute

   Affiliation as printed

   Department of Hematology, Leiden University Medical Center, Leiden, The Netherlands

   Division of Molecular Oncology & Immunology, Oncode Institute, The Netherlands Cancer Institute, Amsterdam, The Netherlands