Journal article
NetTCR-2.0 enables accurate prediction of TCR-peptide binding by using paired TCRα and β sequence data
Department of Health Technology, Technical University of Denmark1
Bayesian Modeling & Molecular Evolution, Bioinformatics, Department of Health Technology, Technical University of Denmark2
Bioinformatics, Department of Health Technology, Technical University of Denmark3
Immunoinformatics and Machine Learning, Bioinformatics, Department of Health Technology, Technical University of Denmark4
T-Cells and Cancer, Experimental & Translational Immunology, Department of Health Technology, Technical University of Denmark5
Experimental & Translational Immunology, Department of Health Technology, Technical University of Denmark6
La Jolla Institute for Allergy and Immunology7
Department of Applied Mathematics and Computer Science, Technical University of Denmark8
Cognitive Systems, Department of Applied Mathematics and Computer Science, Technical University of Denmark9
University of California at San Diego10
...and 0 morePrediction of T-cell receptor (TCR) interactions with MHC-peptide complexes remains highly challenging. This challenge is primarily due to three dominant factors: data accuracy, data scarceness, and problem complexity. Here, we showcase that “shallow” convolutional neural network (CNN) architectures are adequate to deal with the problem complexity imposed by the length variations of TCRs.
We demonstrate that current public bulk CDR3β-pMHC binding data overall is of low quality and that the development of accurate prediction models is contingent on paired α/β TCR sequence data corresponding to at least 150 distinct pairs for each investigated pMHC. In comparison, models trained on CDR3α or CDR3β data alone demonstrated a variable and pMHC specific relative performance drop.
Together these findings support that T-cell specificity is predictable given the availability of accurate and sufficient paired TCR sequence data. NetTCR-2.0 is publicly available at https://services.healthtech.dtu.dk/service.php?NetTCR-2.0.
Language: | English |
---|---|
Publisher: | Nature Publishing Group UK |
Year: | 2021 |
Pages: | 1060 |
ISSN: | 23993642 |
Types: | Journal article |
DOI: | 10.1038/s42003-021-02610-3 |
ORCIDs: | Bentzen, Amalie Kai , Hadrup, Sine R. , Winther, Ole , Jessen, Leon Eyrich , Nielsen, Morten and 0000-0002-3586-4533 |