VDJdb in the pandemic era: a compendium of T cell receptors specific for SARS-CoV-2

Goncharov, Mikhail; Bagaev, Dmitry; Shcherbinin, Dmitrii S.; Zvyagin, Ivan V.; Bolotin, Dmitriy A.; Thomas, Paul G.; Minervina, Anastasia A.; Pogorelyy, Mikhail V.; Ladell, Kristin; McLaren, James E.; Price, David A.; Nguyen, Thi H. O.; Rowntree, Louise C.; Clemens, E. Bridie; Kedzierska, Katherine; Dolton, Garry; Rius, Cristina; Sewell, Andrew K.; Samir, Jerome; Luciani, Fabio; Zornikova, Ksenia V.; Khmelevskaya, Alexandra; Sheetikov, Saveliy Andreevich; Efimov, Grigory A.; Chudakov, Dmitriy M.; Shugay, Mikhail

doi:10.1038/s41592-022-01578-0

Cited by 62 publications

(77 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The benchmark data set was derived from the VDJdb database (downloaded on 23/06/2022). VDJdb is a curated database of TCRs with known antigen specificities [2]. Only those TCR-epitope pairs with paired alpha-beta chain data were collected.…”

Section: Methodsmentioning

confidence: 99%

Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report

Meysman

Barton

Bravi³

et al. 2022

Preprint

View full text Add to dashboard Cite

Many different solutions to predicting the cognate epitope target of a T-cell receptor (TCR) have been proposed. However several questions on the advantages and disadvantages of these different approaches remain unresolved, as most methods have only been evaluated within the context of their initial publications and data sets. Here, we report the findings of the first public TCR-epitope prediction benchmark performed on 23 prediction models in the context of the ImmRep 2022 TCR-epitope specificity workshop. This benchmark revealed that the use of paired-chain alpha-beta, as well as CDR1/2 or V/J information, when available, improves classification obtained with CDR3 data, independent of the underlying approach. In addition, we found that straight-forward distance-based approaches can achieve a respectable performance when compared to more complex machine-learning models. Finally, we highlight the need for a truly independent follow-up benchmark and provide recommendations for the design of such a next benchmark.

show abstract

Section: Methodsmentioning

confidence: 99%

Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report

Meysman

Barton

Bravi³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The major public data resources for TCR-pMHC binding data are VDJdb (3), IEDB (4), McPAS-TCR (5), ImmuneCODE (6), TBAdb (7) and 10X Genomics (8), which all contain TCR CDR3 β chain information.…”

Section: Introductionmentioning

confidence: 99%

Performance comparison of TCR-pMHC prediction tools reveals a strong data dependency

Deng

Ly²,

Abdollahi³

et al. 2022

Preprint

View full text Add to dashboard Cite

The interaction of T-cell receptors with peptide-major histocompatibility complex molecules plays a crucial role in adaptive immune responses. Currently there are various models aiming at predicting TCR-pMHC binding, while a standard dataset and procedure to compare the performance of these approaches is still missing. In this work we provide a general method for data collection, preprocessing, splitting and generation of negative examples, as well as comprehensive datasets to compare TCR-pMHC prediction models. We collected, harmonized, and merged all the major publicly available TCR-pMHC binding data and compared the performance of five state-of-the-art deep learning models (TITAN, NetTCR, ERGO, DLpTCR and ImRex) using this data. Our performance evaluation focuses on two scenarios: 1) different splitting methods for generating training and testing data to assess model generalization and 2) different data versions that vary in size and peptide imbalance to assess model robustness. Our results indicate that the five contemporary models do not generalize to peptides that have not been in the training set. We can also show that model performance is strongly dependent on the data balance and size, which indicates a relatively low model robustness. These results suggest that TCR-pMHC binding prediction remains highly challenging and requires further high quality data and novel algorithmic approaches.

show abstract

“…3). The large degree of overlap between expanded and scAIM-TCRseq-con rmed clonotypes indicates that comparison of serial blood TRB repertoires may be a suitable surrogate for resource-intensive AIM-or peptide-HLA oligomer-scTCRseq to assign TRB sequences as vaccine antigenspeci c. AIM-scTCRseq data from this report substantially augment databases 50 used to match to estimate T cell responses to vaccines, as reported for adenovirus-based SARS-CoV-2 products 51,52 . While limited data suggest that intramuscular mRNA vaccination alone can result in nasal S-speci c CD8 + T cells 11 , more research is required to assess the contribution of a priming or breakthrough infection 12 for the emplacement of SARS-Cov2-speci c T RM in the nose.…”

Section: Discussionmentioning

confidence: 85%

“…1) and AIM-scTCRseq (Fig. 2, 3) clonotypes for TRB assigned to other antigens in public references 50 . Fewer than 0.1% of these TCRs matched EBV or cytomegalovirus (CMV)-assigned CDR3 sequences.…”

Section: Discussionmentioning

confidence: 99%

CD8+ T cell clonotypes from prior SARS-CoV-2 infection predominate during the cellular immune response to mRNA vaccination

Ford¹,

Mayer-Blackwell²,

Jing³

et al. 2022

Preprint

View full text Add to dashboard Cite

Almost three years into the SARS-CoV-2 pandemic, hybrid immunity is highly prevalent worldwide and more protective than vaccination or prior infection alone. Given emerging resistance of variant strains to neutralizing antibodies (nAb), it is likely that T cells contribute to this protection. To understand how sequential SARS-CoV-2 infection and mRNA-vectored SARS-CoV-2 spike (S) vaccines affect T cell clonotype-level expansion kinetics, we identified and cross-referenced TCR sequences from thousands of S-reactive single cells against deeply sequenced peripheral blood TCR repertoires longitudinally collected from persons during COVID-19 convalescence through booster vaccination. Successive vaccinations recalled memory T cells and elicited antigen-specific T cell clonotypes not detected after infection. Vaccine-related recruitment of novel clonotypes and the expansion of S-specific clones were most strongly observed for CD8+ T cells. Severe COVID-19 illness was associated with a more diverse CD4+ T cell response to SARS-CoV-2 both prior to and after mRNA vaccination, suggesting imprinting of CD4+ T cells by severe infection. TCR sequence similarity search algorithms revealed myriad public TCR clusters correlating with human leukocyte antigen (HLA) alleles. Selected TCRs from distinct clusters functionally recognized S in the predicted HLA context, with fine viral peptide requirements differing between TCRs. Most subjects tested had S-specific T cells in the nasal mucosa after a 3rd mRNA vaccine dose. The blood and nasal T cell responses to vaccination revealed by clonal tracking were more heterogeneous than nAb boosts. Analysis of bulk and single cell TCR sequences reveals T cell kinetics and diversity at the clonotype level, without requiring prior knowledge of T cell epitopes or HLA restriction, providing a roadmap for rapid assessment of T cell responses to emerging pathogens.

show abstract

VDJdb in the pandemic era: a compendium of T cell receptors specific for SARS-CoV-2

Cited by 62 publications

References 10 publications

Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report

Benchmarking solutions to the T-cell receptor epitope prediction problem: IMMREP22 workshop report

Performance comparison of TCR-pMHC prediction tools reveals a strong data dependency

CD8+ T cell clonotypes from prior SARS-CoV-2 infection predominate during the cellular immune response to mRNA vaccination

Contact Info

Product

Resources

About