A robust deep learning workflow to predict CD8 + T-cell epitopes.

BACKGROUND: T-cells play a crucial role in the adaptive immune system by triggering responses against cancer cells and pathogens, while maintaining tolerance against self-antigens, which has sparked interest in the development of various T-cell-focused immunotherapies. However, the identification of antigens recognised by T-cells is low-throughput and laborious. To overcome some of these limitations, computational methods for predicting CD8 + T-cell epitopes have emerged. Despite recent developments, most immunogenicity algorithms struggle to learn features of peptide immunogenicity from small datasets, suffer from HLA bias and are unable to reliably predict pathology-specific CD8 + T-cell epitopes. METHODS: We developed TRAP (T-cell recognition potential of HLA-I presented peptides), a robust deep learning workflow for predicting CD8 + T-cell epitopes from MHC-I presented pathogenic and self-peptides. TRAP uses transfer learning, deep learning architecture and MHC binding information to make context-specific predictions of CD8 + T-cell epitopes. TRAP also detects low-confidence predictions for peptides that differ significantly from those in the training datasets to abstain from making incorrect predictions. To estimate the immunogenicity of pathogenic peptides with low-confidence predictions, we further developed a novel metric, RSAT (relative similarity to autoantigens and tumour-associated antigens), as a complementary to 'dissimilarity to self' from cancer studies. RESULTS: TRAP was used to identify epitopes from glioblastoma patients as well as SARS-CoV-2 peptides, and it outperformed other algorithms in both cancer and pathogenic settings. TRAP was especially effective at extracting immunogenicity-associated properties from restricted data of emerging pathogens and translating them onto related species, as well as minimising the loss of likely epitopes in imbalanced datasets. We also demonstrated that the novel metric termed RSAT was able to estimate immunogenic of pathogenic peptides of various lengths and species. TRAP implementation is available at: https://github.com/ChloeHJ/TRAP . CONCLUSIONS: This study presents a novel computational workflow for accurately predicting CD8 + T-cell epitopes to foster a better understanding of antigen-specific T-cell response and the development of effective clinical therapeutics.

Evaluating performance of existing computational models in predicting CD8+ T cell pathogenic epitopes and cancer neoantigens.

T cell recognition of a cognate peptide-major histocompatibility complex (pMHC) presented on the surface of infected or malignant cells is of the utmost importance for mediating robust and long-term immune responses. Accurate predictions of cognate pMHC targets for T cell receptors would greatly facilitate identification of vaccine targets for both pathogenic diseases and personalized cancer immunotherapies. Predicting immunogenic peptides therefore has been at the center of intensive research for the past decades but has proven challenging. Although numerous models have been proposed, performance of these models has not been systematically evaluated and their success rate in predicting epitopes in the context of human pathology has not been measured and compared. In this study, we evaluated the performance of several publicly available models, in identifying immunogenic CD8+ T cell targets in the context of pathogens and cancers. We found that for predicting immunogenic peptides from an emerging virus such as severe acute respiratory syndrome coronavirus 2, none of the models perform substantially better than random or offer considerable improvement beyond HLA ligand prediction. We also observed suboptimal performance for predicting cancer neoantigens. Through investigation of potential factors associated with ill performance of models, we highlight several data- and model-associated issues. In particular, we observed that cross-HLA variation in the distribution of immunogenic and non-immunogenic peptides in the training data of the models seems to substantially confound the predictions. We additionally compared key parameters associated with immunogenicity between pathogenic peptides and cancer neoantigens and observed evidence for differences in the thresholds of binding affinity and stability, which suggested the need to modulate different features in identifying immunogenic pathogen versus cancer peptides. Overall, we demonstrate that accurate and reliable predictions of immunogenic CD8+ T cell targets remain unsolved; thus, we hope our work will guide users and model developers regarding potential pitfalls and unsettled questions in existing immunogenicity predictors.

High-resolution profiling of MHC II peptide presentation capacity reveals SARS-CoV-2 CD4 T cell targets and mechanisms of immune escape.

Understanding peptide presentation by specific MHC alleles is fundamental for controlling physiological functions of T cells and harnessing them for therapeutic use. However, commonly used in silico predictions and mass spectroscopy have their limitations in precision, sensitivity, and throughput, particularly for MHC class II. Here, we present MEDi, a novel mammalian epitope display that allows an unbiased, affordable, high-resolution mapping of MHC peptide presentation capacity. Our platform provides a detailed picture by testing every antigen-derived peptide and is scalable to all the MHC II alleles. Given the urgent need to understand immune evasion for formulating effective responses to threats such as SARS-CoV-2, we provide a comprehensive analysis of the presentability of all SARS-CoV-2 peptides in the context of several HLA class II alleles. We show that several mutations arising in viral strains expanding globally resulted in reduced peptide presentability by multiple HLA class II alleles, while some increased it, suggesting alteration of MHC II presentation landscapes as a possible immune escape mechanism.

HLA-dependent variation in SARS-CoV-2 CD8 + T cell cross-reactivity with human coronaviruses.

The conditions and extent of cross-protective immunity between severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and common-cold human coronaviruses (HCoVs) remain open despite several reports of pre-existing T cell immunity to SARS-CoV-2 in individuals without prior exposure. Using a pool of functionally evaluated SARS-CoV-2 peptides, we report a map of 126 immunogenic peptides with high similarity to 285 MHC-presented peptides from at least one HCoV. Employing this map of SARS-CoV-2-non-homologous and homologous immunogenic peptides, we observe several immunogenic peptides with high similarity to human proteins, some of which have been reported to have elevated expression in severe COVID-19 patients. After combining our map with SARS-CoV-2-specific TCR repertoire data from COVID-19 patients and healthy controls, we show that public repertoires for the majority of convalescent patients are dominated by TCRs cognate to non-homologous SARS-CoV-2 peptides. We find that for a subset of patients, >50% of their public SARS-CoV-2-specific repertoires consist of TCRs cognate to homologous SARS-CoV-2-HCoV peptides. Further analysis suggests that this skewed distribution of TCRs cognate to homologous or non-homologous peptides in COVID-19 patients is likely to be HLA-dependent. Finally, we provide 10 SARS-CoV-2 peptides with known cognate TCRs that are conserved across multiple coronaviruses and are predicted to be recognized by a high proportion of the global population. These findings may have important implications for COVID-19 heterogeneity, vaccine-induced immune responses, and robustness of immunity to SARS-CoV-2 and its variants.

Predicting Cross-Reactivity and Antigen Specificity of T Cell Receptors.

Adaptive immune recognition is mediated by specific interactions between heterodimeric T cell receptors (TCRs) and their cognate peptide-MHC (pMHC) ligands, and the methods to accurately predict TCR:pMHC interaction would have profound clinical, therapeutic and pharmaceutical applications. Herein, we review recent developments in predicting cross-reactivity and antigen specificity of TCR recognition. We discuss current experimental and computational approaches to investigate cross-reactivity and antigen-specificity of TCRs and highlight how integrating kinetic, biophysical and structural features may offer valuable insights in modeling immunogenicity. We further underscore the close inter-relationship of these two interconnected notions and the need to investigate each in the light of the other for a better understanding of T cell responsiveness for the effective clinical applications.

In silico identification of vaccine targets for 2019-nCoV.

Background: The newly identified coronavirus known as 2019-nCoV has posed a serious global health threat. According to the latest report (18-February-2020), it has infected more than 72,000 people globally and led to deaths of more than 1,016 people in China. Methods: The 2019 novel coronavirus proteome was aligned to a curated database of viral immunogenic peptides. The immunogenicity of detected peptides and their binding potential to HLA alleles was predicted by immunogenicity predictive models and NetMHCpan 4.0. Results: We report in silico identification of a comprehensive list of immunogenic peptides that can be used as potential targets for 2019 novel coronavirus (2019-nCoV) vaccine development. First, we found 28 nCoV peptides identical to Severe acute respiratory syndrome-related coronavirus (SARS CoV) that have previously been characterized immunogenic by T cell assays. Second, we identified 48 nCoV peptides having a high degree of similarity with immunogenic peptides deposited in The Immune Epitope Database (IEDB). Lastly, we conducted a de novo search of 2019-nCoV 9-mer peptides that i) bind to common HLA alleles in Chinese and European population and ii) have T Cell Receptor (TCR) recognition potential by positional weight matrices and a recently developed immunogenicity algorithm, iPred, and identified in total 63 peptides with a high immunogenicity potential. Conclusions: Given the limited time and resources to develop vaccine and treatments for 2019-nCoV, our work provides a shortlist of candidates for experimental validation and thus can accelerate development pipeline.