NetMHCpan-4.1 and NetMHCIIpan-4.0: improved predictions of MHC antigen presentation by concurrent motif deconvolution and integration of MS MHC eluted ligand data.
TL;DR: NetMHCpan-4.1 and NetMHCIIpan- 4.0, two web servers created to predict binding between peptides and M HC-I and MHC-II, respectively, exploit tailored machine learning strategies to integrate different training data types, resulting in state-of-the-art performance and outperforming their competitors.
Abstract: Major histocompatibility complex (MHC) molecules are expressed on the cell surface, where they present peptides to T cells, which gives them a key role in the development of T-cell immune responses. MHC molecules come in two main variants: MHC Class I (MHC-I) and MHC Class II (MHC-II). MHC-I predominantly present peptides derived from intracellular proteins, whereas MHC-II predominantly presents peptides from extracellular proteins. In both cases, the binding between MHC and antigenic peptides is the most selective step in the antigen presentation pathway. Therefore, the prediction of peptide binding to MHC is a powerful utility to predict the possible specificity of a T-cell immune response. Commonly MHC binding prediction tools are trained on binding affinity or mass spectrometry-eluted ligands. Recent studies have however demonstrated how the integration of both data types can boost predictive performances. Inspired by this, we here present NetMHCpan-4.1 and NetMHCIIpan-4.0, two web servers created to predict binding between peptides and MHC-I and MHC-II, respectively. Both methods exploit tailored machine learning strategies to integrate different training data types, resulting in state-of-the-art performance and outperforming their competitors. The servers are available at http://www.cbs.dtu.dk/services/NetMHCpan-4.1/ and http://www.cbs.dtu.dk/services/NetMHCIIpan-4.0/.
Citations
More filters
TL;DR: It is shown that SARS-CoV-2 elicits robust, broad and highly functional memory T cell responses, suggesting that natural exposure or infection may prevent recurrent episodes of severe COVID-19.
1,501 citations
TL;DR: In this paper , the authors address whether T cell responses induced by different vaccine platforms (mRNA-1273, BNT162b2, Ad26.COV2.S, and NVX-CoV2373) cross-recognize early SARS CoV-2 variants.
485 citations
02 Jul 2021
TL;DR: In this article, the authors compared SARS-CoV-2-specific CD4+ and CD8+ T-cells against the B.1.7, B.351, P.1, and CAL.
Abstract: The emergence of SARS-CoV-2 variants with evidence of antibody escape highlight the importance of addressing whether the total CD4+ and CD8+ T cell recognition is also affected. Here, we compare SARS-CoV-2-specific CD4+ and CD8+ T cells against the B.1.1.7, B.1.351, P.1, and CAL.20C lineages in COVID-19 convalescents and in recipients of the Moderna (mRNA-1273) or Pfizer/BioNTech (BNT162b2) COVID-19 vaccines. The total reactivity against SARS-CoV-2 variants is similar in terms of magnitude and frequency of response, with decreases in the 10%-22% range observed in some assay/VOC combinations. A total of 7% and 3% of previously identified CD4+ and CD8+ T cell epitopes, respectively, are affected by mutations in the various VOCs. Thus, the SARS-CoV-2 variants analyzed here do not significantly disrupt the total SARS-CoV-2 T cell reactivity; however, the decreases observed highlight the importance for active monitoring of T cell reactivity in the context of SARS-CoV-2 evolution.
404 citations
TL;DR: In this article, two recently emerging mutations in the receptor-binding domain of the SARS-CoV-2 spike protein, L452R and Y453F, are shown to increase spike stability, virus infectivity, and fusogenicity.
316 citations
TL;DR: In this paper, the authors investigated if single dose vaccination, with or without prior infection, confers cross-protective immunity to variants of SARS-CoV-2 vaccine rollout has coincided with the spread of variants of concern.
Abstract: SARS-CoV-2 vaccine rollout has coincided with the spread of variants of concern. We investigated if single dose vaccination, with or without prior infection, confers cross protective immunity to variants. We analyzed T and B cell responses after first dose vaccination with the Pfizer/BioNTech mRNA vaccine BNT162b2 in healthcare workers (HCW) followed longitudinally, with or without prior Wuhan-Hu-1 SARS-CoV-2 infection. After one dose, individuals with prior infection showed enhanced T cell immunity, antibody secreting memory B cell response to spike and neutralizing antibodies effective against B.1.1.7 and B.1.351. By comparison, HCW receiving one vaccine dose without prior infection showed reduced immunity against variants. B.1.1.7 and B.1.351 spike mutations resulted in increased, abrogated or unchanged T cell responses depending on human leukocyte antigen (HLA) polymorphisms. Single dose vaccination with BNT162b2 in the context of prior infection with a heterologous variant substantially enhances neutralizing antibody responses against variants.
255 citations
References
More filters
TL;DR: NetMHCpan-4.0, a method trained on binding affinity and eluted ligand data leveraging the information from both data types, demonstrates an increase in predictive performance compared with state-of-the-art methods when it comes to identification of naturally processed ligands, cancer neoantigens, and T cell epitopes.
Abstract: Cytotoxic T cells are of central importance in the immune system's response to disease. They recognize defective cells by binding to peptides presented on the cell surface by MHC class I molecules. Peptide binding to MHC molecules is the single most selective step in the Ag-presentation pathway. Therefore, in the quest for T cell epitopes, the prediction of peptide binding to MHC molecules has attracted widespread attention. In the past, predictors of peptide-MHC interactions have primarily been trained on binding affinity data. Recently, an increasing number of MHC-presented peptides identified by mass spectrometry have been reported containing information about peptide-processing steps in the presentation pathway and the length distribution of naturally presented peptides. In this article, we present NetMHCpan-4.0, a method trained on binding affinity and eluted ligand data leveraging the information from both data types. Large-scale benchmarking of the method demonstrates an increase in predictive performance compared with state-of-the-art methods when it comes to identification of naturally processed ligands, cancer neoantigens, and T cell epitopes.
1,019 citations
"NetMHCpan-4.1 and NetMHCIIpan-4.0: ..." refers background or methods in this paper
...That is, for each epitope-HLA pair, binding to the HLA was predicted for all overlapping peptides of the source protein using the eluted ligand likelihood prediction score and the FRANK value was reported as the proportion of peptides with a prediction score higher than that of the epitope....
[...]
...0 (12), MixMHCpred (18,21), MHCFlurry (5) and MHCFlurry EL (an unpublished version of MHCFLurry trained with EL SA data, available at GitHub (22))....
[...]
...In short, the NNAlign framework is a singleallele framework permitting the integration of mixed data types (BA and EL) in the model training, which allows information to be leveraged across the different data types, resulting in a boosted predictive power (12,13)....
[...]
...al (12) combined with a comprehensive set of MHC multimer validated epitopes obtained from the IEDB and for MHC class II from Reynisson et al....
[...]
...For the epitope data, the predictive performance was estimated in terms of FRANK (12)....
[...]
TL;DR: It is shown that training with this extended data set improved the performance for peptide binding predictions for both methods, and updated versions of two MHC-II-peptide binding affinity prediction methods, NetM HCII and NetMHCIIpan are presented.
Abstract: Major histocompatibility complex class II (MHC-II) molecules are expressed on the surface of professional antigen-presenting cells where they display peptides to T helper cells, which orchestrate the onset and outcome of many host immune responses. Understanding which peptides will be presented by the MHC-II molecule is therefore important for understanding the activation of T helper cells and can be used to identify T-cell epitopes. We here present updated versions of two MHC-II-peptide binding affinity prediction methods, NetMHCII and NetMHCIIpan. These were constructed using an extended data set of quantitative MHC-peptide binding affinity data obtained from the Immune Epitope Database covering HLA-DR, HLA-DQ, HLA-DP and H-2 mouse molecules. We show that training with this extended data set improved the performance for peptide binding predictions for both methods. Both methods are publicly available at www.cbs.dtu.dk/services/NetMHCII-2.3 and www.cbs.dtu.dk/services/NetMHCIIpan-3.2.
518 citations
Additional excerpts
...2 (23), MixMHC2pred (11), MHCnuggets (24) and DeepSeqPanII (25)....
[...]
TL;DR: A scalable mono‐allelic strategy for profiling the HLA peptidome is implemented and a strategy for systematically learning the rules of endogenous antigen presentation is demonstrated, providing an updated portrait of antigen processing rules.
477 citations
"NetMHCpan-4.1 and NetMHCIIpan-4.0: ..." refers methods in this paper
...Even though one of the main contributions to the improved performance of the prediction methods proposed here (and other recently published methods) is the integration of MS derived EL data, MS data itself contains an inherent bias imposed resulting in for instance overrepresentation of ‘flyable’ (27) and neglecting cysteine-containing peptides (7)....
[...]
TL;DR: A neural network-based machine-learning algorithm leveraging information across multiple receptor specificities and ligand length scales is developed and demonstrated how this approach significantly improves the accuracy for prediction of peptide binding and identification of MHC ligands.
Abstract: Binding of peptides to MHC class I molecules (MHC-I) is essential for antigen presentation to cytotoxic T-cells. Here, we demonstrate how a simple alignment step allowing insertions and deletions in a pan-specific MHC-I binding machine-learning model enables combining information across both multiple MHC molecules and peptide lengths. This pan-allele/pan-length algorithm significantly outperforms state-of-the-art methods, and captures differences in the length profile of binders to different MHC molecules leading to increased accuracy for ligand identification. Using this model, we demonstrate that percentile ranks in contrast to affinity-based thresholds are optimal for ligand identification due to uniform sampling of the MHC space. We have developed a neural network-based machine-learning algorithm leveraging information across multiple receptor specificities and ligand length scales, and demonstrated how this approach significantly improves the accuracy for prediction of peptide binding and identification of MHC ligands. The method is available at www.cbs.dtu.dk/services/NetMHCpan-3.0
.
444 citations
TL;DR: An open-source package for MHC I binding prediction, MHCflurry, which implements allele-specific neural networks that use a novel architecture and peptide encoding scheme and showed competitive accuracy with standard tools, including the recently released NetMHCpan 4.0.
Abstract: Summary Predicting the binding affinity of major histocompatibility complex I (MHC I) proteins and their peptide ligands is important for vaccine design. We introduce an open-source package for MHC I binding prediction, MHCflurry. The software implements allele-specific neural networks that use a novel architecture and peptide encoding scheme. When trained on affinity measurements, MHCflurry outperformed the standard predictors NetMHC 4.0 and NetMHCpan 3.0 overall and particularly on non-9-mer peptides in a benchmark of ligands identified by mass spectrometry. The released predictor, MHCflurry 1.2.0, uses mass spectrometry datasets for model selection and showed competitive accuracy with standard tools, including the recently released NetMHCpan 4.0, on a small benchmark of affinity measurements. MHCflurry's prediction speed exceeded 7,000 predictions per second, 396 times faster than NetMHCpan 4.0. MHCflurry is freely available to use, retrain, or extend, includes Python library and command line interfaces, may be installed using package managers, and applies software development best practices.
293 citations
"NetMHCpan-4.1 and NetMHCIIpan-4.0: ..." refers background in this paper
...0 (12), MixMHCpred (18,21), MHCFlurry (5) and MHCFlurry EL (an unpublished version of MHCFLurry trained with EL SA data, available at GitHub (22))....
[...]