Cross-modality deep learning-based prediction of TAP binding and naturally processed peptide

Besser, Hanan; Louzoun, Yoram

doi:10.1007/s00251-018-1054-6

Cited by 6 publications

(2 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here, the TAP binding ability was predicted through Besser H’s work 41 with default parameters. The output IC 50 value was used as the TAP score (T).…”

Section: Methodsmentioning

confidence: 99%

Neo-intline: integrated pipeline enables neoantigen design through the in-silico presentation of T-cell epitope

Li,

Jing,

Zheng

et al. 2023

Sig Transduct Target Ther

View full text Add to dashboard Cite

Neoantigen vaccines are one of the most effective immunotherapies for personalized tumour treatment. The current immunogen design of neoantigen vaccines is usually based on whole-genome sequencing (WGS) and bioinformatics prediction that focuses on the prediction of binding affinity between peptide and MHC molecules, ignoring other peptide-presenting related steps. This may result in a gap between high prediction accuracy and relatively low clinical effectiveness. In this study, we designed an integrated in-silico pipeline, Neo-intline, which started from the SNPs and indels of the tumour samples to simulate the presentation process of peptides in-vivo through an integrated calculation model. Validation on the benchmark dataset of TESLA and clinically validated neoantigens illustrated that neo-intline could outperform current state-of-the-art tools on both sample level and melanoma level. Furthermore, by taking the mouse melanoma model as an example, we verified the effectiveness of 20 neoantigens, including 10 MHC-I and 10 MHC-II peptides. The in-vitro and in-vivo experiments showed that both peptides predicted by Neo-intline could recruit corresponding CD4+ T cells and CD8+ T cells to induce a T-cell-mediated cellular immune response. Moreover, although the therapeutic effect of neoantigen vaccines alone is not sufficient, combinations with other specific therapies, such as broad-spectrum immune-enhanced adjuvants of granulocyte-macrophage colony-stimulating factor (GM-CSF) and polyinosinic-polycytidylic acid (poly(I:C)), or immune checkpoint inhibitors, such as PD-1/PD-L1 antibodies, can illustrate significant anticancer effects on melanoma. Neo-intline can be used as a benchmark process for the design and screening of immunogenic targets for neoantigen vaccines.

show abstract

“…Here, the TAP binding ability was predicted through Besser H’s work 41 with default parameters. The output IC 50 value was used as the TAP score (T).…”

Section: Methodsmentioning

confidence: 99%

Neo-intline: integrated pipeline enables neoantigen design through the in-silico presentation of T-cell epitope

Li,

Jing,

Zheng

et al. 2023

Sig Transduct Target Ther

View full text Add to dashboard Cite

show abstract

“…Since the majority of peptides presented by MHC class I are 9-mers, we discarded all fragments shorter than nine amino acids long, while fragments longer than nine amino acids were subdivided into all possible overlapping 9-mer combinations. Since the transporter associated with antigen processing (TAP) binding is a needed stage for most MHC-I-binding peptides, we used an online server that applied the model described by Besser and Louzoun [80] to estimate the TAP binding affinity of the predicted peptides. Finally, we used an IEDB MHC I binding prediction tool to predict which peptides would bind more effectively to pig MHC I complex [51].…”

Section: Prediction Of Cd8 + T-cell Epitopesmentioning

confidence: 99%

Computational Analysis of African Swine Fever Virus Protein Space for the Design of an Epitope-Based Vaccine Ensemble

et al. 2020

View full text Add to dashboard Cite

African swine fever virus is the etiological agent of African swine fever, a transmissible severe hemorrhagic disease that affects pigs, causing massive economic losses. There is neither a treatment nor a vaccine available, and the only method to control its spread is by extensive culling of pigs. So far, classical vaccine development approaches have not yielded sufficiently good results in terms of concomitant safety and efficacy. Nowadays, thanks to advances in genomic and proteomic techniques, a reverse vaccinology strategy can be explored to design alternative vaccine formulations. In this study, ASFV protein sequences were analyzed using an in-house pipeline based on publicly available immunoinformatic tools to identify epitopes of interest for a prospective vaccine ensemble. These included experimentally validated sequences from the Immune Epitope Database, as well as de novo predicted sequences. Experimentally validated and predicted epitopes were prioritized following a series of criteria that included evolutionary conservation, presence in the virulent and currently circulating variant Georgia 2007/1, and lack of identity to either the pig proteome or putative proteins from pig gut microbiota. Following this strategy, 29 B-cell, 14 CD4+ T-cell and 6 CD8+ T-cell epitopes were selected, which represent a starting point to investigating the protective capacity of ASFV epitope-based vaccines.

show abstract

NetCleave: An Open-Source Algorithm for Predicting C-Terminal Antigen Processing for MHC-I and MHC-II

Farriol-Duran,

Vallejo-Vallés,

Amengual-Rigo

et al. 2023

Methods in Molecular Biology

View full text Add to dashboard Cite

Antigens presented on the cell surface have been subjected to multiple biological processes. Among them, C-terminal antigen processing constitutes one of the main bottlenecks of the peptide presentation pathways, as it delimits the peptidome that will be subjected downstream. Here, we present NetCleave, an open-source and retrainable algorithm for the prediction of the C-terminal antigen processing for both MHC-I and MHC-II pathways. NetCleave architecture consists of a neural network trained on 46 different physicochemical descriptors of the cleavage site amino acids. Our results demonstrate that prediction of C-terminal antigen processing achieves high accuracy on MHC-I (AUC of 0.91), while it remains challenging for MHC-II (AUC of 0.66). Moreover, we evaluated the performance of NetCleave and other prediction tools for the evaluation of four independent immunogenicity datasets (H2-Db, H2-Kb, HLA-A*02:01 and HLA-B:07:02). Overall, we demonstrate that NetCleave stands out as one of the best algorithms for the prediction of C-terminal processing, and we provide one of the first evidence that C-terminal processing predictions may help in the discovery of immunogenic peptides.

show abstract

Cross-modality deep learning-based prediction of TAP binding and naturally processed peptide

Cited by 6 publications

References 36 publications

Neo-intline: integrated pipeline enables neoantigen design through the in-silico presentation of T-cell epitope

Neo-intline: integrated pipeline enables neoantigen design through the in-silico presentation of T-cell epitope

Computational Analysis of African Swine Fever Virus Protein Space for the Design of an Epitope-Based Vaccine Ensemble

NetCleave: An Open-Source Algorithm for Predicting C-Terminal Antigen Processing for MHC-I and MHC-II

Contact Info

Product

Resources

About