Designing of interferon-gamma inducing MHC class-II binders

Abstract

BackgroundThe generation of interferon-gamma (IFN-γ) by MHC class II activated CD4+ T helper cells play a substantial contribution in the control of infections such as caused by Mycobacterium tuberculosis. In the past, numerous methods have been developed for predicting MHC class II binders that can activate T-helper cells. Best of author’s knowledge, no method has been developed so far that can predict the type of cytokine will be secreted by these MHC Class II binders or T-helper epitopes. In this study, an attempt has been made to predict the IFN-γ inducing peptides. The main dataset used in this study contains 3705 IFN-γ inducing and 6728 non-IFN-γ inducing MHC class II binders. Another dataset called IFNgOnly contains 4483 IFN-γ inducing epitopes and 2160 epitopes that induce other cytokine except IFN-γ. In addition we have alternate dataset that contains IFN-γ inducing and equal number of random peptides.ResultsIt was observed that the peptide length, positional conservation of residues and amino acid composition affects IFN-γ inducing capabilities of these peptides. We identified the motifs in IFN-γ inducing binders/peptides using MERCI software. Our analysis indicates that IFN-γ inducing and non-inducing peptides can be discriminated using above features. We developed models for predicting IFN-γ inducing peptides using various approaches like machine learning technique, motifs-based search, and hybrid approach. Our best model based on the hybrid approach achieved maximum prediction accuracy of 82.10% with MCC of 0.62 on main dataset. We also developed hybrid model on IFNgOnly dataset and achieved maximum accuracy of 81.39% with 0.57 MCC.ConclusionBased on this study, we have developed a webserver for predicting i) IFN-γ inducing peptides, ii) virtual screening of peptide libraries and iii) identification of IFN-γ inducing regions in antigen (http://crdd.osdd.net/raghava/ifnepitope/).ReviewersThis article was reviewed by Prof Kurt Blaser, Prof Laurence Eisenlohr and Dr Manabu Sugai.

Development of HPV16,18,31,45 E5 and E7 peptides-based vaccines predicted by immunoinformatics tools

“…To determine the ability of the selected epitopes to induce interferon-gamma, IFNepitope server (https:// crdd.osdd.net/raghava/ifnepitope/predict.php) was used. In this study, we used Motif and SVM hybrid algorithms (accuracy of 81.39%) and IFN-gamma versus Non IFN-gamma model for prediction (Dhanda et al 2013).…”

In silicodesign of Multi-epitope-based peptide vaccine against SARS-CoV-2 using its spike protein

“…The consensus thus arrived upon can suitably considered as probable sequences that are capable of eliciting the necessary immune response. Apart from this, each of the sequences were filtered based on their predicted antigenicity over Vaxijen [33], allergenicity over AllerTop [34] and Algpred [35], toxicity over ToxinPred [36], capable of eliciting Interferon-gamma over IFN-G [37]. Along with this, the epitope sequences were matched with human proteins to prevent cases of antibody response against a self-antigen with the Multiple Peptide Match Tool.…”

Computational Approach for Screening the Whole Proteome of Hantavirus and Designing a Multi-Epitope Subunit Vaccine

“…Several cytokines such as IL-2, IL-6, IL-10, IFN-γ, and TNF-γ are associated with Hantavirus immune system 93 . Herein, due to the tools unavailability, we only predicted the IL-10, and IFN-γ inducer HTL epitopes through the IL-10Pred server (https://webs.iiitd.edu.in/raghava/il10pred/) 94 and IFNepitope server (http://crdd.osdd.net/raghava/ifnepitope/) 95 , respectively in which the SVM technique was chosen as a prediction method.…”