Advancing therapeutic vaccines for chronic hepatitis B: Integrating reverse vaccinology and immunoinformatics

Table 1 Web servers available for predicting Cytotoxic T Lymphocyte, Helper T lymphocyte, and B-cell epitopes

Web server	Prediction methods	Website link	Ref.
Cytotoxic T lymphocyte epitope
NetCTL 1.2	Uses an ANN approach to combine MHC class I peptide binding, proteasomal C-terminal cleavage, and TAP transport efficiency	https://services.healthtech.dtu.dk/services/NetCTL-1.2/	Larsen et al[69], 2007
NetMHCpan 4.1	Predicts peptide binding to MHC molecules based on quantitative binding affinity and eluted ligands identified by mass spectrometry, using an ANN approach	https://services.healthtech.dtu.dk/services/NetMHCpan-4.1/	Reynisson et al[70], 2020
CTLPred	Predicts CTL epitopes based on T cell epitope patterns, utilizing both ANN and SVM approaches	http://crdd.osdd.net/raghava/ctlpred/	Bhasin and Raghava[71], 2004
Helper T lymphocyte epitope
NetMHCIIpan 4.0	Predicts peptide binding to MHC II molecules (HLA-DR, HLA-DQ, HLA-DP) based on binding affinity and eluted ligands identified by mass spectrometry, using an ANN approach	https://services.healthtech.dtu.dk/services/NetMHCIIpan-4.0/	Reynisson et al[70], 2020
ProPred	Predicts MHC Class II (HLA-DR) binding regions within antigen sequences using QM	http://crdd.osdd.net/raghava/propred/	Singh and Raghava[72], 2001
MARIA	Predicts the likelihood of antigen presentation from a specific gene related to HLA class II alleles, using peptide sequences from mass spectrometry, antigen gene expression levels, and protease cleavage patterns with an ANN approach	https://maria.stanford.edu/	Chen et al[73], 2019
Linear B Cell epitope
ABCpred	Uses an RNN approach that considers peptide length to predict B cell epitopes within antigen sequences	http://crdd.osdd.net/raghava/abcpred/	Saha and Raghava[114], 2006
Bepipred Linear Epitope Prediction 2.0	Uses a random forest algorithm trained on annotated epitopes from antibody-antigen protein structures	https://services.healthtech.dtu.dk/services/BepiPred-2.0/	Jespersen et al[115], 2017
BCEPS	Predicts linear B cell epitopes using an SVM approach based on the tertiary structure of antibody-antigen complexes	http://imbio.med.ucm.es/bceps/	Ras-Carmona et al[116], 2021
SEMA	Applies a transfer learning approach using a pre-trained deep learning model to predict conformational B cell epitopes based on primary antigen sequences and tertiary structures	https://sema.airi.net/	Shashkova et al[117], 2022
LBtope	Uses SVM and Ibk approaches on a large dataset of experimentally validated B cell epitopes and non-epitopes to predict linear B cell epitopes	https://webs.iiitd.edu.in/raghava/lbtope/index.php	Singh et al[118], 2013
Bcepred	Predicts B cell epitopes using physicochemical properties, such as hydrophilicity, flexibility, accessibility, polarity, exposed surface, and turns	http://crdd.osdd.net/raghava/bcepred/	Saha and Raghava[77], 2004
COBEpro	Uses an SVM to predict short peptide fragments within query antigen sequences, calculating an epitope propensity score for each residue	https://scratch.proteomics.ics.uci.edu/	Sweredoski and Baldi[119], 2009
CLBTope	Combines alignment-based and alignment-free machine learning methods to predict B cell epitopes, using epitope and non-epitope sequence composition	https://webs.iiitd.edu.in/raghava/clbtope/	Kumar et al[120], 2024

ANN: Artificial neural network; CTL: Cytotoxic T lymphocyte; HLA: Human leukocyte antigen; Ibk: Instance-based k-nearest neighbor; MHC: Major histocompatibility complex; QM: Quantitative matrix; RNN: Recurrent neural network; SVM: Support vector machine; TAP: Transporter associated with antigen processing.