Table 2.

Epitope prediction or reevaluation on accessible servers.

Programs	Purpose	Tools or procedures of evaluation	Input	Output	Internet accessibility	References
Epitope prediction
MimoDB 2.0	Mimotope database	MySQL relational database	according to menu	Structures visualization, alignments, and so forth	http://immunet.cn/mimodb/	[105]
MimoPro	Maps a group of mimotopes back to a source antigen so as to locate the interacting epitope on the antigen	Branch and bound optimization (analysis of overlapping patches on the surface of a protein)	PDB identifier mimotope sequences	Score + 3D location in antigen	http://informatics.nenu.edu.cn/MimoPro/	[108]
ViPR	Virus pathogen database	Integration of various resources	according to menu	Structures, annotations, and so forth	http://www.viprbrc.org/	[140]
MetaMHC	Prediction of MHC binding epitopes	meta-approach	QS	Four metapredictor scores including MetaSVMp score	http://www.biokdd.fudan.edu.cn/Service/MetaMHC.html	[141]
Epitopia	Prediction of B-cell epitopes	∗MLA ∗MLA + 3D + SA	∗QS ∗PDB ACF	Immunogenicity score + probability score + color scale record on QS	http://epitopia.tau.ac.il	[115]
ElliPro	New structure-based tool for the prediction of antibody epitopes	3D + SA + flx + antigenicity	∗QS ∗PDB ACF	Score + visualized epitope 3D structure and 3D location	http://tools.immuneepitope.org/tools/ElliPro	[114]
MHCPred	Prediction of class II mouse MHC peptide binding affinity	ISC-PLS, SYBYL software package	QS	Binding affinity (pIC50)	http://www.jenner.ac.uk/MHCPred	[129]
Stable epitopes and vaccines
BayesB	SVM-based prediction of linear B-cell epitopes using Bayes Feature Extraction	Residue conservation + position-specific residue propensities	QS	Residue epitope propensity score	http://www.immunopred.org/bayesb/index.html	[120]
OptiTope	Selection of optimal peptides for epitope-based vaccines	Multistep parallel evaluation	MSA of Ag list of ES table: ES, ES related MHC	Fraction of overall immunogenicity covered MHC	http://www.epitoolkit.org/optitope	[142]
PVS	PVS returns a variability-masked sequence, which can be submitted to the RANKPEP server to predict conserved T-cell epitopes.	3D visualization of MSA derived sequence variability “per site”	PDB ACF	3D map of variability fragments with no variable residues and their 3D location in antigen	http://imed.med.ucm.es/PVS/	[143]
PEPVAC	Prediction of MHC I; server can also identify conserved and promiscuous MHC I ligands	PSSM, distance matrix, phylogenic clustering algorithm	Genome, HLA-supertypes	Selected peptide sequences score	http://immunax.dfci.harvard.edu/PEPVAC/	[144]
RANKPEP	Prediction of MHC I and MHC II ligands	Profile comparison	Genome, HLA-supertypes, QS	Selected peptide sequences score	http://www.mifoundation.org/Tools/rankpep.html	[145, 146]

^∗Independent alternatives; Ag: antigen; BFE: Bayes Feature Extraction; flx: flexibility; ES: epitope sequences; HLA: human leukocyte antigens (human MHC); ISC-PLS: iterative self-consistent partial-least-squares based additive method; MHC: alleles of major histocompatibility antigens; MLA: machine learning algorithm; MSA: multiple sequence alignment; PDB: protein data bank; PDB ACF: PDB-derived atom coordinate file; QS: query sequence(s); SA: solvent accessibility; SVM: support vector machines.