REVEAL™ B Cell Epitope Discovery Services

B Cell Epitope Prediction

The REVEAL™ B Cell Epitope Prediction System predicts and ranks B cell epitopes in a protein of interest. Prediction is based on the primary sequence and the three-dimensional structure, when available.

Case Study: Correlation between prediction service and published epitopes

B Cell Linear Epitope Mapping Service

Request Quote

Intelligent consensus prediction methodology

The REVEAL™ B cell epitope prediction system uses a large collection of up to date public domain and commercially licensed scientific databases and software. The strength of the technology lies in a best practise approach that has been developed to enable the output from several different methods to be combined, to generate the best possible consensus prediction.

Unlike combination algorithms that attempt to aggregate the information generated from individual base algorithms mechanistically, we utilize our experience, judgment and any additional information available on the protein sequence of interest to ensure that important data are not overlooked. Automated approaches do not take into account the many exceptions and special cases that so often apply to protein sequences. The REVEAL™ system overcomes this problem by weighting algorithms appropriately and changing the individual parameter settings of methodologies in accordance with experience, in order to produce an intelligent consensus prediction.

Epitope prediction based on primary sequence

The consensus prediction combines several methods, including public and proprietary algorithms that evaluate:

• Antigenicity
• Hydrophobicity
• Hydrophilicity
• Flexibility
• Secondary structure prediction
• Beta-turn structure analysis
• Protein physiochemical properties

These algorithms and their interpretation are applied to primary protein sequences regardless of whether or not a three dimensional structure is available. On request, the top consensus predicted epitopes can be mapped to protein sequences available in public databases and to protease cleavage sites for the protein of interest for analysis of potential degradation.

Epitope prediction based on resolved three-dimensional structure

Where a three dimensional structure is available it can be used to evaluate the characteristics as described above but in addition we can provide the degree of actual epitope surface exposure and association with beta turn structures. This information is used to improve the relative ranking of predicted epitopes. Our reports include detailed three dimensional visualization of predicted epitopes within the available 3-D structure.

Prediction of discontinuous epitopes

If the solved 3-D structure of the target protein is available, the REVEAL™ report includes information about regions that are brought together at the surface of the target molecule to form predicted conformational, or discontinuous epitopes.

Features of the REVEAL™ Rapid B Cell Epitope Discovery System

Sequence Annotation

For projects using only the primary sequence, the client report shows results for all the prediction algorithms used and annotated secondary structure prediction. Specific comments are given on the N- and C-terminal regions, which often contain relevant epitopes, as well as the impact of signal and leader sequences on epitopes. Additionally, when the 3-D structure is available, the report also shows detailed mapping of the predicted epitopes onto the structure.

Analysis of signal and cell sorting sequences
The presence of signal and cell sorting sequences is of particular interest if predicted epitopes fall within them. Detection of such sequences may also provide an indication of where the target protein and its associated epitopes are localized, which can aid the development of strategies for expression or antisera production.

Identification of PEST sequences
Regions rich in the amino acids proline (P), glutamic acid (E), serine (S) and threonine (T) are associated with enhanced intracellular proteolytic processing of the target protein. The hydrophilicity of PEST sequences means that they could form loops or extensions at the surface of a protein. Overlap between a PEST sequence and a predicted epitope can be taken as further evidence in support of the predicted epitope being surface-located on the target molecule. The presence of a PEST sequence will also give some indication of intracellular half-life, which can be relevant if attempting to express the protein of interest or sequences derived from it.

Identification of low complexity regions and conserved protein domains
Predicted epitope regions showing low compositional complexity are generally best avoided if antibody specificity is a priority. The PFAM database is searched to identify whether the predicted epitope regions are part of known conserved protein domain families. Identifying conserved domain regions and low complexity regions can either help to avoid the selection of epitopes that are less likely to be antigenic, or can help to select epitopes for generation of antisera that are more likely to cross react with epitopes on other proteins.

Prediction of transmembrane regions
Predicting transmembrane regions can be particularly important in vaccine target discovery in order to establish whether a protein is presented on the surface of a pathogen and is therefore accessible to the B cell immune response. For this reason the REVEAL™ B Cell Epitope Prediction System includes analysis of the protein for the presence of transmembrane alpha-helices. The location of the transmembrane regions can be predicted and this information is subsequently used to determine whether the predicted consensus epitopes are extra- or intracellular, or could be located in the loops of amino acid sequence between helices.

Peptide-Antibody Production Strategy
All findings from the epitope prediction and related protein annotation algorithms are taken into account in order to advise on the selection of particular antigenic sequences for immunization to generate anti-peptide antibodies. This includes advice on optimizing these sequences to maximize peptide stability, and specific information about conjugation of the sequence to a carrier protein. Regions of low complexity of amino acid composition are identified and are best avoided if antibody specificity is a priority.

A full service report and accompanying data are presented in PDF format on CD.