Using a combined computational-experimental approach to predict antibody-specific B cell epitopes.

Area of Science:

• Immunology and Computational Biology
• Structural Biology and Bioinformatics

Background:

• Antibody epitope mapping is essential for understanding B cell immunity and therapeutic antibody characterization.
• Currently, a lack of widely adopted computational tools hinders B cell epitope prediction, unlike T cell epitope mapping.

Purpose of the Study:

• To develop and validate a computational method for predicting discontinuous B cell epitopes using antibody sequence information.
• To integrate antibody-specific predictions with cross-blocking experiments to enhance epitope prediction accuracy.

Main Methods:

• Developed antibody sequence-based predictions leveraging residue-pairing preferences and interface characteristics.
• Combined computational predictions with cross-blocking experiments to identify overlapping antibody epitopes.
• Validated the approach using X-ray crystallography, peptide ELISA, deuterium exchange, and site-directed mutagenesis against the D8 antigen.

Main Results:

• Demonstrated the ability to identify discontinuous epitopes on antigens solely from antibody sequence data.
• Showcased improved prediction accuracy by integrating antibody-specific computational data with cross-blocking experimental results.
• Successfully mapped conformational B cell epitopes on the D8 antigen with high performance.

Conclusions:

• Antibody sequence-specific computational predictions, when combined with cross-blocking experiments, enable accurate identification of conformational B cell epitopes.
• This integrated approach offers a powerful tool for B cell epitope mapping, advancing therapeutic antibody development and immunological studies.