PLOS Computational Biology: MsgaBpred Model Improves B-Cell Epitope Prediction for Vaccine and Diagnostic Peptide Design
A PLOS Computational Biology paper published April 28 introduced MsgaBpred, a multi-scale graph attention model for predicting B-cell epitopes — short peptide sequences recognized by antibodies. The model achieved high accuracy on cross-species benchmarks and is directly relevant for peptide vaccine antigen design, diagnostic peptide development, and antibody-target discovery. The work joins the broader 2026 wave of AI-driven peptide tooling, complementing the Nature Communications DDA-BERT and CycloSEL papers from earlier in the month.