Abstract
Infectious bronchitis virus (IBV) continues to threaten the poultry industry due to its rapid mutation and genetic diversity. We introduce AstraMEV, a predictive structural immunoinformatic platform that leverages genomic screening of 56 IBV genomes to identify conserved epitopes from the nucleocapsid (N) and spike (S) proteins. Two multiepitope vaccine constructs, SmallTope and BigTope, were designed and refined using AI-driven methods such as RFdiffusion and ProteinMPNN to optimize structural stability and immunogenicity. Molecular dynamics simulations confirmed the constructs' stability and effective binding to chicken Toll-like receptor 4 (chTLR4). Crucially, Poisson-Boltzmann calculations revealed that certain computationally derived variant epitopes significantly improved the binding enthalpy, highlighting key interaction domains. Our comprehensive approach demonstrates the feasibility of leveraging computational immunoinformatics for targeted vaccine design. These findings underscore AstraMEV's robust potential for next-generation IBV vaccines and broader applications in combating rapidly evolving pathogens with significant global impact.