An In Silico Deep Learning Approach to Multi-Epitope Vaccine Design: A Hepatitis E Virus Case Study

Abstract

Hepatitis E Virus (HEV) is a major cause of acute and chronic hepatitis. The severity of HEV infection increases manyfold in pregnant women and immunocompromised patients. Despite the extensive research on HEV in the last few decades, there is no widely available vaccine yet. In the current study, immunoinformatic analyses were applied to predict a multi-epitope vaccine candidate against HEV. From the ORF2 region, 41 conserved and immunogenic epitopes were prioritized. These epitopes were further analyzed for their probable antigenic and non-allergenic combinations with several linkers. The stability of the vaccine construct was confirmed by molecular dynamic simulations. The vaccine construct is potentially antigenic and docking analysis revealed stable interactions with TLR3. These results suggest that the proposed vaccine can efficiently stimulate both cellular and humoral immune responses. However, further studies are needed to determine the immunogenicity of the vaccine construct.