Computational Design of a Hepatitis B Virus Vaccine Using Multiple Sequence Alignment and Modeller v10.5 for 3D Protein Structure Prediction

Background: Hepatitis B is a viral infection affecting the liver, causing severe damage. The disease affects 350-400 million people globally, with the highest rates in Asia, sub-Saharan Africa, and Central Asia. The materials used are the virus's protein, HBsAg, to be studied and to create a vaccine to treat the disease. The research aims to create a vaccine against hepatitis B using in vitro methods and applications like NCBI, MSA, modeler v10.5 program, and Swiss-model web server. Material and Methods: The sequence is obtained using UniProt, and the phylogenetic tree is visualized using MEGA11. The NetSurfP web server predicts secondary structures, and InterPro Scan aligns primary sequences with the homologous structure. Modeller v10.5 is used to predict and create a 3D model for the vaccine. Protein sequences are obtained using NCBI, Discovery Studio software, and VLP-based vaccines to predict the B and MHC I T cell epitope. The allergenicity and antigenicity of vaccines are tested using AlgPred web servers and VaxiJen v2.0. Homology modeling results are obtained and analyzed using the Ramachandran plot. Result: Salient points of this study dealing with the computational design of the Hepatitis B virus vaccine may be summarized as follows: different hepatitis B virus large envelope proteins were subjected to Multiple Sequence Alignment, and Phylogenetic Analysis identified the German sequence as the most recent strain. The secondary and tertiary structure of HBcAg core protein and HBsAg Germany HBV genotype D is predicted. Vaccine design was performed using a German HBsAg strain in BLAST, generating proteins from the selected sequences on Modeller v10.5 and refining the models through multiple templates. Predicted B and T cell epitopes from different web servers were combined into one peptide sequence of the designed vaccine. The allergenicity test using AlgPred forecasted the vaccine as a non-allergen, while VaxiJen demonstrated high antigenicity Conclusion: The Swiss-model web server generated a model for the final vaccine sequence. The Ramachandran plot analysis showed 84.7% of residues in the favored region, indicating a sound structural model. This high percentage of residues in the favored region suggests that the predicted 3D structure of the vaccine candidate is likely to be stable and biologically plausible. The Ramachandran plot is a good guide to the quality of the protein structures: the more residues that fall into the favored regions, the more reliable and generally the refinement of the models.