5.1.1. Reverse vaccinology model
In recent years, vaccine design has undergone extensive evolutions due to reverse vaccinology (RV). In this regard, the desired pathogen genome is first evaluated by bioinformatic analysis and then potential vaccine candidates are identified [38]. Vaxign is the first web-based system which applies the RV algorithm to effectively offer the vaccine candidates for various microbial pathogens. Recently Ong et al. have achieved a new learning method namely Vaxign-ML machine to enhance the resolution of candidate prediction [38]. Using Vaxign RV and then Vaxign-ML systems, they first predicted 6 adhesion protein candidates including S protein and 5 non-structural nasp3, 3CL-pro, nsp8, nsp9, and nsp10 proteins for development of the COVID-19 vaccine. Contrary to previous researches around the COVID-19 vaccine design that focused on the S protein, it was the first time that the nsp3 and nsp8 were also announced as alternative candidates with significant antigenicity scores. Therefore, it seems that the solution to fight against COVID-19 infection is to use a cocktail vaccine that include a set of candidates (nsp3, nsp8 and S proteins) instead of a given antigen (S protein) to elicit a significant protective immunity [39].
A similar study according to in-silico RV strategy tried to render multi-epitope vaccine candidate against SARS-CoV2 infection and evaluated its biological activities by computational methods. They examined three antigens (ORF3a, N and M proteins) with the help of bioinformatic tools to find potential B-T lymphocyte-stimulating epitopes. Eventually, specific domains of the M or NOM protein containing highly scored B and T epitopes was introduced as the main vaccine candidate that established stable conjugates with Toll-like receptor (TLR) 4 and HLA-A-11:01 receptors using the imagery molecular dynamics and docking studies [40]. Therefore, RV seems to guide furthers research to more rapid access to immunogenic antigen cocktails in the design of the COVID-19 vaccine.