Artificial intelligence predicts the immunogenic landscape of SARS-CoV-2: toward universal blueprints for vaccine designs

Brandon Malone; Boris Simovski; Clement Moline; Jun Cheng; Marius Gheorghe; Hugues Fontenelle; Ioannis Vardaxis; Simen Tennoe; Jenny-Ann Malmberg; Richard Stratford; Trevor Clancy

The outbreak of Coronavirus disease 2019 (COVID-19) and its rapid worldwide transmission resulted in the World Health Organization (WHO) declaring COVID-19 as a pandemic and global health emergency. COVID-19 is caused by the novel coronavirus Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) [1] . Like all coronaviridae, SARS-CoV-2 is a positive-sense RNA viruse encapsulated by an envelope, and characterized by an exposed spike glycoprotein (S-protein) that is projected from the viral surface and comprises a large RNA genome [2] . Although the main structural proteins on coronaviridae, such as the S-protein, are reasonably well studied, many of the other proteins are less well characterized. Correcting this gap may be important to improve the design of therapeutic interventions [3] . This particular gap in knowledge is very relevant from the perspective of finding immunogenic targets across the entire virus proteome, in order to guide the design of effective vaccines. The SARS-CoV-2 virus is closely related in sequence identity and receptor binding to SARS-CoV [4, 5] , and therefore it has been purported that one may borrow from this similarity to validate targets in potential vaccines [6, 7] . Much of the emphasis on coronaviridae vaccines to date has focused on antibody responses against the Sprotein, which is the most "antibody exposed" structural protein in the virus.