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 Monte Carlo simulation identified well over 100 different hotspots of length 27, 50 or 100 amino acids, for both AP and IP. Even after filtering for conservation and self-similarity we were left with over 50 different hotpots for both the AP and IP based analyses. In order to develop a blueprint for viable universal vaccine against SARS-CoV-2, it is necessary to 1) cover with fidelity a broad proportion of the human population, and 2) prioritize the selection to even fewer regions (the exact number may depend on the size of the bin and the vaccine platform under consideration). Consequently, we need to identify the optimal constellation of hotspots, or relevant viral segments, that can provide broad coverage in the human population with a limited and targeted vaccine "payload". In order to achieve this aim, we developed and applied (see Materials and Methods) a "digital twin" method, which models the specific HLA background of different geographical populations and used the method to identify optimal clusters of immunogenic epitope hotspots that will induce immunity in the broad human population. A graph-based mathematical optimization approach is then used to select the optimal combination of immunogenic epitope hotspots that will induce immunity in the broad human population. The results of this analysis are shown in Figure 7 . The output clearly identified a subset of hotspots that may be combined to stimulate a robust immune response in a broad global population. An example hotspot for the ORF3a100-150 region is provided in the supplementary data file, which shows the amino acid sequence and its component Class I and Class II epitopes. author/funder. All rights reserved. No reuse allowed without permission.