The global pandemic caused by the SARS-C0V-2 virus continues to spread. Infection with the SARS-CoV-2 causes a disease of variable severity known as COVID-19. It is certain that the viral genome has already mutated and will continue to mutate in unknown directions. These mutations can (1) have no significant impact (they are silent), result in (2) a loss of virulence/function (LOF) or (3) a gain in virulence/function (GOF). Research involving GOF mutations remain especially controversial and highly regulated because accidental release or misuse of a more lethal virus could have catastrophic global effects.
The primary purpose of this project was to evaluate the ability of the DeepNEU machine learning stem-cell simulation platform to enable rapid and efficient assessment of the impact of viral LOF and GOF mutations on SARS-CoV-2 virulence. The data generated from this project confirm that (1) SARS-CoV-2 infection can be simulated in human alveolar type lung cells, (2) these simulated infected lung cells can be used to assess the impact of LOF and GOF mutations in a viral genome, (3) a new and simple four-factor virulence measure, the DeepNEU Case Fatality Rate (dnCFR) based on NSP3, Spike-RDB, N protein and M protein can be used to assess the impact of LOF and GOF mutations and (4) the platform combined with the dnCFR measure successfully identified specific and potentially beneficial mutations (LOF) as well as deleterious mutations (GOF) that potentially increase the virulence of the SARS-CoV-2 virus. We conclude that the DeepNEU platform and the dnCFR measure should be urgently developed, further validated and applied to SARS-CoV-2 and other viral pathogens as important tools for future pandemic preparedness.