Machine intelligence design of 2019-nCoV drugs

Kaifu Gao; Duc Duy Nguyen; Rui Wang; Guo-Wei Wei

S-protein is one of the four proteins in coronavirus, which can be cleaved by a host cell furin-like protease (non-structural protein in coronavirus) into two functional units, S1 and S2. S1 makes up the large receptorbinding domain (RBD) to facilitates virus infection by binding to host receptors and S2 forms the stalk of the spike molecule. 11 Therefore, one possible way to control the infection is to seek protease inhibitors to prevent the S-protein of 2019-nCoV cleaved into S1. Another possible way is to seek specific drugs that have the ability to inhibit RBD binding to host receptors. However, consider the fact that the diversity of CoV is reflected in the variable S-proteins and the gene of S-protein is easy to mutate, [12] [13] [14] [15] it is not economical to design a new drug to prevent the RBD binding to host receptor directly. Moreover, considering the high sequence identity between viral proteases of 2019-nCoV and SARS-CoV, seeking protease inhibitors to treat respiratory diseases induced by SARS and this novel pneumonia will be our first choice. Viral proteases are common targets in dealing with human viruses such as the HIV virus and hepatitis C virus. Protease inhibitors are remarkably effective in blocking the replication of coronavirus, including the SARS and the Middle East respiratory syndrome (MERS), providing a promising foundation for the development of anticoronaviral therapeutics. An overview of SARS-CoV 3CL protease inhibitors has been reported. 16 However, currently, there are no effective anti-SARS-CoV drugs available despite there are more than 3500 publications. It is true that there were only 8422 infected cases and 916 deaths reported for SARS-CoV, which might make the drug development unprofitable. The fact that we are unprepared for another coronaviral pandemic, like the Wuhan coronavirus outbreak, causes the world economy trillion dollars, not to mention the loss of many lives.