Junmei Wang is a scientist, researcher, in the Department of Pharmaceutical Sciences and Computational Chemical Genomics Detection Center at the College of Pharmacy at the University of Pittsburgh, USA. There he has carried out a study to identify how and what drugs already on the market can be used to alleviate and fight the coronavirus. The difference from Wang’s study is that it has used artificial intelligence (AI) and computerized detection to find out in record time, just five days, the potential candidate drugs to be reused against COVID-19.
“It is very efficient to apply computer-aided drug design techniques to quickly identify promising drug reuse candidates, especially after the detailed 3D structures of key viral proteins are resolved,” says the scientist from the University of Pittsburg. And for this, he performed molecular dynamics simulations, followed by binding free energy calculations using a method called MM-PBSA-WSAS, in several drugs: Carfilzomib, Eravacycline, Valrubicin, Lopinavir and Elbasvir.
And what did the computer simulations by Dr. Wang reveal? That carfilzomib, a drug already approved against cancer, specifically against multiple myeloma, and that acts as a proteasome inhibitor, is the most effective of all those that he analyzed against COVID-19. That is, an antitumor is, according to their study, and using the most modern artificial intelligence techniques, the best candidate to treat COVID-19.
“A great application of drug reuse,” Junmei Wang says, “is identifying drugs that were developed to treat other diseases to treat a new disease. Drug reuse can be achieved by performing systematic drug-drug target interaction analyzes, but also using new computational techniques that shorten the analysis time ”. To balance computational efficiency and precision in his research, Wang applied a hieratic strategy that employs different types of scoring functions, both in lead identification and optimization phases.
Computational time
For this study, the amounts of computational time used for each individual task were diverse. A single-core Intel Xeon E5-2683 CPU was used for the docking tests of all approved drugs, which took about 11 hours. The calculations to know the point loads of each studied drug took just over a day, using a GTX-1080 GPU (Graphic Processing Unit). And the MM-PBSA-WSAS calculation consumed another day. Therefore, equipped with a sufficient number of CPUs and GPUs, and current hardware , “we were able to complete the drug reuse tests within four to five days.”
But Dr. Wang also concluded with his study that it is possible, with greater computational capabilities and with artificial intelligence, to carry out these investigations in even less time, and thus be able to identify how existing drugs, and in adequate doses, provide very good options to treat a new disease until a vaccine arrives. A real discovery, and a confirmation, that an antitumor can be used as an antiviral.
Something that has also been confirmed a few days ago in Spain by the National Center for Biotechnology dependent on the CSIC, although for this it has not used artificial intelligence and has used the expertise and intelligence of a team of scientists led by Dr. Luis Enjuanes. The CSIC has reached the same conclusion as the researcher from the University of Pittsburg Junmei Wang, in this case pointing to an antitumor from a Spanish company due to its possible efficiency as an antiviral against COVID-19, something that must now be confirmed with the mandatory tests clinical trials in patients, and which is awaiting authorization from the Spanish Agency for Medicines and Health Products (AEMPS).
