HUAWEI CLOUD and Multiple Research Institutions Jointly Launch a Visualized Drug Screening Platform for COVID-19
Apr 15, 2020
Drug discovery is a long journey, starting from a viral target protein sequencing. Homology modelling and molecular dynamics simulation could be used to predict and optimize the protein 3D structure; virtual screening could then be used to simulate the binding energies of the target protein again millions of drugs, which reduces the drug candidates to hundreds; ADMET & AI models could be used to further prioritize the drug candidates; the prioritized drug candidates are then needed to pass through in vivo experiments, in vitro experiments and clinical trials, before going to market.
The drug screening process
Due to limited computational resources, most drug screening studies only focused on one or two SARS-CoV-2 target proteins. These studies also produce variances because of different tools used and different parameter settings. In order to provide a standard and comprehensive virtual screening database for SARS-CoV-2, HUAWEI CLOUD collaborated with Professors from Huazhong University of Science & Technology, Xi'an Jiaotong University, and Beijing Institute of Genomics. The joint research team identified 21 target proteins of SARS-CoV-2, where virtual screening was used to simulate against all the clinical drugs reported. About 180,000 drug-target simulations were calculated for the 8,500 drugs listed SMDs in the clinical stage. This allows researchers to systematically compare the drug-target binding energies across different targets or different drugs.
To further facilitate researchers to search and visualize the drug screen result, we build a website called "Shennong Project" (https://www.shennongproject.com:11443/#/home), which has an easy to use query interface and 3D molecular visualization interface. By making a few clicks, visitors could interactively investigate into the 3D molecular docking structures.
"Shennong Project" query page and molecular visualization page
"Shennong Project" currently includes small molecule drugs from Drugbank, common compounds from Chinese traditional medicine, and effective ingredients from natural products. This is by far the largest in silico drug screening dataset for SARS-CoV-2 in the world. We hope this website could serve as a good resource for the antiviral drug studies, and also as an education tools for public to learn about drug discoveries. This work is already published in ChemXriv: Systemic in Silico Screening in Drug Discovery for Coronavirus Disease (COVID-19) with an Online Interactive Web Server.
For more information about antiviral drug discovery from HUAWEI CLOUD, please visit:
https://www.huaweicloud.com/intl/en-us/product/eihealth.html.