Supercomputer helped fight coronavirus
An American supercomputer helped find 77 compounds to fight the coronavirus. Now it remains to be seen how effectively they prevent visions from actually penetrating the cells of the human body.
Researchers at the Oak Ridge National Laboratory (ORNL) of the US Department of Energy used the world's most powerful supercomputer, the IBM AC922 Summit, to identify 77 low-molecular compounds that could form the basis of coronavirus drugs. They described the results in an article posted on the ChemRxiv preprint service.
Researchers have modeled over 8,000 compounds that could bind to coronavirus proteins, making it unable to infect host cells.
Of these, they selected 77 most valuable for subsequent experimental studies.
Scientists are interested in how exactly the virus infects the cells of the body. When Chinese researchers sequenced the genome of the virus, it became known that the mechanism of infection is similar to the actions of the causative agent of SARS. Jeremy Smith, director of the ORNL Center for Molecular Physics, suggested that SARS-CoV-2 вЂњdockedвЂќ to the cells in the same way as its counterpart.
Then his colleague, Mykolas Smith, built a model of a protein that forms spikes on the surface of the coronavirus (S-protein). It was based on early studies of the structure of the virus.
вЂњWe were able to develop an accurate computational model based on virus information that has only recently been published in the literature on this virus,вЂќ says Mykolas Smith.
Created in 2018, the Summit supercomputer located in ORNL is capable of performing 200 quadrillion calculations per second вЂ” about a million times more than the average laptop. With it, the researchers modeled the molecular dynamics of the attachment of various compounds to the spikes of the virus.
They were interested in what compounds could prevent the virus from attaching to the cells of the human body.
вЂњUsing Summit, we ranked these compounds based on a number of criteria related to how likely they would adhere to the S-protein spike,вЂќ explains Mykolas Smith.
In total, they selected 77 low molecular weight compounds, both artificial and natural origin. During the simulation, they contacted those areas of the spikes that are used to penetrate human cells вЂ” which means that these compounds can interfere with infection.
Jeremy Smith emphasizes that computing alone is not enough вЂ” an experiment must follow. Now that they have found promising compounds, they still have to study and test how effective they will be in reality.
вЂњSummit was needed to get results quickly,вЂќ says Jeremy Smith. вЂњIt took us a couple of days, and on a regular computer we would have spent months.вЂќ Our results do not mean that we have found a cure for coronavirus. However, we very much hope that our data will serve as the basis for future studies of these compounds. Only then will we find out if any of them has the characteristics necessary to attenuate this virus.вЂќ
Recently, a new, more accurate model of S-protein spikes has appeared, and the team plans to start computing again based on the latest data. They warn that new results may differ from current ones.
In Russia, meanwhile, trials of a coronavirus vaccine began. It was developed at the State Scientific Center for Virology and Biotechnology вЂњVectorвЂќ.
Experts have developed vaccine prototypes based on six different technology platforms. All of them are tested on sensitive laboratory animals, including lower primates.
Ahead is the development of the most promising and safe prototypes, as well as determining the composition, dose, and route of administration of the future vaccine.
Presumably, the vaccine will be introduced in the fourth quarter of this year.
On March 19, Russian scientists succeeded in sequencing the coronavirus gene. According to scientists, the results of the study provide an understanding of how the virus evolves, which can help create a vaccine and drugs against a new infection.