It is still a constant battle for us to win against SARS-CoV-2 Coronavirus. Chinese researchers discovered that once the virus enters our cells, it infects our bodies the same way as Severe Acute Respiratory Syndrome, or SARS, affected 26 countries in 2003. This resemblance sparked the study of a new virus.

Director of University of Tennessee’s Oak Ridge National Laboratory Center for Molecular Biophysics, Jeremy C. Smith, even suggested that the two viruses anchor onto the cell in similar ways.

A postdoctoral researcher from the same department, Micholas Smith, mentioned that they modeled the coronavirus S-protein based on a study published in China Life Sciences.

Coronavirus S-protein - THESIS.PH

Finding a Remedy Using Supercomputer IBM AC922 Summit

A research team from the Department of Energy’s ORNL used the most intelligent and powerful supercomputer, IBM AC922 Summit, to identify 77 small-molecule drug compounds that can enable further studies.

Out of 8,000 compounds simulated on the Summit, they screened potential ones that could bind to the main S-protein, so that the virus would be unable to infect host cells. They ranked the compounds of their interests and published their results on ChemRxiv.

Moreover, a Director’s Discretionary Allocation gave the team computational time to work with the Oak Ridge Leadership Computing Facility’s (OLCF’s) Summit Computers. These perform 200 quadrillion calculations per second, which is far greater than an average laptop. This supercomputer helped to achieve results quickly.

Micholas Smith also contributed more to the research by:

  • stimulating the different compounds that land on the virus’ S-protein spike to conclude if any of them could prevent the spike from sticking on our cells
  • ranking compounds based on how likely they could bind to the S-protein Spike
  • analyzing motions of protein’s particle using GPU-customized GROMACS Code

Fortunately, the researchers found 77 compounds that can interfere with possible infections. During simulations, the said compounds bind to regions where the virus enters the cell, therefore blocking possible infection entrances.

Future Plans for Retesting and Rearranging Compounds

After Science released a new accurate S-Protein model, the team ought to re-run the computational study with the new S-Protein. They further make it a point that it is necessary to test the 77 compounds experimentally before we can determine its usability.

Changing the ranking of chemicals most used, Jeremy noted that it was a good thing that they had the Summit to acquire the simulation results needed in a short amount of time.

Computational Screening is also necessary to introduce candidates for experimental studies verifying the effectiveness of those chemicals in combating the virus that threatens our life. The Laboratory R&D Program funded their research, which uses resources from DOE’s OLCF.

Smith also stresses that their results do not mean that they have found a cure that will solve the pandemic we are experiencing today. Still, they are hopeful that their computational findings will serve as a framework for experimentalists to investigate further these compounds and inform future studies. Only at this point will we know if their results possess the qualities to diminish the virus.

If you were given the chance to use the smartest and fastest supercomputer in the world, what information system would you develop for your thesis?


Please enter your comment!
Please enter your name here