Russian Supercomputer Joins the Fight Against COVID-19

An international research team of scientists – from Russia, Finland, Italy, China, Japan and Canada is using a recently upgraded HPC system at the Joint Supercomputer Center of the Russian Academy of Sciences to develop diagnostics and treatment against COVID-19 coronavirus infection that became the cause of global pandemic.

The Joint Supercomputer Center of the Russian Academy of Sciences (JSCC RAS) provides high priority access to computing resources for research teams studying methods to fight the COVID-19 coronavirus infection. Russian scientists as a part of the multi-national research team use recently upgraded MVS-10P OP cluster system based on high-performance 2nd Generation Intel Xeon Scalable server processors. The international project aims to develop medicine for diagnostics and therapy against the coronavirus contagious disease.

The coronavirus pandemic in 2020 threatens lives of many people and hinders economic and social activity in multiple countries all over the world. As a result, it attracted significant attention of many research groups. Finding treatments to prevent and mitigate negative impact of COVID-19 is the highest priority in the scientific community now. New publications about the virus appear on a regular basis. However, a lot more needs to be learned about coronavirus to develop an effective treatment. Current studies use not only the latest advances in experimental physics, chemistry, and biology to investigate the life cycle of the virus, but also sophisticated simulation methods that require a supercomputing power.

Rapid global spread of COVID-19 coronavirus infection pandemic has shown that there are no clear global emergency response plans against threats to humankind caused by new viruses,” said said Anna Kichkailo, Head of Laboratory For Digital Controlled Drugs and Theranostics at the Krasnoyarsk Federal Science Center. “One of the obvious shortcomings is the lack of technologies for quick development of medicines for diagnostics and therapy. To help solving this problem, an international team of scientists – from Russia, Finland, Italy, China, Japan and Canada – was formed. We all have different competences, knowledge, skills and resources. Our geographically distributed team includes virologists, biologists, chemists, mathematicians and physical scientists. The international cooperation is extremely important to achieve quick progress and rapidly react to the ever-changing situation with global COVID-19 pandemic. We hope that our research will actually help to fight spread of such infections.”

The international team consists of:

  • Laboratory for Digital Controlled Drugs and Theranostics and Laboratory of Physics of Magnetic Phenomena, Kirensky Institute of Physics at the Federal Sience Center, Siberian Branch of Russian Academy of Sciences (KSC RAS, Krasnoyarsk, Russia)
  • Laboratory For Biomolecular and Medical Technology, V.F. Voyno-Yasenetsky Krasnoyarsk State Medical University (KrasSMU, Krasnoyarsk, Russia) – project coordinator
  • Laboratory of Chemical Cybernetics, Department of Chemistry at Lomonosov Moscow State University (MSU, Moscow, Russia)
  • Laboratory for Computer Simulation of Biomolecular Systems and Nanomaterials at N. M. Emanuel Institute of Biochemical Physics Institute, Russian Academy of Sciences (IBCP RAS, Moscow, Russia)
  • Organic Synthesis Laboratory, Institute of Chemical Biology and Fundamental Medical Science, Siberian Branch of the Russian Academy of Sciences (ICBFM SB RAS, Novosibirsk, Russia)
  • Nanoscience Center and Department of Chemistry, University of Jyväskylä, Jyväskylä (Finland)
  • Institute of Endocrinology and Experimental Oncology, CNR, Naples (Italy)
  • The Molecular Science and Biomedicine Laboratory (MBL), Hunan University
    Changsha, Hunan (China),
  • Institute of Molecular Medicine, Shanghai Jiao Tong University, School of Medicine,
    Shanghai (China)
  • Department of Chemical Biology, Xiamen University, Xiamen, Fujian (China),
  • Research Center for Computational Design of Advanced Functional Materials (CD-
    FMat), National Institute of Advanced Industrial Science and Technology (AIST),
    Tsukuba (Japan)
  • Department of Chemistry and Biomolecular Sciences, University of Ottawa,
    Ottawa (Canada)

Computer design of medicine against COVID-19

“We aim to use molecular simulation to create a computer model of a medical drug with selective interaction with receptor-binding domain of Spike protein of SARS-CoV-2 coronavirus strain. The most promising specific binding agents to be used for diagnostics (identification of virus particles in saliva) and development of virus treatment drugs preventing ingress of infection. The results of theoretical calculations and computer simulation will be experimentally tested on proteins, viruses and cells,” — summarizes Anna Kichkailo.

Supercomputer simulations are used to study details of interaction between Spike- protein on coronavirus surface and the human protein ACE2 (angiotensin-converting enzyme 2). ACE2 is known to be the entry point for SARS and SARS-2 coronaviruses. By blocking its interaction with the spike protein, it is possible to reduce virus activity in human body. Massive molecular dynamics and quantum chemistry calculations of virus and human proteins are using to estimate protein binding energies. The results of simulations will be used to design aptamers that will bind with virus proteins better than ACE2. Molecular docking and molecular dynamics methods will be used to create a library of promising aptamers and to estimate the strength of their interaction with virus proteins. Binding energies for the most promising aptamers will be refined with quantum chemistry methods. A lot of supercomputing resources is required to complete all these research stages within limited amount of time.

The need for supercomputers

Developing drugs to mitigate the disease and reduce the risk of the severe complications is one of the most important tasks before coronavirus vaccine will be widely adopted. Computer simulations deliver valuable information on the viral activity on atomic level and they can be used to predict the efficiency of potential drugs. Such calculations are extremely demanding and can be done only with the most powerful supercomputers. HPC systems are widely used in simulations of biochemical processes. The simulations help to reduce the number of experiments that would otherwise be needed to get same results. Leading global pharmaceutical and research centers use molecular modeling at the initial steps of drug development, when a massive number of chemical substances have to be investigated for specific activity.

Experimental data about the coronavirus activity on molecular level is very limited and have been produced in vitro. For example, the viral protein structure corresponds to the crystallized protein and not to a virus in solution. Moreover, there is not enough experimental data on complexes between virus and human proteins or virus proteins and potential drugs. On the other hand, supercomputer calculations can give all the structural data and the details of binding process. Therefore, the computing part is critically important, as well as subsequent experimental verification.

Upgraded MVS-10P OP supercomputer at JSCC RAS

The Joint Supercomputer Center of the Russian Academy of Sciences is one of the most powerful Russian supercomputing centers in the fields of science and education. JSCC RAS staff includes qualified scientists, programmers and engineers. Over 150 groups of researches use JSCC resources for fundamental and applied research tasks.

Total peak performance of JSCC RAS computing facilities exceeds 1.3 Petaflops (petaflops – quadrillion of floating-point operations per second, or 1000 teraflops). Five JSCC RAS cluster systems are included in the Top50 list of the most powerful Russian supercomputers.

After the recent upgrade of MVS-10P OP at the end of 2019 completed by the Ministry of Science and Highest Education of the Russian Federation, its peak performance reached 771 Teraflops (teraflops – trillion of floating-point operations per second). Adding a new segment based on the modern high-performance 2nd Generation Intel Xeon Scalable processors allowed to achieve the significant performance increase of y 209 Teraflops. MVS-10P OP is based on RSC Tornado, an universal ultrahigh-dense and energy efficient platform developed by RSC Group (Russia).

By regularly upgrading JSCC RAS computing resources we get new R&D opportunities, provide RAS and academic researches with powerful resources for various complex fundamental and applied tasks and improve overall efficiency of Russian scientists,” — said Gennady Savin, Academician of RAS and Science Head of the Joint Supercomputer Center of the Russian Academy of Sciences.

Researchers access resources of JSCC RAS using the National Research Network (NICS) of the Ministry of Science and Highest Education of the Russian Federation.

Sign up for our insideHPC Newsletter