A team of Russian scientists from Krasnoyarsk, Novosibirsk and Moscow, in cooperation with foreign colleagues from Finland, China, Japan and Canada, uses the upgraded supercomputer of the Joint Supercomputer Center of the Russian Academy of Sciences (JSC RAS) in the development of medicines for the diagnosis and treatment of the coronavirus infection. JSC RAS performs priority allocation of computing resources to research teams and organizations conducting research to fight COVID-19, which resulted in the global epidemic.

In 2020, the coronavirus pandemic has threatened many human lives and paralyzed the economic and social activity of almost every country in the world. The search for ways to prevent or reduce the negative consequences of the coronavirus infection today is a priority area of scientific research. Efforts of many scientists all over the world are directed towards studying the processes of the viral life cycle: almost every day several new works on this subject are published. This is not surprising, since the development of drugs requires a detailed understanding of the mechanisms of the functioning of the virus. For this, modern research methods are widely used, including complicated calculations and computer modeling in such fields as physics, chemistry and biology.

International project to fight global epidemic

“The rapid spread of the global pandemic of the coronavirus infection COVID-19 has demonstrated the lack of concise schemes and efficient means to quickly respond on a global scale to the arising threats caused by new viral diseases. The need to develop technologies to quickly create medical products for their diagnosis and therapy has brought together research teams from several countries: Russia, Finland, China, Japan and Canada. The importance of such an international cooperation is that we all have different competencies, knowledge, skills and resources. Our geographically distributed team includes virologists, biologists, chemists, mathematicians and physicists. But only by combining our efforts, can we quickly respond to a rapidly changing picture of the global COVID-19 epidemic. Our team hopes that the performed research will be able to provide real help in the fight against the spread of such infections, ” explains Anna Kichkaylo, head of the laboratory of controlled digital drugs and theranostics of the Federal Research Center “Krasnoyarsk Science Center” SB RAS, head of the laboratory of biomolecular and medical technologies of the V.F. Voyno-Yasenetsky Krasnoyarsk State Medical University.

Computer "design" of a medicine against COVID-19

“The idea of our project is to use methods of molecular modeling to make a computer “design”of a medicine which selectively interacts with the receptor-binding domain of the Spike protein of the coronavirus strain SARS-CoV-2. The most promising specific binding agents will be used for the diagnosis (detection of viral particles in saliva) and development of antiviral agents which would block infection. The results of theoretical calculations and computer simulations will then be tested experimentally on proteins, viruses and cells, ” summarizes Anna Kichkaylo.

Within the framework of the project, using the supercomputer simulation, scientists study in detail the interaction of the Spike protein on the surface of coronavirus with its target in a human body, the ACE2 protein (ACE2, Angiotensin-converting enzyme 2). ACE2 serves as an entry gate for SARS and SARS-2 coronaviruses; therefore, blocking the interaction of this protein with the virus is one of the promising approaches to reduce the viral activity in a human body. To estimate the binding energies of these proteins, large-scale molecular-dynamic and quantum-chemical calculations of virus and human protein complexes are carried out. Based on the data obtained, a computer selection of special molecules (aptamers) which bind better with viral proteins than with ACE2 will be performed. Ultimately, a library of aptamers - drug candidates – will be created and their interaction with the viral protein will be characterized using the methods of molecular docking and molecular dynamics. For the most promising aptamers, the binding energies will be refined using quantum chemistry methods. To perform all these stages of scientific research in a short time requires the use of a large amount of computing resources.

What are supercomputer calculations necessary for?

Prior to the start of mass use of the vaccine against coronavirus infection, an important task is the development of medicines which can ease the course of the disease and reduce the risk of serious consequences. Tangible results in such studies can be obtained using computer simulation methods allowing one to study processes at the level of individual molecules. Based on the knowledge gained, it is possible to predict the effectiveness of using various medicines. Such calculations are very time-consuming and can often be carried out only with the help of powerful supercomputers. By modeling biochemical processes on high-performance computing systems, it is possible to quickly obtain necessary data for subsequent full-scale experiments with significantly smaller volumes. This approach is widely used in leading medical-pharmaceutical and research centers at the initial stages of development, since such research is associated with the study of a huge number of candidate compounds in potential drugs for the treatment of specific diseases.

“Experimental data on the virus at the molecular level are very scarce and they were obtained under conditions which are different from reality. For example, the protein structure was obtained for a crystal of the virus protein rather than for a live virus in a solution. Moreover, there is not enough experimental data on the structures of the complex of virus proteins and human cells, as well as that of the virus protein and drug candidates. On the other hand, all these absolutely necessary data on the molecular structure and binding process can be obtained using supercomputer calculations. Therefore, the calculated component is urgent, as well as its subsequent verification by experiment, ” explains Dmitry Fedorov, senior researcher at the Research Center for Computational Design of Advanced Functional Materials, National Institute of Advanced Industrial Science and Technology (Japan).

Upgraded supercomputer at JSC RAS

The Joint Supercomputer Center of the Russian Academy of Sciences is one of the most powerful Russian supercomputer centers for collective use in the field of science and education. The LSC team consists of highly qualified scientists, programmers and engineers. The resources of the Center are used by more than 150 groups of researchers solving fundamental and applied problems.

The total peak performance of the computer systems at JSC RAS is more than 1.3 PFLOPS (petaflops – one quadrillion floating point operations per second, or 1000 teraflops). Five cluster systems of JSC RAS are included in the Top-50 rating of the most powerful Russian supercomputers.
After the latest upgrade of the MVS-10P OP supercomputer performed by the Russian Ministry of Education and Science at the end of 2019 as part of the program for the development of joint centers, its peak performance reached 771 TFLOPS (teraflops - one trillion floating point operations per second).

“Regular modernization of the computing resources of  JSC RAS allows us to give new opportunities for research and development, provide the research teams of RAS and universities with powerful resources to solve various complicated  fundamental and applied problems, as well as to ensure the most effective work of Russian scientists,” says academician of RAS Gennady Ivanovich Savin, supervisor of the Joint Supercomputer Center of the Russian Academy of Sciences.

Researchers have access to the resources of JSC RAS on the basis of the national research computer network of the Russian Ministry of Education and Science (NRCN) which is operated by JSC RAS.