|
O. V. Tinkov, V. Yu. Grigorev, L. D. Grigoreva
Virtual screening and molecular design of
potential SARS-CoV-2 inhibitors
Abstract
According to recent studies, the main Mpro
protease of the SARS-CoV-2 virus, which is the most important target in the
development of promising drugs for the treatment of COVID-19, is evolutionarily
conservative and has not undergone significant changes compared to the main Mpro
protease of the SARS-CoV virus. Many researchers have noted a high similarity
between the binding sites of the main Mpro protease of the SARS-CoV
and SARS-CoV-2 viruses, so in the context of an increasing epidemic, further
research of inhibitors of the main mpro protease of the SARS-CoV virus to fight
COVID-19 seems logical. QSAR methods (Quantitative Structure – Activity
Relationship), which were applied in this paper in the analysis of inhibitors
of SARS-CoV main Mpro protease, can significantly help in reducing
time and financial costs for repositioning and developing new drugs. In the
study using simplex, fractal, HYBOT descriptors, methods Partial Least Squares
(PLS), Random Forest (RF), Support Vectors, Gradient Boosting (GBM), as well as
the OCHEM Internet platform (https://ochem.eu), where
different types of molecular descriptors, machine learning methods are
implemented, QSAR models were built. For the obtained models, a structural
interpretation was performed, which allowed us to identify molecular fragments
that increase and decrease the activity of SARS-CoV inhibitors. The structural
interpretation results were used for the rational molecular design of potential
SARS-CoV-2 inhibitors. The QSAR models were also used for a virtual screening
of 2087 FDA-approved drugs.
Key words: Mpro protease, QSAR,
molecular descriptors, machine learning, structural interpretation.
Copyright (C) Chemistry Dept., Moscow State University, 2002
|
|
