|
O. V. Tinkov, V. Y. Grigorev, L. D. Grigoreva
Prediction of the biotransformation time of
organic compounds by the example of avermectins
Abstract
Currently, the spread of the SARS-CoV-2
coronavirus is a significant problem for all of humanity. One of the promising
agents for the fight against the SARS-CoV-2 coronavirus is ivermectin, which is
a complex of semi-synthetic derivatives of natural avermectins that have been
effectively used in medicine and agriculture for many years as antiparasitic
drugs. However, experimental data on the ecotoxicological assessment of
individual avermectins are still very small. In this regard, the purpose of
this study was to develop a mathematical model that would allow for a reliable
prediction of the biotransformation ability of natural and semi-synthetic
avermectins, as well as to identify the structural fragments of molecules that
most affect the manifestation of this property. The basis for the construction
of the model was a structurally heterogeneous set represented by an organic
compound with experimental values of the biotransformation half-life (KmHL).
Using the web platform OCHEM (https://ochem.eu), in which the calculation of
PyDescriptor descriptors and the Random Forest method, Transformer-CNN are
implemented, a satisfactory (R2test = 0.81) Quantitative
Relationship Structure – Activity (QSAR) model was developed. The calculations
showed that the biotransformation of natural avermectins in fish occurs on
average faster than semi-synthetic ones. In addition, structural fragments that
increase and decrease the rate of biotransformation have been identified.
Key words: macrolides, molecular descriptors,
machine learning, QSAR.
Copyright (C) Chemistry Dept., Moscow State University, 2002
|
|
