Search academic texts

Information × Registration Number 0221U105235, 0117U004134 , R & D reports Title Synthesis, structure, properties and molecular mechanisms of action of new antithrombotic and antiviral agents. popup.stage_title Head Andronati Sergii A., д.х.н. Registration Date 15-06-2021 Organization Physico-Chemical Institute O. Bogatsky National Academy of Sciences of Ukraine popup.description2  An effective method for the synthesis of 6-nitro-2-(4-Boc-piperazin-1-yl)-3H-quinazolin-4-one has been developed. The use of pre-catalyst XantPhos-Pd-G3 leads to the target product with a higher yield, compared with the use of the catalytic system Pd2(dba)3.ССl3-Xantphos. Thus, instead of a five-step scheme for the synthesis of 6-amino-2-(piperazin-1-yl)quinazolin-4-one, two stages were used. Also synthesized: 6-amino-4-phenyl-2-(piperazin-1-yl)quinazoline, -(fluorophenyl)--alanine, -(trifluoromethyl)--alanine and -(fluorophenyl)-γ-aminobutyric acids. Potential DNA intercalators that have an affinity for JNK family protein kinases and have the ability to inhibit the synthesis of proinflammatory cytokines (IL-6 and TNF) in vitro and inhibit the development of inflammation in the carrageenan edema of the paw have been identified. Information was collected and 167 DNA intercalators were investigated. Each molecule contained a polycyclic planar moiety linked to a major amino group to ensure optimal interaction with the nucleic acid. 161 compounds with binding constants lg (Ki) ≥4 were identified, which allowed the use of these compounds to develop a pharmacophore model. As structural filters were used: the number of condensed rings (2-5), H-bond donors (0-3), H-bond acceptors (2-6), the number of rotating bonds (3-12) and molecular weight (268-443). The two-class classification model was constructed using data on the antiviral activity of 117 compounds. SiRMS and Random Forest were used as descriptors of the machine learning method. The data set was randomly divided into training and test samples with a ratio of approximately 4:1. The balanced prediction accuracy for the test sample is 0.78. All developed models are used for screening a data set of 3207605 compounds, which contains databases BioinfoDB (3207317 compounds) and PCIdb (288 virtual compounds). According to the results of stepwise screening, 55 compounds were selected for which the predicted solubility Product Description popup.authors Illyushko Natalia О. Kabanova Tetyana A. Karpenko Oleksandr S. Kornylov Oleksandr Yu. Krysko Andrei А. Krysko Olga L. Kuz'min Viktor Е. Lyakhov Sergii A. Mikhailova Tetyana V. Pavlovs'kyj Viktor I. Shibinskaya Marina O. popup.nrat_date 2021-06-15 Close