ABSTRACT
QSAR studies of a set of previously synthesized azole derivatives tested against human cytomegalovirus (HCMV) were performed using the OCHEM web platform. The predictive ability of the classification models has a balanced accuracy (BA) of 73–79%. The validation of the models using an external test set proved that the models can be used to predict the activity of newly designed compounds with a reasonable accuracy within the applicability domain (BA = 76–83%). The models were applied to screen a virtual chemical library with expected activity of compounds against HCMV. The five most promising new compounds were identified, synthesized and their antiviral activities against HCMV were evaluated in vitro. Two of them showed some activity against the HCMV strain AD169. According to the results of docking analysis, the most promising biotarget associated with HCMV is DNA polymerase. The docking of the most active compounds 1 and 5 in the DNA polymerase active site shows calculated binding energies of -8.6 and -7.8 kcal/mol, respectively. The ligand’s complexation was stabilized by the formation of hydrogen bonds and hydrophobic interactions with amino acids Lys60, Leu43, Ile49, Pro77, Asp134, Ile135, Val136, Thr62 and Arg137.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
All data used to support the results of this study are included in the article, and additional information is provided in the supplementary section. In addition, the data used in this work and developed models are freely available online at OCHEM http://ochem.eu/article/156546.
Supplementary material
Supplemental data for this article can be accessed at: https://doi.org/10.1080/1062936X.2023.2232992