Conformation depends on 4D-QSAR analysis using EC–GA method: pharmacophore identification and bioactivity prediction of TIBOs as non-nucleoside reverse transcriptase inhibitors

Abstract

The electron conformational and genetic algorithm methods (EC-GA) were integrated for the identification of the pharmacophore group and predicting the anti HIV-1 activity of tetrahydroimidazo[4,5,1-jk][1,4]benzodiazepinone (TIBO) derivatives. To reveal the pharmacophore group, each conformation of all compounds was arranged by electron conformational matrices of congruity. Multiple comparisons of these matrices, within given tolerances for high active and low active TIBO derivatives, allow the identification of the pharmacophore group that refers to the electron conformational submatrix of activity. The effects of conformations, internal and external validation were investigated by four different models based on an ensemble of conformers and a single conformer, both with and without a test set. Model 1 using an ensemble of conformers for the training (39 compounds) and test sets (13 compounds), obtained by the optimum seven parameters, gave satisfactory results (R²(training) = 0.878, R²(test)= 0.910, q² = 0.840, q²(ext1) = 0.926 and q²(ext2) = 0.900).

A QSAR Model for Predicting the Anti HIV‐1 Activity of TIBO Derivatives Using the Norm Indexes

“…The statistical results were listed in Table . Comparing with other linear models, the number of data set in the work Akyüz and Sarıpınar was 52, which was smaller than other references listed in Table . The Q 2 of Goodarzi and Freitas's two models were 0.698 and 0.605, respectively, which was lower than our model (0.822).…”

“…In order to evaluate the predictive performance of this model, some other investigations were used for comparison . The statistical results were listed in Table .…”

“…developed a QSAR model for p IC 50 prediction for 82 TIBO derivatives. Akyüz and Sarıpınar developed QSAR models ( R 2 train = 0.878, R 2 test = 0.910) for 52 TIBO derivatives using the electron conformational and genetic algorithm methods (EC–GA). However, most of these previous works are nonlinear models, and the algorithms of these models are not so lucid when based on the CoMSIA, ANN and/or SVM approaches; therefore for these models it might be difficult to be utilized conveniently by other researchers.…”

Two Decades of 4D-QSAR: A Dying Art or Staging a Comeback?

“…The basic operational steps of the combined EC-GA approach are presented in Figure 8 . The ligand 3D structures are constructed and optimized using the semi-empirical (PM3) or quantum chemical calculations (HF/6-311 G ** level in the aqueous medium) [ 82 , 83 , 84 ]. In order to generate the conformational ensemble profile for the investigated series Monte Carlo (MC) randomized search simulations are conducted.…”

The study of dual COX-2/5-LOX inhibitors by using electronic-topological approach based on data on the ligand–receptor interactions

“…Earlier, when forming ETMCs for flexible molecules, it was necessary to carry out complicated calculations. A new approach based on the Genetic Algorithms [36,37] has been applied for the pharmacophoric groups' identification. When ETM and docking are used together, 3D coordinates of a molecule correspond to the most effective position of the molecule within receptor.…”