2D Autocorrelation modeling of the activity of trihalobenzocycloheptapyridine analogues as farnesyl protein transferase inhibitors

Abstract

The inhibitory activity towards farnesyl protein transferase enzyme (FPT) of 49 piperidine substituted trihalobenzocycloheptapyridine analogues (thBCHPs) has been successfully modelled using 2D spatial autocorrelation vectors. Predictive linear and non-linear models were obtained by forward stepwise multilinear regression analysis (MRA) and artificial neural network (ANN) approaches, respectively. A variable selection routine that selected relevant non-linear information from the data set was employed prior to networks training. The MRA model, using three descriptors, was able to explain about 68% data variance. The model showed a linear dependence between the inhibitory activities and autocorrelation coefficients weighted by van der Waals volumes and atomic polarizabilities on the inhibitors molecules. The non-linear approach preserve several characteristics described for the linear one. Three descriptors were selected encoding the same atomic properties, but the new ones were able to explain about 92% data variance. In addition, the ANN model had higher predictive power. Furthermore, inhibitors were well distributed regarding its activity levels in a Kohonen self-organizing map (SOM) built using the input variables of the best neural network.

Bayesian-regularized genetic neural networks applied to the modeling of non-peptide antagonists for the human luteinizing hormone-releasing hormone receptor

“…Despite the obvious stints of information from 2D representations of molecules, 2D autocorrelation descriptors have found successful applications in the performance of QSAR modeling [41,42]. Although more clear information can be obtained from ''traditional'' descriptors, such as log P or pK a , several authors pointed out problems with the definition of a chemistry space using them [43].…”

Genetic neural network modeling of the selective inhibition of the intermediate-conductance Ca2+-activated K+ channel by some triarylmethanes using topological charge indexes descriptors

“…Those evidences, in conjunction with the fact that TCI describe molecular topology and charge transfer through the molecule, support our interest in such descriptors for building highly reliable models for the inhibitory activity of IK Ca blockers. Furthermore, we prefer 2D topological or 3D descriptors over other simpler and more physical descriptors such as octanol/water partition coefficient or molar refractivity because of, in spite of their difficult interpretation, they usually yield outstanding QSAR models when are used in combination with ANNs [15][16][17][18][19][20].…”

“…In this sense, quantitative structure-activity relationships (QSAR) studies have been successfully applied for modeling biological activities of natural and synthetic chemicals [8]. Based on the molecular graph representation of a chemical structure, Graph-Theoretical and Topological methods are included in the most QSAR studies [9][10][11][12][13][14][15][16][17]. In such studies, a set of topological descriptors encoding structural features of the compounds under investigation are computed over the molecular graphs, afterwards multivariated linear or/and non-linear relationships are established between a reduced subset of variables and the biological property.…”

“…Since interactions between a chemical and its biological target are often non-linear, artificial neural network (ANN) methodology rather than multilinear regression analysis (MRA) had been successfully applied in QSAR studies of biological properties in the last decade [15][16][17][18][19][20][21][22][23][24][25]. Besides the non-linearity among biological activities and the computed molecular descriptors, another major problem arises when the number of calculated variables exceeds the number of compounds in the data set, so that one is dealing with an undetermined problem where undesirable overfitting can result [26].…”

Synthesis and Ligand Based 3D-QSAR of 2,3-Bis-benzylidenesuccinaldehyde Derivatives as New Class Potent FPTase Inhibitor, and Prediction of Active Molecules

“…This protein is synthesized in the cell matrix, and transformed gradually, moves into the cell membrane, at this stage, and carry out its original role. 5 Therefore in the stage of structure and activation, it is important to maintain Ras protein in the cell membrane. First stage of Ras protein transforming process is farnesylation reaction by FPTase at cystein side chain of CaaX box.…”