Quantitative structure–activity relationship modeling for predication of inhibition potencies of imatinib derivatives using SMILES attributes

Abstract

Chronic myelogenous leukemia (CML) which is resulted from the BCR-ABL tyrosine kinase (TK) chimeric oncoprotein, is a malignant clonal disorder of hematopoietic stem cells. Imatinib is used as an inhibitor of BCR-ABL TK in the treatment of CML patients. The main object of the present manuscript is focused on constructing quantitative activity relationships (QSARs) models for the prediction of inhibition potencies of a large series of imatinib derivatives against BCR-ABL TK. Herren, the inbuilt Monte Carlo algorithm of CORAL software is employed to develop QSAR models. The SMILES notations of chemical structures are used to compute the descriptor of correlation weights (CWs). QSAR models are established using the balance of correlation method with the index of ideality of correlation (IIC). The data set of 306 molecules is randomly divided into three splits. In QSAR modeling, the numerical value of R2, Q2, and IIC for the validation set of splits 1 to 3 are in the range of 0.7180–0.7755, 0.6891–0.7561, and 0.4431–0.8611 respectively. The numerical result of $${CR}_{p}^{2}$$ CR p 2 > 0.5 for all three constructed models in the Y-randomization test validate the reliability of established models. The promoters of increase/decrease for pIC50 are recognized and used for the mechanistic interpretation of structural attributes.

Models for the No-Observed-Effect Concentration (NOEC) and Maximal Half-Effective Concentration (EC50)

“…These models are based on the correlation weights of molecular features used to calculate the 2D descriptor in the CORAL software (http://www.insilico.eu/coral/ accessed on 11 June 2024). The Monte Carlo method has been used for many other models [20][21][22][23][24][25][26][27].…”

QSAR and molecular docking studies of isatin and indole derivatives as SARS 3CLpro inhibitors

“…Molecular docking is a computational technique for predicting the optimal interaction of two molecules that creates a binding model, typically a small ligand with a protein receptor [18], most commonly used in drug discovery [19]. CORAL is a new software for developing the reliable and predictive QSAR/QSPR models based on SMILES or quasi-SMILES of materials and Monte Carlo optimization [17,20].…”

Genetic descriptor search algorithm for predicting hydrogen adsorption free energy of 2D material

“…Various deep learning-based approaches have been used to predict chemical properties to address computational challenge. Advances in deep learning algorithms and methods for representing the structure of chemical molecules, such as the simplified molecular-input line-entry system (SMILES) 16 – 18 and molecular graph 19 – 21 , have led to significant performance improvements in chemical property prediction. However, deep neural network-based approaches require extensive training datasets to avoid overfitting, and they may not generalize well without sufficient training samples.…”