Predictive QSAR model and clustering analysis of some Benzothiazole derivatives as cytotoxic inhibitors - Issue 3 - research

Sep 28, 2026

Upload by Noah

The study presents an original QSAR modeling and clustering workflow using a dataset of 23 Benzothiazole derivatives evaluated as cytotoxic inhibitors. Descriptor selection is performed via exhaustive search to identify an optimal molecular-descriptor subset. K-means clustering yields three distinct, well-separated clusters, supported by silhouette validation with no misassigned molecules. Molecular descriptors from a physico-chemical class are shown to strongly influence cytotoxic activity, with triﬂuoromethyl-containing molecules exhibiting the strongest effects. Best activity is observed for compounds bearing two triﬂuoromethyl groups in ortho positions, suggested to relate to binding affinity. PLS regression equations show good agreement between fitted and observed activities, with 67% and 80% of variability explained, significant F-statistics, and robust predictive power indicated by R2boots and Q2Loo from bootstrapping and leave-one-out cross-validation.

LanguageEnglish

CategoryGeneral

FormatPDF

Pages10

Reading Time15 min

AccessFree to download

Abstract

Introduction

Materials and methods

Biological data set and molecular optimization

SIFT DESK SDRP Journal of Computational Chemistry & Molecular Modelling (ISSN: 2473-6260) Predictive QSAR model and clustering analysis of some Benzothiazole derivatives as cytotoxic inhibitors DOI: 10.25177/JCCMM.2.3.3 Research Received Date: 21st Jun 2018 Accepted Date: 18th Jul 2018 Published Date:25th Jul 2018 Copy rights: © This is an Open access article distributed under the terms of International License. Samir Kenouche 1*, Dalal Harkati2, Myriem Ghamri2, A. Rahime Chikhaoui 1, Nadjib Melkemi2 1Department of Matter Sciences, Faculty of Exact and Natural Sciences University M.Khider ofBiskra, 07000 Biskra, Algeria 2Research team computatinal and pharmaceutical chemistry LMCE laboratory, University M.Khider of Biskra 07000 Biskra, Algeria. CORRESPONDENCE AUTHOR Samir Kenouche Email: [samir.kenouche@univ-biskra.dz](samir.kenouche@univ-biskra.dz) CONFLICTS OF INTEREST There are no conflicts of interest for any of the authors. CITATION Kenouche Samir, Predictive QSAR model and clustering analysis of some Benzothiazole derivatives as cytotoxic inhibitors(2018)SDRP Journal of Computational Chemistry & Molecular Modelling 2(3) Abstract We propose an original approach dedicated to QSAR modeling and clustering analysis based on a dataset of 23 Benzothiazole derivatives as cytotoxic inhibitors. The choice of relevant molecular descriptors is a key step in QSAR modeling. In this work, model selection by exhaustive search is used to identify the best subset of molecular descriptors. Three distinct clusters have been identiﬁed using Kmeans clustering. Each cluster, groups a homogeneous class of molecules with respect their molecular descriptors. Silhouette analysis, used as cluster validation approach, proves that the molecules are very well clustered and there are no molecules placed in the wrong cluster. Moreover, the results emphasize that the molecular descriptors belonging to physico-chemical class appears to largely inﬂuence the cytotoxic activity of Benzothiazole derivatives. From this classiﬁcation, all molecules with the triﬂuoromethyl group show a strong activity. The best cytotoxic activity was exhibited by compound containing two triﬂuoromethyl groups in ortho position. We suggested that this functional group is correlated to binding afﬁnity. The PLS equations exhibit a good agreement between ﬁtted and observed cytotoxic activities. According to the goodness of ﬁt statistics, 67% and 80% of the variability in cytotoxic activity around its mean are explained. The Fstatistic test revealed the signiﬁcance of both PLS regression equations. The values of bootstrapping correlation coefﬁcient R2boots and leave-one-out cross-validation coefﬁcient Q2Loo revealing that the resulting models have good predictive power and robustness. We concluded that the developed PLS equations can be successfully applied to predict the antiproliferative activity against breast cancer cells lines of Benzothiazole derivatives. Keywords:Benzothiazole derivatives, anticancer activity, principal component analysis, K-means clustering, PLS regression. 1 Introduction After cardiovascular diseases, cancer is the second leading cause of death worldwide. There is a growing body of literature that recognises the importance of cancer prevention from threatening our lives. It is now well established from a variety of studies, that protein tyrosine kinases play a critical role in the development and progression of many kinds of cancer [1] . There are a lot of heterocyclic compounds which have provide enormous potential in the discovery of new chemotherapeutics in drug discovery and development programs,[2, 4] such as benzothiazole type compounds belong to the family of heterocyclic compounds having benzene ring incorporated with ﬁve-membered ring containing nitrogen and sulfur atoms. Because of their biological and pharmacological properties, benzothiazoles were widely used as antitubercular [5, 7], antimicrobial [8

FAQs

What dataset and modeling strategy were used in the study?

The study uses a dataset of 23 Benzothiazole derivatives and applies QSAR modeling with molecular descriptor selection via exhaustive search. It then combines clustering and regression approaches to analyze cytotoxic activity.

How were clusters determined and validated?

Three clusters were identified using K-means clustering. Silhouette analysis was used for cluster validation, showing that molecules are well clustered with no molecules placed in the wrong cluster.

Which molecular feature was linked to stronger cytotoxic activity?

All molecules with the triﬂuoromethyl group showed strong activity. The strongest cytotoxic activity was for compounds with two triﬂuoromethyl groups in ortho positions, suggested to correlate with binding affinity.

Sep 28, 2026

Upload by Noah