Linear Regression QSAR Models for Polo-Like Kinase-1 Inhibitors

Abstract

A structurally diverse dataset of 530 polo-like kinase-1 (PLK1) inhibitors is compiledfrom the ChEMBL database and studied by means of a conformation-independent quantitativestructure-activity relationship (QSAR) approach. A large number (26,761) of molecular descriptorsare explored with the main intention of capturing the most relevant structural characteristics affectingthe bioactivity. The structural descriptors are derived with different freeware, such as PaDEL,Mold2, and QuBiLs-MAS; such descriptor software complements each other and improves the QSARresults. The best multivariable linear regression models are found with the replacement methodvariable subset selection technique. The balanced subsets method partitions the dataset into training,validation, and test sets. It is found that the proposed linear QSAR model improves previouslyreported models by leading to a simpler alternative structure-activity relationship.