Author: Y. Vaishnav, S. Yadav, S. K. Jain, J. P. Panda, A. Mishra, M. Garhewal, R. Pathak, and S. Verma

DOI: http://doi.org/10.31788/RJC.2026.1929617

This study presents a 2D-QSAR analysis of 33 substituted pyrazole-pyrimidine derivatives to identifykeyphysicochemical and structural features influencing their biological activity. Molecular descriptors were generatedusing VLifeMDS (v4.6), with compounds optimised via the MMFF force field. The dataset was divided into training, test, and validation sets employing manual, random, and spherical exclusion techniques to ensure robust model development. The optimal QSAR model, with a constant of 4.2686, incorporated four significant descriptors: HydrogensCount (positive correlation), T_O_O_2, SAMostHydrophobicHydrophilicDistance, and DipoleMoment (negative correlation). This model demonstrated strong statistical validity, with an r² of 0.7562, q² of 0.6095, andapredicted r² of 0.5584, indicating excellent internal and external predictivity. By highlighting the influential physicochemical properties, the model can guide the rational design of more potent derivatives. Notably, amongtheevaluated compounds, XIIE showed the highest predicted activity, underscoring its potential as a lead candidate for anticancer drug development. Overall, the model provides valuable insights into the structure-activity relationship, facilitating the development of novel anticancer agents with improved efficacy.

Keywords: QSAR, Molecular Descriptors, Pyrimidine, Cancer, Multiple Linear Regression.