Accelerating Lung Cancer Drug Discovery Using QSAR Modeling, Molecular Docking, and ADMET Analysis

Computational tools are revolutionizing the process of drug discovery, particularly in the identification of targeted therapies for cancers such as lung cancer. In this study, we employed Quantitative Structure–Activity Relationship (QSAR) modeling to predict the activity of ۲۰۰ chemical compounds against lung cancer cells, specifically focusing on the epidermal growth factor receptor (EGFR). Utilizing a combination of molecular descriptors and machine learning algorithms (Random Forest, Support Vector Machine, and Gradient Boosting), we identified compounds with high predictive efficacy. The Random Forest model performed best, yielding an R² of ۰.۹۲ and a Q² of ۰.۸۵. Top candidates were further evaluated using molecular docking simulations, with the most promising compound demonstrating a binding affinity of -۱۰.۲ kcal/mol with EGFR. ADMET analysis confirmed favorable pharmacokinetic properties and low toxicity risk. These findings underscore the potential of integrated computational methods to streamline anti-cancer drug discovery.

کلیدواژه ها:

QSAR ، anti-cancer drugs ، lung cancer ، EGFR ، machine learning ، molecular docking ، ADMET