2D-QSAR and 3D-QSAR Analyses for EGFR Inhibitors

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Pharmacology

Background:

• Epidermal growth factor receptor (EGFR) is a key target in cancer therapy.
• Developing effective EGFR inhibitors is crucial for cancer treatment.

Purpose of the Study:

• To build robust two-dimensional (2D-QSAR) and three-dimensional (3D-QSAR) quantitative structure-activity relationship models for EGFR inhibitors.
• To predict the inhibitory activity and efficacy of potential EGFR-targeting compounds.

Main Methods:

• Utilized support vector machine (SVM) classifier with feature selection for 2D-QSAR model development.
• Constructed a 3D-QSAR model to predict EGFR inhibitor activity.
• Employed molecular docking to analyze inhibitor-EGFR interactions.

Main Results:

• The 2D-QSAR model demonstrated high predictive accuracy (98.99% tenfold cross-validation, 97.67% independent set).
• The 3D-QSAR model achieved a cross-validated correlation coefficient (q²) of 0.565 and non-cross-validated correlation coefficient (r²) of 0.888.
• Mean absolute errors for the 3D-QSAR model were 0.308 (training) and 0.526 (test) log units.

Conclusions:

• Developed accurate and reliable QSAR models for predicting EGFR inhibitor activity.
• The study provides a valuable computational framework for the design and discovery of novel EGFR inhibitors.
• Molecular docking insights further elucidate the binding mechanisms of these inhibitors with EGFR.