Identification of new potential candidates to inhibit EGF via machine learning algorithm

Area of Science:

• Pharmacology
• Computational Chemistry
• Drug Discovery

Background:

• Drug repositioning offers a cost-effective strategy for identifying novel drug inhibitors.
• Computational methods accelerate drug discovery by bypassing pre-clinical stages.

Purpose of the Study:

• To develop and validate an ensemble computational-experimental approach for drug repositioning.
• To identify existing drugs that can inhibit Epidermal Growth Factor (EGF).

Main Methods:

• A machine learning model (differentiated tree classification) was employed to screen potential drug candidates.
• Molecular docking simulations were performed on selected compounds.
• Experimental validation using MTT assay and In-Cell ELISA was conducted on A431 cells.

Main Results:

• Nine compounds were selected based on machine learning predictions, with seven previously linked to EGF inhibition.
• Salicylic acid and piperazine were identified as potential EGF inhibitors.
• The study confirmed the inhibitory effects of salicylic acid and piperazine on A431 cell growth and EGF signaling.

Conclusions:

• The ensemble computational-experimental approach is effective for drug repositioning.
• Salicylic acid and piperazine show promise as novel EGF inhibitors, warranting further investigation.