Supervised Screening of EGFR Inhibitors Validated through Computational Structural Biology Approaches.

Area of Science:

• Oncology
• Pharmacology
• Computational Biology

Background:

• Human epidermal growth factor receptor (EGFR) overexpression is a key driver of breast cancer (BC) proliferation.
• Targeting EGFR presents a significant therapeutic opportunity for breast cancer treatment.

Purpose of the Study:

• To identify novel therapeutic compounds for breast cancer by targeting EGFR.
• To leverage multiomics data and machine learning for drug discovery.

Main Methods:

• Integrated multiomics data from CCLE, GDSC, and scRNA-seq to identify EGFR as a target and afatinib as a reference drug.
• Employed machine learning models (SVM, ANN, RF) trained on ChEMBL data for drug-like compound screening.
• Conducted computational structural biology assessments to evaluate molecular interactions and dynamics of candidate compounds.

Main Results:

• Afatinib demonstrated superior IC50 values against BC cell lines.
• Machine learning models successfully classified compounds with similar activity to afatinib.
• Four compounds (ChEMBL233324, ChEMBL233325, ChEMBL234580, ChEMBL372692) showed potent EGFR repression.

Conclusions:

• Identified four novel compounds with significant potential as breast cancer therapeutics.
• These compounds exhibit potent repressive action against EGFR, offering new avenues for BC treatment.
• The study highlights the efficacy of integrating multiomics and machine learning in drug discovery for breast cancer.