Quantitative Structure-activity Relationship-based Neural Network Model in Screening Potential Inhibitors from Brown...

Area of Science:

• Computational chemistry
• Drug discovery
• Oncology

Background:

• Oral squamous cell carcinoma (OSCC) is a prevalent head and neck cancer with a poor prognosis.
• Overexpression of the transcription factor RelA is a key driver in OSCC progression.
• Targeting RelA presents a potential therapeutic strategy for OSCC.

Purpose of the Study:

• To identify novel RelA inhibitors for OSCC treatment.
• To develop a predictive quantitative structure-activity relationship (QSAR) model for RelA inhibitors.
• To evaluate natural compounds from brown algae as potential RelA inhibitors.

Main Methods:

• Development of a QSAR-based artificial neural network (multilayer perceptron) model.
• Screening of 1221 brown algae-derived compounds for RelA inhibitory activity.
• Molecular docking and molecular dynamics (MD) simulations to assess binding affinity and stability.

Main Results:

• The QSAR model achieved 91.37% accuracy in classifying active RelA inhibitors (MCC=0.89).
• 1014 out of 1221 screened compounds were identified as potent RelA inhibitors.
• Phlorethopentafuhalol-A exhibited superior binding affinity (-8.45 kcal/mol) compared to known inhibitors and formed stable interactions with RelA via MD simulation.

Conclusions:

• Phlorethopentafuhalol-A is a potential therapeutic candidate for OSCC treatment.
• The developed QSAR model effectively predicts RelA inhibitory properties.
• Natural products from brown algae represent a promising source for novel anti-cancer drug discovery.