Network Pharmacology and Machine Learning Identify Flavonoids as Potential Senotherapeutics.

Area of Science:

• Gerontology
• Computational Biology
• Pharmacology

Background:

• Cellular senescence, characterized by cell cycle arrest and pro-inflammatory SASP, is linked to aging and chronic diseases.
• Senotherapeutics, including senolytics and senomorphics, offer potential treatments for age-related conditions.
• Computational methods like network pharmacology and machine learning accelerate novel drug discovery.

Purpose of the Study:

• To identify potential senotherapeutic compounds using computational approaches.
• To analyze drug-protein interactions related to cellular senescence.
• To classify compounds for therapeutic potential against senescence.

Main Methods:

• Construction of drug-protein interaction networks for cellular senescence.
• Application of machine learning models (Random Forest, SVM, KNN) for compound classification.
• Evaluation of compounds for therapeutic potential and medicinal chemistry properties.

Main Results:

• Identification of 714 compounds with potential senotherapeutic activity.
• Selection of 270 compounds with favorable medicinal chemistry properties.
• Development of an interactive web tool for community access.

Conclusions:

• Flavonoids were the most abundant class among identified senotherapeutics.
• 18 novel flavonoids were identified with no prior reported senotherapeutic activity.
• The study provides a valuable resource for senotherapeutic drug discovery.