Machine Learning-Assisted Drug Repurposing Framework for Discovery of Aurora Kinase B Inhibitors.

Area of Science:

• Biochemistry
• Computational Chemistry
• Pharmacology

Background:

• Aurora kinase B (AurB) is a key regulator of cell division (mitosis).
• Targeting AurB is a promising strategy for cancer treatment.
• There is a need for effective AurB inhibitors in clinical use.

Purpose of the Study:

• To identify potential Aurora kinase B (AurB) inhibitors using a drug repurposing approach.
• To develop and validate a machine learning framework for drug discovery.

Main Methods:

• Utilized a machine learning pipeline combining QSAR modeling, molecular fingerprints, molecular docking, and molecular dynamics (MD) simulations.
• Screened 4680 drugs from the DrugBank database for potential AurB inhibitory activity.

Main Results:

• Identified saredutant, montelukast, and canertinib as potential AurB inhibitors.
• Saredutant, an NK2 antagonist, showed strong binding affinity and stable interactions with AurB.
• Computational analyses confirmed favorable binding energies and molecular dynamics for the identified candidates.

Conclusions:

• The study successfully identified novel drug candidates for AurB inhibition through computational drug repurposing.
• The integrated machine learning methodology is effective for discovering inhibitors against challenging drug targets.
• Saredutant is highlighted as a promising lead compound for further investigation as an AurB inhibitor.