Machine Learning Prediction and Simulation of Drugs Targeting GSK-3β in Breast Cancer

Introduction and Objective: Breast cancer ranks as the second-most prevalent cause of death among women worldwide, with particularly elevated mortality rates in India. Breast cancer’s origin involves biochemical pathway alterations influenced by tumor-inducing proteins. Research has highlighted glycogen synthase kinase-3 beta (GSK-3β) as a crucial protein that regulates the expression of various genes in the cell cycle. Mutations in this protein have a significant impact on cellular development. As a consequence, it triggers aggressive subtypes of breast cancer, such as triple-negative breast cancer. So, the primary aim of this study is to identify novel chemicals targeting GSK-3β using machine learning methods, molecular modeling, and dynamic techniques. Materials and Methods: To achieve the study's objective, small molecules were screened using a Machine Learning (ML) approach, and subsequently, molecular docking and dynamic modelling investigations were conducted to explore interactions between drugs and GSK-3β. Results: The research findings highlighted a specific compound, piperidine, 4-(3,4- dichlorophenyl)-4-[4-(1H-pyrazol-4-yl) phenyl], which exhibited a superior docking score of -9.6 kcal/mol. Piperidine also formed conventional hydrogen bonds with the target protein. Furthermore, the calculated binding free energy of -12.46 kcal/mol suggested that this compound exhibited greater stability compared to commercially available drugs. Conclusion: These promising findings highlight the potential of piperidine and similar small molecules as promising candidates for targeting the tumor-inducing protein GSK-3β. Subsequent investigations, both in vitro and in vivo, will be essential to assess their effectiveness in combating breast cancer.