What is it about?
Acute Myeloid Leukemia (AML) is an aggressive blood cancer driven by the rapid, uncontrolled growth of abnormal white blood cells. Current treatments face challenges with resistance and toxicity, making new therapies essential. In this study, we built a modern computational drug discovery pipeline combining machine learning (Random Forest) and 3D molecular modeling to target four key proteins that fuel AML progression (TK, PKC, MCL-1, and HAT). By screening extensive chemical libraries and simulating how candidate molecules bind to these targets, we identified a promising lead compound with strong multi-target inhibitory potential, high predicted stability, and favorable drug-like safety properties.
Featured Image
Photo by National Institute of Allergy and Infectious Diseases on Unsplash
Why is it important?
Traditional drug discovery typically takes years and millions of dollars before identifying viable compounds. By using machine learning models with over 90% predictive accuracy alongside dynamic molecular simulations, our workflow significantly shortens early-stage screening. More importantly, AML often bypasses single-target drugs; identifying a candidate capable of multi-target inhibition offers a more robust therapeutic strategy to counteract drug resistance and improve survival outcomes.
Perspectives
Drug resistance remains one of the steepest hurdles in managing AML effectively. Our motivation was to demonstrate how accessible computational tools and machine learning can be combined to look beyond single-target therapies and pinpoint versatile lead compounds faster. We hope these findings provide a solid foundation for wet-lab researchers to advance this candidate into experimental testing and accelerate targeted leukemia therapeutics.
Sohith Reddy Korlagunta
Chang Gung University
Read the Original
This page is a summary of: Identification of Leukemia Enzyme Inhibitors by Molecular Modeling and Machine Learning Approaches, Current Chemical Biology, May 2026, Bentham Science Publishers,
DOI: 10.2174/0122127968399382250722063606.
You can read the full text:
Contributors
The following have contributed to this page
