A new method could speed up the process of identifying effective treatments for hard-to-treat neurological conditions. The goal is to predict which drugs are most likely to cross the blood-brain barrier to treat disorders like brain tumors, Alzheimer’s and Parkinson’s disease, and epilepsy. It's been created by researchers at Carle Illinois College of Medicine (CI MED) and Emory University.
The blood-brain barrier is a network of tightly packed cells that protects the brain from infection and other threats by filtering harmful substances before they can reach brain tissue. But this barrier can also limit the effectiveness of drugs intended to combat disorders of the central nervous system. A cross-disciplinary team of researchers from CI MED and the Emory University Department of Neurosurgery is integrating computational chemistry, machine learning, and experimental validation to improve predictions of blood-brain barrier permeability.
“By using computational models to predict which drugs can cross the blood-brain barrier, the study could reduce the time and cost of early drug development. Ultimately, this approach may accelerate the discovery of safe and effective brain-targeted therapies, improving care for people affected by conditions of the central nervous system (CNS),” said CI MED student Megan Lim, first author of the study.
This study introduces a new AI approach that learns from detailed chemical data. The model is first trained using basic molecular properties calculated through physics-based methods, giving it a deeper understanding of how chemicals behave. This foundation helps the AI more accurately predict which compounds might cross the blood–brain barrier, as a first step in identifying potential new drug therapies.
“To our knowledge, no prior work has combined quantum-chemical (QC) properties with transfer-learning approaches specifically for blood-brain barrier permeability prediction in the way we have,” Lim said. “This research could help bring new treatments to patients with neurological diseases more quickly and efficiently.”
Streamlining new drug discovery could be life-changing for patients with difficult-to-treat neurological conditions.
“In neuro-oncology, this research applies to brain tumors, where it can help optimize the delivery of chemotherapeutic or targeted agents across the blood-brain barrier. For Alzheimer’s disease and Parkinson’s disease, it can improve penetration of neuroprotective or disease-modifying compounds into the central nervous system. In epilepsy, it can support the design of antiepileptic drugs that require CNS penetration,” Lim said.
Editor’s note:
Lim’s collaborators include Dr. Wael Mostafa of CI MED and Carle Health; Marybeth Yonk, Kimberly Hoang, Monika Raj, Yuhong Du, Nicholas Boulis, and Kecheng Lei of Emory University’s Department of Neurosurgery; and Annette Molinaro of the University of California, San Francisco Department of Neurosurgery. The team’s research is published in the journal Drug Delivery and Translational Research. The article is available here: https://doi.org/10.1007/s13346-025-02005-5