Computational Design of Drugs for Epilepsy using a Novel Guided Evolutionary Algorithm for Enhanced Blood Brain...

Area of Science:

• Computational chemistry and drug discovery
• Neuroscience and central nervous system (CNS) disorders

Background:

• Epilepsy is a common CNS disorder, but drug development is hindered by the Blood-Brain-Barrier (BBB).
• Designing small-molecule drugs that effectively cross the BBB is crucial for treating CNS conditions like epilepsy, Alzheimer's, and depression.

Purpose of the Study:

• To develop an efficient computational method for designing CNS-active small molecules with high Blood-Brain-Barrier permeability.
• To guide drug design for conditions including epilepsy, addiction, Alzheimer's disease, and tuberculous meningitis.

Main Methods:

• Utilized supervised learning to model Blood-Brain-Barrier (BBB) permeability of small organic molecules.
• Employed a guided evolutionary algorithm, informed by BBB permeability data, to design novel inhibitors.

Main Results:

• Ligands designed via guided evolution demonstrated predicted higher binding affinity and BBB permeability compared to standard evolutionary algorithms.
• Generated drug-like ligands predicted to bind GABA-T with high affinity, possess BBB permeability, and be chemically synthesizable.

Conclusions:

• Guided evolution offers an efficient computational strategy for designing novel CNS drugs.
• The designed ligands show potential as new therapeutic agents for epilepsy, warranting experimental validation.