In Silico drug repurposing pipeline using deep learning and structure based approaches in epilepsy.

Area of Science:

• Computational chemistry
• Neuroscience
• Pharmacology

Background:

• Epilepsy affects many globally, with high recurrence rates, necessitating new treatments.
• Drug repurposing offers a cost-effective strategy for developing novel epilepsy medications.

Purpose of the Study:

• To develop an in-silico drug repurposing pipeline for epilepsy using deep learning and molecular binding affinity.
• To identify potential drug candidates targeting proteins involved in epileptogenesis.

Main Methods:

• Utilized transformer-based deep learning models for drug repurposing.
• Integrated molecular binding affinity calculations for candidate screening.
• Employed molecular dynamics simulations and mechanism analysis for validation.

Main Results:

• Identified numerous candidate inhibitors for 24 epilepsy-associated target proteins.
• Repurposed existing anti-epileptic and psychiatric medications.
• Lomitapide demonstrated high binding affinity for 10 targets, validated through simulations.

Conclusions:

• The developed pipeline effectively identifies potential epilepsy drug repurposing candidates.
• Lomitapide is a promising candidate for epilepsy treatment, warranting further investigation.
• Findings offer new therapeutic insights for central nervous system disorders.