Drug repurposing for obsessive-compulsive disorder using deep learning-based binding affinity prediction models.

Area of Science:

• Computational chemistry and neuroscience
• Application of artificial intelligence in drug discovery

Background:

• Obsessive-compulsive disorder (OCD) is a chronic psychiatric condition characterized by obsessions and compulsive rituals.
• Identifying novel therapeutic targets and drug candidates for OCD remains a significant challenge.

Purpose of the Study:

• To develop and validate deep learning models for predicting molecular interactions with key OCD-related biological targets: SERT, D2, and NMDA.
• To assess the utility of an ensemble deep learning model for screening large molecule databases for potential OCD therapeutics.
• To explore drug repurposing opportunities for OCD using computational methods.

Main Methods:

• Development of three distinct deep learning models to predict molecule binding affinity to SERT, D2, and NMDA targets.
• Creation of an ensemble model by combining the predictions of the individual models.
• External validation of the ensemble model on a large drug database using random sampling.
• Bibliographic validation of high-scoring molecules through case studies to assess relevance to OCD pathophysiology.

Main Results:

• The deep learning models demonstrated predictive capabilities for molecular interactions with SERT, D2, and NMDA.
• The ensemble model achieved robust performance during external validation.
• Bibliographic analysis of top-scoring molecules supported their potential relevance to OCD.
• The study successfully identified molecules with high scores indicating a connection with OCD pathophysiology.

Conclusions:

• Deep learning-based ensemble modeling is a viable strategy for screening molecule databases for potential OCD drug repurposing.
• The developed models can aid in identifying novel therapeutic candidates by predicting interactions with key OCD-related targets.
• This computational approach offers a promising avenue for accelerating drug discovery in psychiatric disorders like OCD.