Predicting Potential SARS-COV-2 Drugs-In Depth Drug Database Screening Using Deep Neural Network Framework SSnet,...

Area of Science:

• Computational chemistry
• Machine learning in drug discovery
• Virology

Background:

• The COVID-19 pandemic necessitated rapid identification of pharmaceutical treatments.
• Existing drug repurposing and de novo drug discovery are crucial for urgent health crises.
• Global cooperation and open-access research models accelerate scientific progress.

Purpose of the Study:

• To introduce and validate a deep neural network-based drug screening method (SSnet).
• To identify existing drugs for COVID-19 repurposing and discover novel therapeutic compounds.
• To create an open-access platform for researchers to access drug screening results.

Main Methods:

• Application of a deep neural network (SSnet) for large-scale drug screening.
• Validation using a molecular docking algorithm on approved drugs.
• Screening of a library comprising 750,000 compounds for de novo drug discovery.
• Development of an open-access web interface for result dissemination.

Main Results:

• SSnet successfully screened a large compound library, identifying potential drug candidates.
• The method validated drug repurposing opportunities for existing pharmaceuticals.
• De novo drug discovery efforts yielded potential ACE2-regulatory compounds.
• An open-access web interface was created to share screening results.

Conclusions:

• SSnet is a powerful machine learning tool for efficient drug discovery and repurposing.
• The combined approach accelerates the identification of molecules with potential therapeutic efficacy against COVID-19.
• Open-access data sharing is vital for rapid response to global health emergencies.