COVID-19 Drug Repurposing: A Network-Based Framework for Exploring Biomedical Literature and Clinical Trials for...

Area of Science:

• Computational biology
• Pharmacology
• Infectious diseases

Background:

• The emergence of SARS-CoV-2 variants necessitates continuous development of vaccines and antiviral treatments.
• Repurposing FDA-approved drugs is a critical strategy for identifying effective COVID-19 therapies.
• Existing research highlights the need for evidence-based drug combinations for Coronavirus treatment.

Purpose of the Study:

• To investigate the hypothesis that FDA-approved drug combinations can serve as candidate treatments for COVID-19.
• To develop and validate a computational framework for identifying such drug combinations.
• To bridge the gap between biomedical literature evidence and clinical trial validation for drug repurposing.

Main Methods:

• A text-mining module was employed to extract drug names from biomedical publications and clinical trial records.
• A network model was constructed to represent associations between extracted drug names.
• A clique similarity algorithm was utilized to identify potential drug combination treatments.

Main Results:

• The computational framework successfully identified potential therapeutic strategies involving two, three, and four drug combinations.
• Specific examples of identified combinations include hydroxychloroquine, doxycycline, and azithromycin.
• The results demonstrate the framework's capability in uncovering evidence-supported drug combinations.

Conclusions:

• The developed framework provides a reliable method for identifying promising drug combinations for COVID-19 treatment.
• The identified drug combinations offer potential therapeutic avenues for managing Coronavirus infections.
• The study validates the hypothesis that a combination of FDA-approved drugs can be a viable candidate for COVID-19 therapy.