Improving drug repositioning accuracy using non-negative matrix tri-factorization.

Area of Science:

• Computational drug discovery
• Pharmacology
• Bioinformatics

Background:

• Drug repositioning accelerates therapeutic development by repurposing existing drugs.
• Identifying novel drug-disease associations is crucial for efficient drug discovery.

Purpose of the Study:

• To introduce and evaluate the IDDNMTF model for predicting drug repositioning opportunities.
• To enhance the precision of identifying new therapeutic uses for existing drugs.

Main Methods:

• Developed the IDDNMTF model integrating multiple datasets for drug-disease association analysis.
• Evaluated model performance using AUC, AUPR, and F1 scores.
• Compared IDDNMTF performance against the NMF model.

Main Results:

• IDDNMTF model performance improved with increased data integration.
• The IDDNMTF model demonstrated superior predictive accuracy compared to the NMF model.
• Data diversity significantly enhances predictive capabilities in drug repositioning.

Conclusions:

• The IDDNMTF model is a promising tool for identifying novel therapeutic applications of existing drugs.
• The model's accuracy and interpretability can accelerate the translation of research findings into clinical practice.
• This approach supports personalized medicine and targeted treatment development.