A network-based drug repurposing method via non-negative matrix factorization.

Area of Science:

• Computational biology
• Pharmacology
• Bioinformatics

Background:

• Drug repurposing offers an efficient alternative to traditional drug discovery.
• Formulating drug repurposing as a recommender system leverages known drug-disease associations to identify novel indications.

Purpose of the Study:

• To present a novel method, NMF-DR, for predicting candidate disease indications for existing drugs.
• To develop a recommender system approach for drug repurposing by integrating diverse data sources.

Main Methods:

• Constructed a heterogeneous drug-disease interaction network by integrating disease similarities, drug-disease associations, and drug similarities.
• Employed an improved non-negative matrix factorization (NMF-DR) method to predict scores for unknown drug-disease pairs, completing the drug-disease adjacency matrix.

Main Results:

• NMF-DR demonstrated superior prediction performance in identifying drug-disease associations compared to existing methods.
• Experimental results validate the effectiveness of the proposed NMF-DR approach.

Conclusions:

• The NMF-DR method provides a powerful tool for drug repurposing and predicting novel drug indications.
• The developed framework enhances the efficiency of discovering new therapeutic uses for existing drugs.