Computational drug repositioning using low-rank matrix approximation and randomized algorithms.

Area of Science:

• Bioinformatics
• Computational Biology
• Drug Discovery

Background:

• Computational drug repositioning is crucial for identifying novel disease treatments.
• Large-scale biological datasets enable advanced computational drug repositioning methods.
• Drug repositioning can be framed as a recommendation system problem using matrix completion.

Purpose of the Study:

• To propose a drug repositioning recommendation system (DRRS) for predicting novel drug indications.
• To integrate diverse data sources and validated drug-disease information.
• To enhance the efficiency and accuracy of drug discovery.

Main Methods:

• Constructing a heterogeneous drug-disease interaction network by integrating drug-drug, disease-disease, and drug-disease networks.
• Representing the network as a large drug-disease adjacency matrix.
• Applying a fast Singular Value Thresholding (SVT) algorithm for matrix completion to predict unknown drug-disease pairs.

Main Results:

• The proposed DRRS improves prediction accuracy compared to existing state-of-the-art methods.
• Experimental results demonstrate the effectiveness of the SVT algorithm in matrix completion.
• Case studies confirm the practical utility of DRRS in identifying potential new drug uses.

Conclusions:

• DRRS offers an efficient and accurate computational approach for drug repositioning.
• The integration of heterogeneous data and matrix completion is a powerful strategy.
• The method holds significant promise for accelerating drug discovery and development.