Drug repurposing based on drug-drug interaction.

Area of Science:

• Pharmacology
• Computational Biology
• Drug Discovery

Background:

• Traditional drug development is high-risk and lengthy.
• Drug repurposing offers a promising alternative.
• Drug-drug interaction data remains underexploited for repurposing.

Purpose of the Study:

• To explore the utility of drug-drug interaction data for drug repurposing.
• To infer drug similarity based on interaction profiles.
• To predict drug physiological effects and targets.

Main Methods:

• Collected 10,835 drug-drug interactions for 1074 drugs.
• Computed drug similarity scores for 700 drugs.
• Constructed a drug association network and applied Markov Clustering Algorithm.

Main Results:

• Clustered 589 drugs into 98 groups, many linked to specific functions.
• The network effectively inferred drug physiological effects.
• Successfully predicted targets for 317 out of 561 drugs with known targets.

Conclusions:

• Drug-drug interaction data is a valuable resource for drug repurposing.
• The developed network effectively infers drug effects and predicts targets.
• This method complements structure-based approaches and demonstrates repurposing feasibility.