Computational Drug Repositioning with Random Walk on a Heterogeneous Network

Area of Science:

• Computational biology
• Pharmacology
• Bioinformatics

Background:

• Drug repositioning accelerates drug discovery by identifying new indications for existing medications.
• Advances in high-throughput technologies generate vast biomedical data, enabling computational approaches for drug repositioning.
• Integrating multi-source data is crucial but challenging for improving drug repositioning accuracy.

Purpose of the Study:

• To propose an efficient computational approach for prioritizing candidate drugs for diseases.
• To enhance drug repositioning accuracy by leveraging heterogeneous biomedical data.

Main Methods:

• Constructed an integrated heterogeneous network incorporating drugs, drug targets, diseases, and disease genes.
• Developed a random walk model to capture global network information.
• Utilized drug targets and disease genes data comprehensively within the network.

Main Results:

• The proposed Random Walk on a Heterogeneous Network for Drug Repositioning (RWHNDR) approach demonstrated superior performance.
• RWHNDR achieved better accuracy in prioritizing candidate drugs compared to state-of-the-art methods.
• The approach effectively exploits multi-source data for improved drug repositioning outcomes.

Conclusions:

• RWHNDR offers an efficient and effective method for computational drug repositioning.
• The approach enhances the identification of novel drug indications by integrating diverse biological data.
• This strategy holds promise for improving the productivity of drug discovery and development pipelines.