Drug target prediction and repositioning using an integrated network-based approach.

Area of Science:

• Bioinformatics
• Computational Biology
• Drug Discovery

Background:

• Discovering novel drug targets remains a significant hurdle in drug development, with a vast number of genes yet to be utilized.
• Existing drug targets are often pleiotropic, involved in multiple diseases, necessitating reliable methods for drug target repositioning.

Purpose of the Study:

• To develop and assess a network-based approach for predicting novel drug targets and repositioning existing ones for specific diseases.
• To evaluate the relevance of network-based methods for predicting drug targets beyond their current applications.

Main Methods:

• Utilized a network-based approach integrating disease gene expression signatures and high-quality interaction networks.
• Developed an algorithm that considers global network topology to identify disease-associated candidates.
• The method predicts both novel, unexploited drug targets and existing targets for new indications.

Main Results:

• Demonstrated high performance in predicting drug targets for specific diseases.
• Successfully identified novel drug targets for scleroderma and various cancers, elucidating their biological underpinnings.
• Showcased the method's capability in identifying unexpected drug target repositioning candidates, exemplified by type 1 diabetes.

Conclusions:

• The presented network-based method effectively predicts valuable drug targets for new and existing diseases.
• This approach facilitates the discovery of first-in-class drugs and optimizes the use of existing drug targets.
• The findings support the application of network analysis in advancing drug discovery and development pipelines.