PPDTS: Predicting potential drug-target interactions based on network similarity.

Area of Science:

• Computational biology
• Bioinformatics
• Drug discovery

Background:

• Identifying drug-target interactions (DTIs) is crucial for drug discovery but remains computationally challenging.
• Existing computational methods often rely on global network similarity, which has limitations.

Purpose of the Study:

• To propose a novel computational method, PPDTS, for predicting DTIs.
• To leverage the local structure of drug-target association networks for improved prediction accuracy.

Main Methods:

• Converted known DTIs into a binary network based on local structure.
• Employed the Resource Allocation algorithm to generate drug-drug and target-target similarity networks.
• Utilized Collaborative Filtering with known topology for similarity scores.
• Developed a linear combination of drug-target and target-drug similarity models for final predictions.

Main Results:

• The proposed PPDTS method significantly outperformed four popular network-based similarity methods.
• Experimental validation on diverse datasets confirmed PPDTS's superior performance.
• Some predicted DTIs were corroborated by entries in UniProt and DrugBank databases.

Conclusions:

• PPDTS offers a more effective approach to predicting DTIs by focusing on local network structures.
• The method shows promise for accelerating the drug discovery pipeline.
• Further validation and application of PPDTS can enhance the identification of potential drug candidates.