Ontology integration to identify protein complex in protein interaction networks

Area of Science:

• Bioinformatics
• Computational Biology
• Systems Biology

Background:

• Protein complex identification is crucial for understanding cellular functions.
• Protein interaction network analysis faces challenges due to unreliable interactions and network complexity.
• Integrating diverse data sources can improve protein complex detection algorithms.

Purpose of the Study:

• To develop a novel semantic similarity method for measuring protein-protein interaction reliability.
• To present a new clustering algorithm (OIIP) for identifying protein complexes using weighted protein-protein interaction networks.

Main Methods:

• Utilized Gene Ontology (GO) annotations to calculate semantic similarity and assess protein-protein interaction reliability.
• Developed a weighted graph representation of protein interaction networks based on interaction reliability.
• Adapted a seeded vertex expansion clustering approach (IPCA) to create the OIIP algorithm.

Main Results:

• The OIIP algorithm was applied to the Saccharomyces cerevisiae protein interaction network.
• OIIP successfully identified numerous well-characterized protein complexes.
• Experimental results demonstrated superior performance of OIIP, indicated by higher F-measure and accuracy compared to existing methods.

Conclusions:

• The developed semantic similarity method effectively measures protein-protein interaction reliability.
• The OIIP algorithm provides an effective approach for identifying protein complexes from weighted protein interaction networks.
• OIIP offers improved accuracy and reliability in protein complex detection, outperforming competing algorithms.