Identifying protein complexes based on the integration of PPI network and gene expression data

Area of Science:

• Computational Biology
• Systems Biology
• Bioinformatics

Background:

• Protein complexes are fundamental to cellular organization and function.
• Understanding protein complex composition is key to deciphering cellular processes.
• Predicting protein functions often relies on identifying their complex partners.

Purpose of the Study:

• To propose a novel algorithm, IPCIPG, for discovering protein complexes.
• To integrate Protein-Protein Interaction (PPI) network data with gene expression data for enhanced complex identification.
• To develop both non-overlapping (IPCIPG-n) and overlapping (IPCIPG-o) cluster detection versions.

Main Methods:

• Developed the IPCIPG algorithm, a local search approach.
• Integrated PPI network topology with gene expression profiles.
• Evaluated IPCIPG on the yeast PPI network against six established algorithms.

Main Results:

• IPCIPG demonstrated superior performance in identifying protein complexes.
• The algorithm identified complexes with significant biological relevance.
• IPCIPG proved more effective, precise, and comprehensive than comparator algorithms (HUNTER, HC-PIN, CMC, SPICi, MOCDE, MCL).

Conclusions:

• IPCIPG is an effective tool for protein complex discovery.
• The integration of PPI networks and gene expression data improves complex identification accuracy.
• IPCIPG offers a robust and comprehensive approach for analyzing cellular organization.