Construction of dynamic probabilistic protein interaction networks for protein complex identification

Area of Science:

• Systems Biology
• Bioinformatics
• Computational Biology

Background:

• High-throughput experiments generate vast protein-protein interaction (PPI) data, enabling the construction of large PPI networks.
• Current systems biology approaches often analyze static PPI networks, overlooking crucial dynamic protein information.
• Dynamic information, such as gene expression data over time, is vital for a comprehensive understanding of cellular processes.

Purpose of the Study:

• To develop a method that integrates dynamic information into protein-protein interaction networks.
• To propose a novel approach for identifying protein complexes using dynamic probabilistic protein networks (DPPN).
• To demonstrate the effectiveness of DPPN in capturing dynamic aspects of protein interactions.

Main Methods:

• Utilized an active probability-based method to determine protein activity levels at different time points.
• Constructed dynamic probabilistic protein networks (DPPN) by integrating gene expression data with static PPI networks.
• Developed and applied a novel algorithm to identify protein complexes within the constructed DPPNs.

Main Results:

• Successfully constructed three DPPNs using yeast PPI and gene expression datasets.
• The proposed method accurately identified many well-characterized protein complexes within the DPPNs.
• The approach effectively leveraged both topological structure and dynamic information inherent in DPPNs.

Conclusions:

• Transitioning from static to dynamic PPI networks is crucial for accurate protein complex identification.
• The developed method provides a robust framework for integrating dynamic information into biological networks.
• This approach has potential applications beyond protein complex identification, including pathway analysis.