An Effective Method for Identifying Functional Modules in Dynamic PPI Networks

Background: Identifying functional modules (FM) in Protein-Protein Interaction (PPI) networks is essential for understanding the organization and evolution of cellular systems. Most current functional module discovery algorithms merely focus on the static PPI network. However, PPI network is dynamic over time and varies under different conditions. Objective: Therefore, discovering functional modules in dynamic PPI networks (DPN) is crucial. In this paper, functional module is defined as the union of a time-line of evolutionary step-modules. A novel StableCore and Adaptive Incremental Algorithm (SCAIA) is developed to discover functional modules in DPN. Method: The SCAIA first detects static step-modules of the first subnetwork and adaptively updates the modular structure of other subnetworks, and then identifies functional modules and their evolutionary trends based on the extracted step-modules of each subnetwork. Results: Extensive results show SCAIA achieves very satisfactory Precision, F-measure and P-value results among the seven functional module discovery algorithms compared in this study. Conclusion: SCAIA performs significantly better than seven methods on discovering accurate and stable functional modules. SCAIA can also track the evolutionary process of functional modules over time, providing insights into the underlying behavior of functional modules for future biological studies.