Network reconstruction based on proteomic data and prior knowledge of protein connectivity using graph theory

Area of Science:

• Systems Biology
• Computational Biology
• Biochemistry

Background:

• Modeling signal transduction pathways is crucial for understanding cellular functions and biochemical microenvironments.
• Existing computational methods for pathway modeling often involve complex optimization or machine learning algorithms.

Purpose of the Study:

• To develop a cell-specific signaling model by integrating prior knowledge networks (PKN) with phosphoproteomic data.
• To introduce a computationally efficient and accurate method for interrogating protein-protein interaction networks.

Main Methods:

• Utilized a breadth-first graph traversal to identify shortest pathways and score proteins within the PKN.
• Integrated experimental data dependencies with PKN using a heuristic formulation to resolve inconsistencies.
• Applied a cross-validation analysis to ensure prediction accuracy.

Main Results:

• The developed algorithm efficiently constructs medium to large-scale signaling networks.
• Demonstrated applicability using a curated interaction graph model for EGF/TNFA stimulation.
• Generated predictive topologies comparable to established Integer Linear Programming (ILP) formulations.

Conclusions:

• The proposed graph theory-based approach provides an efficient and accurate method for interrogating protein-protein interaction networks.
• This approach yields meaningful biological insights and cell-specific signaling models.