Automatic Modeling of Signal Pathways from Protein-Protein Interaction Networks

This paper presents a novel method for recovering signaling pathways from protein-protein interaction networks automatically. Given an undirected weighted protein interaction network, finding signaling pathways is treated as searching for the optimal subnetworks from the network according to some cost function. To approach this optimum problem, an integer linear programming model is proposed in this work to model the signal pathways from the protein interaction network. The numerical results on three known yeast MAPK signal pathways demonstrate the effic iency and effectiveness of the proposed method.