Computational Network Analysis

nloaded geted therapeutics hold tremendous promise in inhibiting cancer cell proliferation. However, targeting ns individually can be compensated for by bypass mechanisms and activation of regulatory loops. Designtimal therapeutic combinations must therefore take into consideration the complex dynamic networks in ll. In this study, we analyzed the insulin-like growth factor (IGF-1) signaling network in the MDA-MB231 cancer cell line. We used reverse-phase protein array to measure the transient changes in the phosphorof proteins after IGF-1 stimulation. We developed a computational procedure that integrated mass acodeling with particle swarm optimization to train the model against the experimental data and infer the wn model parameters. The trained model was used to predict how targeting individual signaling proteins the rest of the network and identify drug combinations that minimally increased phosphorylation of proteins elsewhere in the network. Experimental testing of the modeling predictions showed that optimal ombinations inhibited cell signaling and proliferation, whereas nonoptimal combination of inhibitors sed phosphorylation of nontargeted proteins and rescued cells from cell death. The integrative approach increa described here is useful for generating experimental intervention strategies that could optimize drug combinations and discover novel pharmacologic targets for cancer therapy. Cancer Res; 70(17); OF1–11. ©2010 AACR.