Abstract B2-08: Model order reduction for cell signaling pathways: An investigation of G-protein coupled receptor signaling

Abstract

Abstract G-protein coupled receptors (GPCRs) are frequently overexpressed in human carcinomas, where they are responsible for stimulating the aberrant proliferation of cancer cells. Unfortunately, the mechanisms of this stimulation are poorly understood due to the complexity of the interactions between GPCRs and second-messenger pathways. Robust quantitative methods are therefore necessary to simplify mathematical models of GPCR-mediated signaling. Model order reduction aims to reduce the dimensionality of kinetic models of signaling pathways (such as the GPCR signal transduction system) in order to enable the computational investigation of complex signaling networks. Existing reduction procedures aim to eliminate “slaved species”, which have low net rates of formation. Unfortunately, these methods cannot be applied to reaction networks such as the GPCR system where receptor complexes exhibit high affinities for binding proteins (and therefore have low dissociation constants). This paper aims to develop a novel projection-based model reduction technique for reducing the dimensionality of kinetic models of cell signaling pathways (such as the GPCR signaling cascade). This algorithm (executed at the command-line interface) combines the techniques of proper orthogonal decomposition, trajectory piecewise linearization, and Krylov subspace reduction in an effort to identify a best-fit subspace for the model trajectories that accurately approximates the mapping between the model input and outputs. The command-line tool effectively reduced the dimensionality of a kinetic model of the GPCR network from 16 to 3. In addition, implementing the algorithm revealed strong correlations within two disjoint sets of participating species, suggesting the need to monitor the concentration of only one species from each set during a computational or experimental investigation. The model reduction procedure enables cell biologists to effectively employ kinetic models to analyze the mechanisms of uncontrolled cell proliferation. A comprehensive understanding of these mechanisms will aid pharmacologists in identifying therapeutic targets in cancer cells. Citation Format: Prem M. Talwai. Model order reduction for cell signaling pathways: An investigation of G-protein coupled receptor signaling. [abstract]. In: Proceedings of the AACR Special Conference on Computational and Systems Biology of Cancer; Feb 8-11 2015; San Francisco, CA. Philadelphia (PA): AACR; Cancer Res 2015;75(22 Suppl 2):Abstract nr B2-08.