Incremental Signaling Pathway Modeling by Data Integration
Constructing quantitative dynamic models of signaling pathways is an important task for computational systems biology. Pathway model construction is often an inherently incremental process, with new pathway players and interactions continuously being discovered and additional experimental data being generated. Here we focus on the problem of performing model parameter estimation incrementally by integrating new experimental data into an existing model. A probabilistic graphical model known as the factor graph is used to represent pathway parameter estimates. By exploiting the network structure of a pathway, a factor graph compactly encodes many parameter estimates of varying quality as a probability distribution. When new data arrives, the parameter estimates are refined efficiently by applying a probabilistic inference algorithm known as belief propagation to the factor graph. A key advantage of our approach is that the factor graph model contains enough information about the old data, and uses only new data to refine the parameter estimates without requiring explicit access to the old data. To test this approach, we applied it to the Akt-MAPK pathways, which regulate the apoptotic process and are among the most actively studied signaling pathways. The results show that our new approach can obtain parameter estimates that fit the data well and refine them incrementally when new data arrives.
KeywordsMolecular Species Data Integration Belief Propagation Variable Node Factor Node
Unable to display preview. Download preview PDF.
- 5.Koller, D., Friedman, N.: Probabilistic Graphical Models: Principles and Techniques. The MIT Press, Cambridge (2009)Google Scholar
- 7.Matsuno, H., Tanaka, Y., Aoshima, H., Doi, A., Matsui, M., Miyano, S.: Biopathways representation and simulation on hybrid functional Petri net. Silico Biology 3(3), 389–404 (2003)Google Scholar
- 9.Purvis, J., Radhakrishnan, R., Diamond, S.: Steady-state kinetic modeling constrains cellular resting states and dynamic behavior. PLoS Computational Biology 5(3) (2009)Google Scholar
- 10.Gat-Viks, I., Tanay, A., Raijman, D., Shamir, R.: The factor graph network model for biological systems. In: Miyano, S., Mesirov, J., Kasif, S., Istrail, S., Pevzner, P.A., Waterman, M. (eds.) RECOMB 2005. LNCS (LNBI), vol. 3500, pp. 31–47. Springer, Heidelberg (2005)Google Scholar
- 11.Delcher, A., Kasif, S., Goldberg, H., Hsu, W.: Protein secondary structure modelling with probabilistic networks. In: Proc. Int. Conf. on Intelligent Systems & Molecular Biology, pp. 109–117 (1993)Google Scholar
- 13.Koh, G.: Pathway Models Decomposition and Composition Techniques for Parameter Estimation. PhD thesis, Graduate School of Integrative Sciences, National University of Singapore, Singapore (2008)Google Scholar
- 14.Pearl, J.: Probabilistic reasoning in intelligent systems: networks of plausible inference. Morgan Kaufmann, San Francisco (1988)Google Scholar
- 18.Koh, G., Teong, H.F.C., Clément, M.V., Hsu, D., Thiagarajan, P.: A decompositional approach to parameter estimation in pathway modeling: a case study of the Akt and MAPK pathways and their crosstalk. Bioinformatics 22(14), e271–e280 (2006)Google Scholar
- 19.Gill, P., Murray, W., Wright, M.: Practical Optimization. Academic Press, London (1982)Google Scholar