Original Research ARTICLE

Front. Physiol., 05 July 2011 | http://dx.doi.org/10.3389/fphys.2011.00035

Identification of critical molecular components in a multiscale cancer model based on the integration of Monte Carlo, resampling, and ANOVA

Zhihui Wang1, Veronika Bordas2 and Thomas S. Deisboeck1*
  • 1 Harvard-MIT Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA
  • 2 The University of Texas School of Law, Austin, TX, USA

To date, parameters defining biological properties in multiscale disease models are commonly obtained from a variety of sources. It is thus important to examine the influence of parameter perturbations on system behavior, rather than to limit the model to a specific set of parameters. Such sensitivity analysis can be used to investigate how changes in input parameters affect model outputs. However, multiscale cancer models require special attention because they generally take longer to run than does a series of signaling pathway analysis tasks. In this article, we propose a global sensitivity analysis method based on the integration of Monte Carlo, resampling, and analysis of variance. This method provides solutions to (1) how to render the large number of parameter variation combinations computationally manageable, and (2) how to effectively quantify the sampling distribution of the sensitivity index to address the inherent computational intensity issue. We exemplify the feasibility of this method using a two-dimensional molecular-microscopic agent-based model previously developed for simulating non-small cell lung cancer; in this model, an epidermal growth factor (EGF)-induced, EGF receptor-mediated signaling pathway was implemented at the molecular level. Here, the cross-scale effects of molecular parameters on two tumor growth evaluation measures, i.e., tumor volume and expansion rate, at the microscopic level are assessed. Analysis finds that ERK, a downstream molecule of the EGF receptor signaling pathway, has the most important impact on regulating both measures. The potential to apply this method to therapeutic target discovery is discussed.

Keywords: agent-based model, analysis of variance, multiscale, non-small cell lung cancer, sensitivity analysis

Citation: Wang Z, Bordas V and Deisboeck TS (2011) Identification of critical molecular components in a multiscale cancer model based on the integration of Monte Carlo, resampling, and ANOVA. Front. Physio. 2:35. doi: 10.3389/fphys.2011.00035

Received: 04 February 2011; Accepted: 20 June 2011;
Published online: 05 July 2011.

Edited by:

John Jeremy Rice, Functional Genomics and Systems Biology, USA

Reviewed by:

Eric A. Sobie, Mount Sinai School of Medicine, USA
Feilim Mac Gabhann, Johns Hopkins University, USA

Copyright: © 2011 Wang, Bordas and Deisboeck. This is an open-access article subject to a non-exclusive license between the authors and Frontiers Media SA, which permits use, distribution and reproduction in other forums, provided the original authors and source are credited and other Frontiers conditions are complied with.

*Correspondence: Thomas S. Deisboeck, Complex Biosystems Modeling Laboratory, Harvard-MIT Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital-East, 2301, Building 149, 13th Street, Charlestown, MA 02129, USA. e-mail: deisboec@helix.mgh.harvard.edu