MPI für molekulare Genetik / Department of Vertebrate Genomics |
|Monte-Carlo analysis of an ODE Model of the Sea Urchin Endomesoderm Network.|
|Authors:||Kühn, Clemens; Wierling, Christoph; Kühn, Alexander; Klipp, Edda; Panopoulou, Georgia; Lehrach, Hans; Poustka, Albert J.|
|Research Context:||This work was supported by the Max-Planck Gesellschaft zur Förderung der Wissenschaften e.V. and the European Union by grant LHSG – CT – 2004 – 512092. Clemens Kühn is funded by the Deutsche Forschungsgemeinschaft via the International Research Training Group "Genomics and Systems Biology of Molecular Networks".|
|Date of Publication (YYYY-MM-DD):||2009-08-23|
|Title of Journal:||BMC Systems Biology|
|Journal Abbrev.:||BMC Syst. Biol.|
|Copyright:||2009 Kühn et al; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
|Review Status:||not specified|
|Abstract / Description:||Gene Regulatory Networks (GRNs) control the differentiation, specification and function of cells at the genomic level. The levels of interactions within large GRNs are of enormous depth and complexity. Details about many GRNs are emerging, but in most cases it is unknown to what extent they control a given process, i.e. the grade of completeness is uncertain. This uncertainty stems from limited experimental data, which is the main bottleneck for creating detailed dynamical models of cellular processes. Parameter estimation for each node is often infeasible for very large GRNs. We propose a method, based on random parameter estimations through Monte-Carlo simulations to measure completeness grades of GRNs.
We developed a heuristic to assess the completeness of large GRNs, using ODE simulations under different conditions and randomly sampled parameter sets to detect parameter-invariant effects of perturbations. To test this heuristic, we constructed the first ODE model of the whole sea urchin endomesoderm GRN, one of the best studied large GRNs. We find that nearly 48% of the parameter-invariant effects correspond with experimental data, which is 65% of the expected optimal agreement obtained from a submodel for which kinetic parameters were estimated and used for simulations. Randomized versions of the model reproduce only 23.5% of the experimental data.
The method described in this paper enables an evaluation of network topologies of GRNs without requiring any parameter values. The benefit of this method is exemplified in the first mathematical analysis of the complete Endomesoderm Network Model. The predictions we provide deliver candidate nodes in the network that are likely to be erroneous or miss unknown connections, which may need additional experiments to improve the network topology. This mathematical model can serve as a scaffold for detailed and more realistic models. We propose that our method can be used to assess a completeness grade of any GRN. This could be especially useful for GRNs involved in human diseases, where often the amount of connectivity is unknown and/or many genes/interactions are missing.
|Comment of the Author/Creator:||All authors contributed to the design and coordination of the study. CK performed the computational implementations and prepared the original draft, which was revised by AJP, EK and CW. AK performed experiments. All authors read and approved the final manuscript.|
|External Publication Status:||published|
|Communicated by:||Hans Lehrach|
|Affiliations:||MPI für molekulare Genetik|
|External Affiliations:||1.Humboldt Universität zu Berlin, Institute for Biology, Invalidenstr 42, 10115 Berlin, Germany.|
|You have privileges to view the following file(s):|
|83.pdf [1,00 Mb] [Comment:Open access] |
The scope and number of records on eDoc is subject
to the collection policies defined by each institute
- see "info" button in the collection browse view.