Limits...
MONALISA for stochastic simulations of Petri net models of biochemical systems.

Balazki P, Lindauer K, Einloft J, Ackermann J, Koch I - BMC Bioinformatics (2015)

Bottom Line: The key features of the simulation module are visualization of simulation, interactive plotting, export of results into a text file, mathematical expressions for describing simulation parameters, and up to 500 parallel simulations of the same parameter sets.The open-source distribution allows further extensions by third-party developers.The software is cross-platform and is licensed under the Artistic License 2.0.

View Article: PubMed Central - PubMed

Affiliation: Department of Molecular Bioinformatics, Institute of Computer Science, Cluster of Excellence "Macromolecular Complexes", Johann Wolfgang Goethe-University Frankfurt am Main, Robert-Mayer-Straße 11-15, Frankfurt am Main, 60325, Germany. Pavel.Balazki@gmail.com.

ABSTRACT

Background: The concept of Petri nets (PN) is widely used in systems biology and allows modeling of complex biochemical systems like metabolic systems, signal transduction pathways, and gene expression networks. In particular, PN allows the topological analysis based on structural properties, which is important and useful when quantitative (kinetic) data are incomplete or unknown. Knowing the kinetic parameters, the simulation of time evolution of such models can help to study the dynamic behavior of the underlying system. If the number of involved entities (molecules) is low, a stochastic simulation should be preferred against the classical deterministic approach of solving ordinary differential equations. The Stochastic Simulation Algorithm (SSA) is a common method for such simulations. The combination of the qualitative and semi-quantitative PN modeling and stochastic analysis techniques provides a valuable approach in the field of systems biology.

Results: Here, we describe the implementation of stochastic analysis in a PN environment. We extended MONALISA - an open-source software for creation, visualization and analysis of PN - by several stochastic simulation methods. The simulation module offers four simulation modes, among them the stochastic mode with constant firing rates and Gillespie's algorithm as exact and approximate versions. The simulator is operated by a user-friendly graphical interface and accepts input data such as concentrations and reaction rate constants that are common parameters in the biological context. The key features of the simulation module are visualization of simulation, interactive plotting, export of results into a text file, mathematical expressions for describing simulation parameters, and up to 500 parallel simulations of the same parameter sets. To illustrate the method we discuss a model for insulin receptor recycling as case study.

Conclusions: We present a software that combines the modeling power of Petri nets with stochastic simulation of dynamic processes in a user-friendly environment supported by an intuitive graphical interface. The program offers a valuable alternative to modeling, using ordinary differential equations, especially when simulating single-cell experiments with low molecule counts. The ability to use mathematical expressions provides an additional flexibility in describing the simulation parameters. The open-source distribution allows further extensions by third-party developers. The software is cross-platform and is licensed under the Artistic License 2.0.

No MeSH data available.


Results of a 48 h simulation of the IR model. The IR cycling model was simulated for two days with a typical insulin profile. The basal insulin level is 6·10−11 M. At 09:00 h, 13:00 h and 18:00 h, meal intake stimulates insulin secretion and leads to a peak postprandial concentration of 3.6·10−10 M which returns to its basal state within three hours. The numbers of insulin molecules (upper part) and the number of molecules of different states (lower part) are plotted against the simulated time. Black points represent insulin, red points the free membrane located receptor, blue points the receptor-insulin complex (inactive), green the free receptor in cytosol, yellow the activated phosphorylated IR-insulin complex in cytosol and cyan the phosphorylated receptor-insulin complex in the membrane.
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4496887&req=5

Fig4: Results of a 48 h simulation of the IR model. The IR cycling model was simulated for two days with a typical insulin profile. The basal insulin level is 6·10−11 M. At 09:00 h, 13:00 h and 18:00 h, meal intake stimulates insulin secretion and leads to a peak postprandial concentration of 3.6·10−10 M which returns to its basal state within three hours. The numbers of insulin molecules (upper part) and the number of molecules of different states (lower part) are plotted against the simulated time. Black points represent insulin, red points the free membrane located receptor, blue points the receptor-insulin complex (inactive), green the free receptor in cytosol, yellow the activated phosphorylated IR-insulin complex in cytosol and cyan the phosphorylated receptor-insulin complex in the membrane.

Mentions: Results of a simulation of two days are plotted in Figure 4. The figure shows on top the number of insulin molecules plotted against time. Whereas the number of molecules of different IR states is depicted at the bottom. The PN (Additional file 4) is provided as the MONALISA-project file IR_Model.mlproject. The simulation setup file “IR_Model_params.xml”, (Additional file 5) can be found in the supplement.Figure 4


MONALISA for stochastic simulations of Petri net models of biochemical systems.

Balazki P, Lindauer K, Einloft J, Ackermann J, Koch I - BMC Bioinformatics (2015)

Results of a 48 h simulation of the IR model. The IR cycling model was simulated for two days with a typical insulin profile. The basal insulin level is 6·10−11 M. At 09:00 h, 13:00 h and 18:00 h, meal intake stimulates insulin secretion and leads to a peak postprandial concentration of 3.6·10−10 M which returns to its basal state within three hours. The numbers of insulin molecules (upper part) and the number of molecules of different states (lower part) are plotted against the simulated time. Black points represent insulin, red points the free membrane located receptor, blue points the receptor-insulin complex (inactive), green the free receptor in cytosol, yellow the activated phosphorylated IR-insulin complex in cytosol and cyan the phosphorylated receptor-insulin complex in the membrane.
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4496887&req=5

Fig4: Results of a 48 h simulation of the IR model. The IR cycling model was simulated for two days with a typical insulin profile. The basal insulin level is 6·10−11 M. At 09:00 h, 13:00 h and 18:00 h, meal intake stimulates insulin secretion and leads to a peak postprandial concentration of 3.6·10−10 M which returns to its basal state within three hours. The numbers of insulin molecules (upper part) and the number of molecules of different states (lower part) are plotted against the simulated time. Black points represent insulin, red points the free membrane located receptor, blue points the receptor-insulin complex (inactive), green the free receptor in cytosol, yellow the activated phosphorylated IR-insulin complex in cytosol and cyan the phosphorylated receptor-insulin complex in the membrane.
Mentions: Results of a simulation of two days are plotted in Figure 4. The figure shows on top the number of insulin molecules plotted against time. Whereas the number of molecules of different IR states is depicted at the bottom. The PN (Additional file 4) is provided as the MONALISA-project file IR_Model.mlproject. The simulation setup file “IR_Model_params.xml”, (Additional file 5) can be found in the supplement.Figure 4

Bottom Line: The key features of the simulation module are visualization of simulation, interactive plotting, export of results into a text file, mathematical expressions for describing simulation parameters, and up to 500 parallel simulations of the same parameter sets.The open-source distribution allows further extensions by third-party developers.The software is cross-platform and is licensed under the Artistic License 2.0.

View Article: PubMed Central - PubMed

Affiliation: Department of Molecular Bioinformatics, Institute of Computer Science, Cluster of Excellence "Macromolecular Complexes", Johann Wolfgang Goethe-University Frankfurt am Main, Robert-Mayer-Straße 11-15, Frankfurt am Main, 60325, Germany. Pavel.Balazki@gmail.com.

ABSTRACT

Background: The concept of Petri nets (PN) is widely used in systems biology and allows modeling of complex biochemical systems like metabolic systems, signal transduction pathways, and gene expression networks. In particular, PN allows the topological analysis based on structural properties, which is important and useful when quantitative (kinetic) data are incomplete or unknown. Knowing the kinetic parameters, the simulation of time evolution of such models can help to study the dynamic behavior of the underlying system. If the number of involved entities (molecules) is low, a stochastic simulation should be preferred against the classical deterministic approach of solving ordinary differential equations. The Stochastic Simulation Algorithm (SSA) is a common method for such simulations. The combination of the qualitative and semi-quantitative PN modeling and stochastic analysis techniques provides a valuable approach in the field of systems biology.

Results: Here, we describe the implementation of stochastic analysis in a PN environment. We extended MONALISA - an open-source software for creation, visualization and analysis of PN - by several stochastic simulation methods. The simulation module offers four simulation modes, among them the stochastic mode with constant firing rates and Gillespie's algorithm as exact and approximate versions. The simulator is operated by a user-friendly graphical interface and accepts input data such as concentrations and reaction rate constants that are common parameters in the biological context. The key features of the simulation module are visualization of simulation, interactive plotting, export of results into a text file, mathematical expressions for describing simulation parameters, and up to 500 parallel simulations of the same parameter sets. To illustrate the method we discuss a model for insulin receptor recycling as case study.

Conclusions: We present a software that combines the modeling power of Petri nets with stochastic simulation of dynamic processes in a user-friendly environment supported by an intuitive graphical interface. The program offers a valuable alternative to modeling, using ordinary differential equations, especially when simulating single-cell experiments with low molecule counts. The ability to use mathematical expressions provides an additional flexibility in describing the simulation parameters. The open-source distribution allows further extensions by third-party developers. The software is cross-platform and is licensed under the Artistic License 2.0.

No MeSH data available.