Desktop Example — Calculating Sensitivities

Overview

This example shows you how to set up and calculate sensitivities in the SimBiology desktop. For information on how to calculate sensitivities at the command line, see Command-Line Example — Calculating Sensitivities.

About the Example Model

This example uses the model from Modeling a G Protein Cycle in the SimBiology Model Reference documentation.

This table shows the reactions used to model the G protein cycle and the corresponding rate parameters (rate constants) for each reaction. For reversible reactions, the forward rate parameter is listed first.

About This Example

Yi et al. (2003) show that the rate of G protein inactivation is much lower in the mutant strain (kGd = 0.004) relative to the wild-type strain (kGd = 0.11), which explains the higher levels of activated G protein (Ga) over time as shown in the following figure.

Thus, the active G protein, Ga, is sensitive to the value of the parameter, kGd. Other species or parameters in the model can also affect levels of active G protein. To study the sensitivity of a species to other species or parameters in a model, you can perform sensitivity analysis. Sensitivity analysis lets you compute the time-dependent derivatives of one or more species (Output) relative to either model parameter values or species initial conditions (Input).

First, it might be useful to explore the sensitivities of every species with respect to every parameter in the model. You can later narrow down the results to visualize the sensitivity results for Ga using plots. Thus, you want to calculate the time-dependent derivatives:

Back to Top

Prerequisites

• Opening and Saving the Example Model

• Adding a Task to Calculate Sensitivities

Opening and Saving the Example Model

1. Load the example project at the command line by typing

   sbioloadproject gprotein_norules

   The model is stored in a variable called m1.

2. Open the SimBiology desktop with the model loaded by typing:

   simbiology(m1)

   The SimBiology desktop opens with Yeast_G_Protein_wt.

3. Save the project.

   1. Select File > Save Project As. The Save SimBiology Project dialog box opens.

   2. Specify a name (for example, gprotein_ex) and location for your project and click Save.

Adding a Task to Calculate Sensitivities

To perform sensitivity analysis in a model, first add a model task to calculate sensitivities:

• In the Project Explorer, under Model Session-Yeast_G_Protein_wt, right-click Model Tasks and select Add Task > Calculate sensitivities.

  The Sensitivity Analysis pane opens.

Back to Top

Setting Options for Sensitivity Analysis

1. Add species to the Sensitivity Settings tab:

   1. Select the Component Palette (left pane by default).

   2. Press and hold the Shift key to select all components of the Type species.

   3. Click-drag the selected species components to the table in the Sensitivity Settings tab.

2. In the table, mark all the species components as outputs by selecting their Output check boxes.

3. Add parameters to the Sensitivity Settings tab:

   1. Select the Component Palette (left pane by default).

   2. Press and hold the Shift key to select all components of the Type parameter.

   3. Click-drag the selected parameter components to the table in the Sensitivity Settings tab.

4. In the table, mark all the parameter components as inputs by selecting their Input check boxes.

5. Under Normalization, select Full to facilitate full normalization so the sensitivities can be compared with each other. The Sensitivity Settings tab should resemble the following.

   

6. Save the project by selecting File > Save Project.

For more information about normalization see Normalization in the SimBiology Reference.

Back to Top

Getting Results for Sensitivity Analysis

Click . The Sensitivity Matrix Subplot, which is the default plot for sensitivity analysis, opens.

The plot results show for example that as expected, receptor R is sensitive to values of parameters involved in the receptor-ligand complex formation and receptor degradation.

Back to Top

References

[1] Tau-Mu Yi, Hiroaki Kitano, and Melvin I. Simon. A quantitative characterization of the yeast heterotrimeric G protein cycle. PNAS (2003) vol. 100, 10764–10769.

Back to Top

Sensitivity Analysis

Command-Line Example — Calculating Sensitivities

免费的计算生物学交互式工具包

了解采用 MathWorks 产品分析、可视化及对生物数据和系统进行建模的方式。

索取免费工具包

试用版

试用最新版本的计算生物学产品。

索取试用版本软件，试用最新版本的软件。