BioNessie: A Software Tool for the Simulation and Analysis of Biochemical Networks

by Xu Gu


Introduction

BioNessie is an integrated software platform for the composition, simulation and analysis of biochemical networks which is being developed by the Systems Biology team at the Bioinformatics Research Centre as a part of a large DTI funded project BPS: A Software Tool for the Simulation and Analysis of Biochemical Networks. It simulates the kinetics of models of biochemical networks and provides users with functions of parameter scans and sensitivity analysis. BioNessie is currently being integrated with a number of web services which will allow advanced analysis to be performed on a computational cluster or GRID.

This document is a step-by-step guide to use features of BioNessie to construct, simulate and analyse models of biochemical networks. The first two sections of this document are focused on SBML model construction and simulation. The last section illustrates the use of parameter scans and sensitivity analysis for model analysis.

[Top]


Model Construction

To construct a model of a biochemical network that can be analysed later, following steps are essential and must be carried out in the order as it is introduced below:

Step1: Model Settings

Step 2: Compartment Definition

Step3: Species Definition

Step4: Parameter Definition

Step5: Reaction Input

[Top]


Model Simulation

To simulate a SBML model, go to the tree, right click on the SBML model under the Models folder and select Simulate. This will lead you to the simulation panel. Go to bottom left corner of the panel, define the total time (the starting time point is set as 0 by default; the total time represents the ending point of a time interval over which a simulation will be performed) and steps (the number of time points sampled over the interval) that BioNessie will use for the simulation. You can change step size and total time to observe how the curves look: curves will be getting smoother with an increase in the number of steps, but in the cost of more computational time. You can get the simulation results by double-clicking on the folder simulation in the tree file. There are two ways of generating simulation results: you can either run a simulation first and then save its results later, or run a simulation meanwhile having the results saved which can be achieved by clicking button Save Results before running a simulation (by clicking on Simulate).

In BioNessie, 60 and 120 are used as default values of the total time and time steps for a simulation, but you can also change these values to see how the curves respond. For example, the curves will be getting much smoother with an increase in the number of steps, but in the cost of more computational time. You can also have a particular part of the graph zoomed in or zoomed out, which can be achieved by pressing and holding down the left button on the mouse and draging open a box around the place where you want to zoom in. To zoom out, click the right mouse botton and follow Zoom Out->Both Axes. To go back to the original graph, click the right mouse button and follow Auto Range->Both Axes.

[Top]


Parameter Scans

To perform a parameter scan, go to the tree, right click on the SBML model and select Create Parameter Scan. This will lead you to a dialog. Give an expressive name to this task, and then specify attributes of the scan, which include selecting a parameter to be scanned, defining the minimum (Begin Value) and maximum (End Value) of an interval over which the parameter will be scanned, and specifying the number of different values (Steps) obtained from the interval. For a parameter scan, the model will be sent to the simulator for each value involved in the interval. You will also need to specify the total time and time steps for the simulation. Click OK to finish the scan settings. Now, you can see a panel with information about the scan you just designed at the right side of the window. Before you run a scan, you will have to define a monitor function which BioNessie will apply to compute the final results. To do that, click the tag Monitor Formulas at the top of this panel, give an expressive name and write the formula (which will be in the form of a mathematical expression where players of the formula are connected with mathematical operations). Click Commit Change to save the function, and now you are ready to scan the parameter. To do this, go to bottom right corner of the panel and click Do Local Parameter Scan Now!. Go to the tag Chart Results on the top of this panel, and select a monitor function for the parameter scanned. For each parameter, scanning results will be different with the monitor functions you picked. You will see the scanning results in the graphical form from the display window.

[Top]


Sensitivity Analysis

To activate a sensitivity analysis, go to the tree, right click on the SBML model and select Create Sensitivity. This will lead you to a dialog where you can create a task for the sensitivity analysis. Give an expressive name to the task, and define following parameters for an analysis, which include selecting a parameter to be analysed, giving total time and time steps that the BioNessie will use for the computation. Click OK to complete the settings. Now, you can see a panel with information about the task of the sensitivity analysis you just designed at the right side of the window. Go to right bottom corner of the panel and click Do Sensitivity Analysis. You can view the results by clicking the tag Chart on the top of the panel.

[Top]


Bibliography

  1. Chemical reaction - Wikipedia, the free encyclopedia.
  2. Introduction - Enzyme Kinetics and the Michaelis-Menten equation.

[Top]


This page was last modified on 21 Feb 2007.