|
Experimental data |
  
|
|
Experimental data |
|
Experimental datasets must be the results of time courses expressed as exact or relative values of substance concentrations and may be comprised of several files each including possibly multiple experiments.
The diagram parameters estimation can not be perform until the experimental data are not set. Optimization experiments can be process in the "Experimental data" tab. To add a new experiment, press the button with the symbol plus on it. To define the experiment you can fill the following fields.
| • | Name. The optimization experiment name (must be unique). |
| • | Diagram experiment. The diagram experiment name, which is specified when you need to fit parameters of the diagram modification identified by the diagram experiment. If the value is equal to the empty string the fitting is performed for the diagram without changes. |
| • | Data file. The name of the file with experimental data. The data must be stored in folder BioUML\data_resources\Optimization\Experiments. |
| • | Weight method. The weight method to make all trajectories of each variable have similar importance in the fit. Valid methods for the weight wj calculation are mean: wj = 1/|<xj>|, mean square: wj = (<xj2>)-1/2, and standard deviation: wj = 1/|<xj2>-<xj><xj>|. |
| • | Experiment type. The type of experimental data is time course or steady state. |
| • | Cell line. The parameter for separation of the experiments on the groups. All the fitting parameters declared as local remain global within the same cell line. For the experiments where cell line value is the empty string all local parameters are independently fitted. |
You also need to specified the array of connections between diagram variables and columns of the experimental data table. For this purpose fill the table automatically generated in the experiment tab. The following fields must be identified.
| • | Experiment parameters. The experimental table column names. |
| • | Diagram parameters. The diagram parameters or variables. |
| • | Type. The type of the experimental data column: exact value, relative to initial value or relative to completion value. |
| • | Weight. The weight automatically calculated for each data column. It can be changed by the user. |
To remove some experiments, select them and press the button with the symbol minus.
Example of the experimental data file contents
ID Diagram ID
OS
PLATFORM
TITLE Experimental data title
DESCRIPTION Experimental data description..
SAMPLES
#1 time; Float; NONE; v1
#2 Substance ID; Float; NONE; v2
PROPERTIES
DATA
0 0.0 0.01
1 50.0 0.36
//