Contents - Index


Data Set

 

A data set contains data files, transposition (if any), and reversible changes done in the Data manager such as scaling, inactivation of rows and columns, categorization of samples into test and training samples, categorization of genes and variables into predictors and responses, defining groups and assigning samples to them, and coloring of samples/genes/groups. Pressing Data sets in the left part of the Control panel displays all data sets that belong to the active project. The active project and data set is shown in the status bar at the bottom of the main window

 

      

 

When a new project is created, a data set is also created containing the data file(s) that was loaded into the Control panel. By default this set is called Set1, but it can be changed by ticking the Rename check box to the right in the Control panel, changing the name and pressing Apply. You add more data sets, either by pressing Clone which creates a copy of the data set containing the same data files and settings, or creating an empty data set by pressing the Plus button. The Cross button is used to remove a data set. The settings of a data set cannot be stored seperately, but is saved within your project when you export it. By pressing the Pencil button you can add comments to your data sets, e.g. including which settings you have made in the Data manager.

 

A data set can contain multiple data files. When the settings of a data set are changed with the Data manager or by transposition, the changes apply to all data files in the active data set. Data files within one data set are allowed to have different dimensions, but most analyses that handle multiple data files require that they have the same dimensions. Also, managing files in the data set will be inactivated if the data files have different dimensions. If GenEx has disabled the data managing due to different dimensions of the data files in a data set, make sure that the present files in the data set was not transposed when the new file was added. You can see whether a data set is transposed or not in the status bar in the main window. If they where transposed, remove the new file, transpose the older files back to their original form by pressing the Transpose button once more, and add the new file to the data set again. You can easily switch between data files in the active data set in the drop-down list at the top of the main window or under Data files in the Control panel.

 

If you want to view the managed data, press the View data button at the top of the main window. This displays the data with all settings done in the Data manager, and if you have transposed the data you will see that as well. 

 

To test the influence of a suspected outlier sample, clone the data set and inactivate the suspected outlier in the new data set using the Data Manager. Analyze both data sets and compare the results.