Box and Whiskers Plot
Choose the data set in the Control panel, and which data file to plot either in the Control panel or in the drop-down list in the top of the main window. You can define the box colors by first transposing the data set, then editing the gene colors in the Data manager under the Colors & Symbols tab, and transposing the data again. Press the Box and Whiskers button in the upper right part of the main window, or select it in the View menu (main window). The x-axis represents the different columns and is on a categorical scale.
The data in each column is represented as a box with whiskers indicating the median, 1st and 3rd quartiles, maximum and minimum values that are non-outliers, and outliers. The median is shown as a dashed line across the box. The bottom and top of the box represents the 1st and 3rd quartiles (Q1 and Q3) of the data respectively. Outliers are defined as data points larger than [Q3 + c(Q3-Q1)] and smaller than [Q1 - c(Q3-Q1)]. The constant c is by default 1.5, but can be changed by pressing the Edit button in the chart window, opening the genes one by one, and changing the Length under the Format tab. Under the ExtrOut tab you can choose to make extreme outliers Visible by ticking the check box. Extreme outliers are defined as larger than [Q3 + 2c(Q3-Q1)] and smaller than [Q1 - 2c(Q3-Q1)], and are by default not shown in the plot. The whiskers extend to the smallest and largest values that are not defined as outliers.
In the status bar of the chart window, the project name, data set, and data file are given, and it is indicated if the data is transposed or not. By pressing the Edit button in the chart window, you can customize the plot by adding a title, changing background color and much more.
Normally, samples are arranged in rows and genes in columns which results a box and whiskers plot for each gene describing the variation of samples within each gene. If the data is transposed, it is the other way around: it describes the variation of genes for each sample.
Scaling has large effect on the appearance of plots. Watch the Autoscaling tutorial at www.multid.se to learn more.