The correlation analysis

The correlation analysis can be used to assess similarity between genome heatmap track with a target data track. The target data track could be one of following cases:

Correlation is computed by Spearman's rank method, which is rank-based and more robust for data of different scales. The correlation coefficients are plotted as horizontal color bars on left of genome heatmap:

To select the target data track for correlation, go to its control panel ("Statistics & analysis" → "Correlation"):

To select a heatmap track, press first button. The heatmap track selection box will appear:

Click and select one track from the track selection box, and correlation will be run against it. Name of the selected track will be printed in control panel:

To select a genomic feature track and run correlation with its density data, click second button in the panel. The genomic feature track selection box will appear:

More conveniently, you can always right-click on a target track and select "Correlation" option from menu for this purpose.

Once correlation is running, right click on the correlation coefficent bar plot to get a context menu:

Choose the "Configure" option, a panel appears in floating toolbox to let you configure the bar plot:

Click one color blob in the panel, the color palette appears:

By choosing a color from palette, the correlation bar plot will update accordingly.


Last modifed: 1/2/2012