Circos > Documentation > Tutorials > Recipes > Automating Heatmaps

Use the latest version of Circos and read Circos best practices—these list recent important changes and identify sources of common problems.

If you are having trouble, post your issue to the Circos Google Group and include all files and detailed error logs. Please do not email me directly unless it is urgent—you are much more likely to receive a timely reply from the group.

Don't know what question to ask? Read Points of View: Visualizing Biological Data by Bang Wong, myself and invited authors from the Points of View series.

9 — Recipes

16. Automating Heatmaps

Before reading this tutorial, make sure that you understand how dynamic configuration parameters work (see Configuration Files Tutorial) and have read through the Automating Tracks Tutorial.

For this tutorial, I have created an image with 100 heat map tracks. The data for these tracks are gene densities computed across differently sized windows (0.5-50 Mb). The higest resolution file is data/8/17/genes.0.txt which samples density every 500kb. The lowest resolution file is data/8/17/genes.99.txt which samples density every 50,000kb (1/100th resolution of genes.0.txt).

The gene densities were designed so that each heat map interval occupies the same number of pixels along the map's circumference.

changing heat map color

The color of the heat map is specified using a list of colors

color = red,green,blue

or a color list

color = spectral-11-div

For more about color lists, see the Configuration Files Tutorial.

The track counter can be used to dynamically change the color scheme. For example, as the track counter increases from 0 to 99, the definition

color = eval(sprintf("spectral-%d-div",remap_round(counter(plot),0,99,11,3)))

will assign a list to a track based on the counter. The assignment will range from spectral-11-div for the outer-most track, progressing through spectral-10-div, ..., and end at spectral-3-div for the inner-most track.

You can combine multiple color maps. Here, an orange sequential color list is added to a reversed blue sequential one.

color = eval(sprintf("blues-%d-seq-rev,oranges-%d-seq-rev",

Given the reduced resolution of the inner-most track, reducing the number of colors in its heat map can make the figure more legible.

adjusting log scaling

The scale_log_base parameter controls how heat map values are mapped onto colors. The default value of this parameter is scale_log_base = 1, which corresponds to a linear mapping. For more details about this parameter, see the Heat Map Tutorial.

Keeping the color list constant, but varying the scale_log_base, you can increase the dynamic range of color sampling for small values (if log_scale_base < 1) or large values (if log_scale_base > 1).

color          = spectral-11-div
# 0.05, 0.10, 0.15, ..., 5.00
scale_log_base = eval(0.05*(1+counter(plot)))