Re-creating a heatmap

Question

[cbio]

I am attempting to re-create a heatmap found in this article (Figure 3 B, the MNase-seq one). They rank genes / TSS by length of the nucleosome free regions surrounding the TSS.

I would like to do something similar using DNase-seq data. I have created many heatmaps before, but I am not typically sorting genes / TSS this way. Sorting by length of -1 nuclesome to the +1 nucleosome.

How would I go about creating a heatmap like this? How do I apply my DNase-seq data to find these regions, and then rank them by their length?

Answer 1

[Ryan Dale]

If you're comfortable with Python, you can use metaseq or deepTools to get a NumPy array of DNase signal over TSSs. You would then need to come up with a way of detecting -1 and +1 nucleosomes in each row of the array for sorting purposes.