ajdusting scales when plotting heatmap with bigwigs

[robel.te...@gmail.com]

I already have big wig files for multiple ChIP seq data on different TFs and histone marks. The different data present different levels of values. Some of them go from 0 to 100 others could go from 0 to 500 (just for an example). I'm used to NGSplot where I could tell the program to do the scale for each track of heatmap (meaning heatmap of each TF or histone mark). Meaning that the colour scale can be adjusted on each mark depending on the maximum and minimal value (option local scaling).
My problem is that some marks have high values while others have low, so the ones that have big values can easily mask the lower valued marks. I've tried multiple approaches to resolve this problem.
I've tried to calculate the Z score for each value (z_score= (value-average)/sd) on each mark, but your tool doesn't take a file written by R, even though I make sure that the format is the same.
Then I've tried to plot each mark separately. How ever since I needed the same gene order:
-I've run computMatrix on all the bw files
-did ComputeMatrixOperations subset each sample from the computeMatrix output matrix
-I plotted heatmap for the first bw by asking to give the ordered bed file.
-Used the ordered bed file to sort by using computeMatrix Operations sort
and the result is just frustrating as it indicates me that the individual matrices are missing one cluster

"""The computeMatrix output is missing the 'cluster_8' region group. It has set([u'genes']) but the specified regions have ['cluster_8', 'cluster_9', 'cluster_6', 'cluster_7', 'cluster_4', 'cluster_5', 'cluster_2', 'cluster_3', 'cluster_1', 'cluster_10']."""

I've no more ideas to resolve this problem. Can you help me out.

Thanks

[Devon Ryan]

I suggest manually setting the `--zMin` and `--zMax` scores to set the
range to something appropriate for each sample. If you provide them with
a list of values (e.g., `--zMin 0 100 20 -80`) then you'll note that the
colorbar on the right side will be automatically removed and replaced
with one at the bottom of each heatmap. If you need to, you can also
provide a list to `--yMin` and `--yMax`.

Devon

--
Devon Ryan, PhD
Bioinformatician / Data manager
Bioinformatics Core Facility
Max Planck Institute for Immunobiology and Epigenetics
Email: dpry...@gmail.com

[Roast Teaser]

Thanks, I will do that. 

[barb....@gmail.com]

Did you solve this? I am struggling with the same issue now.