Deeptools plotHeatmap values

Question

[saha.shagnik]

Can someone please help me with the table of values generated from plotHeatmap? I have generated 4 clusters using k-means method and obtained a heatmap for ChIP-seq data for control and treatment. I have also taken the the table of values underlying the heatmap. How do I interpret it? I wish to extract the data for each control and treatment genes for the 4 clusters, i.e., I wish to extract 4 tables corresponding to the 4 clusters from the plotHeatmap values table in an Ubuntu terminal. Can someone help me out with interpreting the main table and the extraction procedure?

Answer 1

[2nelly]

I think that you mean the values from computeMatrix function.

You should have as many lines as the regions you provide. Then values start from the 7th column. The first 6 columns are region info related. The total number of the columns depends on the binSize and the size before and after the region of your interest.

For instance for one sample, if you calculated the signal + and - 500kb from the center of your regions with bin size of 10kb, then for every sample you will get 100 columns (50 for - and 50 for +) .

Lets say that you have 10 samples. In that case the total number of columns will be 6+(100*10).

I hope that helps.

[saha.shagnik]

Thank you for your response! I managed to interpret the data table. How do I extract the data for each cluster (i used k-means clustering and divided into 4 clusters) in a separate table using grep/awk or anything else in Ubuntu terminal? I am new to these tools!

[2nelly]

Of course you can use awk. Also use of cut -f , will give you the corresponding columns for every sample.