Visualizing average read coverage across features with varying length

Question

[rxzlmn]

Hi,

I am trying to figure out a way to plot read densities over features of varying length. I have a set of ~100 genomic features that are between ~3kb to 7kb in length, and I want to plot per-base coverage or raw read coverage across those elements on the same x scale, i.e. from 0 to 100% of feature length.

The only tool so far I found is ngsplot, but it is kind of limited as it is stand-alone and does not have many options.

Is there a way to do that with existing packages or base tools in R?

[Devon Ryan]

For "base tools", are we including just writing your own functions? This shouldn't be too terrible to do yourself (it's just a matter of smoothing things so interpolation makes sense). Have a look at the approx() function.

[rxzlmn]

I think that's exactly what I was looking for! Many thanks

Answer 1

[dariober]

Have a look at the deeptools suite in particular computeMatrix.

Something like:

computeMatrix scale-regions -R exons.bed -S coverage.bw -out exons.mat.gz --outFileNameData profile.tab -B 100 -A 100

will put in profile.tab the raw numbers you can use in R.

[rxzlmn]

Thanks, I'll try that!