How to determine quantitative differences of histone enrichment in enhancer regions

Question

[dally]

I have a single cell line where I have histone ChIP-Seq data for a control and data for the same cell line with treatment.

I then identified possible enhancer regions using the PARE program, and found that my control data came up with approximately 36k possible enhancer sites, while my treatment data came up with 34k.

I then ran bedtools to find sites that were similar between both sets.

I am now interested in finding quantitative differences between histone / TF signal at these enhancer sites. I ran a quick heatmap and metagene analysis and while they look how I expect them too, it isn't very easy to visually see this type of information. Is there a R package or a command line program that allows me to test whether one treated histone mark is more enriched in an enhancer site than my control treated histone mark?

Hopefully this makes sense. I'm still rather new to computational biology. Thank you.

[vivekbhr]

Do you mean you have Control (Untreated) Input and ChIP, and treated Input and ChIP? (so in total 4 samples?)

[dally]

Yes this is correct. Four samples in total.

[vivekbhr]

Nice. then I have an answer..

Answer 1

[vivekbhr]

You can try DiffBind to get enrichment of Treatment sample over the Control sample, for the enhancer regions, taking care of the input as well..

For better visualization of enrichment you can also try DeepTools. Try plotting input normalized Treated sample over input normalized Control sample, for your enhancer regions. (BamCompare -> ComputeMatrix -> plotHeatmap with --perGroup option).

[dally]

I've been giving DiffBind a try, but can't seem to figure it out. Any good tutorials besides the vignette? If not i'll keep trying. I'm not completely sure how to build a model which is obviously important.