How to interpret plot from igv_session.xml file

Question

[Chris]

Hi all,

I want to check if a gene is of interest in the open chromatin region so after typing its name, I got this plot with peaks. So do peaks mean open chromatin region? And if we have no peak mean close region? Thank you so much! There are 2 diseased and 2 controls in this data. Could I tell if the gene in diseased or control is more open than the other in this case?

Answer 1

[Istvan Albert]

You need to read more about what the data represents. The peaks that you have most likely represent regions where a protein was bound.

So these regions are an indication of a factor being present at that location. Most often these factors bind to open chromatin.

[Chris]

Thank you for your reply! This is ATAC data but not ChIP seq, so what kind of protein may bind in this case? Would you suggest me a resource to read on this question? Let me know if you need more information to answer if the gene of interest is different in chromatin accessibility between two conditions.

[Istvan Albert]

ATAC stands for "Assay for Transposase-Accessible Chromatin" so these regions are where the transposes can insert adapters.

In all cases, the methods don't actually measure the wether the chromatin is open or closed, but measure a quantity that correlates with it being open or closed

[Chris]

I can see that diseased samples have more peaks on the right side of the plot than the control samples. So the gene in diseased samples is more open than the gene in the control sample, is that correct? Does each peak represent for an exon? I am not sure why we have both bigwig and broadpeak file when open igv_session.xml. Would you explain a little bit about the meaning of them in this case? Thank you so much!

[Istvan Albert]

you would have to perform a differential binding study to compare the samples.

what you state makes sense and is intuitively correct but should be backed up with numbers

search for ATAC seq differential analysis methodologies

also do note that the IGV session is just a visualization method, it is is not an end result, it contains data someone chose to display

[Chris]

Thank you for your comment! I did perform differential accessibility analysis using DiffBind but I got 21k genes which I don't know what is wrong. Could you have a suggestion?