Overlapping cut and tag H3k27 peaks and ATAC-Seq peaks
1
0
Entering edit mode
4 weeks ago
new ▴ 10

Hello

First post & new to this. I have peaks from cut & tag for H3K27ac (SEACR) and also for ATAC-Seq (MACS2)

I'd like to overlap them to get a file of overlapping peaks.

What is the best way to do this?

Thank you

ATAC-Seq cutNtag • 486 views
ADD COMMENT
0
Entering edit mode

In R one would use findOverlaps() or subsetByOverlaps() from the GenomicRanges package.

ADD REPLY
1
Entering edit mode
ADD COMMENT
0
Entering edit mode

Ok. Thanks. Would you increase the size of the peaks first using bedtools slops?

ADD REPLY
0
Entering edit mode

I would follow a standard ATAC-seq/ChIP-seq pipeline and include that within the MACS2 peak calling as discussed here:

ATAC-seq peak calling with MACS

ADD REPLY
0
Entering edit mode

Yes, I have done that. And used SEACR for the Cut & Tag peaks. Called narrow peaks for ATAC-Seq. Intersect with bedtools is not an issue, but don't know if there is anything I am not thinking of re broad peaks vs narrow peaks. Thanks anyway.

ADD REPLY
0
Entering edit mode

One consideration is the amount of overlap you're looking for. Default behavior will take any amount of overlap. You can change this to require a greater amount or to allow a gap between the peaks.

I prefer a stricter threshold for overlapping peaks, where in this case you might require at least 25-50% of the ATAC peak is overlapped.

Also with bedtools pay attention to what you are outputting. In general, if you provide the ATAC-peaks as -a and K27ac as -b then the output will be the original ATAC-peaks that are marked by K27ac. I imagine this is your desired output, but if you wanted to see K27ac peaks that contained accessible regions, then you would input the K27ac peaks as -a.

Also, the default behavior is to output a line from -a for every overlap found in -b. Sometimes this isn't really an issue, but for example if you had a really large ATAC peaks covered by two separate K27ac peaks (assuming ATAC peaks are input as -a), then you would get the same ATAC-seq peak twice in the output file. I usually use the -u flag to keep a one line per peak ratio.

This is all obvious in their documentation which is comprehensive, and sorry if it's all obvious to you, but it wasn't obvious to me originally.

ADD REPLY
1
Entering edit mode

Thank you - it wasn't obvious to me, and is very helpful. Thanks!

ADD REPLY

Login before adding your answer.

Traffic: 2843 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6