Question

How find correlations between ChIP-seq tracks?

0

Entering edit mode

5.8 years ago

a.rex ▴ 350

I have two ChIP-seq tracks - specifically H3K4me3 and H3K27me3. I have used DeepTools to calculate coverage of each of the marks around annotated TSS's.

How can I correlated between the two tracks at particular points around the TSS? i.e. I want an easy way to find genes that are represented at the TSS by both marks....

Cheers!

ChIP-Seq analysis • 1.8k views

ADD COMMENT • link updated 5.8 years ago by Rashedul Islam ▴ 470 • written 5.8 years ago by a.rex ▴ 350

score 3 · Answer 1 · 2018-07-25

3

Entering edit mode

5.8 years ago

ejm32 ▴ 450

You can use deeptools' multiBamSummary/multiwigSummary with the BED-file sub command and then use plotCorrelation

ADD COMMENT • link 5.8 years ago by ejm32 ▴ 450

score 1 · Answer 2 · 2018-07-25

1

Entering edit mode

5.8 years ago

dariober 14k

See if this script, localEnrichmentBed.py, is of any help. See also TPM for ChIP-seq normalization

ADD COMMENT • link 5.8 years ago by dariober 14k

score 1 · Answer 3 · 2018-07-26

1

Entering edit mode

5.8 years ago

Ian 6.0k

Seqminer (https://sourceforge.net/projects/seqminer/) is old (2012), but allows you to download the genes from the different clusters you observe, something I have not managed to do with Deeptools.

ADD COMMENT • link 5.8 years ago by Ian 6.0k

score 1 · Answer 4 · 2018-07-26

From tracks, if it is wig or bigwig file, it is difficult to find whether "genes that are represented at the TSS by both marks".

If you have access to bam file, call the peaks and to intersect of the peaks to TSS.

To see the correlation, if you have bam file, count the number of mapped read around TSS and correlate.