Peak splitting with MACS2

[Mariam Okhovat]

Hi,

I am interested in splitting my ChIP-seq peaks and I am using macs2 to call peaks. The first issue is that my peaks are not reported as a bed file, but rather an xls file. Also, I don't get a wig directory and when I try to run the --wig or -w command it is not recognized. How can I create these files in order to run the peak splitter.

As an alternative I have tried the "searching for subpeak summits" argument but I get a strange output (copied below). Please note the redundant end and start locations I get. 

Is this normal?

chr start end length abs_summit pileup -log10(pvalue) fold_enrichment -log10(qvalue) name

DP001225 37110 38802 1693 37196 23.00 3.88662 2.17128 1.80028 /scratch/02030/mariamok/DP/me1-stampy-filter.bam.summits.150_peak_1a

DP001225 37110 38802 1693 38159 29.00 7.66922 3.03233 4.64792 /scratch/02030/mariamok/DP/me1-stampy-filter.bam.summits.150_peak_1b

DP001225 37110 38802 1693 38553 20.00 3.44985 2.13192 1.49557 /scratch/02030/mariamok/DP/me1-stampy-filter.bam.summits.150_peak_1c

DP001225 39140 39926 787 39224 24.00 5.18357 2.53800 2.72921 /scratch/02030/mariamok/DP/me1-stampy-filter.bam.summits.150_peak_2a

DP001225 39140 39926 787 39470 22.00 4.28164 2.33496 2.07672 /scratch/02030/mariamok/DP/me1-stampy-filter.bam.summits.150_peak_2b

DP001225 49544 49899 356 49641 57.00 6.03728 1.95897 3.40313 /scratch/02030/mariamok/DP/me1-stampy-filter.bam.summits.150_peak_3

DP001225 92818 94406 1589 92897 139.00 4.06026 1.38781 1.94649 /scratch/02030/mariamok/DP/me1-stampy-filter.bam.summits.150_peak_4a

DP001225 92818 94406 1589 93607 600.00 10.28946 1.31543 6.70696 /scratch/02030/mariamok/DP/me1-stampy-filter.bam.summits.150_peak_4b

DP001225 92818 94406 1589 93980 639.00 40.96196 1.79016 35.43500 /scratch/02030/mariamok/DP/me1-stampy-filter.bam.summits.150_peak_4c

Thanks a lot,

Mariam

[Ian Donaldson]

I take it you got this using --call=summits?

The redundancy comes from the same region having multiple summits.  Have you looked at them on a browser to see if they make sense?

Ian

--
You received this message because you are subscribed to the Google Groups "MACS announcement" group.
To unsubscribe from this group and stop receiving emails from it, send an email to macs-announcem...@googlegroups.com.
To post to this group, send email to macs-ann...@googlegroups.com.
Visit this group at http://groups.google.com/group/macs-announcement.
For more options, visit https://groups.google.com/d/optout.

[Mariam Okhovat]

Hi Ian,

Yes I got them using -call-summit.

I'm not sure what you mean by "the same region having multiple summits"? You mean MACS finds multiple peaks around that area and maps the summit to the same exact base? I have looked at the results on UCSC and it looks ok to me as far as I can tell. It's a broad peak with some local summits here and there.

BTW, have you ever tried the peak splitter (described here: https://groups.google.com/forum/#!searchin/macs-announcement/peak$20split$20macs/macs-announcement/YnFNgmxLzog/bt7zObF7ExIJ) with MACS2?

M

[Ian Donaldson]

I meant that one of the regions found by MACS, e.g. DP001225:37110-38802 of length 1693 contains three sub peaks at 37196 'peak_1a' , 38159 'peak_1b' and 38553 'peak_1c'.

I thought that Mali's script was incorporated into MACS2 to achieve this, but I might be wrong there.

Ian

[Mariam Okhovat]

Oh, I didn't even notice the difference in absolute summits. I was thrown off by the similar peak range. Ok, that clarifies.

Do you know what argument will run the peak splitter in MACS2?

Thanks a bunch,

M