I have a weird problem I ran macs2 to read peaks and got a list of peaks and peak summits The cmd was:
macs2 callpeak -t IP.bam -c input.bam -f BAM -g 994080837 --bw 100 --keep-dup all --nomodel -n WT -B -q 0.01
I was looking at IGV to see the peaks and I see areas indicated as summits were there is NO coverage. Nothing! Now I know IGV doesn't show all the reads - but I've never seen empty areas! Also, and this bothers me even more, in the areas where there is coverage on both IP and input, there are tines where the input coverage is much higher than IP...
EDIT: The genome is mouse (mm10) - size modified to transcriptome size
I am using the same reference for IGV and the alignment
I double checked the reference, macs cmd and everything I did up to that point
The files I have in IGV is the bam file
I also did pileup - the pileup shows no reads there
The only thing I can think is that the algorithm extends the peak there and sets the summit arbitrarily I am confused
Help!
Anecdotal descriptions are not really helpful, please show plots / images.
There is not much to show The IGV plot is empty - no reads...
Please show a screenshot
https://ibb.co/2qG9grQ
Inconsistent annotations? Are both the same release, same source?
yes, same release, same source, same file
Does not sound like a macs issue, probably inconsistency between the chromosome names that the IGV expects and your BAM. This is a custom genome size you use above there, what is the organism and does IGV support it? Please try to be precise. Are there any reads at all anywhere across the genome in the IGV? In any case, please show a screenshot. Did you load the BAM files or Bigwigs?
The files used in IGV and to create the BAM files are the same file. The genome used is mm10 (mouse) the coverage in other areas is normal, there are peaks that look OK I loaded BAM files
https://ibb.co/2qG9grQ
Check for alignment coordinates in your BAM file and then find the regions in IGV.
I did The bam matches the IGV, the macs results do not