nomarlizing sequencing depth when use macs2 bdgbroadpeaks

Question

[langya]

Hey

I was wondering when I use macs2 bdgbroadpeaks to call broad peaks on bedgraph file (which is converted from bigwig file that is converted from bam file by our collaborator, don't ask me why they did this and they couldnt find the original bam file), how do I normalize the sequencing depth by million like macs2 callpeak options. And i notice the only output from this is the bed file:

macs2 bdgbroadcall -i fileNameExample.bw.bedGraph --o-prefix fileNameExample.bw.bedGraph

[Devon Ryan]

It's expecting the input file to be appropriately normalized, if I remember correctly. If you know the original depth (or can at least eye-ball it) then you could use wiggletools to scale the file until you get something reasonable.

[Friederike]

they couldnt find the original bam file

can they find the FASTQ files? If not, they'll be in trouble upon submission anyway as they are going to be required by most journals to submit the raw data to a repository

You'll most likely be much better off to re-do the alignment and do things properly