So I have been calling peaks on ATAC-seq data using --nomodel and --shift -35 --extsize 75. I have been using-f BED on filtered and shifted (+4 -5 offset recommended) bed files coming from bam files (bedtools bamtobed). As my data is paired-end, I've read it would be a good practice using -f BAMPE. So I filter and shifted my bam files but I'm not sure about the parameters to use when using -f BAMPE.
According to the documentation --shift is going to be 0 because of using BAMPE. What about --nomodel and --extsize? They are not neccesarry either? I wonder if MACS2 is reading TLEN from the bam files and then an extension to be used as fragment size is no longer neccesary. And the bam should be sorted by name or macs2 can recognize mate pairs either way?
Hope somebody can comment on this because and I find myself a little bit confused interpreting this Thank you!