macs2 pileup command gives different values for BAM vs BED

[Aarthi Mohan]

Hi all,

I have generated pileup using the 'macs2 pileup' command. I have tried in BAM and BED mode as well, and the results are different. The experiment is Paired-end RNAPol2 ChIP-seq, and for now I am treating them as single-end for testing pileup values, 

Is there anything 'pileup' will do specially when it detects BAM mode? Like duplciate removal, SAM flag filtering?

Here are the results from

1) BAM: macs2 pileup -i T2_Ser25_L001_trimmed_CP_pos_sorted.bam   --extsize 10 -o T2_Ser25_L001_trimmed_CP_pos_sorted_10_bam.bdg

chr1    0       758     0.00000

chr1    758     768     1.00000

chr1    768     795     0.00000

chr1    795     799     1.00000

chr1    799     805     2.00000

chr1    805     809     1.00000

chr1    809     867     0.00000

chr1    867     877     1.00000

chr1    877     1043    0.00000

chr1    1043    1053    1.00000

2) BED: macs2 pileup -i T2_Ser25_L001_trimmed_CP_pos_sorted.bed  --extsize 10 -o T2_Ser25_L001_trimmed_CP_pos_sorted_10_bed.bdg

chr1    0       745     0.00000

chr1    745     755     1.00000

chr1    755     758     0.00000

chr1    758     768     1.00000

chr1    768     788     0.00000

chr1    788     790     1.00000

chr1    790     795     2.00000

chr1    795     798     3.00000

chr1    798     799     2.00000

chr1    799     800     3.00000

Appreciate any help!

Thanks,

Aarthi

[Aarthi Mohan]

Ok, I figured it out. macs2 pileup will only choose the first mate for extension, when a BAM file is used. 

Cheers,

Aarthi