Wig Files- Macs Vs. Igvtools Count In Focal Amplification

Question

[snarkyscientist]

I have been testing WIGs made from the same ChIP-Seq bam file but using two different tools with two different results:

1) Macs with -w -S --space=50 --keep-dup=1

2) igvtools count with -e 160 -w 50

The cell type I'm interrogating has a well-known focal amplification event of about 1 megabase. In this region, the Macs WIG has a plateau of signal across the whole region but the igvtools WIG has distinct peak signals with a slight increase in background noise. Is there a known reason these should differ so drastically, despite using 1 read per position in Macs?

Answer 1

[Ming Tommy Tang]

MACS removes the duplicated reads when make wig file?

[snarkyscientist]

Yep. Chalk this up to my inexperience with igvtools. I somehow thought that "count" was removing duplicates.