ChIPSeq: Unusual narrow peaks

[aritakum]

HI,
I am looking at some ChIPSeq data that is rather unusual. The forward and reverse reads form deep thin stacks. Their width is mostly just above the read length of 75 bp. The 'peaks' (pseudo-peaks?) are distributed well across the mouse genome. Have any of you see this? What could be the reason? This experiment is investigating transcription factor binding ....


Thanks!
Jarus

[ffinkernagel]

Hm. Sounds to me like a very low complexity library - I have seen one or two of those, but can only speculate about the reason.
Perhaps very low amounts of input dna with too many PCR cycles?

[ETHANol]

Sounds like a low complexity issue but it would be helpful if you posted some images of what the data looks like.

[aritakum]

RE: ChIPSeq: Unusual narrow peaks

Attached is a screen shot of what I typically see through out the genome

[steinmann]

Yes, thats what PCR artifacts from low complexity libraries look like.

[aritakum]

Thank you all. I think its time to get back to re-doing experiment. I guess, Data analysis-wise there is not much to do...