expectations of high quality chip seq experiments

[hawainpanda]

I will be doing some chip-seq expeirments in the near future and wanted to know about what are some expected controls that are required to produce high quality chip seq results.

one thing in particular I've noticed is that people will compare chipseq with knockdown of the particular protein being used for chip. Therefore in kd samples, peaks are greatly decreased. On the other hand I've seen many other people who do not do this who pub in high end journals.

KD of the particular protein I'm studying is extremely difficult and not easily reproducible. and thus would be unlikely feasable for me to do chipseq.

On the other hand I was thinking during validation of certain peaks, I could try kd my protein and using qpcr to determine the validity of these peaks.

Anybody have suggestions/
thanks!

[mudshark]

comparing to target knockdown is not at all a standard control in ChIP.
you should rather sequence the chromatin input and ideally perform IP with two different antibodies.

[ETHANol]

mudshark is 100% correct on his three point.
1) knock-down is not a standard control
2) Control to input signal
3) Use two different antibodies to two different epitopes (critical). Only count peaks that appear in both antibodies. I see some papers were authors only use one antibody. This is not good.