ChIP-seq controls question - (Feb/25/2015 )
Hi all,
I've a question about ChIP-seq and how to use the controls.
I've already done the ChIP and it's been sequenced, so far so good.
Now, one way or another the IGG controls are not available. What we are left with are the two antibodies of interest, a histone modification (closed chromatin) antibody, and no-antibody (input undergone the ChIP experiment - apart for the antibody part).
-All of the above in duplicates - also, the chip was done with a magnetic beads-kit.
The big question is, do we use the no-antibody for noise removal ? Or do we use the closed chromatin, knowing that by removing all the peaks coming from it we remove both the noise and possible coinciding peaks?
This is a much debated issue in my lab at the moment, so any help/suggestion/reference to a paper would be extremely appreciated!
Thanks a lot in advance
IP
I'm no expert but we have used the input (no antibody) for noise removal and it has worked nicely. We carried out some basic normilisations using a simplistic workflow we designed but we also used the MACs peak caller to identify statistically valid peaks..we also work with histone proteins. Why don't you just set up two normalisation experiments. One with closed chromatin and one with input and compare peak enrichment in both samples.
Thanks a lot for your reply j47027
I am planning to give both a try, however the experiment has not been done with this cell line and antibodies before-it is a custom made cell line- so I wouldn't have a good enough point of reference to confirm which dataset would be the "correct" one (other than the qpcr validations that would be done at the next stage).
That is why I am looking for a paper that shows using either of the two strategies for noise removal, the no antibody/input or another antibody - like the histone.
You say that you have used the no-antibody in your lab; is this something that you have published? Or did you find a previous publication that made you make this choice?
Thanks again!
IP