How do you determine the linear range of ChIP and input samples? - (Jan/09/2008 )
Hello,
I am trying to do a ChIP analysis for a transcription factor binding to a promoter element.
When I do linear range analysis, obviously input DNA will give lower Ct cycle than same dilution for ChIP DNA. My question is that how much should be the Ct difference between ChIP and input DNA for further analysis. Most papers say that they do a linear range analysis of input DNA. Should'nt it also be done for ChIP DNA?
I would be thankful if some experienced person can give me some advise on this.
-yogini mathur-
QUOTE (yogini mathur @ Jan 9 2008, 01:11 PM)
Hello,
I am trying to do a ChIP analysis for a transcription factor binding to a promoter element.
When I do linear range analysis, obviously input DNA will give lower Ct cycle than same dilution for ChIP DNA. My question is that how much should be the Ct difference between ChIP and input DNA for further analysis. Most papers say that they do a linear range analysis of input DNA. Should'nt it also be done for ChIP DNA?
I would be thankful if some experienced person can give me some advise on this.
I am trying to do a ChIP analysis for a transcription factor binding to a promoter element.
When I do linear range analysis, obviously input DNA will give lower Ct cycle than same dilution for ChIP DNA. My question is that how much should be the Ct difference between ChIP and input DNA for further analysis. Most papers say that they do a linear range analysis of input DNA. Should'nt it also be done for ChIP DNA?
I would be thankful if some experienced person can give me some advise on this.
I think what would be more important for you to do is:
1) Compare the Ct difference between your IP and input at your region of interest and at some negative control region (where you don't expect your factor to bind).
2) Compare the Ct difference between a mock IP (pre-immune IgG, peptide blocked antibody, or beads alone) and your input at your region of interest and the negative control region.
If the IP/input ratio is significantly higher at your region of interest than at your negative control region AND if this difference between the regions is larger than what you get with the difference in mock IP/input ratio at the two regions (though you would expect that the mock/input ratio to be the same at both regions this is sometimes not the case so you have to control for it), then you can say that your factor is binding your region of interest.
Let me know if I've explained this badly and I can try again.
-KPDE-
Thank you KPDE for the very nice explanation.
I'll keep that in mind.
-yogini mathur-