3 replies to this topic

[andrey1225]

Is it acceptable to present fold enrichment over negative control primers after normalizing for input? Something like this:

2^-[(Ct IP - Ct input)_{target gene} - (Ct IP - Ct input)_{negative control}], where the fold enrichment of the negative control region is set to 1.

I'm running an in vivo ChIP on brain tissue and have a lot of variability between samples if I normalize to input. But, the data cleans up after I normalize to a negative control region not associated with my protein of interested (H3K9ac). I think this accounts for potential differences in cross-linking or elution of DNA, but my advisor is worried that this isn't a publication-quality approach to ChIP analysis. Any advice?

Thanks for your help.

[chabraha]

This is pretty much what I do for my ChIPs........works very well. Specifically, I subtract the %Input from my IgG IP from the %Input of my specific IP...............do this for all regions tested then divide each adjusted % Input by the negative PCR regions adjusted %Input..............doing this way accounts for technical variations across ChIPs from different days............especially the variations you can get when shearing the DNA using the old school sonicatiors which arent excatly most precise tools for shearing.

[andrey1225]

chabraha, on 03 April 2012 - 02:04 PM, said:

This is pretty much what I do for my ChIPs........works very well. Specifically, I subtract the %Input from my IgG IP from the %Input of my specific IP...............do this for all regions tested then divide each adjusted % Input by the negative PCR regions adjusted %Input..............doing this way accounts for technical variations across ChIPs from different days............especially the variations you can get when shearing the DNA using the old school sonicatiors which arent excatly most precise tools for shearing.

Thanks. I'm characterizing how a promoter changes in response to a drug. I'm using a single negative control region for H3K4me3 and H3K9ac, but need to use a different control region for the repressive histone marks H3K9me and H3K27me (I was thinking the actin promoter, since it's constitutively active and likely not associated with repressive histone marks). Is there an issue using two sets of negative control primers for analysis?

[chabraha]

not really...........you just need to show significant enrichment at your positve control region for your mark of interest over the negative control region..........and show that the drug has minimal/no influence on these regions