parameters for normalisation in ChIP - how to represent ChIP data (Jan/06/2009 )
I am trying to work on histone modifications over my gene of interest. I have read many recent papers regarding ChIP data caliberation like %input, No ab, IgG Ab, and Internal control Primers for QPCR normalisation.
But i could not understand from any paper that
1) how %input is calculated,
2) we use actin primers as +ve ctrl in h3k4me2(i.e. modification for transcriptional activation) and -ve ctrl in h3k9me2 (i.e. modification for transcriptional repression),Now what control primers should be used for +ve ctrl in h3k9me2(i.e. modification for transcriptional repression)
Third one is related to personal expertise-
3)How many reactions do we need to actually study ONE modification under ONE situation(i.e. ctrl vs. exptal conditions),so that we dont miss any parameter of analysis mentioned above and any if left? actually I need to study a full panel of Histone modifications and that too over a time course so I need to be well planned.
And yes, currently i am using MaxHaring's paper in PLANT METHODS as baseline for caliberating my Ab and input DNA.Any more suggestions??????????????
Any help is most welcome.
for 1) - are you doing semi-quantitative or real-time PCR? For real-time PCR, calculating the '% of Input' is literally that, i.e. taking the amount of product in your 'input' reaction [Input] and the amount of product in your 'IP' reaction [IP]. %Input = [IP]/[Input]*100. Make sure to take into consideration dilution factors and/or how much material you used in each reaction. For example, if you used 20ul chromatin as Input and 100ul chromatin per IP, then resuspend your DNA in the same volumes and use 2ul each 1:100 'Input' or undiluted 'IP' per PCR reaction, your % of Input would be calculated as: %Input = 1/5 * [IP]/[Input] (*100 / 100, which is 1). I would assume it works the same way for semi-quantitative, just instead of the Ct values you use the arbitrary fluorescence units or whatever the output is.
for 2) - a suitable positive control for heterochromatin of most sorts would be telomeric sequence or most other form of repeat sequences, such as for example satellite 2 (in mammals at least). We also find that the 5S rDNA loci show steady levels of (most) histone modifications...
for 3) - considering that ChIP results are prone to differences owing to a wide range of experimental and biological variables, I'd do at least three independent repeat experiments for each condition/modification/time point. If the standard-deviation is too high, you should probably add a fourth run.
Hope that helps, good luck with your experiments!