Search
Question: normalization of CHIP-seq samples (ChIP and INPUT)
0
5 months ago by
Bogdan490
Palo Alto, CA, USA
Bogdan490 wrote:

Dear all,

please could I have your advise on the following : shall we have multiple samples from ChIP experiments (where the proteins A, B, C were immuno-precipitated on the chromatin) and INPUT DNA (from the samples A, B, and C), would we normalize them all together (i.e. ChIP + INPUT) in edgeR, for example ?  Many thanks.

-- bogdan

modified 5 months ago by Aaron Lun19k • written 5 months ago by Bogdan490

If you haven't already, you should check out the csaw package, both the User's Guide and the associated publication.

Dear Ryan, thank you for your reply. Here the question is not very much about TMM normalization (implemented in csaw or edgeR); I would be interested to know  whether I could normalize in the same procedure both ChIP samples and INPUT_DNA samples. Sorry if I missed the discussion in the manual. Any suggestions are welcome.

0
5 months ago by
Aaron Lun19k
Cambridge, United Kingdom
Aaron Lun19k wrote:

Well, it's fairly easy to normalize the ChIP sample to the input sample for each protein, see Section 4.2 of the csaw user's guide. However, I would question the sensibility of normalizing ChIP samples for different proteins, as this sounds like a precursor to some sort of comparison of binding intensities between different proteins. The scientific value of this analysis is questionable and the results seem hard to interpret. For example, what does differential binding mean if you have different proteins involved? How do you deal with differences in antibody efficiency?

Dear Aaron, great to hear from you, and thank you for suggestions.

On a side note, thought I shall ask (following some old postings) : can we obtain the normalized counts by using also CPM function (below). Many thanks !

libSizes <- as.vector(colSums(x))

y <- DGEList(counts=x, group=group, lib.size=libSizes)

## NORMALIZATION :

y <- calcNormFactors(y)

## OBTAINING THE NORMALIZED COUNTS with CPM function () :

counts.per.m <- cpm(y, normalized.lib.sizes=TRUE)