Differential binding ChIP-seq/RNA-seq analysis

Question

[dcheng1]

Hi! I want to do some cell-type-specific analysis using ChIP-seq. I want to find a list of ChIP-seq peaks locations that are specific to cell types.

I know there are lots of good tools can compare differential binding, such as Diffbind, diffrep, csaw. But it seems those tools can only compare 2 groups.

I'm just wondering if there is a tool that can do differential analysis for more than 3 groups? Any advice would be greatly appreciated!!

[Devon Ryan]

Where did you see that CSAW can only do 2 group comparisons? There's no reason that I know of that it should have that restriction.

[dcheng1]

I noticed in the user guide of CSAW, they use two groups, each with 2 replicates. So I assume they can only compare 2. But I not sure about this. Thanks for your clarification! I will try CSAW with 3 groups and see how it works.

Answer 1

[Gordon Smyth]

csaw can handle any experimental design that edgeR can, i.e., it can handle any number of groups and any number of explanatory factors.

To quote from the csaw User's Guide (page 41):

A generalized linear model (GLM) will be fitted to the counts for each window using the specified design [22]. This provides a general framework for the analysis of complex experiments with multiple factors. Readers are referred to the user’s guide in edgeR for more details on parametrization.