Differential Histone Chip-Seq Analysis Without Replicates

Question

[Diana]

Hi everyone,

I have histone chip-seq data for 3 developmental stages. I want to find out the differential histone acetyl marks between these 3 stages. I've been trying to work with DiffBind but unfortunately I have no replicates so I cant use that. Does anyone have any suggestions as to how I can do this without replicates?Are there any tools out there or packages in R that deal with this problem. I know there is Bedtools but anything else?

Thanks a lot!!!!

Answer 1

[shenli.sam]

You can use diffReps. It is an automated pipeline for ChIP-seq differential analysis. Since you don't have replicates, you can choose G-test (recommended) or Chi-square test. diffReps has implemented them all.

Answer 2

[sirenbeserker]

This page: Get the most out of your chip-seq experiments, contains a list of quantitative tools that look like what you are after. I'm doing similar things with my ChIP-seq data, good luck!

Answer 3

[Ian]

I realise this is an old question, but one i have recently been considering.

A recent article summarised the problem nicely for me and highlights the two main strategies for identifying regions to compare in differential binding analysis.

Use of windows:

E.g. diffReps

This does not require peak calling to have been perfomed. As histone peak profiles can be narrow and broad and everything in between the use of windows seems to be a better option.

Use of pre-called peaks:

E.g. MAnorm

If you have confidence in the peak calling results then this is a popular option. To my mind it is not so useful for diffuse histone marks were peak calling is often not as accurate as with transcription factors or narrow histone marks.