Chipseq data with Mock, Input And IP

Question

[Sudhir Jadhao]

Hello Everyone,

I have ChIPseq data with two control MOCK and Input.

I have called peaks(using MACS2) by two way as given below:

1. IP Vs Input
2. IP Vs Mock

But for downstream analysis I want to select True positive peaks or how should I filter these two peaks conditions for downstream analysis

Answer 1

[Ian]

Given your experimental set up the easiest route is to compare the peaks from comparison 1 against the peaks found in comparison 2. By compare I mean compare the overlap of binding region coordinates, using bedtools intersect or GALAXY intersect.

If you use MACS2 for peak calling and use the SPMR parameter you can visualise the read coverage normalised to millions of reads in UCSC browser or IGB or IGV.

My personal opinion is that the mock antibody (IgG?) is not a good control for ChIP-seq. This was discussed in a very informative ChIP-seq paper; P919 (last paragraph), Kiddler and Zhao 2011 (http://www.ncbi.nlm.nih.gov/pubmed/?term=21934668).