ChIP-seq heatmaps using binding site of an specific protein as a reference

Question

[piecesofalegohouse]

Hello, I'm trying to realize if it is possible to perform a heatmap with ChIP-seq data of different proteins (i.e. A, B, C) using the binding site of another protein as a reference (i.e. X), this in order to check if proteins A, B, C correlate or are enriched around the binding site of X. I have only created TSS plots with deepTools, but I haven't figured out how to generate the heatmaps that I need.

I will appreciate your help a lot.

[ATpoint]

Alternatively, if you want to work in R/BioC: Using EnrichedHeatmap for visualization of NGS experiments

Answer 1

[colin.kern]

You can use the same command from deepTools (plotHeatmap) that you used to make the TSS plot. When you run computeMatrix reference-point to generate the matrix that plotHeatmap takes as input, you can specify a bed file for the binding sites you want to generate the heatmap around. I'm not sure what method you're intending to use to get the binding sites. It could be peak summits from a peak caller using ChIP-seq for those proteins, or the result of some motif finder such as HOMER or Meme.

[piecesofalegohouse]

Thak you so much! Let me see if I understood, I must generate a file containing the binding sites of protein X, and use it as a reference with all of my other protein files

[colin.kern]

Yes, that's correct.