ChIP-exo Tutorials

Also "ChIP-nexus"

Read more:

ChIP-Seq/HT-ChIP/ChIP-exo/Mint-ChIP (Enseqlopedia) ChIP-exo 1.0 (Rhee 2012) ChIP-exo 5.0 (Rossi 2018)

Which tutorial should I start with?

We like for users unfamiliar with ScriptManager to start with the heatmaps and composite plots tutorial to get a quick feel for the types of figures you can make with your data. If you have not called peaks, eventually you may need to in order to have RefPTs for your publication Tag

Which RefPTs should I use?

It is common practice to use peak calls for general heatmap generation but with ChIP-exo being base-pair resolution data and peak callers having imperfect resolution, a more exact and high resolution reference point can give you more interesting insights into your data.

For ssTFs, there are usually weblogos representing the position-specific nucleotide preferences of the TFBS. If you are ChIP-ing one of these targets, you have the opportunity to get some higher resolution RefPTs by obtaining every instance of some motif in the genome (use FIMO) to use as a RefPT for your heatmap and composite plots. You can even intersect your motif instances to keep the ones that overlap for a peak call to get the set of "bound" sites (obligatory disclaimer on thresholding caveats here). For ChIP-exo data, this will give you a better chance at observing the actual bp-resolution crosslinking profiles of a target.

How should I handle replicates?

This is an important question as there are many options available to you but some may create some circular reasoning when you test your hypothesis. One thing we like to do to avoid this is to call peaks or do whatever manipulations you want to build your RefPTs all based on one replicate. Then when you pileup and observe signal enrichment of the data, you can plot a different replicate that was held out from any RefPT-building to fairly perform comparisons.

In other instances, you may need as much signal as you can get and you could consider merging your BAM files.