Best Way To Graphically Compare Chip-Seq Or Dnase-Seq Biological Replicates Of Read Enrichment/Peaks Across The Genome?

Question

[margiezilla]

I have bam/bedgraph files of read enrichment across the genome from ChIP-seq and DNase-seq biological replicates. I also have bed files of called peak regions of these samples as well.

I'm trying to figure out the best way to graphically compare the replicates by plotting them on two different axes to see if there the data looks similar across the genome.

I am thinking of graphing them in R but I am not sure if there is already a package in R/Bioconductor that does this or another tool and I do not yet know how to treat genome coordinates in R.

Any advice or suggestions would be greatly appreciated. Thanks

[QVINTVS_FABIVS_MAXIMVS]

You can also do a log transformation on the data to plot them on the same axis. There are functions within R that allow you to plot using two different y axes, but i don't know them on top of my head.

Answer 1

[QVINTVS_FABIVS_MAXIMVS]

If you have bedgraph files you can upload them with the reference genome to IGV (http://www.broadinstitute.org/igv/). The interactive genomics viewer implements a GUI and is pretty easy to navigate from.

If you really want to do it in R you can plot the genome positions on the x axis and then use whatever data are in you bedgraph column on the y axis. I'm assuming this is preliminary, you shouldn't see anything too striking but you should get a general feel of the data and plan next steps to filter it.

Although, if you want to compare ChIP-seq data among treatments you could normalize then rank the signals for each treatment. Like this figure

from Whyte et al. 2013 Cell. Then you can see where most of the reads are coming from, and then take a closer look at those possibly interesting sites.

I bring this up because looking at a global trends genome wise (matters the size of the genome of course) usually exhibits a lot of noise that makes it difficult to tease out the data.

Have Fun!