Good peak finders for methylation data?

[Heisman]

Hey all,

I'm back in the nextgen sequencing game for this summer. Working with methylation data for the first time. I have a series of cases and controls of MBD-seq data and would like to find differentially methylated peaks/regions between the two, and additionally I want to simply be able to call peaks in any given sample. Are there any standard peak callers out there that are good for this type of thing (or is there a good review that covers this to some degree)? I would very much appreciate any input.

[qiangsong]

A good review of genome-wide methylation profiling techniques, including MBD-seq, is

Analysing and interpreting DNA methylation data by Christoph Bock published in Nature Reviews Genetics (http://www.nature.com/nrg/journal/v1...l/nrg3273.html)

As far as I understand, the analysis of MBD-data should be similar to ChIP-Seq, expect
in MBD-Seq one may need to normalize for CpG density etc.

Hey all,

I'm back in the nextgen sequencing game for this summer. Working with methylation data for the first time. I have a series of cases and controls of MBD-seq data and would like to find differentially methylated peaks/regions between the two, and additionally I want to simply be able to call peaks in any given sample. Are there any standard peak callers out there that are good for this type of thing (or is there a good review that covers this to some degree)? I would very much appreciate any input.

[Heisman]

That is most definitely relevant to my interests! Thanks a ton; I'll go through that tomorrow for sure.

If anyone else has feedback feel free to chime in, of course.

[Heisman]

Alright all, I have fairly thoroughly gone through a decent amount of literature for DMR determination. I'm still curious about peak calling.I have case and control samples (ie, cancer vs. healthy), but otherwise no "background" sample.

It seems like a lot of papers use MACS, with some other programs such as USeq possibly being better in review comparisons with ChIP-Seq data. I have thus far only found one peak caller specifically written with methylation enrichment data in mind (http://www.ncbi.nlm.nih.gov/pubmed/?...eq+Technology\), called BALM, and there they show they can call distinct peaks that are convoluted with better resolution than other callers such as MACS. There are other ways to calculate DMRs without calling peaks but I'm specifically interested in peak calling for the purpose of this thread.

For those of you who have done this and are more knowledgeable than me, have you used BALM, or have you found any other caller that would noticeably be more applicable than MACS?

[asiniard]

Hi Heisman, Did you ever find a solution? I'm looking for something similar now. Thanks~!