I have developed a DNA methylation algorithm (COHCAP: City of Hope CpG island Analysis Pipeline) that is useful for the following applications:
1) Illumina methylation arrays (450k and 27k)
2) Targeted BS-Seq
In particular, the algorithm is designed to assist with integration of DNA methylation and gene expression (any expression platform should be OK), but the algorithm will also work if you only have DNA methylation data.
The COHCAP algorithm was published relatively recently in Nucleic Acids Research: http://nar.oxfordjournals.org/content/41/11/e117.long
The tool is open-source and the code can be downloaded here: http://sourceforge.net/projects/cohcap/
The COHCAP package contains a manual and demo data, and there are also the following resources to assist with using COHCAP:
COHCAP Wiki: http://sourceforge.net/p/cohcap/wiki/Home/
COHCAP Discussion Board: http://sourceforge.net/p/cohcap/discussion/
If you have the appropriate type of data, I would encourage you to give it a try. I developed the algorithm out a need to have a way to describe methylation patterns for CpG islands (most 450k/27k analysis focuses only on individual CpG sites). Even now, I think IMA is the only other algorithm that can handle region-based analysis for 450k data, and I use this tool for my own analysis (of the relevant data types)
As you may notice, I actively check the discussion board and promptly respond to questions (within ~1 day). If you need any help using COHCAP, please let me know and I will do whatever I can to help you!
