Kasper Daniel Hansen | 4 Dec 03:10 2012

[package announcement] bumphunter

The 'bumphunter' package has just been accepted to Bioconductor (we
anticipate binary build being available in around 36 hours).  Bump
hunting is a general methodology we have successfully used to analyze
various types of DNA methylation data.  It is more formally described
in [1]. Variants of the code are already included in the charm package
and the bsseq package; we anticipate both package to eventually depend
on this new package.

The package is meant to provide a backend for developers wishing to
utilize (a variant) of the method on their own data.  Usually, some
adaptation needs to be done on any given data type and we envision
various assay specific packages to use this backend (ie. it is not
really meant for casual users).

In addition, the package contain annotateNearest, a function we find
very useful for further analysis of GenomicRanges::nearest.

[ Now that bumphunter is available, we plan to release our version for
the Illumina 450k platform in the minfi package "really soon".  We
just need to finish up the documentation, tidy up the user interface a
bit, and move code from our internal repository into the public
version of minfi. ]

On behalf of the authors,
Kasper D Hansen
Nathan-McKusick Institute of Genetic Medicine
Dept of Biostatistics
Johns Hopkins University

[1] Jaffe, A. E. et al. Bump hunting to identify differentially
methylated regions in epigenetic epidemiology studies. International
Journal of Epidemiology 41, 200–209 (2012).

Bioconductor mailing list
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor