start=end or fixed wiedth

[TJ Sou]

Hi

The methylkit manual and my own results always give start=end position of the methylated region analyzed (or I can specify a fixed for all genome regions). Seems that there is no point in output start and end columns if they are always = or set by user. 

I appreciate your help understanding what is the definition of the start and end in the output. Is it just the first position of the region differentially methylated? or is it the middle position? or is it the first position of the window tested? and is there a way to obtain an estimate of the width of the methylated region tested?

Many thanks for your help

TJ

[Alexander Blume]

Hi TJ,

The default analysis mode in methylKit will be CpG-base-centric, thus having the same start and end position is expected. The default outcome will be a single differentially methylated cytosine (DMC), not a region (DMR). 

However, you may perform regional analysis over regions of interest (https://bioconductor.org/packages/devel/bioc/vignettes/methylKit/inst/doc/methylKit.html#41_Regional_analysis) or genomic tiles (https://bioconductor.org/packages/devel/bioc/vignettes/methylKit/inst/doc/methylKit.html#35_Tiling_windows_analysis) to find DMRs.  You can also aggregate neighbouring DMCs with similar levels of methylation difference into DMRs using segmentation (https://bioconductor.org/packages/devel/bioc/vignettes/methylKit/inst/doc/methylKit.html#57_methSeg():_segmentation_of_methylation_or_differential_methylation_profiles).

Best,
Alex