Hi,

I'm a big fan of bsseq because it is exactly the data structure I need for a package I maintain (methylSig) that tests for differential methylation (at CpGs and over regions).

As of the Bioc 3.7 release it appears that a check in the BSseq constructor was added (I don't see a reference to it in the NEWS file).

    if(any(width(gr) != 1))
        stop("'gr' needs to have widths of 1")

It was incredibly convenient to be able to create BSseq objects that aggregated CpGs over regions so that there was still a unified methylation object to access and manipulate.

I would like to request that this check be removed to enable BSseq objects with ranges greater than 1bp. Of course, I understand there may be intra-package consistency reasons for the change, but I figured it was worth a shot to ask because I think there is value in BSseq objects that represent methylation over regions.

A quick note, I also posted a GitHub issue because I wasn't sure which got checked faster (https://github.com/hansenlab/bsseq/issues/68)

Thanks, Raymond

That seems a reasonable request, Raymond. Let's continue the discussion on the GitHub issue (I prefer that for more code-focused issues).