bsseq - access for methylation data

Question

[bsmith030465]

I was trying to use the bsseq package to analyze methylation. I was able to make a bsseq object but cannot seem to get the methylation values from the object. For example, I create an object (tdmr) with 4 samples:

tdmr

An object of type 'BSseq' with 27027 methylation loci 4 samples has not been smoothed

The object seems to contain the data for the 4 samples, but when I convert this into a GRanges object I can't see the data:

gr <- granges(tdmr)

gr GRanges object with 27027 ranges and 0 metadata columns:

       seqnames                 ranges strand
          <Rle>              <IRanges>  <Rle>
   [1]     chrX     [3045200, 3045200]      *
   [2]     chrX     [3045201, 3045201]      *
   [3]     chrX     [3045249, 3045249]      *
   [4]     chrX     [3045250, 3045250]      *
   [5]     chrX     [3045279, 3045279]      *
   ...      ...                    ...    ...

[27023] chrX [155614920, 155614920] *

[27024] chrX [155615020, 155615020] *

[27025] chrX [155615021, 155615021] *

[27026] chrX [155615169, 155615169] *

[27027] chrX [155615170, 155615170] *

---

seqinfo: 450 sequences from an unspecified genome; no seqlengths

Have I converted the bsseq object incorrectly? Or do I need to do something else?

thanks!

Answer 1

[hoseintoossi]

The GRanges object is shared between the samples.

 getCoverage(tdmr,type="M")

will give the methylated read count and

getCoverage(tdmr)

will give you overall coverage for each of these samples over these ranges.

sampleNames(tdmr)

gives the sample names.

Answer 2

[bsmith030465]

Thanks, yes I realize I can get it through the getCoverage functions, but should it be in the granges object too (as data)?

...sorry, new to GRanges objects!

Also, when I try to intersect this with genes/exons, I get nothing:

=================

$ library(TxDb.Hsapiens.UCSC.hg38.knownGene)

$ tx <- TxDb.Hsapiens.UCSC.hg38.knownGene

$ genegr <- genes(tx)

$ promUp <- promoters(tx,upstream = 2000)

$ xx <- intersect(granges(tdmr),genegr)

$ xx

GRanges object with 0 ranges and 0 metadata columns: seqnames ranges strand <rle> <iranges> <rle>

---

seqinfo: 455 sequences (1 circular) from hg38 genome

How should I be doing this to get overlaps between the methylation loci and genes?