[BioC] Coverage Question: GenomicRanges

Lakshmanan Iyer laxvid at gmail.com
Fri Mar 9 18:29:07 CET 2012


I am calculating the coverage on a subset of a gapped aliment that obtained
using "subsetByOverlaps" between a "GappedAlignments" and a "GRanges" object
#> class (Dendaligns)
#[1] "GappedAlignments"
#attr(,"package")
#[1] "GenomicRanges"
#> class (p3UTRsChr)
#[1] "GRanges"
#attr(,"package")
#[1] "GenomicRanges"
DendfiltData2 <- subsetByOverlaps (Dendaligns, p3UTRsChr);
#> class (DendfiltData2)
#[1] "GappedAlignments"
#attr(,"package")
#[1] "GenomicRanges"
#
# Find coverage
Dendcov <- coverage(DendfiltData2)
#
# Find islands of coverage with min value of 1
Dendislands <- slice (Dendcov, lower=1);
#
# convert the island ranges in to a data frame
DendislandsData <- as.data.frame (ranges(Dendislands[[chr]]) );
#
# extract the scores of the coverage in a list
Dendisland_scores <- viewApply (Dendislands[[chr]], function (x)
as.integer(x))

When I look at the scores, the first island contains a series of 1 (45 of
them in my case).
However, when I visualize the same regions in IGV with the raw bam file
(see attached picture), I see that the counts are > 1 for bases after first
10 bases.

I am wondering where the counting becomes different from summing up the
reads at any position!

Thanks for any thoughts
-best
-Lax
-------------- next part --------------
A non-text attachment was scrubbed...
Name: igv_panel.png
Type: image/png
Size: 6317 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/bioconductor/attachments/20120309/93fcaa7e/attachment.png>


More information about the Bioconductor mailing list