Hi Marc, Herve,
I also noticed this behaviour:
library(TxDb.Hsapiens.UCSC.hg19.knownGene)
txdb <- TxDb.Hsapiens.UCSC.hg19.knownGene
seqlevels(txdb, force=TRUE) <- c("chr1", "ch2")
oldLevs <- seqlevels(txdb)
seqlevels(txdb, force=TRUE) <- "chr1"
seqlevels(txdb, force=TRUE) <- oldLevs
Error in .seq
Yes sorry to Mike, Michael, Kasper, Julian, and all those that
have been hit by the refactoring of the subsetting code in
IRanges/GenomicRanges. The main bulk of this refactoring is over
now and things should be much more stable. The major goals of this
refactoring were to simplifiy the code, fi
Yes, works for me now.
Thanks,
Mike
On Sep 18, 2013, at 8:47 PM, Valerie Obenchain wrote:
> Hi Michael,
>
> This looks like an artifact from reorganizing the subsetting code in
> IRanges/XVector. The operation works for me with the most recent versions
> from devel. I'm assuming it now work
Note that currently it's kind of inconsistent anyway because it doesn't
look at the seqlevels that are in use. For example:
library(TxDb.Hsapiens.UCSC.hg19.knownGene)
txdb <- TxDb.Hsapiens.UCSC.hg19.knownGene
Trying to drop chrUn_gl000249 (which I know is not in use):
> seqlevels(txdb) <-
Hi Kasper,
On 09/17/2013 06:05 PM, Kasper Daniel Hansen wrote:
Nice. I thought you had made something like this. What would be
extremely useful is a function that just takes a GRanges and return a
fixed version - less typing, and easier to understand what goes on.
sortSeqlevels() only sorts
Hi Michael,
This looks like an artifact from reorganizing the subsetting code in
IRanges/XVector. The operation works for me with the most recent
versions from devel. I'm assuming it now works for you?
Valerie
On 09/13/2013 02:33 AM, Michael Love wrote:
hi,
I came across an error in trying
I actually considered this, but I opted to do it this way just for the
sake of being consistent (which was my whole mission for implementing
seqlevels in here in the 1st place). Now I could make it more
convenient here and break consistency with how it is used elsewhere, but
what do people pre
Hi Marc,
Wouldn't it make sense to just ignore the 'force' arg when
dropping the seqlevels of a TranscriptDb?
The 'force' argument is FALSE by default and this prevents
seqlevels<- to shrink GRanges or other vector-like objects
when the user tries to drop seqlevels that are in use.
Internally se
I would second this. If we can already figure out the style, then
converting between 3 of the most common ones should be easy, at least for
the major organisms. Robert has also proposed in the past that rather than
converting between names, somehow Seqinfo has a dictionary of aliases, like
most gen
Thanks, I will predict such a function (sorting and fixing) will be
considered an extremely handy convenience function by 136% (114%-165%) of
current and future GenomicRanges users. I think you should try to include
it in the next release.
Best,
Kasper
On Wed, Sep 18, 2013 at 11:59 AM, Hervé Pag
For the archives: this issue has been resolved for the time being, although
the change in R-3.0.2 is different from R-devel.
Kasper
On Tue, Sep 17, 2013 at 10:22 AM, Ramon Diaz-Uriarte wrote:
>
>
>
> kasperdanielhan...@gmail.com writes:
>
> > I have reported this on R-devel. I think it is a bu
11 matches
Mail list logo
