[BioC] flip strand information in GappedAlignments or GRangesList Object

[Hervé Pagès]

Hi Stefanie,

On 03/30/2012 07:09 AM, Michael Lawrence wrote:
> On Fri, Mar 30, 2012 at 5:54 AM, Steve Lianoglou<
> mailinglist.honeypot at gmail.com>  wrote:
>
>> Hi,
>>
>> On Fri, Mar 30, 2012 at 8:42 AM, Kathi Zarnack<zarnack at ebi.ac.uk>  wrote:
>>> Hi,
>>> I think you have to add vector() to strand(ga), since the strand
>> information
>>> comes as Rle which does not work in the ifelse() test directly. At least
>>> this is my experience with GRanges objects.
>>
>> It's true that sometimes you get bitten by Rle's popping up where you
>> didn't expect them when you try and index things and need to do some
>> as.vector() mojo here and there, but in this case it should actually
>> work (depending on your version of R/bioc, I guess).
>>
>> In my case (running R-2.15 RC), the Rle "surprise" doesn't catch you
>> here, but it's still a good idea to keep in mind.
>>
>>
> Thanks, Steve. ifelse() has worked on Rle for a very long time. Btw, this
> issue usually comes up when trying to extract from an ordinary vector with
> [. Since we can't / don't want to define methods on base generics for base
> data structures, we need to define new generics. In the case of [, one can
> use seqselect() as an alternative.
>
> Thanks for pointing that out!

And in case you wonder how to do this on a GRangesList object, the
idiom is:

   (1) Unlist:

         unlisted <- unlist(grl, use.names=FALSE)

   (2) Applies Steve's ifelse() code to 'unlisted' (GRanges object).

   (3) Relist:

         grl2 <- relist(unlisted, grl)

The "unlist/transform/relist" idiom is a very efficient/convenient way
to transform CompressedList objects in general (not only GRangesList
objects, which are only a particular case of CompressedList objects).
However, it does NOT work for any transformation performed in (2),
but only for a transformation where:
     (a) the input/output have the same class
     (b) and they have the same length (and the i-th element in the
         output corresponds to the i-th element in the input, e.g. rev()
         does not satisfy this).

Note that relist() looses the original elementMetadata() but you can
always propagate it "by hand":

   elementMetadata(grl2) <- elementMetadata(grl)

Maybe we should modify the "relist" methods to do this automatically,
after all it propagates the names so why not the elementMetadata...

Cheers,
H.

>>
>> -steve
>>
>>> Best regards,
>>> Kathi
>>>
>>>
>>> On 30/03/12 13:37, Steve Lianoglou wrote:
>>>>
>>>> Hi,
>>>>
>>>> On Fri, Mar 30, 2012 at 6:04 AM, Stefanie<stefanie.tauber at univie.ac.at>
>>>>   wrote:
>>>>>
>>>>> Dear list,
>>>>>
>>>>> having a GappedAlignments or GRangesList object at hand,
>>>>> what is the quickest way to flip strand signs?
>>>>>
>>>>> So for each entry, I want to flip "-" to "+" and vice versa.
>>>>
>>>> Let's assume that your GappedAlignments object is called `ga`, I think
>>>> this should work, no?
>>>>
>>>> R>   strand(ga)<- ifelse(strand(ga) == '+', '-', '+')
>>>>
>>>> To the devs:
>>>>
>>>> For some reason, `example(GappedAlignments)` is throwing the following
>>>> error on me, so I can't actually test at the moment:
>>>>
>>>> Error in elementLengths(rglist(x)) :
>>>>    error in evaluating the argument 'x' in selecting a method for
>>>> function 'elementLengths': Error in .Call(.NAME, ..., PACKAGE =
>>>> PACKAGE) :
>>>>    Incorrect number of arguments (6), expecting 4 for
>>>> 'cigar_to_list_of_IRanges_by_alignment'
>>>>
>>>> sessionInfo() pasted below.
>>>>
>>>> HTH,
>>>>
>>>> -steve
>>>>
>>>> R version 2.15.0 RC (2012-03-24 r58823)
>>>> Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)
>>>>
>>>> locale:
>>>> [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
>>>>
>>>> attached base packages:
>>>> [1] stats     graphics  grDevices utils     datasets  methods   base
>>>>
>>>> other attached packages:
>>>> [1] Rsamtools_1.7.41     Biostrings_2.23.6    GenomicRanges_1.7.40
>>>> [4] IRanges_1.13.34      BiocGenerics_0.1.14
>>>>
>>>
>>> --
>>> Dr. Kathi Zarnack
>>> Luscombe Group
>>> European Bioinformatics Institute
>>> Wellcome Trust Genome Campus
>>> Hinxton, Cambridge
>>> CB10 1SD, UK
>>> tel +44 1223 494 526
>>>
>>
>>
>>
>> --
>> Steve Lianoglou
>> Graduate Student: Computational Systems Biology
>>   | Memorial Sloan-Kettering Cancer Center
>>   | Weill Medical College of Cornell University
>> Contact Info: http://cbio.mskcc.org/~lianos/contact
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>>
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor


-- 
Hervé Pagès

Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N, M1-B514
P.O. Box 19024
Seattle, WA 98109-1024

E-mail: hpages at fhcrc.org
Phone:  (206) 667-5791
Fax:    (206) 667-1319