All answers are helpful.  Tim I agree that the uncertainty of the state
assignments should be
readily available in principle.  Michael/Martin, bridging to the
AnnotationHub via GenomicFiles
concepts makes sense.  I am not sure we need new infrastructure at the
moment.

On Wed, Aug 12, 2015 at 6:22 PM, Martin Morgan <[email protected]>
wrote:

> On 08/12/2015 02:01 PM, Vincent Carey wrote:
>
>> It seems to me we may need a class to manage related annotation
>> structures.  For example, the chromImpute segmentations of the genome
>> defined for various cell types.  I would like to be able to take a region
>> of the genome (say a SNP) and ask how the state varies across cell types.
>>
>> AnnotationHub will provide access to cell-type specific GRanges but there
>> is no container that I can think of that would coordinate these as
>> analogous
>> to different "samples".
>>
>
> A somewhat different strategy is to manage the cached (or in some cases
> remote, as with bigWig) files and associated ranges, along the lines of
>
>     library(GenomicFiles)      # aarg, 'Biobase::cache &
> AnnotationHub::cache !
>     library(AnnotationHub)
>     register(SerialParam())    # turn off parallel eval for development
>
>     hub = AnnotationHub()
>     hublet = query(hub, c("files i'm", "interested in"))
>     fls = cache(hublet)        # cache (if need) and return local path to
> files
>     ## alternative, e.g., rtracklayer::bigWig supports remote query
>     ## urls = hublet$sourceurls
>
>     rngs = GRanges("chr10", IRanges(c(100000, 200000), width=1))
>     gf = GenomicFiles(rngs, fls) # use this to manage
>
>     MAP = function(rng, fl) import(BEDFile(fl), which=rng)$name
>     REDUCE = unlist
>     xx = reduceFiles(gf, MAP=MAP, REDUCE=REDUCE)
>     mcols(rngs) = simplify2array(xx)
>     ## or SummarizedExperiment(list(my=simplify2array(xx)), rowRanges=rngs)
>
>
>
>> Am I missing something?
>>
>>         [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> [email protected] mailing list
>> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>>
>>
>
> --
> Computational Biology / Fred Hutchinson Cancer Research Center
> 1100 Fairview Ave. N.
> PO Box 19024 Seattle, WA 98109
>
> Location: Arnold Building M1 B861
> Phone: (206) 667-2793
>

        [[alternative HTML version deleted]]

_______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Reply via email to