[BioC] Using annotation maps as environments in packages

[Antti Honkela]

Hi Marc,

I think there is something deeper going on in that the overloaded  
"get" generic function for the ProbeAnnDbBimap class does not work the  
way one would assume when called from a function inside a package.  
Unfortunately my understanding of intricacies of R packages,  
namespaces and generic functions does not help any further.

The suggested getAnnMap() does not solve that, as changing the  
function (in the package) to
getAnnotation <- function(gene, map, chip) {
   require(annotate)

   env <- getAnnMap(map, chip)
   show(env)
   return (get(gene, env=env))
}

leads to the same result:
 > getAnnotation("1419728_at", "SYMBOL", "mouse430a2")
Loading required package: annotate
Loading required package: AnnotationDbi
Loading required package: Biobase

Welcome to Bioconductor

   Vignettes contain introductory material. To view, type
   'openVignette()'. To cite Bioconductor, see
   'citation("Biobase")' and for packages 'citation(pkgname)'.

Loading required package: org.Mm.eg.db
Loading required package: DBI
SYMBOL map for chip mouse430a2 (object of class "ProbeAnnDbBimap")
Error in get(gene, env = env) : invalid 'envir' argument
--------------------------------------------------------------------

Note the output of show(env), i.e. we are getting the correct map but  
the get() call still fails.

At the same time doing the same on the command line works:
------------------------------------------------
 > map <- getAnnMap("SYMBOL", "mouse430a2")
 > map
SYMBOL map for chip mouse430a2 (object of class "ProbeAnnDbBimap")
 > get("1419728_at", map)
[1] "Cxcl5"
------------------------------------------------


Antti


On Mar 4, 2010, at 18:42 , Marc Carlson wrote:

> Hi Antti,
>
> You didn't give a lot of detail, but this message probably means that
> you did not load the Annotation package before calling (in the 3rd
> case). You can put a call to require() in your function to make the
> specific annotation package be loaded, but this is probably NOT the  
> best
> way to handle that.  It is probably a much better idea at this point  
> for
> you to look at the getAnnMap() function from the annotate package.
>
> library(annotate)
> ?getAnnMap
>
> As you read the manual page, you will see that getAnnMap is used to
> return the map (what you are calling the envir) so that you can do  
> this
> sort of thing in a more generic way.  That is, you specify the package
> name and the mapping you want, and getAnnMap will try to load any  
> needed
> packages etc. to make sure that its there for you.
>
> Hope this helps,
>
>
>  Marc
>
>
>
>
> On 03/04/2010 10:18 AM, Antti Honkela wrote:
>> Hi all,
>>
>> I have a weird problem when I try to write a function in a package
>> that would use annotation maps (e.g. mouse430a2SYMBOL) as an
>> environment that the function works on its own breaks when moved into
>> a package.
>>
>> The maps can usually be used as environments without problems and the
>> following works fine:
>>> library(mouse430a2.db)
>>> get("1419728_at", env=mouse430a2SYMBOL)
>>
>> The same is true if I define an own function to access the map, and
>> this runs without problems:
>>> getAnnotation <- function(gene, env) { return (get(gene, env=env)) }
>>> getAnnotation("1419728_at", env=mouse430a2SYMBOL)
>>
>> However, if I move the same function definition to an independent
>> package, the same call results in an error:
>>> getAnnotation("1419728_at", env=mouse430a2SYMBOL)
>> Error in get(gene, env = env) : invalid 'envir' argument
>>
>> Any ideas how to fix or circumvent this?
>>
>> (Note that in the sessionInfo below, testpkg is a minimal package  
>> only
>> containing the above function definition.)
>>
>>
>> Best regards,
>>
>> Antti
>>
>> ---------------------------------------------------------
>>> sessionInfo()
>> R version 2.10.1 (2009-12-14)
>> i386-apple-darwin9.8.0
>>
>> locale:
>> [1] en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8
>>
>> attached base packages:
>> [1] stats     graphics  grDevices utils     datasets  methods   base
>>
>> other attached packages:
>> [1] mouse430a2.db_2.3.5 org.Mm.eg.db_2.3.6  RSQLite_0.7-3
>> DBI_0.2-4           AnnotationDbi_1.8.1 Biobase_2.6.0
>> [7] testpkg_0.0-1
>>
>> loaded via a namespace (and not attached):
>> [1] affy_1.24.2          affyio_1.14.0        preprocessCore_1.8.0
>>
>>
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>
>

-- 
Antti Honkela
Antti.Honkela at tkk.fi   -   http://www.cis.hut.fi/ahonkela/