[BioC] error in limma read.maimages(files, source="imagene")

Gordon K Smyth smyth at wehi.EDU.AU
Fri Feb 3 12:13:30 CET 2006


> Date: Thu, 02 Feb 2006 12:44:24 +0100
> From: Georg Otto <georg.otto at tuebingen.mpg.de>
> Subject: [BioC] error in limma read.maimages(files, source="imagene")
> To: bioconductor at stat.math.ethz.ch
>
> Hi,
>
> I am working with two-colour data generated by Imagene image
> analysis. I encountered a problem in limma 2.4.9
>
> I try to read in intensity data (according to the User's guide section
> 4.4) like this:
>
>> targets<-readTargets()
>> targets
>   SlideNumber  FileNameCy3  FileNameCy5 Cy3 Cy5
> 1           1 slide1-g.txt slide1-r.txt mut  wt
> 2           2 slide2-g.txt slide2-r.txt  wt mut
>> files<-targets[,c("FileNameCy3", "FileNameCy5")]
>> file.path="../quantification/"
>> myfun<-function(x) as.numeric(x[, "Flag"] < 1)
>> RG<-read.maimages(files, source="imagene", path=file.path, wt.fun=myfun)
> Error in read.maimages(files, source = "imagene", path = file.path, wt.fun = myfun) :
>         targets frame doesn't contain FileName column
>
>
> However, calling read.imagene directly works:
>
>
>> RG<-read.imagene(files, path=file.path, wt.fun=myfun)
>
>
> Could this be a bug in imagene 2.4.9?
>
>> sessionInfo()
> R version 2.2.1, 2005-12-20, powerpc-apple-darwin8.4.0
>
> attached base packages:
> [1] "methods"   "stats"     "graphics"  "grDevices" "utils"     "datasets"
> [7] "base"
>
> other attached packages:
>   limma
> "2.4.9"
>
>
> Best,
>
> Georg

Yes it is a bug.  Thanks for reporting it.  I'll fix it for limma 2.4.10.

The problem arises because, from limma 2.4.5, the 'files' argument to read.maimages() is allowed
to be the targets frame, with the file names extracted from the appropriate column.  But with
source="imagene" the 'files' argument is already a data.frame, leading to the confusion.

Best wishes
Gordon



More information about the Bioconductor mailing list