[BioC] subset eset by row broken

Martin Morgan mtmorgan at fhcrc.org
Sat May 13 16:24:05 CEST 2006


Hi Mark --

The basic functionality works in Biobase

> sessionInfo()
Version 2.3.0 Patched (2006-04-24 r37913) 
x86_64-unknown-linux-gnu 

attached base packages:
[1] "tools"     "methods"   "stats"     "graphics"  "grDevices" "utils"    
[7] "datasets"  "base"     

other attached packages:
 Biobase 
"1.10.0" 
> data(sample.exprSet)
> sample.exprSet[1:5,]
Expression Set (exprSet) with 
        5 genes
        26 samples
                 phenoData object with 3 variables and 26 cases
         varLabels
                sex: Female/Male
                type: Case/Control
                score: Testing Score

so the problem is a little deeper. Other elements of your exprSet may
be malformed. The following should both work:

se.exprs(sample.exprSet)[1:5,]
reporterInfo(sample.exprSet)[1:5,]

My guess is that it's the se.exprs that is causing problems. If these
measures are unimportant, then, in your case,

se.exprs(eset) <- matrix(nrow=0,ncol=59)

will replace what you currently have with a matrix that indicates to
exprSet that it has no information. If the se.exprs *are* important,
then we'll need more information about how the exprSet was created, to
figure out where the malformed entry is coming from.

Hope that helps.

Martin
-- 
Bioconductor

"Kimpel, Mark William" <mkimpel at iupui.edu> writes:

> Tonight one of my functions suddenly stopped working and I have
> traced the problem to not being able to subset esets by row as I
> previously could. Thus, eset[1:5,] yields an error "subscript out of
> bounds", whereas, eset[,1:5] does not. See below for a concrete
> example with output and sessionInfo(). Thanks, Mark
>
>> eset
> Expression Set (exprSet) with 
>         119 genes
>         59 samples
>                  phenoData object with 4 variables and 59 cases
>          varLabels
>                 sampleNames: read from file
>                 Strain: read from file
>                 Region: read from file
>                 Animal: read from file
>> eset[1:5,]
> Error in eset[1:5, ] : subscript out of bounds
>> eset[,1:5]
> Expression Set (exprSet) with 
>         119 genes
>         5 samples
>                  phenoData object with 4 variables and 5 cases
>          varLabels
>                 sampleNames: read from file
>                 Strain: read from file
>                 Region: read from file
>                 Animal: read from file
>> exprs(eset)[1:5, 1:5]
>                WS02R111.CEL WS02R112.CEL WS02R113.CEL WS02R114.CEL WS02R115.CEL
> L07074_at         10.218309    10.270285    10.502409    10.735885    10.374594
> U53873cds_at       6.589952     6.720901     6.503514     6.347453     6.507239
> M85193_at          9.341253     9.159996     9.116925     9.184803     9.170827
> U72349_at          8.169101     7.628983     8.136620     7.878559     8.049834
> rc_AA800265_at     6.603026     6.621253     5.902590     6.437841     6.394334
>> sessionInfo()
> Version 2.3.0 (2006-04-24) 
> i386-pc-mingw32 
>
> attached base packages:
> [1] "splines"   "tools"     "methods"   "stats"     "graphics"  "grDevices" "utils"     "datasets"  "base"     
>
> other attached packages:
> genefilter     rgu34a   multtest   survival       affy     affyio    Biobase    RWinEdt 
>   "1.10.1"   "1.12.0"    "1.9.5"     "2.24"   "1.10.0"    "1.0.0"   "1.10.0"    "1.7-4"
> Mark W. Kimpel MD 
>
>  
> Official Business Address:
>  
> Department of Psychiatry
> Indiana University School of Medicine
> Biotechnology, Research, & Training Center
> 1345 W. 16th Street
> Indianapolis, IN  46202
>  
> Preferred Mailing Address:
>  
> 15032 Hunter Court
> Westfield, IN  46074
>  
> (317) 490-5129 Home, Work, & Mobile
> 1-(317)-536-2730 FAX
>
> _______________________________________________
> 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



More information about the Bioconductor mailing list