Hi,

I'm trying to (re)barcode some human 1.0 st arrays and am encountering a problem I did not have earlier. I used to use media_polish as summarization method but when I try to run those with my current files I get the following error:
 eset <- frma(affyRaw,input.vecs=hugene.1.0.st.v1frmavecs, summarize="median_polish", target="core")
Error in exprs[match(map[, 2], rownames(exprs)), , drop = FALSE] :
  object of type 'closure' is not subsettable
 

Using "median" as summarization method works fine, but this would alter the workflow we use and I would like to stay consistent.

The following is the script up to the point of error I'm using:
library('oligo')
library(pd.hugene.1.0.st.v1)
library("hugene.1.0.st.v1frmavecs")
library("frma")
data("hugene.1.0.st.v1frmavecs")
data("hugene.1.0.st.v1barcodevecs")
celFiles = list.celfiles()
affyRaw = read.celfiles(celFiles)
eset <- frma(affyRaw,input.vecs=hugene.1.0.st.v1frmavecs, summarize="median", target="core")

Any idea what might go wrong here?

R version 3.2.2 (2015-08-14)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 14.04.3 LTS

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C             LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats4    parallel  stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] hugene10sttranscriptcluster.db_8.3.1 org.Hs.eg.db_3.1.2                   AnnotationDbi_1.30.1                
 [4] GenomeInfoDb_1.4.2                   hugene10stv1cdf_2.16.0               affy_1.46.1                         
 [7] BiocInstaller_1.18.4                 frma_1.20.0                          hugene.1.0.st.v1frmavecs_1.1.0      
[10] pd.hugene.1.0.st.v1_3.14.1           RSQLite_1.0.0                        DBI_0.3.1                           
[13] oligo_1.32.0                         Biostrings_2.36.4                    XVector_0.8.0                       
[16] IRanges_2.2.7                        S4Vectors_0.6.5                      Biobase_2.28.0                      
[19] oligoClasses_1.30.0                  BiocGenerics_0.14.0                 

loaded via a namespace (and not attached):
 [1] affxparser_1.40.0     MASS_7.3-44           GenomicRanges_1.20.6  splines_3.2.2         zlibbioc_1.14.0       bit_1.1-12           
 [7] foreach_1.4.2         tools_3.2.2           ff_2.2-13             iterators_1.0.7       preprocessCore_1.30.0 affyio_1.36.0        
[13] codetools_0.2-14   

--------------------------

A short update.
I just tested the code with an older R and bioconductor version. It works, but gives a couple of warnings about deprecated functions.

Warning messages:
1: 'isIdCurrent' is deprecated.
Use 'dbIsValid' instead.
See help("Deprecated")
2: 'isIdCurrent' is deprecated.
Use 'dbIsValid' instead.
See help("Deprecated")
3: 'isIdCurrent' is deprecated.
Use 'dbIsValid' instead.
See help("Deprecated")
4: 'isIdCurrent' is deprecated.
Use 'dbIsValid' instead.
See help("Deprecated")

The session Info for this run is:

R version 3.0.1 (2013-05-16)
Platform: x86_64-unknown-linux-gnu (64-bit)

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=C                 LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods  
[8] base     

other attached packages:
 [1] frma_1.14.0                    hugene.1.0.st.v1frmavecs_1.0.0
 [3] pd.hugene.1.0.st.v1_3.8.0      RSQLite_1.0.0                 
 [5] DBI_0.3.1                      oligo_1.26.6                  
 [7] Biostrings_2.30.1              XVector_0.2.0                 
 [9] IRanges_1.20.7                 Biobase_2.22.0                
[11] oligoClasses_1.24.0            BiocGenerics_0.8.0            
[13] BiocInstaller_1.12.1          

loaded via a namespace (and not attached):
 [1] affxparser_1.34.2     affy_1.40.0           affyio_1.30.0        
 [4] bit_1.1-12            codetools_0.2-14      ff_2.2-13            
 [7] foreach_1.4.2         GenomicRanges_1.14.4  iterators_1.0.7      
[10] MASS_7.3-26           preprocessCore_1.24.0 splines_3.0.1        
[13] stats4_3.0.1          tools_3.0.1           zlibbioc_1.8.0 

So some things seems to have been changed, that cause problems. For now, I'll just do the first part of the computation in the old R version and copy the remainder over to the newer R version for further computation.
 

Thanks for pointing out this bug. It should be fixed in frma version 1.21.1.

A couple related points:

1. In frma median and median_polish are the same (probe effects have been pre-estimated and frozen). The frma function now mentions this when the user chooses summarize="median_polish". This summarization option is included to mimic the rma function options. 

2. If your next step is to run the barcode function on the frma preprocessed data, you should be using the default version of frma unless you've created your own barcode vectors.

Please let me know if you run into any other problems.

Best,

Matt

Hi Matt,

Thanks for the info. I read that part about median and median polish being the same in the documentation but admittedly I was not sure, whether it meant, that it was the same IF "probe effects have been pre-estimated and frozen" by the user before or whether they were generally the same
About using the defaults: So you should not select a specific summarization option?

Best,

Thomas

If you are planning to use the default version of the barcode function (specifically the precomputed means and variances), you should use the default version of the frma function. This is because the barcode vectors were computed using the default version of frma.