[BioC] Nimblegen CGH ringo to snapCGH

[Joern Toedling]

Hello,

I am not familiar with snapCGH, but looking at your code excerpt and the
function processCGH I see a few likely candidates for generating that,
admittedly non-informative, error message.

If you type the name of the function without brackets like that
> processCGH
you can check in R what it does. Also look at its manual page to see possible
problems. So for example, your call RG.MA is not necessary as the first thing
the function does in the case that the input is an RGList is to call MA.RG.
And your input objects seems to require a data.frame 'genes' with a column
called 'ID'. Furthermore, you may need to adjust the argument
'maxChromThreshold' of the function processCGH for your call.

In general, if you want to find out at exactly what position a function call
breaks, you can use the 'debug' method:
> debug(processCGH)
> ma2 <- processCGH(rg1)
### and when you are done
> undebug(processCGH)

This way, pressing Enter, you go step by step through the function to find out
at which point it breaks.

My guess is that there still could be a problem with the chromosome names, you
may want to check the output of
> table(as.integer(Chr))

And a last remark:
MAList objects can be subsetted like matrices, which allows to you write the
part "removal of non-chromosome probes" in a much shorter way:

> ma <- ma[ grep('CHR', ma$genes[,'ID']) , ]

Sorry for the slightly generic response. I am sure that the authors of snapCGH
will be able to provide more detailed help.

Regards,
Joern



On Tue, 20 Apr 2010 15:42:21 +0200, Chris Fenton wrote
> Has anyone gotten nimblegen CGH data from Ringo into snapCGH.
> 
> I have looked at
> https://stat.ethz.ch/pipermail/bioconductor/2006-October/014724.html
> 
> but it seems that Nimblegen has yet again changed format. For 
> instance I find no file remotely named ORDER_ID3051_SampleKey.txt. I 
> do have a file sample_key.txt """ ARRAY_ID        CY3_SAMPLE_NAME
CY5_SAMPLE_NAME
> 39450A01_290310 patient_f       male_ref
> """
> 
> but I doubt looking at the code this is the right file. 
> I tried a small hack with little luck below.
> 
> Any help greatly appreciated.
> 
> Chris
> 
> """"
> 
> library('Ringo') 
> library('snapCGH') 
> exRG     <- readNimblegen('samples.txt', 'spots.txt')
> ma       <- preprocess(exRG, method='nimblegen', returnMAList=TRUE) 
> ma$design <- c(-1, -1)
> 
> # remove non chromosomal probes 
> probes   <- grep('CHR', ma$genes[,'ID']) 
> ma$genes <- ma$genes[probes,]
> ma$M     <- ma$M[probes,]
> ma$A     <- ma$A[probes,]
> 
> #parse chromosomes from gene name 
> Chr        <- gsub('CHR','', ma$genes[,'ID']) 
> Chr        <- gsub('FS.*','', Chr)
> 
> # stolen from read.clonesinfor 
> indX       <- which(Chr == "X" | Chr == "x")
> indY       <- which(Chr == "Y" | Chr == "y")
> Chr[indX]  <- 23
> Chr[indY]  <- 24
> rg1 <- RG.MA(ma) 
> rg1$genes <- data.frame(rg1$genes, 'Position'=rg1$genes[,
>  'POSITION'], 'Chr'=as.numeric(Chr))
> 
> ma2 <- processCGH(rg1)
> 
> """"
> 
> I get
> 
> Error: subscript out of bounds
> 
> on the processCGH.
> 
> sessionInfo()
> R version 2.10.1 (2009-12-14) 
> i686-pc-linux-gnu
> 
> locale:
> [1] C
> 
> attached base packages:
> [1] splines   grid      stats     graphics  grDevices utils
> datasets 
> [8] methods   base
> 
> other attached packages:
>  [1] snapCGH_1.16.0     aCGH_1.22.0        multtest_2.2.0
> survival_2.35-7   
>  [5] cluster_1.12.1     GLAD_2.6.0         DNAcopy_1.20.0
> tilingArray_1.24.0
>  [9] pixmap_0.4-10      Ringo_1.10.0       Matrix_0.999375-33
> lattice_0.17-26   
> [13] limma_3.2.1        RColorBrewer_1.0-2 Biobase_2.6.1
> 
> loaded via a namespace (and not attached):
>  [1] AnnotationDbi_1.8.1  DBI_0.2-5            MASS_7.3-4          
>  [4] RSQLite_0.8-2        affy_1.24.2          affyio_1.14.0       
>  [7] annotate_1.24.1      genefilter_1.28.2    preprocessCore_1.8.0
> [10] strucchange_1.4-0    tools_2.10.1         vsn_3.14.0          
> [13] xtable_1.5-6
> 

---
Joern Toedling
Institut Curie -- U900
26 rue d'Ulm, 75005 Paris, FRANCE
Tel. +33 (0)156246927