[BioC] Spot matrix manglings?

[Jean Yee Hwa Yang]

Hi Barry,

If you specify your layout and the data seprately. The marray packages do
assume you are reading your data file in certain order.  But it looks like
you are reading a typical Gal file.   If you are reading a gal file
from GenePix, could you try the following and 

gal <- read.Galfile("gene.txt", info.id=1:6)  ## this way, all row and col
                                                 information are store.
object <- gal$layout
Gnames <- gal$gnames
tmp <- data.frame(GR = maGridRow(object), GC = maGridCol(object), SR =
	maSpotRow(object), SC = maSpotCol(object), 
	maInfo(Gnames))
tmp[1:10,]  ## Check that the output is the same as your input

Alternative, if you send me a small samples, I am happy to try it out to
determine if it's a bug in our package.  

Cheers

Jean

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Jean Yee Hwa Yang			 jean at biostat.ucsf.edu
 Division of Biostatistics,		   Tel: (415) 476-3368
 University of California,		   Fax: (415) 476-6014
 500 Parnassus Avenue, MU 420-W,  San Francisco, CA 94143-0560
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

On Fri, 14 Mar 2003, Barry Henderson wrote:

> I have noticed that BioC is not maintaining the original spot matrix
> coordinates in my original data files (I'm accessing the coordinates via
> maSpotRow and maSpotCol).  Reading the vignettes I see that the
> marrayLayout Class "computes" layout parameters.  Is it normal behavior
> for the marrayLayout methods to reorder spot coordinates within a
> subgrid????   Or am making an error in reading the data into BioC,
> spitting the coordinates out?
>  
> I'm concerned since I was was going to export the data compiled from a
> large set of individual data files using BioC into an ordered file for
> analysis elsewhere.  As I will be doing spatial normalization I am
> concerned about this reordering.  It appears as though it should not be
> a problem since I the ordering appears to be occurring only within
> subgrids.   
>  
> An example of the input and output matrices are pasted below as well as
> the code used to read the data into BioC.
>  
> thanks in advance
>  
> Barry
>  
>  
>  
> 	Starting Grid Location	 			BioC Computed
> Grid locations	 	
> MetaRow	 MetaColumn	 Row	 Column	 GeneID	 	MR	 MC
> Row	 Col	 GeneID	
> 1	 1	 1	 1	 AF191028	 	1	 1
> 1	 1	 AF191028	
> 1	 1	 1	 2	 AF247559	 	1	 1
> 1	 2	 X56062	
> 1	 1	 1	 3	 AF168390	 	1	 1
> 1	 3	 NM_015453	
> 1	 1	 1	 4	 X14212	 	1	 1	 1
> 4	 NM_007204	
> 1	 1	 1	 5	 U91966	 	1	 1	 1
> 5	 XM_031044	
> 1	 1	 1	 6	 X58149	 	1	 1	 1
> 6	 NM_005481	
> 1	 1	 1	 7	 AF198054	 	1	 1
> 1	 7	 AF247559	
> 1	 1	 1	 8	 AF159803	 	1	 1
> 1	 8	 AF168390	
> 1	 1	 1	 9	 AF159801	 	1	 1
> 1	 9	 X14212	
> 1	 1	 1	 10	 X56062	 	1	 1	 1
> 10	 U91966	
> 1	 1	 1	 11	 NM_015453	 1	 1	 1
> 11	 X58149	
> 1	 1	 1	 12	 NM_007204	 1	 1	 1
> 12	 AF198054	
> 1	 1	 1	 13	 XM_031044	 1	 1	 1
> 13	 AF159803	
> 1	 1	 1	 14	 NM_005481	 1	 1	 1
> 14	 AF159801	
>  
>  
> 
> exp.layout <- read.marrayLayout(fname=file.path(datadir, "genes.txt"),
> ngr= 4, ngc = 4, nsr = 12, nsc = 14, skip=0, ctl.col= 6)
> 
> ctl <- rep("Control", maNspots(exp.layout))
> 
> ctl[maControls(exp.layout) != "Control"] <- "normal"
>  
> maControls(exp.layout) <- factor(ctl)
> 
>  
> 
> #Load Sample data from samples.txt
> 
>  
> 
> exp.samples <- read.marrayInfo(file.path(datadir, "samples.txt"))
> 
>  
> 
> #Load gene names from gal file
> 
>  
> 
> exp.gnames <- read.marrayInfo(file.path(datadir, "genes.txt"), info.id =
> 5:6, labels = 5, skip = 0)
> 
>  
> 
>  
> 
> # Read data files....
> 
>  
> 
> fnames <- dir(path = datadir, pattern = paste("*", "Rinput", sep = "."))
> 
> exp.raw <- read.marrayRaw(fnames, path = datadir, name.Gf = "GreenSig",
> name.Gb = "GreenBkg", name.Rf = "RedSig", name.Rb = "RedBkg", layout =
> exp.layout, gnames = exp.gnames, targets = exp.samples)
> 
> 
> 	[[alternate HTML version deleted]]
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor
>