[BioC] RE: Dye Swap Experiments

[Gordon Smyth]

At 08:02 PM 11/07/2003, michael watson (IAH-C) wrote:
>Hi Gordon
>
> >>Also, is there way of generating the matrix of data easily from say a
> >>GenePix file?
> >>I find this difficult as GenePix just has a single co-ordinate for a
> >>sub-grid called
> >>"Block" and there is nothing in the actual file to say how many rows and
> >>columns there are.
>
> >Do you mean generate the layout parameters? If you have a GenePix
> >allocation list (gal) file in your working directory with extension ".gal",
> >then you can type
>
> >gal <- readGAL()
> >layout <- getLayout(gal)
>
> >to extract the layout parameters.
>
>
>No.  What I mean is I have an RGList object from read.maimages.  And I 
>have a layout object defining my layout.  But if I do:
>
>imageplot(RG,layout=lout)
>
>I get an error message.  So how do I generate a matrix that can be used 
>with imageplot (ie how do i generate the FIRST argument for imageplot, 
>given that I have an RGList and an MAList if I want, for example, an 
>imageplot of ratios)???

The first argument in imageplot should be just the data you want to plot, 
not the whole object. RGList and MAList objects are just simply lists with 
added attributes, so you can extract any components you need using the list 
'$' operator. To get an imageplot of the (un-normalized) log-ratios of the 
first array,

     MA <- MA.RG(RG)
     imageplot(MA$M[,1],lout)

To plot the normalized log-ratios you would of course use MA <- 
normalizeWithinArrays(RG,lout) instead of MA.RG(RG).

Cheers
Gordon

>Thanks
>Mick
>
>-----Original Message-----
>From: Gordon Smyth [mailto:smyth at wehi.edu.au]
>Sent: 20 June 2003 17:32
>To: michael watson (IAH-C)
>Cc: bioconductor at stat.math.ethz.ch
>Subject: RE: Dye Swap Experiments
>
>
>At 11:32 PM 20/06/2003, michael watson (IAH-C) wrote:
> >Hi Gordon
> >
> >I'm still playing with limma and just came across another oddity you may
> >or may not be aware of!
> >
> >In the help for imageplot(), your example is this:
> >
> >M <- rnorm(8*4*16*16)
> >imageplot(M,layout=list(ngrid.r=8,ngrid.c=4,nspot.r=16,nspot.c=16))
> >
> >Here we see that the array consists of 8 rows of sub-grids and 4 columns.
> >
> >However, if I execute this command, i get a plot with 4 rows of subgrids
> >and 8 columns...?
>
>I deliberately chose to plot arrays on their side, i.e., the "top" of the
>array is on the left of the screen. That is why the help document says that
>the first spot on the array is on the bottom left of the plot. I choose to
>do this because most computer screens are wider than high.
>
> >Also, is there way of generating the matrix of data easily from say a
> >GenePix file?
> >I find this difficult as GenePix just has a single co-ordinate for a
> >sub-grid called
> >"Block" and there is nothing in the actual file to say how many rows and
> >columns there are.
>
>Do you mean generate the layout parameters? If you have a GenePix
>allocation list (gal) file in your working directory with extension ".gal",
>then you can type
>
>gal <- readGAL()
>layout <- getLayout(gal)
>
>to extract the layout parameters.
>
> >However, even with user input, it would be good to be able to convert from
> >a list to a matrix
> >
> >Thanks
> >Mick
>
>Gordon