[BioC] Re: AFFY package for R, image of cell file turned upside down?

[Laurent Gautier]

On Wed, Oct 02, 2002 at 02:09:00PM +0200, Jesper Ryge wrote:
> Hi laurent
> 
> I am using the affy package for R.  I just have one question.  I tried to 
> read a cell file and plot it with image(cel.file).  That works fine, except 
> the image is turned upside down!  It is a rat neuro chip, so  there is a 
> control sequence in the top of the chip that reads "genechip Rat neurochip" 
> if the spike-ins hybridize OK.   But this is now shown in the bottom of the 
> image an upside down?  How is that?
> 
> Jesper Ryge
> 


Hi,

I CC'ed this to bioconductor as it might be of interest to others.

The chip image seems flipped with the Hu6800 too
(my benchmark since it is the first Affymetrix cheap I met)

When a CEL files is parsed, the intensity values are stored in a
matrix (using the X and Y to index the row and column respectively).
You may want to read the affy.pdf vignette around the page 18 to know
more.

However, there is something I realized quite recently (and I am
about to address for the next version of the package).
There is something like follows in the header of
CEL files:
-----------
Axis-invertX=0
AxisInvertY=0
swapXY=0
-----------

I never found files where it was set to one, so the code is not
looking closely at the values of these properties. I would appreciate
reports about CEL where the values change.

In the meanwhile I'll add a little something to make the parser yell
if the values change (just in case). 'should be in BioC by this
week-end.


Hopin' it helps,



Laurent





-- 
--------------------------------------------------------------
Laurent Gautier			CBS, Building 208, DTU
PhD. Student			DK-2800 Lyngby,Denmark	
tel: +45 45 25 24 89		http://www.cbs.dtu.dk/laurent