[BioC] Questions on getting information from affy's CEL, CDF or 1lq file s

[Dennis Kostka]

On Wednesday 12 March 2003 18:28, Wang, Hui wrote:
> Greetings,
>
> I am exploring the affy packages and I have trouble to get information I
> need (part is because I am not so farmilar with the language and the
> package). I have the following questions, your help will be really
> appreciated.
>
>
> 1)how to get the coordinates of each probe intensities? For example, in the
> .CEL file, there is x, y and intensity columns corresponding to each
> probes. I used read.celfile to read information. If there seems no way to
> get the x, y columns which are the corrdinate of the probes (there is no
> method provided from Class Cel). Are there other functions to use?  Similar
> question to .CDF file.

for the CEL file coordinates:
i would calculate them from the the index of the expression slot of the 
AffyBatch object (call it abo) like this:
        i2xy <- function(i){cbind((i-1)%%nrow(abo) , (i-1)%/%ncol(abo))}
so that i2xy(a) should return the x and y coordinate of a (starting from 0 as 
in the CEL file). important is that a is the index of the intensity you're 
interested in, not hte intensity itself.  x and y might be swapped, but you 
could check that in the file.

> 2). How to extract non-control, non-qc probe information?

i.e. like in the bg.correct.mas function:
        pm.i <- unique(unlist(indexProbes(abo,"pm")))
returns the indices of perfect match probes, where abo is the AffyBatch 
object. for the mismatches you just have to exchange "pm" to "mm".

> 3) for many of the functions listed under affy package, there is no help
> file. For a simple example, read.affybatch, or write.celfile etc. are there
> other place to look?

typing 
        >?read.affybatch 
works at the installation here. i don't know about help files, but if there is 
one called something like "AffyBatch-class.something" that's where i'd look.

dennis