[BioC] Using combineReplicates in computeRunningMedians() function in Ringo

[Joern Toedling]

Hi Paul,

please see the answers below.

On Fri, 12 Feb 2010 16:03:13 +0000, Paul Geeleher wrote
> Hi list,
> 
> I'm trying to use computeRunningMedians() to combine my replicate
> arrays using the modColumn and combineReplicates arguements. My code
> looks something like this:
> 
> pData(X)$Cy5 <- targets$Cy5
> smoothComb <- computeRunningMedians(X, modColumn="Antibody",
> winHalfSize=400, min.probes=3, probeAnno=pA, modColumn="Cy5",
> combineReplicates=TRUE)

Looking at the function call, indeed there is the argument 'modColumn=' twice.
For once, the R error message really indicated the source of the problem.

> Basically I'm attaching my column of interest to pData of X and
> running the smoothing. It however give me this error:
> 
> Error in computeRunningMedians(X, modColumn = "Antibody",
>  winHalfSize = 400,  :  formal argument "modColumn" matched by 
> multiple actual arguments
> 
> or if I remove "modColumn='Cy5'" it gives this error:
> 
> Chromosome 10 ...
>  ... Error: subscript out of bounds

Well, if you remove that, there still is the argument 'modColumn="Antibody"'.
Do you have a column called 'Antibody' in your pData of X?
If not, then that's the reason for this error. I admit that I probably need to
add a check to see if the value for 'modColumn' actually exists before
returning such an error.
If there is a column called 'Antibody' in pData (X), then something strange is
going on here and I would need more information to look into this, starting
with the output of sessionInfo().

Regards,
Joern

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