[BioC] RMA and NA values

[Laurent Gautier]

Benjamin Haibe-Kains wrote:
> Hi,
> 
> In order to compare different types of affy chips, I construct manually 
> AffyBatch object with NA values for the probes which are not in common.
> Unfortunately RMA doesn't deal with NA values. Is it possible to let the 
> NA values during the normalization in the next release of RMA ? Or any 
> kind of NA value management ?

Since you go for the creation of chimeric AffyBatch objects, what about 
creating an appropriate chimeric cdfenvs as well ?
This means that only the probes you are interested in (i.e., non NA) are 
found such a cdfenv.

(the packs altcdfenvs, makecdfenv, and matchprobes are likely contain 
code or examples or interest)



L.


> 
> Thanks a lot
>