[BioC] if / when the bug was fixed in RMA function eset <- rma()?

[Ben Bolstad]

I wouldn't necessarily classify this as a bug. More as a difference of
opinion between implementers. However, it has been possible since early
January to duplicate the results of expresso via the command rma() using
the parameter bgversion=2 (with slight rounding differences). Or
alternatively via expresso you can use bg.method="rma2" and this will
duplicate the background used in rma() (exactly).

Thanks,

Ben








> Dear all,
> 
> I have a question considering the bug in the RMA function eset <- rma() that was highlighted (by Hu Wei) in the end of January this year. He found that the results from this rma streamline version differed from the results obtained using the individual commands: eset<. expresso(), normalize.method=quantiles, bg.method="rma", pmcorrect.method= "rma" etc. etc.
> 
> I need to know if / when was this bug fixed? (Have used eset <- rma() for analysis just recently....)
> Or do I have to use the eset <- justRMA (Data) command instead?
> 
> Best regards
> Anna

-- 
Ben Bolstad <bolstad at stat.berkeley.edu>