[BioC] Problem with Expresso and RMA

[McGee, Monnie]

I am sorry.  I know better than to leave out such important information.
 
I am using R on a Macintosh Powerbook G4.  I was using R through an X-windows interface (XFree86 4.3.0).  The version of affy I am running is 1.5.8 and the Biobase version is 1.5.0.
 
________________________________

From: Vincent Carey 525-2265 [mailto:stvjc at channing.harvard.edu]
Sent: Tue 2/15/2005 9:33 AM
To: McGee, Monnie
Cc: bioconductor at stat.math.ethz.ch
Subject: Re: [BioC] Problem with Expresso and RMA



To help you diagnose this we will need to know what version of R
you are running, and what versions of affy and Biobase you have

packageDescription("affy")$Version

it would also be nice if you told us what platform you are using

> Dear BioC Users,
>
> I am analyzing data from an AffyBatch object (that I've called
> "allarray") with the following features:
>
>  > print(allarray)
> AffyBatch object
> size of arrays=1164x1164 features (63515 kb)
> cdf=HG-U133_Plus_2 (54675 affyids)
> number of samples=6
> number of genes=54675
> annotation=hgu133plus2
>
> Note that there are no replicates in this array.  In other words, the 6
> samples are 6 different arrays with 6 different RNA treatments.  I'm
> trying to understand which genes are differentially expressed among the
> 6 arrays.
>
> I wanted to use RMA normalization, so I typed rma(allarray) and
> received the following error
>  > eset1 <- rma(allarray)
> Background correcting
> Normalizing
> Calculating Expression
> Error in sigToEnv(signature, fdef) : Trying to get slot "signature"
> from an object of a basic class ("NULL") with no slots
>
> I tried to do the same thing using expresso and received the same
> error.  Here's what I typed and the response of bioconductor:
>  > eset1
> <-
> expresso(allarray,bgcorrect.method="rma",normalize.method="quantiles",pm
> correct.method="pmonly",summary.method="medianpolish")
> background correction: rma
> normalization: quantiles
> PM/MM correction : pmonly
> expression values: mas
> background correcting...done.
> normalizing...Error in sigToEnv(signature, fdef) : Trying to get slot
> "signature" from an object of a basic class ("NULL") with no slots
>
> I read an archived message with said that it was not a good idea to use
> median polish without replicated arrays, and that it would be better to
> use rlm.  However, I can't seem to get an eset so that I can use rlm.
>
> The funny thing is, just last week, expresso was working fine.  I was
> able to obtain an eset with no errors, using the same AffyBatch with
> the same choices for all the methods.
>
> I would appreciate some help.
>
> Thanks,
> Monnie
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
>