[BioC] Problem with tkWidgets with 2.5.0 on Mac OS Tiger

[Martin Morgan]

John --



John Cornell <cornell at uthscsa.edu> writes:

> Thank you Martin:
> The problem with tkWidgits also occurs with the Windows version.  

Yes, the widget code is complicated to debug and hence difficult to
transform to work with the AnnotatedDataFrame object that replaces
phenoData. As I think I mentioned, this code is not specific to
operating system.

This has come up several times, how is the widget used -- reading
files from disk?  entering phenoData interactively? editing data
frames to be transformed to phenoData? If one of these uses
predominated, and especially if that use was 'reading files from
disk', then this might make it more feasible to update tkWidgets.

> I am encountering problems with other Affy tools, as well. For example, when I
> attempted to demonstrate the MAplot with data I had already analyzed, I got
> the following error message.
>> MAplot(baboon.raw,col=densCols(exprs(baboon.raw)),cex=.6,cex.main=.7)
> Error in assayData(object) : no slot of name "assayData" for this object of
> class "AffyBatch"

This is a version mis-match or, if baboon.raw has been serialized
(saved to disk), an object that needs to have

> baboon.fresh = updateObject(baboon.raw)

to transform it to the current version. It is still possible and
perhaps even likely that packages are expecting to work with the
old-style AffyBatch, but many work with the new style.

> At this point, we are rolling back to 2.4 on all systems until these
> and other unknown bugs are fixed in 2.5 and BioC 2.0
> Cheers,
> John E. Cornell, Ph.D.
> Professor
> Department of Medicine
> Department of Epidemiology and Biostatistics
> University of Texas Health Science Center, San Antonio
> 7703 Floyd Curl Drive
> San Antonio, Texas 78229-3900
> ----------------
> "Expectation, hope, intention towards possibility ... this is not only a basic
> feature of human consciousness, but ... a basic determination within reality
> as a whole."
> -- Ernst Bloch
> -- The Principle of Hope, vol. 1
>
>   --------------------------------------------------------------------------
>
> From: Martin Morgan <mtmorgan at fhcrc.org>
> Date: Fri, 29 Jun 2007 16:23:15 -0500
> To: "Cornell, John E" <cornell at uthscsa.edu>
> Cc: Bioconductor <bioconductor at stat.math.ethz.ch>
> Conversation: [BioC] Problem with tkWidgets with 2.5.0 on Mac OS Tiger
> Subject: Re: [BioC] Problem with tkWidgets with 2.5.0 on Mac OS Tiger
> Unfortunately, this means that tkWidgets cannot be used to
> ReadAffy. This is a limitation of affy / Biobase, and not an issue
> with your OS.  You'll have to provide the information through the
> command line.
> Martin
> John Cornell <cornell at uthscsa.edu> writes:
>> Hi Folks:
>>
>> I upgraded to R 2.5.0 for Mac OS X 10.4. I am using the widget=T option in
>> ReadAffy( ) to read in a set of CEL files. The process starts fine but
>> terminates when I click finish in the first frame. The following error
>> message is printed.
>>
>>
>> Error in read.AnnotatedDataFrame(sampleNames = sampleNames, widget = TRUE) :
>>     sorry, tkWidgets support not available for read.AnnotatedDataFrame
>>
>> Anyone know how to solve this problem.
>>
>> Cheers,
>>
>> John E. Cornell, Ph.D.
>> Professor
>> Department of Medicine
>> Department of Epidemiology and Biostatistics
>> University of Texas Health Science Center, San Antonio
>> 7703 Floyd Curl Drive
>> San Antonio, Texas 78229-3900
>>
>> ----------------
>>
>> "Expectation, hope, intention towards possibility ... this is not only a
>> basic feature of human consciousness, but ... a basic determination within
>> reality as a whole."
>>
>> -- Ernst Bloch
>> -- The Principle of Hope, vol. 1
>>
>>
>>
>>       [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
> --
> Martin Morgan
> Bioconductor / Computational Biology
> http://bioconductor.org
>

-- 
Martin Morgan
Bioconductor / Computational Biology
http://bioconductor.org