[BioC] cellHTS2 and channelNames

[Wolfgang Huber]

Dear Mark

thank you for your feedback and for reporting out this bug. Joseph Barry (the cellHTS2 maintainer) pointed out that the method that implements `channelNames<-` for NChannelSet objects (which is defined in cellHTS2) is incompatible with the way NChannelSet manages itself; this bad behaviour was probably introduced a long time ago (predating Joe's involvement) and apparently not noticed because it is rarely used.

For the time being, I recommend that you do not use the `channelNames<-` method, and rather try to already create the object with names that please you. Let us know if you need help with that.

We will remove the  `channelNames<-` generic and method from the cellHTS2 package. We will see whether it is possible/worthwhile to add repaired versions of these to the Biobase package (which is where the NChannelSet class is defined and where this generic function and method definition really belong.)

Hope this helps, thanks for your feedback and patience again -

	Wolfgang


On May 5, 2013, at 7:53 PM, Mark Dane <markdane08 at gmail.com> wrote:

> Hi Wolfgang,
> 
> Thank you. I am learning this at several levels and I appreciate your help.
> 
> When I made the changes you provided, it renamed the channels within the cellHTS object but then order them alphabetically. For instance,
> 
>> channelNames(x) = c('All#','G1/S','G2','M','Aneuploid','Apoptotic','cPARP','Envision')
>> channelNames(x)
> [1] "All#"      "Aneuploid" "Apoptotic" "cPARP"     "Envision"  "G1/S"     
> [7] "G2"        "M"        
> 
> This ordering shows up in the Plate List tab of the output of writeReport. The file names and their data are in the correct order but the channel names are not. This makes it appear that the channel names are for the incorrect channel data. 
> 
> The boxplots in Plate Summaries are also reordered. Here they do represent the correct data. This seems to be similar outcome to a posting by Juliane Siebourg on 2012-06-05.
> 
> Is gmane.org the best place to  search for previously answered cellHTS questions? I appreciate your direct mail answer. I'm just unsure where the proper response (like this one) should go.
> 
> thank you,
> 
> Mark
> 
> On May 5, 2013, at 9:02 AM, Wolfgang Huber <whuber at embl.de> wrote:
> 
>> Dear Mark
>> 
>> try 
>> 
>> channelNames(x) = c('All#','G1/S','G2','M','Aneuploid','Apoptotic','cPARP','Envision')
>> 
>> See also Section 2.1 of http://cran.r-project.org/doc/manuals/r-release/R-intro.html
>> 
>> Best wishes
>> 	Wolfgang
>> 
>> On May 4, 2013, at 11:03 pm, Mark Dane <markdane08 at gmail.com> wrote:
>> 
>>> Hi,
>>> 
>>> I'm using cellHTS2 for a large siRNA screening project and want to rename the channels. The cellHTS instance I created has the default names:
>>> 
>>>> channelNames(x)
>>> [1] "Channel 1" "Channel 2" "Channel 3" "Channel 4" "Channel 5" "Channel 6"
>>> [7] "Channel 7" "Channel 8"
>>> 
>>> The documentation in version 2.22.0 of Package 'cellHTS2' is: 
>>> 
>>> channelNames<-(object, value)
>>> signature(object = "cellHTS", value = "character"). Replace the channel names in object.
>>> 
>>> The way I interpret this, I think the following code should work:
>>> 
>>> #create new names for the channels
>>> chnames=c('All#','G1/S','G2','M','Aneuploid','Apoptotic','cPARP','Envision')
>>> 
>>> #Assign the new names to the channels
>>> channelNames(x,chnames)
>>> 
>>> The result is:
>>> 
>>> Error in .local(object, ...) : 
>>> unused argument (c("All#", "G1/S", "G2", "M", "Aneuploid", "Apoptotic", "cPARP", "Envision"))
>>> 
>>> Can you please explain the proper way to change the channel names?
>>> 
>>> 
>>> thank you for your help,
>>> 
>>> Mark Dane
>>> Oregon Health and Sciences University
>>> 
>>> 
>>> Version info:
>>> version.string R version 3.0.0 (2013-04-03)
>>> cellHTS2_2.24.0
>>> 	[[alternative HTML version deleted]]
>>> 
>>> _______________________________________________
>>> Bioconductor mailing list
>>> Bioconductor at r-project.org
>>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>>> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>> 
>