[BioC] class label

Sat Dec 18 19:37:44 CET 2010

On 12/18/2010 09:45 AM, Peevi Ijkl wrote:
> Hello martin
> i know that multtest works..but for it to work correctly i need to
> create a class label right..which am not able to ..i am attching the
> original txt file that i created..it is tab delimited..
> ID    disease
> GSM469509.CEL    A
> GSM469510.CEL    A
> GSM469511.CEL    A
> GSM469512.CEL    A
> GSM469514.CEL    A
> GSM469518.CEL    A
> GSM469520.CEL    A
> GSM469522.CEL    A
> GSM469523.CEL    A
> GSM469508.CEL    H
> GSM469513.CEL    H
> GSM469515.CEL    H
> GSM469516.CEL    H
> GSM469517.CEL    H
> GSM469519.CEL    H
> GSM469521.CEL    H
> A stands for atopic,H for healthy normal
> now i want to create a class label..can u help?

Hi peevi -- I know you're trying, and in some ways I'm being less
helpful than I could be. But this isn't helpful -- attachments are
removed by the mailing list software, you could have read this data in
through several different ways (especially stringsAsFactors=TRUE or
FALSE could be important), it doesn't relate to your original question
(there is no column 'Diabetic', and no variable that it would seem
reasonable to compare to "y") and we don't know what function you are
trying to use or from which library it is supposed to come from.

To return to my previous suggestion, here's what I'd guess, cut and
pasted from my R session

> library(multtest)
> lbl <- pd$Diabetic
> m <- matrix(runif(10 * length(pd$Diabetic)), 10)
> str(lbl)
 Factor w/ 2 levels "n","y": 2 2 2 2 1 1 1 1
> res <- mt.teststat(m, as.integer(lbl=="y"))

but what do you actually do and see? It could also be that you are using
an older version of software (though I doubt that's the case here) and
so we encourage (a) making sure that you are using current software by
following the instructions to update packages at

  http://localhost:3000/install/index.html#update-bioconductor-packages

and (b) providing the output of sessionInfo() after the error has
occurred, and in an R session that is used only to provide the minimal
example. Here's mine

> sessionInfo()
R version 2.12.0 Patched (2010-11-21 r53647)
Platform: x86_64-unknown-linux-gnu (64-bit)

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8
 [5] LC_MONETARY=C              LC_MESSAGES=en_US.UTF-8
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C
 [9] LC_ADDRESS=C               LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

other attached packages:
[1] multtest_2.6.0 Biobase_2.10.0

loaded via a namespace (and not attached):
[1] MASS_7.3-9      splines_2.12.0  survival_2.36-2 tools_2.12.0

Martin

> 
> peevi
> 
> 
> ------------------------------------------------------------------------
> *From:* Martin Morgan <mtmorgan at fhcrc.org>
> *To:* Peevi Ijkl <ipeevi at yahoo.com>
> *Cc:* bioconductor <bioconductor at stat.math.ethz.ch>
> *Sent:* Sat, December 18, 2010 11:02:23 AM
> *Subject:* Re: [BioC] class label
> 
> On 12/18/2010 07:56 AM, Peevi Ijkl wrote:
>> Hello
>> ok my phenodata(target.txt) looks something like this
>>
>> 1.CEL  A
>> 1.CEL  A
>> 3.CEL  A   
>> 4.CEL  H
>> 5.CEL  H
>> 6.CEL  H
>>
>>  i loaded the above into my R using
>> library(affy)
>> library(simpleaffy)
>> eset<-read.AnnotatedDataFrame("target.txt", header=T, row.names=1,
> sep="\t" )
>>
>> to view my data:
>> pData(eset)
>>
>> then i did normalization on it
>>
>> Rmaeset<-rma(eset)
>>
>> now i want to perform multtest..but first i have to create class
> label..can
>> anyone tell me how do i do tht??
> 
> Hi peevi -- You need to be more helpful to us, and stick with your
> original question. Likely the problem is that your pd$Diabetic variable
> is not what you think it is, e.g., a factor when you're expecting it to
> be a character, or of the wrong length, or ???. But how are we to know?
> You have to show it to us. And you can't really say 'multtest doesn't
> work', because we know that
> 
>   example(multtest)
> 
> *does* work. And since even simple emails like this can take five
> minutes to write, you have to make it easy for us to see what your
> problem might be, so no sense giving us a dozen lines of code -- reduce
> your real problem to something that shows what the issue is, in just a
> few lines, and nothing more.
> 
> 
> Martin
> 
>>
>> peevi
>>
>>
>>
>>     
>>     [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at r-project.org <mailto:Bioconductor at r-project.org>
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
> 
> 
> -- 
> Computational Biology
> Fred Hutchinson Cancer Research Center
> 1100 Fairview Ave. N. PO Box 19024 Seattle, WA 98109
> 
> Location: M1-B861
> Telephone: 206 667-2793
> 

-- 
Computational Biology
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N. PO Box 19024 Seattle, WA 98109

Location: M1-B861
Telephone: 206 667-2793