[BioC] new user trying to create a cdf package for custom array

James W. MacDonald jmacdon at med.umich.edu
Mon Aug 28 22:25:03 CEST 2006


Hi Donald,

Schwartz, Donald wrote:
> I am a new R user. I am trying to create a cdf environment for a
> custom affy array I have installed the cdf package and loaded the
> makecdfenv library:
> 
> utils:::menuInstallLocal() package 'makecdfenv' successfully unpacked
> and MD5 sums checked updating HTML package descriptions
> 
>> library(makecdfenv)
> 
> 
> I have made sure that the cdf file is in the working directory.
> 
> But when I try to make.cdf.package, the response is to want a
> species. When I add the species in a trial, the following is what I
> get:
> 
> proteasea520066f =
> make.cdf.package("protease520066f",species=Homo_sapiens) Error in
> make.cdf.package("protease520066f", species = Homo_sapiens) : object
> "Homo_sapiens" not found
> 
> Note that this array is a very unique dual species array. It has
> probes for both H sapiens and M musculus. Could that be the issue?

No. The issue here is that you are not quoting 'Homo_sapiens', so R 
thinks it is an object in your workspace rather than a text string. When 
it can't find this object it gives the error you see.

Also, you don't say what OS you are on, but from the menuInstallLocal() 
above, I think you are probably on a Windows box. Unfortunately, 
installing a source package (which is what make.cdf.package() makes) is 
a non-trivial exercise that might be less than fun for a new user.

You might rather use make.cdf.env(), which doesn't require you to 
install anything. However it can be tricky as well. Two things to remember:

1.) You have to name your env correctly, or affy won't be able to find 
it. The name will be the _exact_ name of the cdf, without the .cdf part. 
Assuming that there are no capital letters in the name, you would do this:

protease520066f <- make.cdf.env("protease52066f.cdf", species  = 
"Homo_sapiens")

2.) An env is not persistent like a package. In other words, if you 
don't save your workspace (or better yet, save the env alone; see 
?save), you will have to make it every time you need it. If you do 
save() it, you will also have to load() it any time you want to use it.

Best,

Jim


> 
> Does the cdf ("channel file") need to be in a particular format? I
> have two formats, one is humanly readable (Version=GC3.0) when opened
> in notepad and the other is pretty much gibberish past the probe set
> names at the top of the file.
> 
> Don
> 
> 
> 
> 
> 
> ------ Confidentiality Notice: This email message, including any
> at...{{dropped}}
> 
> _______________________________________________ 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


-- 
James W. MacDonald, M.S.
Biostatistician
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109
734-647-5623


**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues.



More information about the Bioconductor mailing list