[BioC] Working flow_own CEL file reading problem

James W. MacDonald jmacdon at med.umich.edu
Sat May 21 04:31:57 CEST 2005


Junshi Yazaki wrote:
> Hi Jim,
> 
> Thank you very  much for your suggestion.
> I tried make environment again like attached txt as you said.

I think you have a serious misunderstanding about what you are doing 
here. I will try to point some things out, but you really need to be 
doing some close reading of the available documentation.

> 
>>>> source("http://www.bioconductor.org/getBioC.R")
>>>> getBioC("all")

This installs *all* of the Bioconductor packages. This is probably a bit 
excessive, but the main problem here is it appears you think you have to 
re-install the Bioconductor packages every time you do anything with 
your computer. Once you have them installed, they are there for your 
use. This is about the same as installing Microsoft Office every time 
you want to open a Word document.

>>>> library(makecdfenv)
>>>> Library(affy)
>>>> make.cdf.package ("arabidopsistlgF.cdf")
>>>
>>>
>>> And move to Terminal on my Mac,
>>>
>>>> R CMD INSTALL arabidopsistlgfcdf

At this point you have installed the cdfenv, but it is *not* the correct 
name.

> 
> 
> back to R
> 
>>>>  source("http://www.bioconductor.org/getBioC.R")
>>>>  getBioC()

Re-installing just the default packages. Please stop doing this...

>>>>  library(affy)
>>>> Data <- readAffy()
>>
>>
>>  >cleancdfname(cdfName(Data))
>> [1] "arabidopsistlgf4xcdf"

Here you have established that the name the affy package is looking for 
is arabidopsistlgf4xcdf. Note that this is *different* from the name of 
the cdfenv you installed above.

>>  > cleancdfname("arabidopsistlgf4xcdf")
>> [1] "arabidopsistlgf4xcdfcdf"
>>  > library(makecdfenv)
>>  >make.cdf.package("arabidopsistlgF.cdf", 
>> packagename="arabidopsistlgf4xcdf")

Here you are making a package with the correct name, but unfortunately 
not ever installing it.

>> Reading CDF file.
>> Creating CDF environment
>> Wait for about 0 dots
>>
>>>  cel.files=list.files(pattern=".CEL$")
>>>  data=ReadAffy(filenames=cel.files)
>>>  eset <- rma(Data)

Since you have not installed the package with the corrected name, the 
affy package goes to the Bioconductor website and tries to find the 
correct cdfenv. Since this is a custom chip, it can't find anything and 
errors out.


>>
>> Note: You did not specify a download type.  Using a default value of: 
>> Source
>> This will be fine for almost all users
>>
>> Error in getCdfInfo(object) : Could not obtain CDF environment, 
>> problems encountered:
>> Specified environment specified did not contain arabidopsis_tlgF_4x
>> Library - package arabidopsistlgf4xcdf not installed
>> Data for package affy did not contain arabidopsis_tlgF_4x
> 
> Bioconductor - arabidopsistlgf4xcdf not available
> 
> I think this command "make.cdf.package("arabidopsistlgF.cdf", 
> packagename="arabidopsistlgf4xcdf")" does work.
> But according to above message, I can not make cdf environment 
> (Bioconductor - arabidopsistlgf4xcdf not available). Could you please 
> help me what should I do more?

No. You *are* making the cdf environment, but not installing the one 
with the correct name. You need to read the available documentation more 
closely and try to understand what it is you are trying to do. It 
appears you are simply trying different things in the hope that 
something will work.

Creating packages and installing them can be fairly confusing the first 
time you do it. I remember being a bit overwhelmed the first time I 
tried it, and thinking that the documentation was a bit sparse. However, 
  all the information you need to do this is available to you in the 
help files for makecdfenv, and in 'R Installation and Administration'. 
It is well worth the time spent to read (maybe several times) all this 
documentation before going any further, and making sure you understand 
exactly what you are doing.

Good luck,

Jim


> 
> Thank you,
> Junshi


-- 
James W. MacDonald
University of Michigan
Affymetrix and cDNA Microarray Core
1500 E Medical Center Drive
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