[BioC] how to revert to an older limma version?

[Maya Bercovich]

Hi Maya --

I am glad that Sys.setlocale("LC_ALL", "en_US") allowed you to read
most files.

I apologize for the delayed response, I was away with no access to my email or the internet.


>> RG1 <-read.maimages(targets.RG1$Filename,columns=list(Rf="F635Median",Gf="F532Median",Rb="B635Median",Gb="B532Median"), path=path)
>
> Read /mnt/lifestore/Biotech/DannyS_Shared/Users/Revital/ssdp//B12Z0471_A.gpr
>
> (...all the other .gpr files are read, until: )
>
> Read /mnt/lifestore/Biotech/DannyS_Shared/Users/Revital/ssdp//B12J0504_A.gpr
>
> Error in readGenericHeader(fullname, columns = columns) :
>
>         Specified column headings not found in file

I guess the obvious question is: are the column names present in the
.gpr file? The files are processed in the order given in
targts.RG1$Filename, so you can figure out which file is causing
problems, open it in a text editor and look for the column names that
you provide. Do NOT use your text editor's 'Save' function, as this
could make unexpected changes to the file.

The problem is now solved. It turned out that the file that caused the problems was originally created in Windows. It was re-created in a Linux text editor and it's fine! 
To sum up, the first problem was an R locale problem and the second a simple "dos to Unix".
Thank you so much for your help, it saved our PhD student's project.


Two e-mail tips:

1. Use 'reply all' or similar to reply to the list. This keeps the
email in 'threads' that make it possible to follow different issues
(see the archive at
https://stat.ethz.ch/pipermail/bioconductor/2007-July/thread.html for
examples of this)

2. With 'Reply all', most email clients will insert a '>' at the start
of each line. This is very helpful to follow different conversations,
for example:

>> First email
> Second email, responding to first email
Third email, responding to first


Thanks for that, too.

Best,

Maya

Best of luck,
-- 
Martin Morgan
Bioconductor / Computational Biology
http://bioconductor.org