[BioC] Problems when access elements in mogene10stprobeset.db

[Martin Morgan]

Peng Yu wrote:
> On Sat, Aug 8, 2009 at 12:37 AM, Martin Morgan<mtmorgan at fhcrc.org> wrote:
>> Peng Yu wrote:
>>> Hi,
>>>
>>> I removed the directory that my own package installations are in, that
>>> is where 'mogene10stprobeset.db' was. The I found that there were
>>> 'mogene10stprobeset.db' in the system already. Therefore, I run my
>>> scripts mogeneID.R, which now gives me the correct results. My
>>> original problem was because that there are packages installed in the
>>> system and packages installed in my home directory. Somehow, there are
>>> some conflicts in between.
>> It is not likely to be a conflict -- R works happily with user and
>> system libraries. Likely the package(s) in your local library were
>> installed incorrectly or were corrupted during use.
>>
>>> The reason that I have my own installations is that our system admin
>>> may not install packages in time. So I have to install the packages in
>>> my own directory, when he is not available. However, he may later
>>> install the same packages in the systems. I am wondering what is the
>>> correct way to maintain the consistencies between the two
>>> installations. For example, two questions are
>>>   1. how to figure what packages are installed in both places?
>>  pkgs = installed.packages()
>>  dups = pkgs[duplicated(pkgs[,c("Package", "Version")]), "Package"]
>>
>>>   2. how automatically uninstall all the packages in my directory that
>>> are already installed in both places?
>>  remove.packages(dups)
>>
>> I would not do this 'automatically'.
> 
> Is there a reason why you would not do this 'automatically'?

Because the problems are unlikely to trace to duplicate packages
installed, but to package installation or other more subtle issues that
I would like to understand.

Martin
> 
> Regards,
> Peng
> 
> _______________________________________________
> 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