[BioC] Building BioC 2.2 with R 2.7.0

[Atro Tossavainen]

Thanks all for your help so far.

>    My guess is that Kasper is pretty close, but that the real culprit is 
> RSQLite. But without a complete log of what did and did not install we 
> won't know.
> 
>    So, if you could, please try
>   library(RSQLite)

Loading RSQLite on both Solaris 8 and x86 Linux (RHEL4) produces the
expected results:

Loading required package: DBI
(back to prompt)

>    and
> 
>   library(GO)

Error in library(GO) : there is no package called 'GO'

That's in fact what it says on the other platforms as well where the
whole installation of Bioconductor seemed to work properly.  But now
that GO has finished installing on the other two, I see that

---------
The package 'GO' is deprecated and will not be supported in the future. 

Instead we strongly reccomend that you should start using the 'GO.db' package.
---------

Oh well.  However, trying to execute

library(GO.db)

fails on the two platforms as described earlier:

> library(GO.db)
Loading required package: AnnotationDbi
Loading required package: Biobase
Loading required package: tools

Welcome to Bioconductor

  Vignettes contain introductory material. To view, type
  'openVignette()'. To cite Bioconductor, see
  'citation("Biobase")' and for packages 'citation(pkgname)'.

Loading required package: DBI
Loading required package: RSQLite
Error in sqliteExecStatement(con, statement, bind.data) : 
  RS-DBI driver: (error in statement: no such table: metadata)
Error : .onLoad failed in 'loadNamespace' for 'GO.db'
Error: package/namespace load failed for 'GO.db'

(The error message is identical on both.  Loading GO.db works fine
 on IRIX and x86_64 RHEL4, as stated before.)

>   Depending on the outcome of that, you might then try to install as 
> described on the download instructions - via biocLite (as you may also 
> have version mismatches, which we can't know about from the output given).

I'm installing from scratch.  I always do.  The directory that I
installed R 2.7.0 into has no previous content at all, so although
I've got a working copy of an earlier version of R and associated
Bioconductor somewhere, they're elsewhere and should not be being
looked into at all.  I wouldn't even dream of trying to upgrade over
a known working configuration, that's suicidal. :-)

>    If you did not use biocLite please do, it is intended to help avoid 
> just these issues.

I did precisely that:

source("http://bioconductor.org/biocLite.R")
biocLite()

and NOTHING else to begin with.

> > Here it seems like you GO.db installation is wrong. This is not a 
> > dynamically linked issue, but something else. My guess is that you have 
> > some version mismatch.

It shouldn't be possible.

I'll try to reinstall GO.db on the affected platforms with

biocLite("GO.db")

and see how it goes.  It works now.  I don't understand why, but
it's good enough.

>    This looks a lot like you did not reinstall genefilter, or if you 
> did, then perhaps you did not use biocLite.

I did (assuming "biocLite()" does), and I did.

>    Using ldd on
> R/2.7.0/lib/R/library/genefilter/libs/genefilter.so
>    might reveal what library is missing

All library references in genefilter.so resolve properly.

>   biocLite("genefilter")
> 
>   and then see if it fails to install properly - report back if it 
> fails, with complete details of what you typed and what you got - and we 
> might be able to help.

I'll try that and report back.

-- 
Atro Tossavainen (Mr.)               / The Institute of Biotechnology at
Systems Analyst, Techno-Amish &     / the University of Helsinki, Finland,
+358-9-19158939  UNIX Dinosaur     / employs me, but my opinions are my own.
< URL : http : / / www . helsinki . fi / %7E atossava / > NO FILE ATTACHMENTS