[Bioc-sig-seq] BSgenome installation error from SVN

Sean Davis seandavi at gmail.com
Fri Feb 19 16:41:15 CET 2010


On Fri, Feb 19, 2010 at 10:26 AM, Nicolas Delhomme <delhomme at embl.de> wrote:
> Hi folks, just to clarify.
>
> 1) I'm extensively working with Rsamtools and this package is in very active
> development (Thanks Martin!) which means that I do not want to wait for the
> package to be available through biocLite (which for me is 1 and 1/2 day
> after any bug correction has been published on the mailing list). So, thanks
> Sean, but I need the SVN checkout , so a life with a production R and
> biocLite will remain an utopia for me... Luckily, I have no problem dealing
> with the dependancies by myself.

Cool.  So just use R-devel (see below) and know that the SVN version
of any package may be broken.  Even in devel, the build process can be
helpful because if a package version is available via biocLite(), then
the version actually built; this is not guaranteed with an SVN
checkout, though most authors are as careful as possible to check in
working code.

> 2) Kasper, maybe I was naive, but well, I expected that patching a
> "production" R with the latest development patch (the R-release.diff.gz)
> would result in the same as using a "development" R. I mean, that is the
> whole purpose of that diff file, right?

Actually, that diff file is just to patch the release version of R; it
does not end up being equal to R-devel, I don't think.  In any case,
he final major version of R that you need (currently) to be truly
using R-devel is 2.11 in your sessionInfo() and yours is reported as
2.10.

> So, either I might have been expecting too much, or (thanks Arnaud for
> making me feel less alone and, well, possibly not that crazy ;-) ) that
> might be an R related issue (r51142 fails for me, r51149 works for you).
>
> Since I've "faith" in R, I got the latest R-release.diff.gz and I've patched
> my R 2.10.1 again. I'll let you know if that fixes the issue. Otherwise,
> I'll just get the R-devel; it's not like the download bandwidth would be an
> issue anyway.

Actually, since bioc-devel is being developed on R-devel (currently
2.11), there is a possibility (especially for packages that use C code
and rely on the underlying R API) that a previous version of R will
not work with a bioc-devel package because of changes in R that occur
between releases (and also in the development of R).

I'm not sure what is going on here, specifically, but Kasper's advice
to stick with R-devel when using Bioc devel packages (via SVN or
biocLite()) is a good rule-of-thumb.  After getting there, it will be
easier to see what issues might remain.

Sean

> ---------------------------------------------------------------
> Nicolas Delhomme
>
> High Throughput Functional Genomics Center
>
> European Molecular Biology Laboratory
>
> Tel: +49 6221 387 8426
> Email: nicolas.delhomme at embl.de
> Meyerhofstrasse 1 - Postfach 10.2209
> 69102 Heidelberg, Germany
> ---------------------------------------------------------------
>
>
>
> On 19 Feb 2010, at 15:30, Kasper Daniel Hansen wrote:
>
>> You should _always_ use R-devel when using Bioc-devel.
>>
>> Kasper
>>
>> On Fri, Feb 19, 2010 at 9:20 AM, Droit Arnaud <Arnaud.Droit at ircm.qc.ca>
>> wrote:
>>>
>>> Hi Nicolas,
>>>
>>> I had the same problem with the BSgenome packages from SVN.
>>> I tried different installation and desinstallation with different version
>>> of BSgenome, Biostrings, IRanges....
>>>
>>> Finally, I resolved my problem by installing the latest build of R-dev
>>> (version.string R version 2.11.0 Under development (unstable) (2010-02-18
>>> r51149).
>>>
>>> Good luck,
>>>
>>> Arnaud.
>>>
>>>> sessionInfo()
>>>
>>> R version 2.11.0 Under development (unstable) (2010-02-18 r51149)
>>>
>>> x86_64-apple-darwin10.2.0
>>>
>>> locale:
>>>
>>> [1] fr_CA.UTF-8/fr_CA.UTF-8/C/C/fr_CA.UTF-8/fr_CA.UTF-8
>>>
>>> attached base packages:
>>>
>>> [1] stats     graphics  grDevices utils     datasets  methods   base
>>>
>>> other attached packages:
>>>
>>> [1] BSgenome_1.15.7    Biostrings_2.15.21 IRanges_1.5.47
>>>
>>> loaded via a namespace (and not attached):
>>>
>>> [1] Biobase_2.7.4
>>>
>>>
>>>
>>>
>>> Le 2010-02-19 à 08:51, Nicolas Delhomme a écrit :
>>>
>>> Hi all,
>>>
>>> Since a few days, I get the following error message when trying to
>>> install BSgenome from my svn checkout directory (R CMD INSTALL --
>>> preclean --clean BSgenome):
>>>
>>>  installing *source* package ‘BSgenome’ ...
>>> ** R
>>> ** inst
>>> ** preparing package for lazy loading
>>> Error in conformMethod(signature, mnames, fnames, f, fdef, definition) :
>>>  in method for ‘start<-’ with signature ‘x="GRanges"’: formal
>>> arguments (check = "NA") omitted in the method definition cannot bein
>>> the signature
>>> Error : unable to load R code in package 'BSgenome'
>>> ERROR: lazy loading failed for package ‘BSgenome’
>>>
>>> The details about the checkout are (svn info BSgenome/):
>>>
>>> Path: BSgenome
>>> URL: https://hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/BSgenome
>>> Repository Root: https://hedgehog.fhcrc.org/bioconductor
>>> Repository UUID: bc3139a8-67e5-0310-9ffc-ced21a209358
>>> Revision: 44766
>>>
>>> My current sessionInfo:
>>>
>>> R version 2.10.1 Patched (2010-02-14 r51142)
>>> x86_64-unknown-linux-gnu
>>>
>>> locale:
>>>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
>>>  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8
>>>  [5] LC_MONETARY=C              LC_MESSAGES=en_US.UTF-8
>>>  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C
>>>  [9] LC_ADDRESS=C               LC_TELEPHONE=C
>>> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
>>>
>>> attached base packages:
>>> [1] stats     graphics  grDevices utils     datasets  methods   base
>>>
>>> other attached packages:
>>> [1] BSgenome_1.15.4    Biostrings_2.15.21 IRanges_1.5.47
>>>
>>> loaded via a namespace (and not attached):
>>> [1] Biobase_2.7.4 tools_2.10.1
>>>
>>> Any idea what is happening there? Instead of patching my R with the R-
>>> release.diff.gz, should I install R-devel?
>>>
>>> Thanks for any help,
>>>
>>> Cheers,
>>>
>>> ---------------------------------------------------------------
>>> Nicolas Delhomme
>>>
>>> High Throughput Functional Genomics Center
>>>
>>> European Molecular Biology Laboratory
>>>
>>> Tel: +49 6221 387 8426
>>> Email: nicolas.delhomme at embl.de<mailto:nicolas.delhomme at embl.de>
>>> Meyerhofstrasse 1 - Postfach 10.2209
>>> 69102 Heidelberg, Germany
>>>
>>> _______________________________________________
>>> Bioc-sig-sequencing mailing list
>>>
>>> Bioc-sig-sequencing at r-project.org<mailto:Bioc-sig-sequencing at r-project.org
>>> >
>>> https://stat.ethz.ch/mailman/listinfo/bioc-sig-sequencing
>>>
>>> _______________________________________________
>>> Bioc-sig-sequencing mailing list
>>> Bioc-sig-sequencing at r-project.org
>>> https://stat.ethz.ch/mailman/listinfo/bioc-sig-sequencing
>>>
>
> _______________________________________________
> Bioc-sig-sequencing mailing list
> Bioc-sig-sequencing at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioc-sig-sequencing
>



More information about the Bioc-sig-sequencing mailing list