DEseq2 load failure after successful install
1
0
Entering edit mode
@afreedman405-12964
Last seen 7.5 years ago
Harvard University, Cambridge, Massachu…

After updating my R version to 3.4.0 on a mac, after successfully re-installing sundry bioconductor modules, including DESeq2, when I load DESeq2 I get the following:

library("DESeq2")
Loading required package: S4Vectors
Loading required package: stats4
Loading required package: BiocGenerics
Loading required package: parallel

Attaching package: ‘BiocGenerics’

The following objects are masked from ‘package:parallel’:

    clusterApply, clusterApplyLB, clusterCall, clusterEvalQ, clusterExport, clusterMap, parApply, parCapply, parLapply, parLapplyLB, parRapply, parSapply, parSapplyLB

The following objects are masked from ‘package:stats’:

    IQR, mad, sd, var, xtabs

The following objects are masked from ‘package:base’:

    anyDuplicated, append, as.data.frame, cbind, colMeans, colnames, colSums, do.call, duplicated, eval, evalq, Filter, Find, get, grep, grepl, intersect, is.unsorted, lapply, lengths, Map, mapply, match,
    mget, order, paste, pmax, pmax.int, pmin, pmin.int, Position, rank, rbind, Reduce, rowMeans, rownames, rowSums, sapply, setdiff, sort, table, tapply, union, unique, unsplit, which, which.max, which.min


Attaching package: ‘S4Vectors’

The following object is masked from ‘package:base’:

    expand.grid

Loading required package: IRanges
Loading required package: GenomicRanges
Loading required package: GenomeInfoDb
Loading required package: SummarizedExperiment
Loading required package: Biobase
Welcome to Bioconductor

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

Loading required package: DelayedArray
Loading required package: matrixStats

Attaching package: ‘matrixStats’

The following objects are masked from ‘package:Biobase’:

    anyMissing, rowMedians


Attaching package: ‘DelayedArray’

The following objects are masked from ‘package:matrixStats’:

    colMaxs, colMins, colRanges, rowMaxs, rowMins, rowRanges

The following object is masked from ‘package:base’:

    apply

Error: package or namespace load failed for ‘DESeq2’ in dyn.load(file, DLLpath = DLLpath, ...):
 unable to load shared object '/Users/adamfreedman/Library/R/3.4/library/genefilter/libs/genefilter.so':
  dlopen(/Users/adamfreedman/Library/R/3.4/library/genefilter/libs/genefilter.so, 6): Library not loaded: /usr/local/gfortran/lib/libgfortran.3.dylib
  Referenced from: /Users/adamfreedman/Library/R/3.4/library/genefilter/libs/genefilter.so
  Reason: image not found
>

Any idea what's going on here?

 

Thanks,

Adam Freedman

deseq2 • 4.1k views
ADD COMMENT
0
Entering edit mode

Actually, it looks like I'm having the same problem with edgeR too. Both ~/Library/R/3.4/library/edgeR and Library/R/3.4/library/DESeq2 directories exist, and they aren't empty either. And the error messages indicate the problem is not, as with a previous post, trying to load from a library directory from the older version. I have not seen the problem with other bioconductor modules installed on 3.4 thus far.

ADD REPLY
0
Entering edit mode

Can you post sessionInfo() ?

This is sometimes helpful for resolving problems.

ADD REPLY
0
Entering edit mode
@herve-pages-1542
Last seen 4 hours ago
Seattle, WA, United States

Hi Adam, Michael,

It's a problem with some of the Mac binaries we produce. If a package like genefilter has a shared object (e.g. genefilter.so) that is linked to .dylib files located in /usr/local/gfortran/lib -- which is where the gfortran libs are located on our build machines -- then these links need to be replaced with links to the .dylib files shipped with R.

I'm working on a fix. Once our binaries are fixed, we will bump the versions of the affected packages so the new binaries will propagate to our public repos. I'll post here again when this is done (in 48 hours or so).

Thanks for your patience and sorry for the inconvenience.

H.

 

ADD COMMENT
0
Entering edit mode

37 Mac binary packages are currently affected by this problem in release. I fixed the script we use on our Mac builders to generate the package binaries and bumped the version of all the affected packages. This is commit 129423 in the 3.5 branch:

svn log -r 129423 -v https://hedgehog.fhcrc.org/bioconductor/branches/RELEASE_3_5/madman/Rpacks
------------------------------------------------------------------------
r129423 | hpages@fhcrc.org | 2017-05-05 12:32:27 -0700 (Fri, 05 May 2017) | 11 lines
Changed paths:
   M /branches/RELEASE_3_5/madman/Rpacks/CGEN/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/CNAnorm/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/CNPBayes/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/DESeq2/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/DNAcopy/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/GRENITS/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/GeneSelectMMD/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/GeneticsPed/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/LEA/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/LVSmiRNA/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/Linnorm/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/NetPathMiner/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/NuPoP/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/SC3/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/SIMLR/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/SNPRelate/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/affyPLM/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/edgeR/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/flipflop/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/gcatest/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/genefilter/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/iClusterPlus/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/impute/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/inveRsion/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/lfa/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/lpsymphony/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/maanova/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/netReg/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/networkBMA/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/oligo/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/pRoloc/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/preprocessCore/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/qpgraph/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/scde/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/scran/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/sigPathway/DESCRIPTION
   M /branches/RELEASE_3_5/madman/Rpacks/signeR/DESCRIPTION

Bump version in order to propagate new Mac binary

End users who don't have the gfortran compiler on their machine cannot
load the current Mac binary because the shared object in it has links
to .dylib files located in /usr/local/gfortran/lib (e.g. libgfortran.3.dylib,
libquadmath.0.dylib, etc...). This is where Coudert's gfortran 6.1 (the
Fortran compiler officially used by the CRAN folks) is installed on
Bioconductor Mac builder veracruz2. In the new binary, these links are
replaced with links to the .dylib files shipped with the official Mac
binary for R 3.4.0 available on CRAN.

------------------------------------------------------------------------

If everything goes as expected, the new binaries will be made during the next build run and will become available via biocLite() in about 24 hours.

Cheers,

H.

ADD REPLY
0
Entering edit mode

thank you Hervé

ADD REPLY
0
Entering edit mode

The new Mac binaries are available e.g.

  https://master.bioconductor.org/packages/3.5/bioc/html/genefilter.html

Update with biocLite().

H.

 

ADD REPLY

Login before adding your answer.

Traffic: 517 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6