[Bioc-devel] Changes S4Vectors etc.
Martin Morgan
martin.morgan at roswellpark.org
Thu Feb 9 13:19:09 CET 2017
On 02/09/2017 06:25 AM, Hervé Pagès wrote:
> Hi Leonard,
>
> mmh... I can't reproduce this and it doesn't show up on the build report
> either. But I got another strange error when I tried to *install*
> SGSeq: something about TxFeatures not being able to extend GenomicRanges
> because of incompatible type for the elementMetadata slot.
>
> I got rid of it by re-installing GenomicRanges from svn.
> I guess I had some stalled class definition somewhere in a cache.
>
> Looks like maybe you also have a stalled class definition somewhere
> in a cache for one of SGFeatureCounts's parent classes. Not sure which
> one though :-/
>
> Maybe I forgot to bump a package version somewhere after I renamed
> characterORNULL -> character_OR_NULL? Re-installing S4Vectors, IRanges,
> GenomicRanges, and SummarizedExperiment directly from svn might help.
See also https://support.bioconductor.org/p/92201/#92205 where
GeneSetCollection extends 'list', which extends 'vector'
GSEABase Depends: on annotate which Depends: on AnnotationDbi which
Imports: S4Vectors.
S4Vectors defines a class union on vector_OR_factor (previously
vectorORfactor). Consequently, GeneSetCollection extends vector_OR_factor.
Class definitions are cached at package installation time, so GSEABase
class definitions became out-of-date when S4Vectors was updated.
A conservative approach is to bump the version of all packages that
depend on S4Vectors, even GSEABase which is three steps away from
S4Vectors. I believe that this involves
db = available.packages(repos=BiocInstaller::biocinstallRepos()[1])
revdeps = tools::package_dependencies("S4Vectors", db, recursive=TRUE,
reverse=TRUE)
which is
> length(revdeps$S4Vectors)
[1] 662
packages!
Leonard could find candidates for re-installation with
> mydep = tools::package_dependencies("SGSeq", db, recursive=TRUE)
> mydep$SGSeq[mydep$SGSeq %in% revdeps$S4Vectors]
[1] "IRanges" "GenomicRanges" "Rsamtools"
[4] "SummarizedExperiment" "AnnotationDbi" "Biostrings"
[7] "GenomicAlignments" "GenomicFeatures" "GenomeInfoDb"
[10] "rtracklayer" "XVector" "biomaRt"
[13] "DelayedArray"
Martin
>
> Cheers,
> H.
>
> On 02/08/2017 04:18 PM, Leonard Goldstein wrote:
>> Hi Hervé,
>>
>> It looks like there have been some changes in bioc-devel (S4Vectors etc.)
>> that break the SGSeq package (see below). I'm not sure whether this is
>> something that needs to be addressed in SGSeq or its dependencies. I'd be
>> grateful for any pointers. Thanks for your help.
>>
>> Leonard
>>
>> --
>>> example(makeSGFeatureCounts, "SGSeq")
>>
>> mkSGFC> sgfc <- makeSGFeatureCounts(sgf_pred, si,
>> mkSGFC+ matrix(0L, length(sgf_pred), nrow(si)))
>> Error in checkSlotAssignment(object, name, value) :
>> assignment of an object of class "NULL" is not valid for slot
>> 'NAMES' in
>> an object of class "SGFeatureCounts"; is(value, "characterORNULL") is not
>> TRUE
>>>
>>> sessionInfo()
>> R Under development (unstable) (2017-02-06 r72129)
>> Platform: x86_64-apple-darwin13.4.0 (64-bit)
>> Running under: OS X El Capitan 10.11.6
>>
>> locale:
>> [1] C
>>
>> attached base packages:
>> [1] stats4 parallel stats graphics grDevices utils datasets
>> [8] methods base
>>
>> other attached packages:
>> [1] SGSeq_1.9.1 SummarizedExperiment_1.5.5
>> [3] DelayedArray_0.1.4 Biobase_2.35.0
>> [5] Rsamtools_1.27.12 Biostrings_2.43.4
>> [7] XVector_0.15.2 GenomicRanges_1.27.22
>> [9] GenomeInfoDb_1.11.9 IRanges_2.9.18
>> [11] S4Vectors_0.13.13 BiocGenerics_0.21.3
>>
>> loaded via a namespace (and not attached):
>> [1] igraph_1.0.1 Rcpp_0.12.9
>> AnnotationDbi_1.37.2
>>
>> [4] magrittr_1.5 zlibbioc_1.21.0
>> GenomicAlignments_1.11.9
>> [7] BiocParallel_1.9.5 lattice_0.20-34 tools_3.4.0
>>
>> [10] grid_3.4.0 DBI_0.5-1 digest_0.6.12
>>
>> [13] Matrix_1.2-8 GenomeInfoDbData_0.99.0 rtracklayer_1.35.5
>>
>> [16] bitops_1.0-6 RUnit_0.4.31 biomaRt_2.31.4
>>
>> [19] RCurl_1.95-4.8 memoise_1.0.0 RSQLite_1.1-2
>>
>> [22] compiler_3.4.0 GenomicFeatures_1.27.6 XML_3.98-1.5
>>
>>>
>>
>> [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> Bioc-devel at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>>
>
This email message may contain legally privileged and/or...{{dropped:2}}
More information about the Bioc-devel
mailing list