[BioC] zero-row result breaks select() on PolyPhen.Hsapiens.* and SIFT.Hsapiens.*

Valerie Obenchain vobencha at fhcrc.org
Tue Nov 26 01:12:55 CET 2013


Hi Robert,

Here is an update on the new SIFT/PolyPhen annotations.

The SIFT tool has been abandoned and a few of the original maintainers 
have started the PROVEAN tool. (See http://provean.jcvi.org/index.php 
for details.) The PROVEAN group has provided pre-computed SIFT and 
PROVEAN scores for dbSNP 137 which I've packaged up as 
SIFT.Hsapiens.dbSNP137. This should be available via biocLite() by the 
end of the week and requires VariantAnnotation 1.9.13.

The PolyPhen group has a new download of pre-computed scores but they 
have completely changed the format and volume. It will take some time to 
package those up. Right now, adapting ensemblVEP to support multiple 
versions and to remove the perl script burden is a higher priority than 
the pre-computed PolyPhen scores. I have this on the TODO but it will be 
awhile before I can get to it.

Valerie

On 09/24/2013 12:17 PM, Valerie Obenchain wrote:
> I will update the SIFT and PolyPhen databases for the upcoming release.
>
> Valerie
>
>
> On 09/23/2013 02:21 PM, Robert Castelo wrote:
>> hi Valerie,
>>
>> On 9/23/13 9:41 PM, Valerie Obenchain wrote:
>>> Hi Robert,
>>>
>>> Thanks for reporting this. Now fixed in VariantAnnotation 1.7.47.
>>>
>> great! thanks for the quick fix.
>>
>>> Have you looked at the ensemblVEP package? It's a wrapper to Ensembl's
>>> Variant Effect Predictor tool. We encourage the use of ensemblVEP
>>> instead of the SIFT and PolyPhen databases because it accesses the
>>> most current information. As you know, the SIFT and PolyPhen dbs are
>>> becoming dated and we don't have plans to package newer versions.
>>>
>>> emsemblVEP requires that you download and install the script located
>>> here,
>>>
>>> http://uswest.ensembl.org/info/docs/tools/vep/script/vep_download.html
>>>
>>> The variant_effect_predictor.pl executable must be in your path. Let
>>> us know if you have trouble with the install/setup.
>> yes, i looked at it, and i think it is a great solution for analysis of
>> a few hundred variants as it needs to acces the internet to download the
>> information. However, i'm working on a package that eventually needs to
>> annotate a few thousand variants and i find the dependency on an
>> external perl script that the end user must install, somewhat troubling.
>> let me know if you have suggestions about this.
>>
>> for software packages that need to efficiently access SIFT and PolyPhen
>> annotations from R, freezing the data regularly is, in my opinion, a
>> much better solution. i was actually going to ask you if you could
>> update these two packages. As much as you want to keep an up to date
>> version of the SNPloc.Hsapiens.* or TxDb.* packages, i'd do it for SIFT
>> and Polyphen, unless there's some licensing issue that prevents this, as
>> it happens now with OMIM.
>>
>> cheers,
>> robert.
>>
>>> Valerie
>>>
>>> On 09/20/2013 05:25 PM, Robert Castelo wrote:
>>>> Dear list,
>>>>
>>>> interrogating the TxDb.Hsapiens.UCSC.hg19.knownGene package with no
>>>> result gives the following expected result:
>>>>
>>>> library(TxDb.Hsapiens.UCSC.hg19.knownGene)
>>>> select(TxDb.Hsapiens.UCSC.hg19.knownGene, keys="dummy",
>>>> keytype="GENEID", cols="SYMBOL")
>>>> [1] GENEID
>>>> <0 rows> (or 0-length row.names)
>>>>
>>>> however, when i try the same with the annotation packages
>>>> PolyPhen.Hsapiens.dbSNP131 and SIFT.Hsapiens.dbSNP132, the select
>>>> instruction breaks into an error:
>>>>
>>>> library(SIFT.Hsapiens.dbSNP132)
>>>> library(PolyPhen.Hsapiens.dbSNP131)
>>>>
>>>> select(SIFT.Hsapiens.dbSNP132, keys=c("dummy"))
>>>> Error in data.frame(RSID = unlist(rsid), PROTEINID =
>>>> unlist(protein_id),  :
>>>>    arguments imply differing number of rows: 1, 0
>>>>
>>>> select(PolyPhen.Hsapiens.dbSNP131, keys="dummy")
>>>> Error in `*tmp*`$RSID : $ operator is invalid for atomic vectors
>>>>
>>>> i guess these two annotation packages should work analogously to
>>>> TxDb.Hsapiens.UCSC.hg19.knownGene, and give just a 0-row data.frame
>>>> object, right?
>>>>
>>>> these errors reproduce also with the current devel version of BioC,
>>>> please find below both sessionInfo() outputs.
>>>>
>>>> cheers,
>>>> robert.
>>>>
>>>> =====RELEASE====
>>>> R version 3.0.1 (2013-05-16)
>>>> Platform: x86_64-apple-darwin10.8.0 (64-bit)
>>>>
>>>> locale:
>>>> [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
>>>>
>>>> attached base packages:
>>>> [1] parallel  stats     graphics  grDevices utils     datasets
>>>> methods base
>>>>
>>>> other attached packages:
>>>>   [1] TxDb.Hsapiens.UCSC.hg19.knownGene_2.9.2 GenomicFeatures_1.12.3
>>>>   [3] AnnotationDbi_1.22.6 Biobase_2.20.1
>>>>   [5] PolyPhen.Hsapiens.dbSNP131_1.0.2 SIFT.Hsapiens.dbSNP132_1.0.2
>>>>   [7] RSQLite_0.11.4 DBI_0.2-7
>>>>   [9] VariantAnnotation_1.6.7 Rsamtools_1.12.4
>>>> [11] Biostrings_2.28.0 GenomicRanges_1.12.5
>>>> [13] IRanges_1.18.3 BiocGenerics_0.6.0
>>>> [15] vimcom_0.9-8 setwidth_1.0-3
>>>> [17] colorout_1.0-0
>>>>
>>>> loaded via a namespace (and not attached):
>>>> [1] biomaRt_2.16.0     bitops_1.0-6       BSgenome_1.28.0
>>>> RCurl_1.95-4.1     rtracklayer_1.20.4
>>>> [6] stats4_3.0.1       tools_3.0.1        XML_3.95-0.2 zlibbioc_1.6.0
>>>>
>>>>
>>>>
>>>> =====DEVEL=====
>>>> R version 3.0.1 (2013-05-16)
>>>> Platform: x86_64-apple-darwin10.8.0 (64-bit)
>>>>
>>>> locale:
>>>> [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
>>>>
>>>> attached base packages:
>>>> [1] parallel  stats     graphics  grDevices utils     datasets
>>>> methods base
>>>>
>>>> other attached packages:
>>>>   [1] TxDb.Hsapiens.UCSC.hg19.knownGene_2.9.2 GenomicFeatures_1.13.40
>>>>   [3] AnnotationDbi_1.23.23 Biobase_2.21.7
>>>>   [5] PolyPhen.Hsapiens.dbSNP131_1.0.2 SIFT.Hsapiens.dbSNP132_1.0.2
>>>>   [7] RSQLite_0.11.4 DBI_0.2-7
>>>>   [9] VariantAnnotation_1.7.46 Rsamtools_1.13.41
>>>> [11] Biostrings_2.29.19 GenomicRanges_1.13.44
>>>> [13] XVector_0.1.4 IRanges_1.19.37
>>>> [15] BiocGenerics_0.7.5 vimcom_0.9-8
>>>> [17] setwidth_1.0-3 colorout_1.0-0
>>>>
>>>> loaded via a namespace (and not attached):
>>>> [1] biomaRt_2.17.3      bitops_1.0-6        BSgenome_1.29.1
>>>> RCurl_1.95-4.1      rtracklayer_1.21.12
>>>> [6] stats4_3.0.1        tools_3.0.1         XML_3.95-0.2 zlibbioc_1.7.0
>>>>
>>>> _______________________________________________
>>>> Bioconductor mailing list
>>>> Bioconductor at r-project.org
>>>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>>>> Search the archives:
>>>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>>>
>>
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor



More information about the Bioconductor mailing list