[BioC] annaffy annotation problem (hgu133plus2)

[sdurinck at ebi.ac.uk]

Hi Jo,

You could do a local install of the Ensembl biomart database.  Normally
the latest Ensembl release is downloadable from
http://www.ebi.ac.uk/BioMart/install.html
but I am not able to access this right now. If you then change in biomaRt
the host to your local host you should be able to use it.  You need to do
this in the package R code but I'll update the package later this week so
you can do this from within R.

best,
Steffen

> the biomaRt package is very nice,
> i also downloaded it and tested it, but as i am behind a (****) http
> proxy i am not able to connect to the according servers at ensembl. i
> assume the package connects directly through a SQL connection to the
> database... would it be possible to use a web service?
>
> cheers, jo
>
> Quoting Steffen Durinck <sdurinck at ebi.ac.uk>:
>
>> Hi,
>>
>> You could also try using the biomaRt package (see developmental
>> packages).  biomaRt uses direct queries to the latest release of
>> Ensembl to get annotation information.  Ensembl does an independent
>> mapping of the affy probe sequences to the genome sequences.  This is
>> what I got for your affy id:
>>
>> > library(biomaRt)
>> > mart<-martConnect()
>> -  Connected to ensembl_mart_29, ensembl_mart_29, ensembl_mart_29 and
>> ensembl_mart_29 -
>> > getGene(id="208078_s_at",array="hgu133plus2",mart=mart)
>> An object of class "martTable"
>> Slot "id":
>> [1] "208078_s_at"
>>
>> Slot "table":
>> $symbol
>> [1] "SNF1LK"
>>
>> $description
>> [1] "Probable serine/threonine-protein kinase SNF1LK (EC 2.7.1.37).
>> [Source:Uniprot/SWISSPROT;Acc:P57059]"
>>
>> $band
>> [1] "q22.3"
>>
>> $chromosome
>> [1] "21"
>>
>> $start
>> [1] 43658828
>>
>> $end
>> [1] 43671436
>>
>> $martID
>> [1] "ENSG00000142178"
>>
>>
>> Cheers,
>> Steffen
>>
>> Dipl.-Ing. Johannes Rainer wrote:
>>
>>> hi,
>>>
>>> i am wondering how the annotation of the Affymetrix chips are
>>> done... i know how the AnnBuilder package works and i also used it,
>>> but as the starting point for any annotation is the GenBank id for
>>> which the probe set codes i would like to know from where this
>>> mapping information comes. are those packages built using the probe
>>> set to genbank mapping that can be downloaded from affymetrix, or
>>> the (sometimes) newer annotation that is available through
>>> Affymetrix NetAffx or some other source?
>>>
>>> I am really interested in this, because we got a little annotation
>>> problem with the annotation for the hgu133plus2 chip (version
>>> 1.7.0). one of our interesting genes in our study was the probe set
>>> 208078_s_at which is mapped using the hgu133plus2 package (and the
>>> annotation from Affymetrix) to NM_030751 (TCF8). we made real time
>>> PCR to confirm its regulation but this TCF8 was not regulated at
>>> all. so i went back to NetAffx tool and saw that Affymetrix gave a
>>> new annotation for this probe set, all 11 probes perfectly match
>>> with the gene NM_173354 (SNF1LK) (i wonder how Affymetrix could
>>> annotate this gene to the TCF8 gene).
>>>
>>> if the annotation packages are built using the downloadable probe
>>> set to genbank mapping files from Affymetrix (which may never
>>> updated and corrected for errors by Affymetrix) i believe that there
>>> may be more annotation errors in this package like the one we found.
>>>
>>> cheers, jo
>>>
>>> _______________________________________________
>>> Bioconductor mailing list
>>> Bioconductor at stat.math.ethz.ch
>>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>>>
>>
>>
>
>
>
>