[BioC] topTable (fit) annotation

[Sean Davis]

Hi, Jing.

NCBI GEO maintains two types of GPL records.  The normal variant is
just supplied by the submitter.  However, when a GEO Series is curated
by NCBI GEO into a GEO DataSet (GDS), they create a so-called
"Annotation GPL".  These have a relatively standard set of columns.  I
have not made the change to GEOquery yet to grab this annotation GPL
when getting Series Matrix files.  But, you can get them yourself by
specifying:

gplannot = getGEO("GPL96", AnnotGPL=TRUE)

You can always replace the feature data of the ExpressionSets with the
information in the retrieved Annotation GPL.

I hope that is clear.

Sean


On Tue, Aug 30, 2011 at 5:01 PM, Jing Huang <huangji at ohsu.edu> wrote:
> Dear All members,
>
> I have been extracting data from GEO (GEO package) and do some analysis on them by using limma package. What I discover is the components of topTable(fit) are different from the dataset GDS and GSE.
>
> If the data is from GDS, then the colnames of topTable (fit) looks like this.
>
>> colnames(topTable(fit))
> [1] "ID"                    "Gene.title"            "Gene.symbol"
>  [4] "Gene.ID"               "UniGene.title"         "UniGene.symbol"
>  [7] "UniGene.ID"            "Nucleotide.Title"      "GI"
> [10] "GenBank.Accession"     "Platform_CLONEID"      "Platform_ORF"
> [13] "Platform_SPOTID"       "Chromosome.location"   "Chromosome.annotation"
> [16] "GO.Function"           "GO.Process"            "GO.Component"
> [19] "GO.Function.1"         "GO.Process.1"          "GO.Component.1"
> [22] "CTRL"                  "HIF1a"                 "HIF2a"
> [25] "HIF1a2a"               "AveExpr"               "F"
> [28] "P.Value"               "adj.P.Val"
>
> If the data is from GSE, then the   colnames of topTable(fit) looks like this:
>
>>colnames(topTable(fit)
>
> [1] "ID"        "mir210"    "CTRL2"     "AveExpr"   "F"         "P.Value"   "adj.P.Val"
>
> I am trying to add some term into this table by doing following one by one: the data is generated by Affymetrix human U133 platform:
>
>>Library(hgu133plus2.db)
>>x=hgu133plus2SYMBOL
>>y=topTable(fit)
>>y$SYMBOL=unlist(as.list(x[y$ID]))
>
> It works but I need to add ENTREZID,SYMBOL,CHR, CHRloc, and GO annotations as well.  I like to have the topTable more like the topTable(fit) generated at top by data GEO GDS data
>
> I am wondering if there is an easy way to annotate all once.
>
> In addition, I am having a trouble to annotate GO term.
>
> Many Thanks
>
> Jing
>
>
>
>
>        [[alternative HTML version deleted]]
>
> _______________________________________________
> 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
>