[BioC] How to do clustering

[Seth Falcon]

"ssls sddd" <ssls.sddd at gmail.com> writes:

> Thanks Seth! I got a chance to check with nsFilter in genefilter package but
> there is some problem with my dataset.
>
> My codes are:
>
> library(genefilter)
>
> ans <- nsFilter(as.matrix(x), require.entrez = FALSE, require.symbol =
> FALSE, require.GOBP = FALSE, require.GOCC = FALSE, require.GOMF = FALSE,
> remove.dupEntrez = FALSE, var.func = IQR, var.cutoff = 0.75, var.filter =
> TRUE)
>
> ans$eset
> ans$filter.log
>
> and the error message is :
>
> error in function (classes, fdef, mtable)  :
>        unable to find an inherited method for function "nsFilter", for
> signature "matrix"

Yes, nsFilter only supports ExpressionSet objects -- this is clearly
stated in the man page for the function.

> And I did the following:
>
>> library(genefilter)
>> args(nsFilter)
> function (eset, require.entrez = TRUE, require.symbol = TRUE,
>    require.GOBP = FALSE, require.GOCC = FALSE, require.GOMF = FALSE,
>    remove.dupEntrez = TRUE, var.func = IQR, var.cutoff = 0.5,
>    var.filter = TRUE)
> NULL
>> showMethods("nsFilter")
> Function: nsFilter (package genefilter)
> eset="ExpressionSet"
>
> Perhaps my matrix x is not compatible with "ExpressionSet"? Any suggestions
> on this?

A matrix object in R is certainly not equivalent to an ExpressionSet
object.  I would recommend reading over the first vignette in the
Biobase package: _An introduction to Biobase and ExpressionSets_.  You
can get to it by doing:

   library("Biobase")
   openVignette(packge="Biobase")
   ## choose the first item

+ seth

-- 
Seth Falcon | Computational Biology | Fred Hutchinson Cancer Research Center
http://bioconductor.org