Gsva Analysis In R: Unable To Identify Method For Function

Question

[k.nirmalraman]

Hi All,

I am trying to do gene set variance analysis using GSVA package for different samples. I have a list of signification genes as ExpressionSet and a GeneSet

  > eset<-new("ExpressionSet",exprs=as.matrix(scaledcounts))

  > eset
   ExpressionSet (storageMode: lockedEnvironment)
   assayData: 25658 features, 21 samples 
   element names: exprs 
    : none
   phenoData: none
   featureData: none
   experimentData: use 'experimentData(object)'
   Annotation: none

and my Gene list is

GeneIDs<-c("abc","cda","xcv",.....)

which I am conversiting to a GeneSet

`HpoGS<-GeneSet(GeneIDs,geneIdType=NullIdentifier(),setName="Hpo")

I am getting the following error message, which I am failing to understand with my limited knowledge in oo programming.

`gsva<-gsva(eset,HpoGS)
 Error in (function (classes, fdef, mtable)  : 
 unable to find an inherited method for function ‘gsva’ for signature ‘"ExpressionSet", "GeneSet", "missing"’

I am wondering does it have anything to do with the way, I am creating the expressionSet or GeneSet? Any help to address this would be highly appreciated. Thanks a lot in advance!

[David W]

I don't know anything about gvsa, but a first step here would be to use showMethods('gsva'), which should show the combinations of of classes gvsa can take (referred to as a signature in the error message)

[Michael]

First step is ?gsva

Answer 1

[Michael]

According to the documentation of GSVA:

gset.idx.list Gene sets provided either as a list object or as a GeneSetCollection object.

So, you have provided a GeneSet object, but either a list or GeneSetCollection object is required. The GeneSetCollection class is part of the package GSEA base and described in its documentation.

Making a list of genes is probably even easier.

Answer 2

[smacneil88]

I am not sure what gene sets you are using, but if you are using GeneSets from the Molecular Signatures Database just read them in as a list.

geneSets="c2.symbols.gmt"
geneSets=readList(geneSets)

As far as your expression data goes, just read it in as Matrix. GSVA can also take a matrix.

microarrayData <- read.delim("SomeData.txt", check.names=FALSE, row.names=1)
microarrayData= as.matrix(microarrayData)

gsva_analysis <- gsva(microarrayData, ggeneSets, min.sz=1, max.sz=Inf, verbose=TRUE, rnaseq=FALSE)$es.obs