Dear list,
I have gene expression measurements obtained by PCR on 11 genes,
tabulated as a data matrix.
I'm attempting to use GSA package to distinguish any significant changes
in these genes as a pathway.
My response variable is binary, 0=no disease, 1=disease.
I have read the PCR data into R as follows:
data <-
read.delim("CD4PCR.txt",header=TRUE,row.names=1,sep="\t",dec=".",fill=TR
UE)
x<-as.matrix(data)
dim(x)
(11,37)
=11 genes
=37 samples (20 no disease, 17 disease)
this code:
set.seed(100)
y <-c(rep(0,20),rep(1,17))
genenames<-as.character(data$Gene.Symbol)
geneset<-as.character(rownames(x))
GSA.obj<-GSA.func(x,y, genenames, geneset, resp.type="Two class
unpaired")
returns this error:
Error in 1:max(ngenes, na.rm = TRUE) : result would be too long a vector
In addition: Warning message:
In max(ngenes, na.rm = TRUE) :
no non-missing arguments to max; returning -Inf
Could someone explain the error?
I have performed this analysis with the globaltest package without
problems.
SessionInfo()
R version 2.9.1 (2009-06-26)
i386-pc-mingw32
with thanks
----------------------------------------
David
[[alternative HTML version deleted]]