Dear list members,
I am analyzing Affymentrix gene expression data and would like to
apply the R package, VarSelRF to identifying small sets of genes that could
be used for diagnostic purpose.
Basically, the data matrix is composed of 22277 rows (genes) and 65 columns
(samples).
I did unsupervised clustering using pvclust to get 4 classes. What I would
like to do is
to get unique genes for each class which can best characterize them.
I did so and had the problem when running the code.
The error message is:
> rf.vs1 <- varSelRF(exprSet, cl, ntree = 200, ntreeIterat = 100,
vars.drop.frac = 0.2)
error in randomForest.default(x = xdata, y = Class, ntree = ntree, mtry mtry, :
length of response must be the same as predictors
My code is:
library(varSelRF)
exprSet <- as.matrix(read.table('varSelRF_x.txt',header = FALSE))
cl <- factor(c(rep("C", 2), rep("D", 2),
rep("B", 1), rep("A", 1), rep("D",
1), rep("B", 1), rep("C", 2), rep("B", 1),
rep("D", 1), rep("A", 1),
rep("D", 2), rep("B", 1),rep("A",
1),rep("B", 1), rep("D", 1),rep("B",
1),rep("C", 1),rep("D", 2),rep("C",
2),rep("B", 2),rep("D", 1),rep("C",
1),rep("D", 1),rep("D", 1),rep("C",
1),rep("B", 1),rep("C", 1),rep("A",
1),rep("C", 1),rep("B", 1),rep("D",
3),rep("D", 1),rep("C", 1),rep("B",
2),rep("D", 1),rep("D", 1),rep("B",
2),rep("D", 1),rep("B", 1),rep("C",
1),rep("D", 1),rep("B", 3),rep("D",
5),rep("B", 1),rep("D", 2),rep("B",
1),rep("D", 1)))
rf.vs1 <- varSelRF(exprSet, cl, ntree = 200, ntreeIterat = 100,
vars.drop.frac = 0.2)
rf.vs1
plot(rf.vs1)
Would you like to give me some suggestions which could result in the error
message?
Thank you very much and I am looking forward to your reply!
Best Regards,
Alex
[[alternative HTML version deleted]]
