R help - Nov 2007 - Doing PCA

Hi Fellow R enthusiasts
 I have managed to reshape my data using a much shorter script than before.
Woohoo
However now I have new problems. The code is below. There are no problems with
the create matrix section.
The problem code is highlighted in bold. I am trying to do PCA on the data. 
Here are the errors.

Error1 
code :  OGSscaled = rangescale(OGS)
error message : Error in dim(newX) <- c(prod(d.call), d2) : 
  dims [product 47960] do not match the length of object [43600]

Error2
I tried to do PCA without rangescaling
code :OGSpca <- prcomp(OGS, center=FALSE)
error message : Error in svd(x, nu = 0) : infinite or missing values in
 'x'
 

CODE
######################################################
###########  Create matrix ###########################
######################################################

# load packages
require(reshape)
source("rangescale.r")

#Open the csv file
OGSdata <- read.table("MG3199.csv",sep=",",header=TRUE)



#create matrix
x.m <- melt(OGSdata, measure.var="pct")
OGS <- cast(x.m, mci ~ sample)

######################################################
###########  PCA  ####################################
######################################################

#scale profiles
OGSscaled = rangescale(OGS)

#do PCA
result = prcomp(OGS, center=FALSE)

#obtain scores matrix
scores=result$rotation

#PC1 vs PC2 plot
plot(scores[,1], scores[,2], xlab="PC1", ylab="PC2")

#add labels (0.005 and 0.003 offset to aviod obscuring points

Kind Regards
Amit Patel




      ___________________________________________________________

now.

	[[alternative HTML version deleted]]