[R-SIG-Mac] trouble running Bioconductor

Olga Vitek ovitek at stat.purdue.edu
Fri Aug 11 03:33:40 CEST 2006


Dear All,

I am having difficulty running Limma: every time I run the code R
crashes, and I can't explain the reason for the problem. Could someone
point me out to where to look for trouble? I attach the code and the error
messages below.

Thanks a lot in advance!
Olga

ERROR MESSAGE:
---------------
 *** caught bus error ***
address 0x2c, cause 'invalid alignment'

Traceback:
 1: sum(out$residuals^2)
 2: lm.series(M, design = design, ndups = ndups, spacing = spacing,
weights = weights)
 3: lmFit(eset, design)
 4: NCOL(fit$coefficients)
 5: contrasts.fit(lmFit(eset, design), contr.matrix)
 6: ebayes(fit = fit, proportion = proportion, stdev.coef.lim =
stdev.coef.lim)
 7: eBayes(contrasts.fit(lmFit(eset, design), contr.matrix))


CODE:
------
makeFit <- function(eset) {
	design <- model.matrix(~0+factor(c(eset$stimulation)))
	dimnames(design)[[2]] <- c("Control", "Insulin")
	contr.matrix <- makeContrasts(Insulin-Control, levels=design)
	eBayes( contrasts.fit( lmFit(eset, design), contr.matrix ) )
}
limmaFit <- makeFit(data$eset)
rm(makeFullFit)

SPECS:
--------
Mac OS X V.10.4.7, 1GHz PowerPC G4, 1.25GB memory
R Version 2.3.1 (2006-06-01)
> package.version("Biobase")
[1] "1.10.1"
> package.version("limma")
[1] "2.7.9"



More information about the R-SIG-Mac mailing list