[BioC] ANOVA performance

Isaac Neuhaus isaac.neuhaus@bms.com
Wed, 19 Feb 2003 12:02:26 -0500

Is there a recommended package to run ANOVA. The generic functions in R perform
very poorly with large microarray datasets which preclude us from running
permutation analyses. (Please note below the way I am implementing it for a
one-way anova for three levels). Are there plans to implement some code in C
passing affy objects?


my.anova <- function(x) {
 gr <- gl(3,3,9, label=c(1:3))
 wg <- as.numeric(x)
 anova(lm(wg ~ gr))

data <- read.table("file", header=T, sep = "\t", row.names=1)
res <- array()
res <- apply(data, 1, my.anova)
res <- data.frame(res, row.names = row.names(data))
write.table(res, quote=FALSE, col.names=FALSE, sep="\t", file="fileout")

file has this format
A    B    C    D    E    F    G    H    I
a    x    x    x    x    x    x    x    x    x
b    x    x    x    x    x    x    x    x    x
c    x    x    x    x    x    x    x    x    x

	[[alternate HTML version deleted]]