R help - Jun 2007 - Wilcoxon test on data matrix

Dear everyone,
I am trying to do a Wilcoxon one-sided test on my gene expression data.
These are the data i have in R:
data.matrix (matrix, numeric) containing all gene expression data (42
rows=genes,  42 columns=tumors), no column header or row names
data.cl (vector, numeric) consisting of 42 0's and 1's to indicate class
0 or class 1 for each column in data.matrix

I want to do a Wilcoxon one-sided test on the data from class 0 versus the data
from class 1, for each row (gene) of the data set.
My first try:

#to make separate matrices for both classes:
data.matrix.0 <- data.matrix[,data.cl==0] 
data.matrix.1 <- data.matrix[,data.cl==1]

# to run the wilcox.test function for each row:
rawp <- apply(data.matrix.0, 1, wilcox.test, y=data.matrix.1,
alternative="less")


The result of printing rawp is:
$`1`

	Wilcoxon rank sum test with continuity correction

data:  newX[, i] and data.matrix.1 
W = 7585, p-value = 1
alternative hypothesis: true location shift is less than 0 


$`2`

	Wilcoxon rank sum test with continuity correction

data:  newX[, i] and data.matrix.1 
W = 6700, p-value = 0.9983
alternative hypothesis: true location shift is less than 0 


Etcetera for each row of the data matrix. 
I can get the p value for one row (gene) using:
rawp.1 <- rawp$'1'$p.value

But how can I get these p-values in one list? I have tried:
rawp <- NULL
for (i in 1:42) {
	a <- paste("'", i, "'", sep="")
	rawp <- rbind(rawp, test$a$p.value)
 }

but that doesn't work (no errors but rawp stays NULL)

There must be an easier way to do a wilcoxon analysis on a matrix!
I'd appreciate your help with this...

Marije


De inhoud van dit bericht is vertrouwelijk en alleen bestemd voor de
geadresseerde(n). Anderen dan de geadresseerde mogen geen gebruik maken van dit
bericht, het openbaar maken of op enige wijze verspreiden of vermenigvuldigen.
Het UMCG kan niet aansprakelijk gesteld worden voor een incomplete aankomst of
vertraging van dit verzonden bericht.

The contents of this message are confidential and only intended for the eyes of
the addressee(s). Others than the addressee(s) are not allowed to use this
message, to make it public or to distribute or multiply this message in any way.
The UMCG cannot be held responsible for incomplete reception or delay of this
transferred message.

wilcox.test(data.matrix[i,] ~ data.cl)$p.value

will return the p.value for row 'i'.

With a loop:

   pv = rep(0,42)
   for (i in 1:42) { pv[i]=wilcox.test(data.matrix[i,] ~ data.cl)$p.value }

Is this what you wanted?

Christophe




On 6/14/07, Booman, M <m.booman@path.umcg.nl>
wrote:
>
> Dear everyone,
> I am trying to do a Wilcoxon one-sided test on my gene expression data.
> These are the data i have in R:
> data.matrix (matrix, numeric) containing all gene expression data (42
> rows=genes,  42 columns=tumors), no column header or row names
> data.cl (vector, numeric) consisting of 42 0's and 1's to indicate
class 0
> or class 1 for each column in data.matrix
>
> I want to do a Wilcoxon one-sided test on the data from class 0 versus the
> data from class 1, for each row (gene) of the data set.
> My first try:
>
> #to make separate matrices for both classes:
> data.matrix.0 <- data.matrix[,data.cl==0]
> data.matrix.1 <- data.matrix[,data.cl==1]
>
> # to run the wilcox.test function for each row:
> rawp <- apply(data.matrix.0, 1, wilcox.test, y=data.matrix.1,
> alternative="less")
>
>
> The result of printing rawp is:
> $`1`
>
>         Wilcoxon rank sum test with continuity correction
>
> data:  newX[, i] and data.matrix.1
> W = 7585, p-value = 1
> alternative hypothesis: true location shift is less than 0
>
>
> $`2`
>
>         Wilcoxon rank sum test with continuity correction
>
> data:  newX[, i] and data.matrix.1
> W = 6700, p-value = 0.9983
> alternative hypothesis: true location shift is less than 0
>
>
> Etcetera for each row of the data matrix.
> I can get the p value for one row (gene) using:
> rawp.1 <- rawp$'1'$p.value
>
> But how can I get these p-values in one list? I have tried:
> rawp <- NULL
> for (i in 1:42) {
>         a <- paste("'", i, "'",
sep="")
>         rawp <- rbind(rawp, test$a$p.value)
> }
>
> but that doesn't work (no errors but rawp stays NULL)
>
> There must be an easier way to do a wilcoxon analysis on a matrix!
> I'd appreciate your help with this...
>
> Marije
>
>
> De inhoud van dit bericht is vertrouwelijk en alleen bestemd voor de
> geadresseerde(n). Anderen dan de geadresseerde mogen geen gebruik maken van
> dit bericht, het openbaar maken of op enige wijze verspreiden of
> vermenigvuldigen. Het UMCG kan niet aansprakelijk gesteld worden voor een
> incomplete aankomst of vertraging van dit verzonden bericht.
>
> The contents of this message are confidential and only intended for the
> eyes of the addressee(s). Others than the addressee(s) are not allowed to
> use this message, to make it public or to distribute or multiply this
> message in any way. The UMCG cannot be held responsible for incomplete
> reception or delay of this transferred message.
>
> ______________________________________________
> R-help@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>

-- 
Christophe Pallier (http://www.pallier.org)

	[[alternative HTML version deleted]]