[R] apply a function to loop through

[Weiwei Shi]

Hi,

first of all, your Scores are factors instead of numeric, you need to change it;

I made a new myDat for test purpose by

myDat = as.data.frame(rbind(myDat, myDat))
myDat[16:30, 3] <- "gene2"

> myDat
   tissue            Scores   Grp
a       A  3.01494535196933 gene1
b       A  2.99647624484379 gene1
c       A  3.00527533284709 gene1
d       A  3.02321059636917 gene1
e       A  3.04694197334289 gene1
f       B  11.1464974692841 gene1
g       B  12.2372838904939 gene1
h       B  11.7277801841221 gene1
i       B  11.6405683605147 gene1
j       B  11.1631961720026 gene1
k       C  0.67528704974662 gene1
l       C  1.21000950157251 gene1
m       C 0.843722400594721 gene1
n       C 0.881706314004343 gene1
o       C  1.43670211710054 gene1
a1      A  3.01494535196933 gene2
b1      A  2.99647624484379 gene2
c1      A  3.00527533284709 gene2
d1      A  3.02321059636917 gene2
e1      A  3.04694197334289 gene2
f1      B  11.1464974692841 gene2
g1      B  12.2372838904939 gene2
h1      B  11.7277801841221 gene2
i1      B  11.6405683605147 gene2
j1      B  11.1631961720026 gene2
k1      C  0.67528704974662 gene2
l1      C  1.21000950157251 gene2
m1      C 0.843722400594721 gene2
n1      C 0.881706314004343 gene2
o1      C  1.43670211710054 gene2
> mytry <- by(myDat, INDICES=as.factor(myDat[,3]), FUN=function(x) {
+ pairwise.wilcox.test(as.numeric(as.character(x$Scores)),
+
+ x$tissue,
+
+ p.adjust.method = "none")$p.value
+ })
> mytry
as.factor(myDat[, 3]): gene1
            A           B
B 0.007936508          NA
C 0.007936508 0.007936508
------------------------------------------------------------
as.factor(myDat[, 3]): gene2
            A           B
B 0.007936508          NA
C 0.007936508 0.007936508


HTH,

Weiwei

On 7/17/07, Hai Lin <kevinvol2002 at yahoo.com> wrote:
> Dear R users,
>
> I have a dataset generated as follows,
>
> myDat <- data.frame(matrix(c(rep(LETTERS[1:3],
> each=5),
>                              rnorm(5,mean=3,sd=0.03),
>                              rnorm(5,12,1),
>                              rnorm(5,1,0.5)),
>                            ncol=2,
>
> dimnames=list(c(letters[1:15]),
>
> c("tissue","Scores"))))
> myDat$Grp <-c("gene1")
>
> There is one level "gene1" in $Grp in my data step.
> I'd like to do pairwise.wilcox.test on $tissue while
> going throug $Grp if there are more levels with gene2,
> gene3.
>
> I tried to loop through $Grp using apply with an error
> message "Error in sort(unique.default(x), na.last =
> TRUE) :
>         'x' must be atomic".
>
> mytry <-  apply(as.matrix(myDat),
>                 1,
>
> function(Grp)pairwise.wilcox.test(Grp$Scores,
>
> Grp$tissue,
>
> p.adjust.method = "none")$p.value)
>
> I could not find any similar stuffs in forum. Could
> anyone here give a hand?
>
> Thanks a bunch.
>
> kevin
>
> ______________________________________________
> R-help at 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.
>


-- 
Weiwei Shi, Ph.D
Research Scientist
GeneGO, Inc.

"Did you always know?"
"No, I did not. But I believed..."
---Matrix III