Hi Jorge,
Thank you.
Regarding the "group" variable, I am not sure how I can make my two groups -
such that there will be 1013 Wilcox p-values from 1013 rows. Each row has 2
groups where Wilcox test is to be conducted : Group A (samples 1-3), Group B
(samples 4-6) of the dataframe. That is, in simple form, something like this
:
wilcox.test(groupA_samples of row i ~ groupB_samples of row i, ........)
regards,
Chintanu
==================================================
On Fri, Sep 9, 2011 at 3:32 PM, Jorge I Velez <jorgeivanve...@gmail.com>wrote:
> Hi Chintau,
>
> For A, try
>
> # some data
> set.seed(123)
> X <- matrix(rnorm(1013*6), ncol = 6)
> group <- rep(0:1, each = 3)
>
> # option 1
> result <- apply(X, 1,
> function(x) wilcox.test(x ~ group, paired = TRUE)$p.value)
> result
>
> # option 2
> res <- rep(NA, NROW(X))
> for(i in 1:NROW(X)){
> res[i] <- wilcox.test(X[i, ] ~ group, paired = TRUE)$p.value
> }
> res
>
> # are equal?
> all.equal(result, res)
>
> For B, see http://cran.r-project.org/other-docs.html
>
> HTH,
> Jorge
>
>
> On Fri, Sep 9, 2011 at 1:14 AM, Chintanu <> wrote:
>
>> Hi,
>>
>> Two queries, if I may ask:
>>
>> A. I wish to retrieve the p-values from wilcox test for EACH gene ---
>> such
>> that Samples 1-3 is considered as one group while the rest remains in the
>> other group.
>>
>> There is 1013 genes in the dataframe as attempted to exemplify in the
>> table
>> below.
>>
>>
>> |Sample 1 | Sample2 | sample3 | SDample4 | sample 5 |
>> sample 6
>> _________________________________________________________________________
>> Gene 1 | 2.4 | 3.5 | 4.2 | 6.2
>> | 2.4 | 6.3
>> Gene 2 | 6.4 | 6.5 | 2.2 | 8.2
>> | 4.4 | 2.3
>> ----
>> -----
>> Gene 1013 | 1.4 | 2.5 | 3.2 | 4.2 |
>> 5.4 | 7.3
>>
>>
>> Accordingly, the following function is made.
>>
>> funn <- function (x,y)
>> {
>>
>> for (i in 1:1013)
>> {
>>
>> wil <- wilcox.test(as.numeric(x), as.numeric (y), paired=TRUE);
>> print (wil);
>> }
>>
>> return (wil);
>>
>> }
>>
>>
>> However, I think it is somewhat incorrect as commands like the following
>> generates several warnings (and it gives more than just the p-value).
>>
>>
>> funn (c(file [,1]:file [,3]), c(file[,4]:file [,6])) # "file" is the
>> dataframe.
>>
>>
>> B. Any good books/weblink you could advise which focus more on the
>> programming side of r, such as function-building ? The few books that I
>> have
>> gone thorugh have very less on such aspect - they focus more on
>> statistical
>> aspects and use of built-in functions.
>>
>> Thank you.
>>
>> Cheers,
>> Chintanu
>>
>> [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> R-help@r-project.org 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.
>>
>
>
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org 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.
