[BioC] Bug in mt.sample.rawp (multtest), R 2.4.1,

shirley zhang shirley0818 at gmail.com
Tue May 22 19:55:20 CEST 2007


If B = 1, it should return the t-value and p-value of unpermuted data,
which means the result is equal to t.test for one row. I found
mt.sample.teststat returns the right t-value, but mt.sample.rawp does
not. Please see the following.

permt<-mt.sample.teststat(golub[50,],golub.cl,B=1)
>permt =  3.200749

permp<-mt.sample.rawp(golub[50,],golub.cl,B=1)
>permp = 1


Furthermore, I could not understand the following histgram figure when
running the example code from help
>permp<-mt.sample.rawp(golub[1,],golub.cl,B=1000)
>hist(permp)


BTW, since mt.teststat(multtest) only return the t-value, could
anybody tell me a function to get the rawp-value from a microarray
data frame but Not a Permutation unadjusted p-values like in mt.maxT?

Thanks
---------------------------------------
> sessionInfo()
R version 2.4.1 (2006-12-18)
i386-pc-mingw32

locale:
LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
States.1252;LC_MONETARY=English_United
States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252

attached base packages:
[1] "splines"   "tools"     "stats"     "graphics"  "grDevices" "utils"
[7] "datasets"  "methods"   "base"

other attached packages:
multtest survival  Biobase
"1.14.0"   "2.30" "1.12.2"



More information about the Bioconductor mailing list