This function can be used to carry out score test inference with GPUs. Currently the function supports two test statistics: the Cox score and rank score tests (Jung et al, 2005), for right-censored time-to-event outcomes.
scoregpu(y, event, markers, test, B=0, stand=TRUE, pval=TRUE, index=FALSE, scale=FALSE)The name of the outcome variable.
In the case of survival analysis, event is the name of the event indicator.
Expression matrix. Probes are along rows.
specifies the test to be performed. Possible options are
cox (Cox score test)
and npcox (Cox rank score test).
number of permutations.
If TRUE, the squared test statistic is returned.
If TRUE, the p-value is returned (stand has to be FALSE).
If TRUE, the sample indexes at each permutation are returned.
If TRUE, markers are centered.
This function returns a data frame with one column of test statistics.
Jung, S.-H., Owzar K., George, S.L. (2005) A multiple testing procedure to associate gene expression levels with survival.Statistics in Medicine. 24, 20, 3077--88.
Shterev, I.D., Jung, S.-H., George S.L., Owzar K. permGPU: Using graphics processing units in RNA microarray association studies. BMC Bioinformatics 2010, 11:329.
# NOT RUN {
set.seed(123)
n<-100
K<-3
x1<-matrix(rnorm(n*K),K,n)
x2<-matrix(rnorm(n*K),K,n)
x3<-matrix(rnorm(n*K),K,n)
otime<-rexp(n)
event<-rbinom(n,1,0.8)
rownames(x1)<-paste("g",1:K,sep="")
colnames(x1)<-paste("p",1:n,sep="")
rownames(x2)<-paste("g",1:K,sep="")
colnames(x2)<-paste("p",1:n,sep="")
rownames(x3)<-paste("g",1:K,sep="")
colnames(x3)<-paste("p",1:n,sep="")
x=list(x1,x2,x3)
library(foreach)
# carry out analysis with npcox test
foreach(i=1:length(x))
# }
# NOT RUN {
<!-- %do% scoregpu(otime,event,x[[i]],"npcox") -->
# }
# NOT RUN {
# carry out analysis with npcox test using B=10 permutations
foreach(i=1:length(x))
# }
# NOT RUN {
<!-- %do% scoregpu(otime,event,x[[i]],"npcox", B=10) -->
# }
# NOT RUN {
# }
Run the code above in your browser using DataLab