cor.rep.bootci {GeneNT}R Documentation

Bootstrap Confidence Interval for multivariate correlation estimator from replicated gene microarray data

Description

This function compute bootstrap confidence interval for correlation estimation from replicated gene microarray data.

Usage

cor.rep.bootci(x, y=NULL, m, G, alpha)

Arguments

x x is the gene microarray data in matrix OR data.frame format.
y y (optional). Should be used when x, y are vectors.
m m is the number of replicates for each independent observations. Note: m*G == nrow(dat).
G G is the number of the genes. Note: m*G == nrow(dat).
alpha alpha is the significance level.

Details

This function computes bootstrap confidence interval of the new multivariate correlation estimator at given significance level. It is used together with cor.rep(), and cor.rep.pv() whenever necessary.

Value

This function returns matrices of upper and lower bounds corresponding to the matrix returned by cor.rep().

Author(s)

Dongxiao Zhu (http://www-personal.umich.edu/~zhud)

References

Zhu, D., Li, Y., and Hero, AO. Estimating gene expression correlation from replicated microarray data - A multivariate approach. {it Submitted}

See Also

cor.rep

Examples

# load GeneNT and GeneTS library
library(GeneNT)
d0 <- rnorm(8)
#parameters are set to small values for quick demo purposes
for(l in 2:3)
d0 <- rbind(d0, rnorm(8))
d0<- t(d0)
M <- cor.rep(d0, m = 4, G= 2)
M.bootci <- cor.rep.bootci(d0, m = 4, G= 2, alpha = 0.05)

[Package GeneNT version 1.3 Index]