entropyTest: One-sided two-sample test for entropy comparison

Description

A one-sided two-sample test compares the entropy of a (high-dimensional) multivariate random variable between two groups. The test is one-sided: one group is a priori suspected to have a larger entropy. The null distribution is obtained via an efficient permutation resampling algorithm.

Usage

entropyTest(Y, id, nPerm = 1000, method = "normal", k0 = 1, k1 = 1, 
center = TRUE, lowCiThres=0.10, ncpus=1, verbose=FALSE)

Arguments

(High-dimensional) matrix. Rows are assumed to represent the samples, and columns represent the samples' genes or traits.

An indicator variable for the two groups to be compared. The groups should be coded as 0 and 1. There is an asymmetric interest in the groups: the group indicated by 1 is believed to exhibit a larger entropy.

nPerm

Number of permutations.

method

Distributional assumption under which entropy is to be estimated.

k-nearest neighbor parameter for group comprising of samples indicated by a zero in the indicator variable id.

k-nearest neighbor parameter for group comprising of samples indicated by a one in the indicator variable id.

center

Logical indicator: should the columns of Y be centered around zero?

lowCiThres

A value between 0 and 1. Determines speed of efficient p-value calculation. If the probability of a p-value being below lowCiThres is smaller than 0.001 (read: the test is unlikely to become significant), the permutation analysis is terminated and a p-value of 1.00 is reported.

ncpus

Number of cpus used for the permutations.

verbose

Logical indicator: should intermediate output be printed on the screen?

Value

Object of entTest-class.

References

Van Wieringen, W.N., Van der Vaart, A.W. (2011), "Statistical analysis of the cancer cell's molecular entropy using high-throughput data", Bioinformatics, 27(4), 556-563.

Van Wieringen, W.N., Van de Wiel, M.A., Van der Vaart, A.W. (2008), "A test for partial differential expression", Journal of the American Statistical Association, 103(483), 1039-1049.

Examples

Run this code

# load data
data(pollackGE16) 
Y <- exprs(pollackGE16) 

# assign samples to groups
id <- sample(c(0,1), 41, replace=TRUE) 

# perform testing and print test results
testRes <- entropyTest(t(Y), id, nPerm = 5, method="knn") 
summary(testRes)