simesCT {rOCEAN}R Documentation

Closed testing with Simes

Description

Calculates five parameters from closed testing with Simes local tests based on raw data. These parameter are unique per data/alpha-level combination and do not depend on feature sets. Calculation may be somewhat long depending on the size of data sets and PC configurations.

Usage

simesCT(om1, om2, mps, m, alpha = 0.05)

Arguments

om1, om2

Two omics data sets where rows are features and columns are samples.

mps, m

Optional, pre-calculated matrix/vector of pairwise associations and the size. To save time in calculation of parameters, a threshold such as the type I error may be applies to remove larger p-values. If a threshold is used, size of matrix and m will not match. m should always be the size of the matrix of associations (number of features in om1 X number of features in om2).

alpha

type I error rate, default value is 0.05.

Value

Vector of integers: grand H value, concentration p-value, size of concentration set z, size of the original pair-wise associations matrix and the type I error level used in calculations.

References

See more details in "Hommel's procedure in linear time" doi:10.1002/bimj.201700316.

Examples


#number of feature per omic data set
n_cols<-100
n_rows<-120

#random matrix of p-values
set.seed(1258)
pvalmat<-matrix(runif(n_rows*n_cols, min=0, max=1)^6, nrow=n_rows, ncol=n_cols)

#calculate CT parameters
gCT<-simesCT(mps=pvalmat, m=nrow(pvalmat)*ncol(pvalmat))


[Package rOCEAN version 1.0 Index]