uc.dist {GSDA}

U-centered Distance Matrix

Description

The function calculates the U-centered distance matrix for a variable.

Usage

uc.dist(X, dmeth = "me")

Arguments

X

vector, matrix, or data.frame to compute a distance matrix

dmeth

Distance method to use, options include "oe" for overall Euclidean, "me" for marginal Euclidean, "om" for overall Manhattan, "mm" for marginal Manhattan, "ct" for categorical, and "st" for censored survival time.

Details

A distance matrix is first calculated for a scale or data frame of a variable. The distance matrix is then centered according to U-centering formula on page 6 of arXiv 1902.03291 paper.

Value

For distance methods "oe", "om", "ct", and "st", one matrix of overall distances computed using data from all variables. For distance methods "me" and "mm", an array of distance matrices, one distance matrix per variable.

Author(s)

Xueyuan Cao [e-mail] and Stanley Pounds [e-mail]

References

Cao X and Pounds S (2021) Gene-Set Distance Associations (GSDA): A Powerful Tool for Gene-Set Association Analysis.

Zhu C, Yao S, Zhang X and Shao X. Distance-based and RKHS-based Dependence Metrics in High Dimension. arXiv:1902.03291

See Also

U.center, prep.gsda

Examples

data(target.aml.expr)
target.aml.expr=sqrt(target.aml.expr)
target.aml.expr=t(target.aml.expr)
oe.dist=uc.dist(target.aml.expr,"oe") # overall Euclidean