comb_cor_p: com_cor_p

In goodarzilab/Ribolog: Logistic-regression-based analysis of ribosome profiling data

com_cor_p

Description

Function to combine correlated p-values

Usage

comb_cor_p(p, w, cm)

Arguments

p	Array (nxm matrix) of p-values. n: number of features e.g. genes or transcripts, m: number of tests being combined.
w	Array (nxm matrix) of weights. n: number of features e.g. genes or transcripts, m: number of tests being combined.
cm	Correlation matrix of z-values or any other test statistics of individual tests that are being combined (mxm). m: number of samples.

Details

This is based on the the thoery expounded in a 2003 paper by Kepher Makambi.

Value

A matrix with three columns: [1] M (combined test statistic) [2] nu (degree of freedom of chi-sq distribution) [3] p.com (the combined or meta-analytical p-value)

References

Makambi, K. 2003. Weighted inverse chi-square method for correlated significance tests. Journal of Applied Statistics, 30(2): 225-234.

goodarzilab/Ribolog documentation built on Oct. 7, 2022, 10:14 p.m.