max2: Compute the squared maximum of two series of pValues

In jumentib/highmed: High Dimensionnal Mediation Analysis

Description

This function compute the squared maximum of two series of pValues from the multivariate_EWAS() function. The objective of this function is to test all the markers and to determine which could be potential mediators in the exposure-outcome association.

Usage

max2(pval1, pval2, diagnostic.plot = F, ...)

Arguments

pval1	vector of pValues (p*1) of exposure.
pval2	vector of pValues (p*1) of ouctome.
diagnostic.plot	if TRUE the histogram of the p-values together with the estimate of eta0 null line is plotted. This is useful to visually check the fit of the estimated proportion of null p-values.
...	argument of the fdrtool function from the fdrtool package

Details

The pValue is computed for each markers following this formula

pV = max(pVal1, pVal2)^2

This quantity eta0, i.e. the proportion eta0 of null p-values in a given vector of p-values, is an important parameter when controlling the false discovery rate (FDR). A conservative choice is eta0 = 1 but a choice closer to the true value will increase efficiency and power - see Benjamini and Hochberg (1995, 2000) and Storey (2002) for details. We use the fdrtool package to transform pValues into qValues, which allows us to control the FDR.

Value

an object with the following attributes:

- a pValue for each markers

- a qValue for each markers

- the eta0 of the set of pValues

Author(s)

Basile Jumentier

Examples

library(highmed)

# Run multivariate_EWAS

res <- multivariate_EWAS(X = example$X, Y = example$Y, M = example$M, K = 5)

# Run max2

res <- max2(pval1 = res$pValue[, 1], pval2 = res$pValue[, 2])

# Manhattan plot

plot(-log10(res$pval), main = paste0("Eta0 = ", round(res$eta0, 3)))
abline(h = -log10(0.05 / ncol(example$M)))

jumentib/highmed documentation built on Sept. 3, 2020, 2 p.m.