pval2qval: Strimmer et al.'s fdrtool-based q-values

In mutoss: Unified multiple testing procedures

Description

The function pval2qval takes a vector of p-values and estimates for each case the tail area-based FDR, which can be regarded as a p-value corrected for multiplicity. This is done by calling the fdrtool function. If a cutoff is supplied, a vector of rejected hypotheses will be returned as well.

Usage

pval2qval(pValues, cutoff)

Arguments

pValues	Numeric vector of p-values to be used.
cutoff	The positive FDR cutoff for rejection. Hypotheses with qValues smaller then cutoff will be rejected.

Value

A list containing:

qValues	A numeric vector with one q-value for each hypothesis.
rejected	A logical vector indicating rejection/retention for each hypothesis when cutoff is supplied.

Author(s)

JonathanRosenblatt

References

Strimmer, K. (2008). fdrtool: a versatile R package for estimating local and tail area-based false discovery rates. Bioinformatics 24: 1461-1462.
Storey, J. D. (2003) The Positive False Discovery Rate: A Bayesian Interpretation and the q-Value. The Annals of Statistics 31(6): 2013-2035

Examples

pvals<- runif(1000)^2
pval2qval(pvals)
pval2qval(pValues=pvals, cutoff=0.1)

mutoss documentation built on May 2, 2019, 5:56 p.m.