structSSI (version 1.2.1)

GBH-class: Manage Group Benjamini-Hochberg Outputs

Description

This defines a class GBH for managing outputs from the Group Benjamini-Hochberg procedure. This object makes it easy to print, summarize, and plot the results of the testing procedure.

Prints the entire table of adjusted p-values and their associated FDR adjusted significance levels, together with the estimated proportions of null hypotheses, within each group.

Shows results from multiple testing via GBH. Also supplies the estimated proportion of null hypothesis within each group and a table of counts of adjusted significance across groups.

Show results of testing hypothesis, sorted according to GBH adjusted significance, shape coded according to group membership, and color coded according to pre and post GBH p-value adjustment.

Usage

# S4 method for GBH
initialize(.Object, ...)
# S4 method for GBH
show(object)
# S4 method for GBH
summary(object)
# S4 method for GBH,ANY
plot(x, title = "GBH Adjustment", ...)

Arguments

.Object

Dummy to initialize S4 class

...

Any other arguments are accepted, but they will be ignored.

object

A GBH object whose hypotheses we want to summarize.

x

A GBH object whose p-values to plot.

title

The name added to the top of the plot. Defaults to 'GBH Adjustment'.

Slots

p.vals

Object of class 'data.frame'. Each row corresponds to an individual hypothesis. The different columns correspond to, * hypothesisIndex: The index of the current hypothesis in the unadjp vector * hypothesisName: The name of the current hypothesis, from the names of the unadjp vector * unadjp: The unadjusted p-values input from unadjp * adjp: The adjusted p-values, after the GBH adjustment. * group: The group to which the original hypothesis belonged * significance: A code for the significance of each hypothesis

pi0

Object of class 'numeric'. The proportion of null hypotheses within each group. This is either known a priori or estimated adaptively from the unadjusted p-values.

adaptive

Object of class 'logical'. An indicator of whether the proportion pi0 was estimated adaptively from the data or known a priori.

alpha

Object of class 'numeric'. The level at which the FDR is controlled, during the GBH procedure.

Examples

# These are the unadjusted p-values corresponding to the outcome of some
# multiple testing experiment. The first 500 hypotheses are null and the last
# 1500 are true alternatives.

unadjp <- c(runif(500, 0, 0.01), runif(1500, 0, 1))
names(unadjp) <- paste('Hyp: ', 1:2000)

# These are the unadjusted p-values corresponding to the outcome of some
# multiple testing experiment. The first 500 hypotheses are null and the last
# 1500 are true alternatives.
unadjp <- c(runif(500, 0, 0.01), runif(1500, 0, 1))
names(unadjp) <- paste('Hyp: ', 1:2000)

# Here there are two groups total we have randomly assigned hypotheses to these
# two groups.
group.index <- c(sample(1:2, 2000, replace = TRUE))

# Perform the GBH adjustment.
result <-  Adaptive.GBH(unadjp, group.index, method = 'storey')

# A summary of the GBH adjustment
summary(result)