p.adjust {POSTm}

Adjust P-values for Multiple Comparisons

Description

Given a set of p-values, returns p-values adjusted using one of several methods.

Given an object of class POST, returns the POST and SO p-values adjusted using the specified method(s).

Usage

p.adjust(p, ...)

## S3 method for class 'POST'
p.adjust(p, ..., method = p.adjust.methods, n = length(x = p), alpha = 0.05)

Arguments

p

An object of class POST. The value object returned by post().

...

Ignored.

method

A character vector object. One or more correction methods. must be one or more from {"holm", "hochberg", "hommel", "bonferroni", "BH", "BY", "fdr", "none", "TSBH"}.

n

number of comparisons, must be at least ‘length(p)’; Per ?stats::p.adjust – only set this (to non-default) when you know what you are doing!

alpha

A numeric object. A nominal type I error rate, or a vector of error rates, used for estimating the number of true null hypotheses in the two-stage Benjamini & Hochberg procedure ("TSBH"). Default is 0.05. NOTE: This definition does not conform to stats::p.adjust.

Details

Extends the p.adjust() method of stats and incorporates the TSBH method available through multtest::mt.raw2adjp.

Value

A list, the contents of which will depend on the type of analysis. If a tree object was provided for the analysis, the list will contain

adjPOST

A matrix object. The adjusted p-values for the POST p-values. The rows correspond to each OTU; the columns to each specified adjustment method.

adjSO

A matrix object. The adjusted p-values for the single OTU p-values. The rows correspond to each OTU; the columns to each specified adjustment method.

If no tree object was provided, only adjSO as described above will be returned.

Examples

data("POSTmData")

y <- as.integer(x = metadata[,"GC"] == "BV")
X <- metadata[,"mRace"]

result <- post(y = y, 
               X = X, 
               OTU = otu[,1:20], 
               tree = otutree,
               cValues = seq(0,0.05,by=0.01))

 p.adjust(p = result, method = c("BH","BY"))