pvalRRHO: Compute the significance of the overlap between two lists

Description

Computes the significance of the agreements between lists as returned by RRHO using resampling.

Usage

pvalRRHO(RRHO.obj, replications, stepsize=RRHO.obj$stepsize, FUN= max)

Arguments

RRHO.obj

The output object of the RRHO function.

replications

The number of samples to be taken from the distribution of the aggregated test statistic.

stepsize

Controls the resolution of the test: how many items between any two overlap tests (i.e., netween any two $i$-s and two $j$-s.)

FUN

The function aggregating infomation from the whole overlap matrix into one summary statistic. Typically the $min$ pvalue, or $max$ on $-log(pval)$ scale.

Value

pval: The FWER corrected significance of observed aggregated pvalue.
FUN.ecdf: The simulated sampling distribution of the aggregated pvalues.
FUN: The matrix aggregation function used. typicall max for minimal p-value.
n.items: Length of lists.
stepsize: See RRHO
replications: The number of simulation replications.
call: The function call.

Details

The distribution of $FUN(-log(pval))$ is computed using resampling. The aggregating function will typically be the max function, corresponding to the maximal -log(pvalue), i.e., the most significant agreement over all sublists. The distribution is computed by resampling pairs of null sequences, computing the significances of all the overlaps as done in the reference, applying the aggregating function supplied by the user, and returning the permutation based significance.

Examples

Run this code

list.length <- 100
list.names <- paste('Gene',1:list.length, sep='')
gene.list1<- data.frame(list.names, sample(list.length))
gene.list2<- data.frame(list.names, sample(list.length))
RRHO.example <-  RRHO(gene.list1, gene.list2, alternative='enrichment')
pval.testing <- pvalRRHO(RRHO.example,50)