metarank: Meta-rank Analysis

In cchng/MetaAnalysis: Meta-analysis

Description

A meta-analysis approach using meta-ranks.

Usage

metarank(merged)

Arguments

merged

A data frame with raw p-values of genes (rows) in each study (columns).

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

## The function is currently defined as
function (merged) 
{
    k <- apply(merged[, c(2:ncol(merged))], 1, function(x) {
        length(which(!is.na(x)))
    })
    merged <- merged[which(k >= 3), ]
    within.rank <- apply(merged[, c(2:ncol(merged))], 2, function(x) {
        rank(x, na.last = "keep")
    })
    merged$Ave.Rank <- rowMeans(within.rank, na.rm = TRUE)
    sqindex <- (within.rank - merged$Ave.Rank)^2
    merged$Heterogeneity <- rowSums(sqindex, na.rm = TRUE)
    merged$Meta.Rank <- rank(merged$Ave.Rank)
    merged.ord <- merged[order(merged$Meta.Rank), ]
    return(merged.ord)
  }

cchng/MetaAnalysis documentation built on May 29, 2019, 2:34 p.m.