diagplot.avg.ftd: Create average False (or True) Discovery curves

In pmoulos/metaseqr: An R package for the analysis and result reporting of RNA-Seq data by combining multiple statistical algorithms.

Description

This function creates false (or true) discovery curves using a list containing several outputs from diagplot.ftd.

Usage

    diagplot.avg.ftd(ftdr.obj, output = "x11",
        path = NULL, draw = TRUE, ...)

Arguments

ftdr.obj	a list with outputs from diagplot.ftd.
output	one or more R plotting device to direct the plot result to. Supported mechanisms: "x11" (default), "png", "jpg", "bmp", "pdf" or "ps".
path	the path to create output files.
draw	boolean to determine whether to plot the curves or just return the calculated values (in cases where the user wants the output for later averaging for example). Defaults to TRUE (make plots).
...	further arguments to be passed to plot devices, such as parameter from par.

Value

A named list with two members: the first member (avg.ftdr) contains a list with the means and the standard deviations of the averaged ftdr.obj and are used to create the plot. The second member (path) contains the path to the created figure graphic.

Author(s)

Panagiotis Moulos

Examples

p11 <- 0.001*matrix(runif(300),100,3)
p12 <- matrix(runif(300),100,3)
p21 <- 0.001*matrix(runif(300),100,3)
p22 <- matrix(runif(300),100,3)
p31 <- 0.001*matrix(runif(300),100,3)
p32 <- matrix(runif(300),100,3)
p1 <- rbind(p11,p21)
p2 <- rbind(p12,p22)
p3 <- rbind(p31,p32)
rownames(p1) <- rownames(p2) <- rownames(p3) <-
    paste("gene",1:200,sep="_")
colnames(p1) <- colnames(p2) <- colnames(p3) <-
    paste("method",1:3,sep="_")
truth <- c(rep(1,40),rep(-1,40),rep(0,20),
    rep(1,10),rep(2,10),rep(0,80))
names(truth) <- rownames(p1)
ftd.obj.1 <- diagplot.ftd(truth,p1,N=100,draw=FALSE)
ftd.obj.2 <- diagplot.ftd(truth,p2,N=100,draw=FALSE)
ftd.obj.3 <- diagplot.ftd(truth,p3,N=100,draw=FALSE)
ftd.obj <- list(ftd.obj.1,ftd.obj.2,ftd.obj.3)
avg.ftd.obj <- diagplot.avg.ftd(ftd.obj)

pmoulos/metaseqr documentation built on Dec. 29, 2020, 5:56 a.m.