plotEvalROC: Visualize error rate curves and associated summary statistics

In bvieth/powsimR: Power Simulations for RNA-sequencing

Visualize error rate curves and associated summary statistics

Description

This function plots the results of evaluateROC for assessing relative operating characteristic curves and summary statistics.

Usage

plotEvalROC(evalRes,
cutoff=c('liberal', 'conservative'),
Annot=TRUE)

Arguments

evalRes	The output of evaluateROC.
cutoff	Character vector defining whether the "liberal" or "conservative" FDR control is considered.
Annot	A logical vector. If TRUE, a short figure legend under the plot is included.

Value

A ggplot object.

Author(s)

Beate Vieth

Examples

## Not run: 
# estimate gene parameters
data("CELseq2_Gene_UMI_Counts")
estparam_gene <- estimateParam(countData = CELseq2_Gene_UMI_Counts,
                               readData = NULL,
                               batchData = NULL,
                               spikeData = NULL, spikeInfo = NULL,
                               Lengths = NULL, MeanFragLengths = NULL,
                               RNAseq = 'singlecell', Protocol = 'UMI',
                               Distribution = 'NB', Normalisation = "scran",
                               GeneFilter = 0.1, SampleFilter = 3,
                               sigma = 1.96, NCores = NULL, verbose = TRUE)
# define log2 fold change
p.lfc <- function(x) sample(c(-1,1), size=x,replace=T)*rgamma(x, shape = 1, rate = 2)
# set up simulations
setupres <- Setup(ngenes = 10000, nsims = 10,
                  p.DE = 0.1, pLFC = p.lfc,
                  n1 = c(20,50,100), n2 = c(30,60,120),
                  Thinning = NULL, LibSize = 'equal',
                  estParamRes = estparam_gene,
                  estSpikeRes = NULL,
                  DropGenes = FALSE,
                  sim.seed = 34269, verbose = TRUE)
# run simulation
simres <- simulateDE(SetupRes = setupres,
                     Prefilter = "FreqFilter", Imputation = NULL,
                     Normalisation = 'scran', Label = 'none',
                     DEmethod = "limma-trend", DEFilter = FALSE,
                     NCores = NULL, verbose = TRUE)
# evaluation
evalrocres <- evaluateROC(simRes = simres,
                          alpha.type = "adjusted",
                          MTC = 'BH', alpha.nominal = 0.05,
                          raw = FALSE)
# plot evaluation
plotEvalROC(evalRes = evalrocres, cutoff = "conservative", Annot = TRUE)

## End(Not run)

bvieth/powsimR documentation built on Aug. 19, 2023, 7:48 p.m.