plotSD: Estimate And Plot Median Standard Deviation

In rnaseqcomp: Benchmarks for RNA-seq Quantification Pipelines

Description

For each cell line in each pipeline, the standard deviation of detrend logsignals are calculated for individual features. Then, loess smooth on standard deviation are plotted stratified by detrended log signals for select cell line. The median of standard deviation at three different levels of detrend logsignals are reported.

Usage

plotSD(dat, constant = 0.5, loessspan = 1/3, thresholds = c(1, 6),
  plotcell = 1, ...)

Arguments

dat	A rnaseqcomp S4 class object.
constant	A numeric pseudo-constant to be added on all the signals before transferred to log scale. (default: 0.5)
loessspan	A numeric number indicating span used for loess smooth. Details see loess.smooth function. (Default: 1/3)
thresholds	A vector of two numbers define cutoffs for three levels of detreded log signals. (default: c(1, 6))
plotcell	1 or 2 indicating which cell line will be plotted. If values other than 1 and 2, both cell lines will be plotted. This value won't affect estimation for both cell lines. (default: 1)
...	Parameters for base function plot.

Value

plot	SD plots of quantification pipelines for selected cell line by plotcell.
list	A list of two matrices of median and standard error of standard deviations.

Examples

data(simdata)
condInfo <- factor(simdata$samp$condition)
repInfo <- factor(simdata$samp$replicate)
evaluationFeature <- rep(TRUE, nrow(simdata$meta))
calibrationFeature <- simdata$meta$house & simdata$meta$chr == 'chr1'
unitReference <- 1
dat <- signalCalibrate(simdata$quant, condInfo, repInfo, evaluationFeature,
calibrationFeature, unitReference, calibrationFeature2 = calibrationFeature)
plotSD(dat)

rnaseqcomp documentation built on Nov. 8, 2020, 8:04 p.m.