ECDF: Create ECDF plot of a sample against a reference

In tnaake/MatrixQCvis: Shiny-based interactive data-quality exploration for omics data

Create ECDF plot of a sample against a reference

Description

The function ECDF creates a plot of the empirical cumulative distribution function of a specified sample and an outgroup (reference). The reference is specified by the group argument. The row-wise (feature) mean values of the reference are calculated after excluding the specified sample.

Usage

ECDF(se, sample = colnames(se), group = c("all", colnames(colData(se))))

Arguments

se	SummarizedExperiment object
sample	character, name of the sample to compare against the group
group	character, either "all" or one of colnames(colData(se))

Details

Internal use in shinyQC.

The function ECDF uses the ks.test function from stats to perform a two-sample Kolmogorov-Smirnov test. The Kolmogorov-Smirnov test is run with the alternative "two.sided" (null hypothesis is that the true distribution function of the sample is equal to the hypothesized distribution function of the group).

The exact argument in ks.test is set to NULL, meaning that an exact p-value is computed if the product of the sample sizes is less than 10000 of sample and group. Otherwise, asymptotic distributions are used whose approximations might be inaccurate in low sample sizes.

Value

gg object from ggplot2

Examples

## create se
set.seed(1)
a <- matrix(rnorm(1000), nrow = 100, ncol = 10, 
    dimnames = list(seq_len(100), paste("sample", seq_len(10))))
a[c(1, 5, 8), seq_len(5)] <- NA
cD <- data.frame(name = colnames(a), type = c(rep("1", 5), rep("2", 5)))
rD <- data.frame(spectra = rownames(a))
se <- SummarizedExperiment(assay = a, rowData = rD, colData = cD)

ECDF(se, sample = "sample 1", group = "all")

tnaake/MatrixQCvis documentation built on Nov. 4, 2024, 7:42 a.m.