testZeroes-method: testZeroes: Test for differential proportions of zero values

In waddR: Statistical tests for detecting differential distributions based on the 2-Wasserstein distance

Description

Test for differential proportions of zero expression between two conditions for a specified set of genes

Usage

testZeroes(x, y, these = seq_len(nrow(x)))

## S4 method for signature 'matrix,vector,ANY'
testZeroes(x, y, these = seq_len(nrow(x)))

## S4 method for signature 'SingleCellExperiment,SingleCellExperiment,vector'
testZeroes(x, y, these = seq_len(nrow(x)))

Arguments

x	matrix of single-cell RNA-sequencing expression data with genes in rows and samples (cells) in columns
y	vector of condition labels
these	vector of row numbers (i.e. gene numbers) employed to test for differential proportions of zero expression. Default is seq_len(nrow(dat))

Details

Test for differential proportions of zero expression between two conditions that is not explained by the detection rate using a (Bayesian) logistic regression model. Adapted from the scDD package (Korthauer et al. 2016).

Value

A vector of (unadjusted) p-values

References

Korthauer et al. (2016).

Examples

x1 <- c(rnorm(100,42,1), rnorm(102,45,3))
x2 <- c(rnorm(100,0,1), rnorm(102,0,2))
dat <- matrix(c(x1,x2), nrow=2, byrow=TRUE)
condition <- c(rep(1,100), rep(2,102))
# test over all rows
testZeroes(dat, condition)
# only consider the second row
testZeroes(dat, condition, these=c(2))

waddR documentation built on Nov. 8, 2020, 8:32 p.m.