matrix_wilcox: Vectorized wilcox rank-sums test

In YosefLab/FastProjectR: Functional interpretation of single cell RNA-seq latent manifolds

Vectorized wilcox rank-sums test

Description

Given indices in cluster_ii, compute the ranksums test on every column for values in cluster_ii vs the rest.

Usage

matrix_wilcox(ranks, cluster_ii, check_na = FALSE, check_ties = FALSE)

Arguments

ranks	matrix of ranks, each column representing a separate variable
cluster_ii	numeric vector - indices denoting the group to be compared
check_na	- whether or not to check for NA values (slows computation)
check_ties	- whether or not to check for ties in ranks (slows computation)

Value

pval - numeric vector, pvalue for each column

stat - numeric vector, test statistic (AUC) for each column

YosefLab/FastProjectR documentation built on June 13, 2024, 4:20 p.m.