#' Compute all pairwise trait correlations
#'
#' Computes pairwise correlations for all traits
#' using either \link[WGCNA]{cor} (faster) or
#' \link[stats]{cor} (slower).
#' @inheritParams get_obsm
#' @inheritParams stats::cor
#' @inheritParams WGCNA::cor
#' @returns A single-cell object with a graph.
#'
#' @keywords internal
calc_cor <- function(obj,
keys = NULL,
graph_key = NULL,
assay = NULL,
slot = NULL,
transpose = FALSE,
method = "pearson",
fill_na = NULL,
use = "all.obs",
nThreads = 1,
verbose = TRUE) {
# devoptera::args2vars(calc_cor)
#### Extract relevant matrix ####
if (!is.null(keys)) {
X <- get_obsm(
obj = obj,
keys = keys,
n = 1,
verbose = verbose)
#### Important! must transpose ####
X <- to_sparse(obj = X, transpose = TRUE)
} else {
X <- get_x(
obj = obj,
assay = assay,
slot = slot,
transpose = transpose,
n = 1,
verbose = verbose)
}
if (!is.null(fill_na)) X[is.na(X)] <- fill_na
#### Compute corr ####
if (is_installed(pkg = "WGCNA")) {
messager("Computing r with WGCNA.", v = verbose)
Xcor <- WGCNA::cor(x = X,
method = method,
use = use,
nThreads = nThreads,
verbose = verbose)
} else {
messager("Computing r with stats.", v = verbose)
Xcor <- stats::cor(x = X,
use = use,
method = method)
}
return(Xcor)
}
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.