Compute global bivariate spatial correlation — global_biv

Compute bivariate global spatial correlation. You can use either Lee's L (default) or bivariate Moran's I statistics.

global_biv_spatcor(
  seurat,
  var1 = NULL,
  var2 = NULL,
  sim = 49,
  lag = 1,
  method = c("lee", "moran"),
  alt = "two.sided",
  layer = "data",
  seed = 347548
)

Arguments

seurat: a Seurat object containing Kandinsky data (`KanData()`)
var1: character string or vector specifying the first variable(s) to use to compute bivariate spatial correlation.
var2: character string or vector specifying the second variable(s) to use to compute bivariate spatial correlation
sim: number of Monte Carlo simulations to be run for estimating Moran/Lee coefficients significance
lag: integer value indicating the extent of cell/spot neighbours to be considered to calculate Moran/Lee statistics. `lag = 1` indicates that only 1st order neighbours will be considered, while `lag=2` indicates that all neighbours of each 1st order neighbour will be also considered for each spot/cell, and so on.
method: character string specifying which method to use to estimate spatial correlation. Must be one of the following: "lee" (default), "moran".
alt: a character string specifying the alternative hypothesis, Must be one of the following: "two.sided" (default), "greater" or "less".
layer: character string indicating which Seurat layer will be considered to search for the variable of interest
seed: numeric, random seed for reproducibility. Default is set to 347548

Value

a data.frame reporting the chosen spatial statistic coefficient with the associated p value for each possible combination between variables specified with `var1` `var2` parameters.