Neighbourhood Enrichment Analysis

Tests whether specific phenotype pairs are spatially enriched or depleted relative to a random permutation baseline.

Usage

NeighbourhoodEnrichment(object, radius, n_perm = 100L, seed = NULL)

Arguments

object

An SpatialCellData-class object.

radius

Numeric. Neighbourhood radius.

n_perm

Integer. Number of permutations. Default 100.

seed

Integer or NULL. Random seed.

Value

A data frame with columns from, to, observed, mean_expected, z_score, and p_value.

Examples

set.seed(1)
counts <- matrix(rnorm(200), nrow = 100,
                 dimnames = list(NULL, c("CD3", "CD8")))
coords <- data.frame(x = runif(100, 0, 500), y = runif(100, 0, 500))
meta <- data.frame(cell_id = 1:100, sample_id = "s1",
  phenotype = sample(c("A", "B"), 100, replace = TRUE))
obj <- CreateSpatialObject(counts, coords, meta)
NeighbourhoodEnrichment(obj, radius = 50, n_perm = 10)
#>   from to observed mean_expected    z_score   p_value
#> 1    A  A       58          68.0 -0.9977852 0.3183835
#> 2    B  A       79          77.9  0.1268387 0.8990681
#> 3    A  B       79          77.9  0.1268387 0.8990681
#> 4    B  B       92          84.2  0.5196665 0.6032960