Skip to contents

For each declared source set of cell types, finds the Euclidean distance from every source cell to its nearest target cell (via CrossNNDistance()) and reports the number of source cells, the median distance and the percentage (or fraction) within each radius (inclusive). Optionally adds the summed area of all cells (for example Voronoi territories) as tissue area.

Usage

SummarizeNearestDistances(
  cells,
  from,
  to,
  x = "x",
  y = "y",
  type = "type",
  radius = NULL,
  percent = TRUE,
  area = NULL,
  area_divisor = 1
)

Arguments

cells

Cell table with coordinate and type columns.

from

Named list (or named character vector) of source sets: output label to cell types. An unnamed character vector uses each type as its own set. Sets must be disjoint from each other and from to.

to

Character vector of target cell types.

x, y, type

Column names in cells.

radius

Optional numeric radii; one output row per set and radius.

percent

Report percentages (TRUE) or fractions.

area

Optional column summed over all cells into area.

area_divisor

Divisor applied to the summed area (e.g. 1e6 for square micrometres to square millimetres).

Value

A data frame with from, n, n_target, median_distance, radius, within and optionally area.

Examples

set.seed(1)
cells <- data.frame(x = runif(200, 0, 100), y = runif(200, 0, 100),
                    type = sample(c("t", "a", "b"), 200, replace = TRUE))
SummarizeNearestDistances(cells, from = list(A = "a", B = "b"), to = "t",
                          radius = 10)
#>   from  n n_target median_distance radius   within
#> 1    A 61       75        5.009221     10 95.08197
#> 2    B 64       75        5.734158     10 89.06250