Construct Spatial Metacells from Single-Cell Spatial Data — GetSpatialMetacells • SpatialEcoTyper

This function computes spatial metacells from single-cell spatial transcriptomics. Spatial metacells represent aggregated expression profiles for spatially proximal cells. The function utilizes k-nearest neighbor (KNN) weighting to aggregate expression profiles from neighboring cells and generate metacell expression profiles for each cell type.

Usage

GetSpatialMetacells(
  normdata,
  metadata,
  X = "X",
  Y = "Y",
  CellType = "CellType",
  spotCoord = NULL,
  k = 20,
  radius = 50,
  grid.size = 50,
  min.cells.per.region = 1,
  ncores = 4
)

Arguments

normdata: A matrix or data frame representing normalized gene expression data, where rows correspond to genes and columns correspond to cells.
metadata: A data frame containing metadata associated with each cell. Must include spatial coordinates (e.g., X and Y) as well as cell type annotations. The row names of the metadata must match the column names of the normdata.
X: Character string specifying the column name in metadata containing the X spatial coordinates.
Y: Character string specifying the column name in metadata containing the Y spatial coordinates.
CellType: Character string specifying the column name in metadata containing the cell type annotations.
spotCoord: A data frame containing the spatial coordinates (X and Y) of each spatial neighborhood, with neighborhood identifiers as row names. If NULL, a regular grid of spatial neighborhoods is generated automatically, spaced `grid.size` apart.
k: Integer specifying the number of nearest spatial neighbors for constructing metacells.
radius: Numeric value specifying the radius (in units of spatial coordinates) within which neighboring cells are considered.
grid.size: numeric, spacing between adjacent SN centers when `spotCoord` is auto-generated.
min.cells.per.region: Integer specifying the minimum number of cells required in each spatial neighborhood to compute metacells.
ncores: Integer specifying the number of CPU cores to use for parallel processing (default: 4).

Value

A matrix containing the spatial metacell expression profiles, with rows representing individual genes and columns representing metacells.

Examples

library(data.table)
library(Seurat)
library(SpatialEcoTyper)
scdata <- fread("https://spatialecotyper.stanford.edu/inc/inc.public.vignettes.php?file=Melanoma1_subset_counts.tsv.gz",
                sep = "\t",header = TRUE, data.table = FALSE)
rownames(scdata) <- scdata[, 1]
scdata <- as.matrix(scdata[, -1])
normdata = NormalizeData(scdata)
scmeta <- read.table("https://spatialecotyper.stanford.edu/inc/inc.public.vignettes.php?file=Melanoma1_subset_scmeta.tsv",
                      sep = "\t",header = TRUE, row.names = 1)

# Construct spatial metacells around B cells
snmeta = scmeta[scmeta$CellType=="B", c("X", "Y")]
metacells <- GetSpatialMetacells(normdata = normdata,
                                 metadata = scmeta,
                                 X = "X", Y = "Y",
                                 CellType = "CellType",
                                 spotCoord = snmeta)
head(metacells[, 1:10])
#> 6 x 10 Matrix of class "dgeMatrix"
#>          HumanMelanomaPatient1__cell_4572..B
#> PDK4                                       0
#> TNFRSF17                                   0
#> ICAM3                                      0
#> FAP                                        0
#> GZMB                                       0
#> TSC2                                       0
#>          HumanMelanomaPatient1__cell_4599..B
#> PDK4                                0.000000
#> TNFRSF17                            0.000000
#> ICAM3                               3.585688
#> FAP                                 0.000000
#> GZMB                                0.000000
#> TSC2                                0.000000
#>          HumanMelanomaPatient1__cell_4634..B
#> PDK4                                0.000000
#> TNFRSF17                            0.000000
#> ICAM3                               2.297794
#> FAP                                 0.000000
#> GZMB                                0.000000
#> TSC2                                0.000000
#>          HumanMelanomaPatient1__cell_4647..B
#> PDK4                               0.0000000
#> TNFRSF17                           0.9508847
#> ICAM3                              4.6057927
#> FAP                                0.0000000
#> GZMB                               0.0000000
#> TSC2                               0.9333217
#>          HumanMelanomaPatient1__cell_4648..B
#> PDK4                                0.000000
#> TNFRSF17                            0.000000
#> ICAM3                               3.330895
#> FAP                                 0.000000
#> GZMB                                0.000000
#> TSC2                                0.000000
#>          HumanMelanomaPatient1__cell_4652..B
#> PDK4                                0.000000
#> TNFRSF17                            0.000000
#> ICAM3                               3.661183
#> FAP                                 0.000000
#> GZMB                                0.000000
#> TSC2                                0.000000
#>          HumanMelanomaPatient1__cell_4694..B
#> PDK4                                0.000000
#> TNFRSF17                            0.000000
#> ICAM3                               2.143066
#> FAP                                 0.000000
#> GZMB                                0.000000
#> TSC2                                0.000000
#>          HumanMelanomaPatient1__cell_4723..B
#> PDK4                               0.0000000
#> TNFRSF17                           0.6030263
#> ICAM3                              1.9151089
#> FAP                                0.0000000
#> GZMB                               2.6278250
#> TSC2                               0.0000000
#>          HumanMelanomaPatient1__cell_4728..B
#> PDK4                                0.000000
#> TNFRSF17                            0.000000
#> ICAM3                               2.713943
#> FAP                                 0.000000
#> GZMB                                0.000000
#> TSC2                                0.000000
#>          HumanMelanomaPatient1__cell_4732..B
#> PDK4                                0.000000
#> TNFRSF17                            1.137039
#> ICAM3                               2.299928
#> FAP                                 0.000000
#> GZMB                                0.000000
#> TSC2                                0.000000