Compute a graph layout

Calculates a graph layout for a component's edgelist, and outputs a list with the bipartite graph, layout, and antibody counts per A pixel.

Usage

ComputeLayout(object, ...)

# S3 method for tbl_graph
ComputeLayout(
  object,
  layout_method = c("pmds", "wpmds", "fr", "kk", "drl"),
  dim = 2,
  normalize_layout = FALSE,
  project_on_unit_sphere = FALSE,
  k = 0,
  pivots = 100,
  seed = 123,
  custom_layout_function = NULL,
  custom_layout_function_args = NULL,
  ...
)

# S3 method for CellGraph
ComputeLayout(
  object,
  layout_method = c("pmds", "wpmds", "fr", "kk", "drl"),
  layout_name = NULL,
  dim = 2,
  normalize_layout = FALSE,
  project_on_unit_sphere = FALSE,
  k = 0,
  pivots = 100,
  seed = 123,
  custom_layout_function = NULL,
  custom_layout_function_args = NULL,
  ...
)

# S3 method for MPXAssay
ComputeLayout(
  object,
  layout_method = c("pmds", "wpmds", "fr", "kk", "drl"),
  layout_name = NULL,
  dim = 2,
  normalize_layout = FALSE,
  project_on_unit_sphere = FALSE,
  k = 0,
  pivots = 100,
  seed = 123,
  verbose = TRUE,
  custom_layout_function = NULL,
  custom_layout_function_args = NULL,
  cl = NULL,
  ...
)

# S3 method for CellGraphAssay
ComputeLayout(
  object,
  layout_method = c("pmds", "wpmds", "fr", "kk", "drl"),
  layout_name = NULL,
  dim = 2,
  normalize_layout = FALSE,
  project_on_unit_sphere = FALSE,
  k = 0,
  pivots = 100,
  seed = 123,
  verbose = TRUE,
  custom_layout_function = NULL,
  custom_layout_function_args = NULL,
  cl = NULL,
  ...
)

# S3 method for CellGraphAssay5
ComputeLayout(
  object,
  layout_method = c("pmds", "wpmds", "fr", "kk", "drl"),
  layout_name = NULL,
  dim = 2,
  normalize_layout = FALSE,
  project_on_unit_sphere = FALSE,
  k = 0,
  pivots = 100,
  seed = 123,
  verbose = TRUE,
  custom_layout_function = NULL,
  custom_layout_function_args = NULL,
  cl = NULL,
  ...
)

# S3 method for Seurat
ComputeLayout(
  object,
  assay = NULL,
  layout_method = c("pmds", "wpmds", "fr", "kk", "drl"),
  layout_name = NULL,
  dim = 2,
  normalize_layout = FALSE,
  project_on_unit_sphere = FALSE,
  k = 0,
  pivots = 100,
  seed = 123,
  verbose = TRUE,
  custom_layout_function = NULL,
  custom_layout_function_args = NULL,
  ...
)

Arguments

object: An object
...: Additional parameters passed to other methods
layout_method: The method for calculating the graph layout: weighted Pivot MDS ("wpmds"), Pivot MDS (pmds), Fruchterman-Reingold ("fr"), Kamada-Kawai ("kk"), "drl".
dim: An integer specifying the dimensions of the layout (2 or 3)
normalize_layout: Logical specifying whether the coordinate system should be centered at origo and the coordinates scaled such that their median length (euclidean norm) is 1.
project_on_unit_sphere: Should the resulting layout be projected onto a unit sphere?
k: The size of the neighborhood from which to pool counts from in the UPIA antibody count table. 0 is recommended.
pivots: Only used for "wpmds" and "pmds" layout algorithms. See ?layout_with_pmds for details
seed: Set seed for reproducibility
custom_layout_function: A custom function for layout computation. The function should take a tbl_graph object and a dim value as input and return a matrix of dimensions NxD, where N is the number of nodes and D is equal to dim. Note that this will override the layout_method.
custom_layout_function_args: A list of arguments passed to custom_layout_function. The dim is automatically passed to custom_layout_function and should not be included in custom_layout_function_args.
layout_name: The name of the computed layout. If this name is not given, the layout_method will be used as the name.
verbose: Print messages
cl: A cluster object created by makeCluster, or an integer to indicate number of child-processes (integer values are ignored on Windows) for parallel evaluations. See Details on performance in the documentation for pbapply. The default is NULL, which means that no parallelization is used.
assay: Name of assay to compute layouts for

Value

An object containing a graph layout

Examples

library(pixelatorR)
library(dplyr)

pxl_file <- system.file("extdata/five_cells",
                        "five_cells.pxl",
                        package = "pixelatorR")

# Load example data
seur <- ReadMPX_Seurat(pxl_file)
#> ✔ Created a 'Seurat' object with 5 cells and 80 targeted surface proteins

# Load 1 cellgraph
seur <- LoadCellGraphs(seur, cells = colnames(seur)[1],
                       load_as = "Anode", force = TRUE)
#> →    Loading CellGraphs for 1 cells from sample 1
#> ✔ Successfully loaded 1 CellGraph object(s).

# Get CellGraph
cg <- CellGraphs(seur)[[colnames(seur)[1]]]

# Get tbl_graph object
tbl_graph <- slot(cg, name = "cellgraph")

# Compute layout for a tbl_graph
layout <- ComputeLayout(tbl_graph, layout_method = "fr")
layout %>% head()
#> # A tibble: 6 × 2
#>       x     y
#>   <dbl> <dbl>
#> 1 -13.9 -3.73
#> 2 -14.3 -1.98
#> 3 -12.6 -3.69
#> 4 -11.5 -3.45
#> 5 -12.9 -4.20
#> 6 -13.5 -2.32


# Compute layout for a CellGraph
cg <- ComputeLayout(cg, layout_method = "fr")


# Compute layout for a CellGraphAssay
cg_assay <- ComputeLayout(seur[["mpxCells"]], layout_method = "fr")
#> ℹ Computing layouts for 1 graphs


# Compute layout for a Seurat object
seur <- ComputeLayout(seur, layout_method = "fr")
#> ℹ Computing layouts for 1 graphs

Usage

Arguments

Value

See also

Examples