Compute network graph embedding using UMAP.

get_network_graph(object, ...)

# S4 method for class 'CSNObject'
get_network_graph(
  object,
  network = DefaultNetwork(object),
  celltypes = NULL,
  graph_name = "module_graph",
  rna_assay = "RNA",
  rna_layer = "data",
  umap_method = c("weighted", "corr", "coef", "none"),
  features = NULL,
  seed = 1,
  verbose = TRUE,
  ...
)

Arguments

object: The input data, a csn object.
...: Additional arguments for umap.
network: Name of the network to use.
celltypes: Celltypes to plot. If NULL, all celltypes are plotted.
graph_name: Name of the graph.
rna_assay: Name of the RNA assay.
rna_layer: Name of the RNA slot to use.
umap_method: Method to compute edge weights for UMAP: * 'weighted' - Correlation weighted by GRN coefficient. * 'corr' - Only correlation. * 'coef' - Only GRN coefficient. * 'none' - Don't compute UMAP and create the graph directly from modules.
features: Features to use to create the graph. If NULL uses all features in the network.
seed: Random seed for UMAP computation
verbose: Print messages.

Value

A CSNObject object.