Generate a diffusion map — run_diffusion

This function is a wrapper around destiny::DiffusionMap function to compute a diffusion map and add it in the Seurat object. Note that it take a lot of time to run on a full expression matrix. It may cause R session to abort... Moreover, diffusion map can be well ran on a reduced space, such as PCA.

Usage

run_diffusion_map(
  sobj,
  input = NULL,
  seed = 1337L,
  verbose = FALSE,
  n_eigs = 50,
  suppress_dpt = TRUE,
  return_dm = FALSE,
  ...
)

Arguments

sobj: A Seurat object (no default)
input: CHARACTER or MATRIX : if it is a character, the function will search of an assay or a reductions in the Seurat object. Otherwise, it is supposed to be a dataframe with cells in rows and dimensions in columns (default to NULL)
seed: INTEGER : the seed to be used by destiny::DiffusionMap (default to 1337L)
verbose: LOGICAL : whether to print messages or not (default to FALSE)
n_eigs: INTEGER : number of eigenvectors/values to compute in the diffusion map (default to 50)
suppress_dpt: LOGICAL : whether to skip calculation of necessary (but spacious) information for DPT in the returned object or not (default to TRUE)
return_dm: LOGICAL : whether to return the dm object or not (default to FALSE)
...: other parameters passed to destiny::DiffusionMap

Value

This function returns the input Seurat object with the new DimReducObject. If return_dm is set to TRUE, returns a list with $sobj containing the Seurat object and $dm containing the dm object. To plot eigenvalues and dimensions, use : ggplot2::qplot(y = destiny::eigenvalues(dm))