Reduces dimensionality (tSNE, UMAP, PCA, or custom)

Source: R/scMethrix_dimensionality.R

Reduces dimensionality (tSNE, UMAP, PCA, or custom)

dim_red_scMethrix(
  scm,
  assay = "score",
  type = c("tSNE", "UMAP", "PCA"),
  var = c("top", "rand"),
  top_var = 1000,
  perplexity = 30,
  verbose = FALSE,
  n_components = 2,
  n_neighbors = 15,
  ...
)

Arguments

scm

scMethrix; Input scMethrix object
assay

string; The assay to use. Default is 'score'
type

string; the type of imputation "tSNE","UMAP", or "PCA"
var

strning; Choose between random CpG sites ('rand') or most variable CpGs ('top'). Default 'top'
top_var

integer; Number of variable CpGs to use. Default 1000 Set it to NULL to use all CpGs (which is not recommended due to memory requirements).
perplexity

numeric; Perplexity parameter (should not be bigger than 3 * perplexity < nrow(X) - 1, see details for interpretation)
verbose

boolean; flag to output messages or not
n_components

integer; Number of components to use
n_neighbors

integer; number of nearest neighbors for UMAP
...

Other arguments that can be passed to Rtsne

Value

scMethrix object with reducedDim assay

Details

Does reduction stuff

See also

plot_dim_red() for plotting

Examples

data('scMethrix_data')