Run UMAP (Uniform Manifold Approximation and Projection)

Usage

RunUMAP2(object, ...)

# S3 method for class 'Seurat'
RunUMAP2(
  object,
  reduction = "pca",
  dims = NULL,
  features = NULL,
  neighbor = NULL,
  graph = NULL,
  assay = NULL,
  layer = "data",
  umap.method = "uwot",
  reduction.model = NULL,
  n_threads = NULL,
  return.model = FALSE,
  n.neighbors = 30L,
  n.components = 2L,
  metric = "cosine",
  n.epochs = 200L,
  spread = 1,
  min.dist = 0.3,
  set.op.mix.ratio = 1,
  local.connectivity = 1L,
  negative.sample.rate = 5L,
  a = NULL,
  b = NULL,
  learning.rate = 1,
  repulsion.strength = 1,
  reduction.name = "umap",
  reduction.key = "UMAP_",
  verbose = TRUE,
  seed.use = 11,
  ...
)

# Default S3 method
RunUMAP2(
  object,
  assay = NULL,
  umap.method = "uwot",
  reduction.model = NULL,
  n_threads = NULL,
  return.model = FALSE,
  n.neighbors = 30L,
  n.components = 2L,
  metric = "cosine",
  n.epochs = 200L,
  spread = 1,
  min.dist = 0.3,
  set.op.mix.ratio = 1,
  local.connectivity = 1L,
  negative.sample.rate = 5L,
  a = NULL,
  b = NULL,
  learning.rate = 1,
  repulsion.strength = 1,
  reduction.key = "UMAP_",
  verbose = TRUE,
  seed.use = 11L,
  ...
)

Arguments

object: An object. This can be a Seurat object, a matrix-like object, a Neighbor object, or a Graph object.
...: Additional arguments to be passed to UMAP.
reduction: The reduction to be used. Default is "pca".
dims: The dimensions to be used. Default is NULL.
features: The features to be used. Default is NULL.
neighbor: The name of the Neighbor object to be used. Default is NULL.
graph: The name of the Graph object to be used. Default is NULL.
assay: The assay to be used. Default is NULL.
layer: The layer to be used. Default is "data".
umap.method: The UMAP method to be used. Options are "naive" and "uwot". Default is "uwot".
reduction.model: A DimReduc object containing a pre-trained UMAP model. Default is NULL.
n_threads: Num of threads used.
return.model: Whether to return the UMAP model. Default is FALSE.
n.neighbors: A number of nearest neighbors to be used. Default is 30.
n.components: A number of UMAP components. Default is 2.
metric: The metric or a function to be used for distance calculations. When using a string, available metrics are: euclidean, manhattan. Other available generalized metrics are: cosine, pearson, pearson2. Note the triangle inequality may not be satisfied by some generalized metrics, hence knn search may not be optimal. When using metric.function as a function, the signature must be function(matrix, origin, target) and should compute a distance between the origin column and the target columns. Default is "cosine".
n.epochs: A number of iterations performed during layout optimization for UMAP. Default is 200.
spread: The spread parameter for UMAP, used during automatic estimation of a/b parameters. Default is 1.
min.dist: The minimum distance between UMAP embeddings, determines how close points appear in the final layout. Default is 0.3.
set.op.mix.ratio: Interpolate between (fuzzy) union and intersection as the set operation used to combine local fuzzy simplicial sets to obtain a global fuzzy simplicial sets. Both fuzzy set operations use the product t-norm. The value of this parameter should be between 0.0 and 1.0; a value of 1.0 will use a pure fuzzy union, while 0.0 will use a pure fuzzy intersection.
local.connectivity: The local connectivity, used during construction of fuzzy simplicial set. Default is 1.
negative.sample.rate: The negative sample rate for UMAP optimization. Determines how many non-neighbor points are used per point and per iteration during layout optimization. Default is 5.
a: The parameter a for UMAP optimization. Contributes to gradient calculations during layout optimization. When left at NA, a suitable value will be estimated automatically. Default is NULL.
b: The parameter b for UMAP optimization. Details see parameter a.
learning.rate: The initial value of "learning rate" of layout optimization. Default is 1.
repulsion.strength: A numeric value determines, together with alpha, the learning rate of layout optimization. Default is 1.
reduction.name: The name of the reduction to be stored in the Seurat object. Default is "umap".
reduction.key: The prefix for the column names of the UMAP embeddings. Default is "UMAP_".
verbose: Whether to print the message. Default is TRUE.
seed.use: The random seed to be used. Default is 11.

Examples

pancreas_sub <- Seurat::FindVariableFeatures(pancreas_sub)
#> Finding variable features for layer counts
pancreas_sub <- RunUMAP2(
  object = pancreas_sub,
  features = SeuratObject::VariableFeatures(pancreas_sub)
)
#> Warning: Layer ‘data’ is empty
#> Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 't': subscript out of bounds
CellDimPlot(
  pancreas_sub,
  group.by = "CellType",
  reduction = "umap"
)
#> Warning: No shared levels found between `names(values)` of the manual scale and the
#> data's fill values.