computeUMAP-methods: Perform UMAP.

In edroaldo/fusca: Framework for Unified Single-Cell Analysis

Perform UMAP.

Description

Dimensionality reduction using UMAP provided by the umap package.

Usage

computeUMAP(
  object,
  num.pcs,
  spread = 1,
  min_dist = 0.3,
  n_neighbors = 30,
  metric = "cosine",
  set_op_mix_ratio = 1,
  local_connectivity = 1,
  repulsion_strength = 1,
  negative_sample_rate = 5,
  seed = 1
)

## S4 method for signature 'CellRouter'
computeUMAP(
  object,
  num.pcs,
  spread = 1,
  min_dist = 0.3,
  n_neighbors = 30,
  metric = "cosine",
  set_op_mix_ratio = 1,
  local_connectivity = 1,
  repulsion_strength = 1,
  negative_sample_rate = 5,
  seed = 1
)

Arguments

object	CellRouter object.
num.pcs	numeric; number of principal components used for dimensionality reduction using UMAP.
spread	numeric; used during automatic estimation of a/b parameters.
min_dist	numeric; determines how close points appear in the final layout.
n_neighbors	numeric; number of nearest neighbors.
metric	character or function; determines how distances between data points are computed.
set_op_mix_ratio	numeric in range [0,1]; determines who the knn-graph is used to create a fuzzy simplicial graph.
local_connectivity	numeric; used during construction of fuzzy simplicial set.
repulsion_strength	numeric; weighting applied to negative samples in low dimensional embedding optimization.
negative_sample_rate	numeric; determines how many non-neighbor points are used per point and per iteration during layout optimization.
seed	numeric; seed.

Value

CellRouter object with the umap slot updated.

edroaldo/fusca documentation built on March 1, 2023, 1:43 p.m.