getUMAP: Uniform Manifold Approximation and Projection(UMAP) algorithm...

In mmkhan19/singleCellTK: Interactive Analysis of Single Cell RNA-Seq Data

Uniform Manifold Approximation and Projection(UMAP) algorithm for dimension reduction.

Description

Uniform Manifold Approximation and Projection(UMAP) algorithm for dimension reduction.

Usage

getUMAP(
  inSCE,
  useAssay = "logcounts",
  reducedDimName = "UMAP",
  n_neighbors = 5,
  n_iterations = 200,
  alpha = 1,
  metric = "euclidean",
  run_pca = TRUE
)

Arguments

inSCE	Input SCtkExperiment object. Required
useAssay	Indicate which assay to use. The default is "logcounts".
reducedDimName	a name to store the results of the dimension reduction coordinates obtained from this method. This is stored in the SingleCellExperiment object in the reducedDims slot. Required.
n_neighbors	specify the number of nearest neighbors. Default is 5.
n_iterations	number of iterations performed during layout optimization. Default is 200.
alpha	initial value of "learning rate" of layout optimization. Default is 1.
metric	distance metric. Default is euclidean, other options are 'manhattan', 'cosine', 'pearson'.
run_pca	run UMAP on PCA components? Default is TRUE.

Value

a SCtkExperiment object with the reduced dimensions updated under reducedDimName specified.

Examples

umap_res <- getUMAP(inSCE = mouseBrainSubsetSCE, useAssay = "counts",
                    reducedDimName = "UMAP", n_neighbors = 3, n_iterations = 200,
                    alpha = 1, metric = "euclidean",  run_pca = TRUE)
reducedDims(umap_res)

mmkhan19/singleCellTK documentation built on March 22, 2022, 7:43 a.m.