Seurat (version 2.3.4)

RunUMAP: Run UMAP

Description

Runs the Uniform Manifold Approximation and Projection (UMAP) dimensional reduction technique. To run, you must first install the umap-learn python package (e.g. via pip install umap-learn). Details on this package can be found here: https://github.com/lmcinnes/umap. For a more in depth discussion of the mathematics underlying UMAP, see the ArXiv paper here: https://arxiv.org/abs/1802.03426.

Usage

RunUMAP(object, cells.use = NULL, dims.use = 1:5, reduction.use = "pca",
  genes.use = NULL, assay.use = "RNA", max.dim = 2L,
  reduction.name = "umap", reduction.key = "UMAP", n_neighbors = 30L,
  min_dist = 0.3, metric = "correlation", seed.use = 42, ...)

Arguments

object

Seurat object

cells.use

Which cells to analyze (default, all cells)

dims.use

Which dimensions to use as input features, used only if genes.use is NULL

reduction.use

Which dimensional reduction (PCA or ICA) to use for the UMAP input. Default is PCA

genes.use

If set, run UMAP on this subset of genes (instead of running on a set of reduced dimensions). Not set (NULL) by default

assay.use

Assay to pull data for when using genes.use

max.dim

Max dimension to keep from UMAP procedure.

reduction.name

dimensional reduction name, specifies the position in the object$dr list. umap by default

reduction.key

dimensional reduction key, specifies the string before the number for the dimension names. UMAP by default

n_neighbors

This determines the number of neighboring points used in local approximations of manifold structure. Larger values will result in more global structure being preserved at the loss of detailed local structure. In general this parameter should often be in the range 5 to 50.

min_dist

min_dist: This controls how tightly the embedding is allowed compress points together. Larger values ensure embedded points are more evenly distributed, while smaller values allow the algorithm to optimise more accurately with regard to local structure. Sensible values are in the range 0.001 to 0.5.

metric

metric: This determines the choice of metric used to measure distance in the input space. A wide variety of metrics are already coded, and a user defined function can be passed as long as it has been JITd by numba.

seed.use

Set a random seed. By default, sets the seed to 42. Setting NULL will not set a seed.

...

Additional arguments to the umap

Value

Returns a Seurat object containing a UMAP representation

References

McInnes, L, Healy, J, UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction, ArXiv e-prints 1802.03426, 2018

Examples

# NOT RUN {
pbmc_small
# Run UMAP map on first 5 PCs
pbmc_small <- RunUMAP(object = pbmc_small, dims.use = 1:5)
# Plot results
DimPlot(object = pbmc_small, reduction.use = 'umap')
# }
# NOT RUN {
# }