Skip to contents

Barnes-Hut implementation of t-Distributed Stochastic Neighbor Embedding (t-SNE)

Source: R/CoreMethods.R
RunTSNE.Rd

Run nonlinear dimensionality reduction using t-SNE with the PCA-transformed consensus matrix as input.

Usage

RunTSNE.SingleCellExperiment(
  object,
  dims,
  dimred.type,
  perplexity,
  dimred.name,
  ...
)

# S4 method for class 'SingleCellExperiment'
RunTSNE(
  object,
  dims = NULL,
  dimred.type = "PCA",
  perplexity = 30,
  dimred.name = "TSNE",
  ...
)

Arguments

object

Object of SingleCellExperiment class.

dims

Dimensions to select from dimred.type. By default NULL, i.e., all the dimensions are selected. Provide a numeric vector to select a specific range, e.g., dims = 1:10 to select the first 10 dimensions.

dimred.type

Dimensional reduction type to use. By default "PCA".

perplexity

Perplexity of t-SNE.

dimred.name

Dimensional reduction name given to the returned t-SNE. By default "TSNE".

...

Parameters to be passed to the Rtsne function. The parameters given should match the parameters accepted by the Rtsne function. Check possible parameters with ?Rtsne::Rtsne.

Value

A SingleCellExperiment object.

Examples

# Import package
suppressPackageStartupMessages(library("SingleCellExperiment"))

# Create toy SCE data
batches <- c("b1", "b2")
set.seed(239)
batch <- sample(x = batches, size = nrow(iris), replace = TRUE)
sce <- SingleCellExperiment(assays = list(logcounts = t(iris[,1:4])),  
                            colData = DataFrame("Species" = iris$Species, 
                                               "Batch" = batch))
colnames(sce) <- paste0("samp", 1:ncol(sce))

# Run PCA
set.seed(125) # to ensure reproducibility for the default 'irlba' method
sce <- RunPCA(object = sce, assay.name = "logcounts", 
              pca.method = "stats", p = nrow(sce))

# Run t-SNE
set.seed(125) # to ensure reproducibility for the default 'irlba' method
sce <- RunTSNE(object = sce, dimred.type = "PCA", check_duplicates = FALSE)

# Plot result 
cowplot::plot_grid(PlotDimRed(object = sce, color.by = "Batch", 
                              legend.nrow = 1),
                   PlotDimRed(object = sce, color.by = "Species", 
                             legend.nrow = 1), ncol = 2)