iCellR (version 1.6.1)

run.pc.tsne: Run tSNE on PCA Data. Barnes-Hut implementation of t-Distributed Stochastic Neighbor Embedding

Description

This function takes an object of class iCellR and runs tSNE on PCA data. Wrapper for the C++ implementation of Barnes-Hut t-Distributed Stochastic Neighbor Embedding. t-SNE is a method for constructing a low dimensional embedding of high-dimensional data, distances or similarities. Exact t-SNE can be computed by setting theta=0.0.

Usage

run.pc.tsne(
  x = NULL,
  dims = 1:10,
  my.seed = 0,
  add.3d = TRUE,
  initial_dims = 50,
  perplexity = 30,
  theta = 0.5,
  check_duplicates = FALSE,
  pca = TRUE,
  max_iter = 1000,
  verbose = FALSE,
  is_distance = FALSE,
  Y_init = NULL,
  pca_center = TRUE,
  pca_scale = FALSE,
  stop_lying_iter = ifelse(is.null(Y_init), 250L, 0L),
  mom_switch_iter = ifelse(is.null(Y_init), 250L, 0L),
  momentum = 0.5,
  final_momentum = 0.8,
  eta = 200,
  exaggeration_factor = 12
)

Arguments

x

An object of class iCellR.

dims

PC dimentions to be used for tSNE analysis.

my.seed

seed number, default = 0.

add.3d

Add 3D tSNE as well, default = TRUE.

initial_dims

integer; the number of dimensions that should be retained in the initial PCA step (default: 50)

perplexity

numeric; Perplexity parameter

theta

numeric; Speed/accuracy trade-off (increase for less accuracy), set to 0.0 for exact TSNE (default: 0.5)

check_duplicates

logical; Checks whether duplicates are present. It is best to make sure there are no duplicates present and set this option to FALSE, especially for large datasets (default: TRUE)

pca

logical; Whether an initial PCA step should be performed (default: TRUE)

max_iter

integer; Number of iterations (default: 1000)

verbose

logical; Whether progress updates should be messageed (default: FALSE)

is_distance

logical; Indicate whether X is a distance matrix (experimental, default: FALSE)

Y_init

matrix; Initial locations of the objects. If NULL, random initialization will be used (default: NULL). Note that when using this, the initial stage with exaggerated perplexity values and a larger momentum term will be skipped.

pca_center

logical; Should data be centered before pca is applied? (default: TRUE)

pca_scale

logical; Should data be scaled before pca is applied? (default: FALSE)

stop_lying_iter

integer; Iteration after which the perplexities are no longer exaggerated (default: 250, except when Y_init is used, then 0)

mom_switch_iter

integer; Iteration after which the final momentum is used (default: 250, except when Y_init is used, then 0)

momentum

numeric; Momentum used in the first part of the optimization (default: 0.5)

final_momentum

numeric; Momentum used in the final part of the optimization (default: 0.8)

eta

numeric; Learning rate (default: 200.0)

exaggeration_factor

numeric; Exaggeration factor used to multiply the P matrix in the first part of the optimization (default: 12.0)

Value

An object of class iCellR.

Examples

# NOT RUN {
demo.obj <- run.pc.tsne(demo.obj, dims = 1:10,perplexity = 20)

head([e-mail])[1:5]

# }