RunICA: Run Independent Component Analysis on gene expression

View source: R/generics.R

RunICAR Documentation

Run Independent Component Analysis on gene expression

Description

Run fastica algorithm from the ica package for ICA dimensionality reduction. For details about stored ICA calculation parameters, see PrintICAParams.

Usage

RunICA(object, ...)

## Default S3 method:
RunICA(
  object,
  assay = NULL,
  nics = 50,
  rev.ica = FALSE,
  ica.function = "icafast",
  verbose = TRUE,
  ndims.print = 1:5,
  nfeatures.print = 30,
  reduction.name = "ica",
  reduction.key = "ica_",
  seed.use = 42,
  ...
)

## S3 method for class 'Assay'
RunICA(
  object,
  assay = NULL,
  features = NULL,
  nics = 50,
  rev.ica = FALSE,
  ica.function = "icafast",
  verbose = TRUE,
  ndims.print = 1:5,
  nfeatures.print = 30,
  reduction.name = "ica",
  reduction.key = "ica_",
  seed.use = 42,
  ...
)

## S3 method for class 'Seurat'
RunICA(
  object,
  assay = NULL,
  features = NULL,
  nics = 50,
  rev.ica = FALSE,
  ica.function = "icafast",
  verbose = TRUE,
  ndims.print = 1:5,
  nfeatures.print = 30,
  reduction.name = "ica",
  reduction.key = "IC_",
  seed.use = 42,
  ...
)

Arguments

object

Seurat object

...

Additional arguments to be passed to fastica

assay

Name of Assay ICA is being run on

nics

Number of ICs to compute

rev.ica

By default, computes the dimensional reduction on the cell x feature matrix. Setting to true will compute it on the transpose (feature x cell matrix).

ica.function

ICA function from ica package to run (options: icafast, icaimax, icajade)

verbose

Print the top genes associated with high/low loadings for the ICs

ndims.print

ICs to print genes for

nfeatures.print

Number of genes to print for each IC

reduction.name

dimensional reduction name

reduction.key

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

seed.use

Set a random seed. Setting NULL will not set a seed.

features

Features to compute ICA on


Seurat documentation built on Nov. 18, 2023, 1:10 a.m.