VST: Variance Stabilizing Transformation

In satijalab/seurat: Tools for Single Cell Genomics

Variance Stabilizing Transformation

Description

Apply variance stabilizing transformation for selection of variable features

Usage

VST(data, margin = 1L, nselect = 2000L, span = 0.3, clip = NULL, ...)

## Default S3 method:
VST(data, margin = 1L, nselect = 2000L, span = 0.3, clip = NULL, ...)

## S3 method for class 'IterableMatrix'
VST(
  data,
  margin = 1L,
  nselect = 2000L,
  span = 0.3,
  clip = NULL,
  verbose = TRUE,
  ...
)

## S3 method for class 'dgCMatrix'
VST(
  data,
  margin = 1L,
  nselect = 2000L,
  span = 0.3,
  clip = NULL,
  verbose = TRUE,
  ...
)

## S3 method for class 'matrix'
VST(data, margin = 1L, nselect = 2000L, span = 0.3, clip = NULL, ...)

Arguments

data	A matrix-like object
margin	Unused
nselect	Number of of features to select
span	the parameter \alpha which controls the degree of smoothing.
clip	Upper bound for values post-standardization; defaults to the square root of the number of cells
...	Arguments passed to other methods
verbose	...

Value

A data frame with the following columns:

• “mean”: ...

• “variance”: ...

• “variance.expected”: ...

• “variance.standardized”: ...

• “variable”: TRUE if the feature selected as variable, otherwise FALSE

• “rank”: If the feature is selected as variable, then how it compares to other variable features with lower ranks as more variable; otherwise, NA

satijalab/seurat documentation built on May 11, 2024, 4:04 a.m.