low_expression_filter: Filter lowly expressed features prior to DGE analysis

In bryancquach/omixjutsu: Utilities for omics analysis tasks

Filter lowly expressed features prior to DGE analysis

Description

Generic method to remove expression features below a given threshold.

Usage

## S3 method for class 'DESeqDataSet'
low_expression_filter(
  dds,
  value_cutoff,
  min_sample_fraction = NULL,
  threshold_variable = NULL,
  use_min_fraction = T,
  approximate = F
)

low_expression_filter(object, ...)

## S3 method for class 'matrix'
low_expression_filter(object, value_cutoff, sample_cutoff, approximate)

## S3 method for class 'DGEList'
low_expression_filter(
  dge_data,
  value_cutoff,
  min_sample_fraction = NULL,
  threshold_variable = NULL,
  use_min_fraction = T,
  approximate = F
)

Arguments

value_cutoff	The minimum value required for a feature for greater than min_sample_fraction proportion of samples to retain the feature.
min_sample_fraction	The minimum proportion of samples with greater than value_cutoff values to retain a feature.
threshold_variable	Name of the binary variable to use for calculating sample fractions.
use_min_fraction	If TRUE, uses the smaller of the sample fractions calculated from the threshold_variable. Otherwise uses the larger of the sample fractions.
approximate	If TRUE, round min_sample_fraction to a number with the hundreths digit equal to '0' or '5'.
object	A DESeqDataSet, DGEList, or matrix object from which to filter row features.

Details

Filters out features considered lowly expressed using a count threshold and sample proportion set by the user. The sample proportion can also be estimated by calculating proportions of values from a binary variable specified by the user.

Value

A filtered DESeqDataSet object if a DESeqDataSet object is provided as input.

A filtered matrix if a matrix is provided as input.

A filtered DGEList object if a DGEList object is provided as input.

bryancquach/omixjutsu documentation built on Jan. 29, 2023, 3:47 p.m.