get_cooks_outliers: Flag outliers using Cook's distance

In bryancquach/omixjutsu: Utilities for omics analysis tasks

Flag outliers using Cook's distance

Description

Flag expression features as outliers based on Cook's distance.

Usage

get_cooks_outliers(dds_fit, f_threshold = 99)

Arguments

dds_fit	A DESeqDataSet object after model fitting.
f_threshold	Percentile of the F distribution used as the cutoff for calling outliers.

Details

Uses differential expression regression model design to specify an F distribution and calculate a value that equates to the user-specified F distribution percentile. This value is used as a Cook's distance cutoff above which features are considered outliers. Since a Cook's distance value is calculated for every feature and sample, the max Cook's distance value across samples for each feature is used to call outliers.

Value

A boolean vector indicating if a feature was influenced by outlier samples.

See Also

get_cooks_cutoff, plot_cooks_per_sample

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