fold_change_int {structToolbox}R Documentation

Fold change for interactions between factors

Description

For more than one factor the fold change calculation is extended to include all combinations of levels (interactions) of all factors. Paired fold changes are not possible for this computation.

Usage

fold_change_int(
  alpha = 0.05,
  factor_name,
  threshold = 2,
  control_group = character(0),
  method = "geometric",
  ...
)

Arguments

alpha

(numeric) The p-value cutoff for determining significance. The default is 0.05.

factor_name

(character) The name of a sample-meta column to use.

threshold

(numeric) The fold change threshold for labelling features as significant. The default is 2.

control_group

(character) The level names of the groups used in the denominator (where possible) when computing fold change. One level for each factor, assumed to be in the same order as factor_name. The default is character(0).

method

(character) Fold change method. Allowed values are limited to the following:

  • "geometric": A log transform and a t-test is used to calculate fold change and estimate confidence intervals. In the non-transformed space this is equivalent to fold change using geometric means.

  • "median": A log transform and the method described by Price and Bonett to calculate fold change and estimate confidence intervals. In the non-transformed space this is equivalent to using group medians to calculate fold change.

The default is "geometric".

...

Additional slots and values passed to struct_class.

Value

A fold_change_int object.

References

Lloyd GR, Weber RJM (2021). struct: Statistics in R Using Class-based Templates. R package version 1.3.1.

Examples

D = MTBLS79_DatasetExperiment()
D=D[,1:10,drop=FALSE]
M = filter_smeta(mode='exclude',levels='QC',factor_name='Class') +
    fold_change_int(factor_name=c('Class','Batch'))
M = model_apply(M,D)

[Package structToolbox version 1.4.3 Index]