Step: ROC Analysis

Perform ROC analysis using glystats::gly_roc(), extract top 10 variables with highest AUC, and plot ROC curves for these variables using glyvis::plot_roc().

Usage

step_roc(pos_class = NULL, plot_width = 5, plot_height = 5)

Arguments

pos_class: A character string specifying which group level should be treated as the positive class. If NULL (default), the second level (alphabetically) will be used as the positive class.
plot_width: Width of the plot in inches. Default is 5.
plot_height: Height of the plot in inches. Default is 5.

Value

A glysmith_step object.

Details

Data required:

exp: The experiment to perform ROC analysis on

Tables generated:

roc_auc: A table containing the ROC AUC values for all variables

Plots generated:

roc_curves: ROC curves for the top 10 variables

AI Prompt

This section is for AI in inquire_blueprint() only.

Include this step if the user explicitly asks for ROC analysis, or if he/she mentions "biomarker(s)" in the prompt.
If the experiment has more than 2 groups but the user wants a specific two-group comparison, ask which two groups to compare and include step_subset_groups(groups = c("A", "B")) before this step.

Examples

step_roc()
#> <step "step_roc()"> ROC analysis