get_stats: Metrics for Predictive Performance

In cobrbra/ICBioMark: Data-Driven Design of Targeted Gene Panels for Estimating Immunotherapy Biomarkers

Metrics for Predictive Performance

Description

A function to return a variety metrics for predictions vs actual values. Works well when piped to straight from get_predictions().

Usage

get_stats(
  predictions,
  biomarker_values,
  model = "",
  threshold = 300,
  metrics = c("R", "AUPRC")
)

Arguments

predictions	(list) A list with two elements, 'predictions' and 'panel_lengths', as produced by the function get_predictions().
biomarker_values	(dataframe) A dataframe with two columns, 'Tumor_Sample_Barcode' and a column with the name of the biomarker in question containing values.
model	(character) The name of the model type producing these predictions.
threshold	(numeric) The threshold for biomarker high/low categorisation.
metrics	(character) A vector of the names of metrics to calculate.

Value

dataframe with 5 columns:

• panel_length: the length of each panel.

• model: the model that produced the predictions.

• biomarker: the name of the biomarker in question.

• stat: the metric values for each panel.

• metric: the name of the metric.

Examples

example_stat <- get_stats(predictions = get_predictions(example_refit_panel,
new_data = example_tables$val), biomarker_values = example_tmb_tables$val,
model = "Refitted T", threshold = 10)

cobrbra/ICBioMark documentation built on May 4, 2023, 2:16 a.m.