pred_intervals function

Produce Error Bounds for Predictions

A function to produce a confidence region for a linear predictor. In upcoming versions will (hopefully) be greatly simplified.

pred_intervals(
  predictions,
  pred_model,
  gen_model,
  training_matrix,
  gene_lengths,
  biomarker_values,
  alpha = 0.1,
  range_factor = 1.1,
  s = NULL,
  max_panel_length = NULL,
  biomarker = "TMB",
  marker_mut_types = c("NS", "I"),
  model = "Refitted T"
)

Arguments

• predictions: (list) A predictions object, as produced by get_predictions().
• pred_model: (list) A predictive model, as produced by pred_first_fit(), pred_refit_panel() or pred_refit_range().
• gen_model: (list) A generative model, as produce by fit_gen_model
• training_matrix: (sparse matrix) A training matrix, as produced by get_tables()$matrix or get_table_from_maf()$matrix.
• gene_lengths: (data frame) A data frame with columns 'Hugo_Symbol' and 'max_cds'. See example_maf_data$gene_lengths, or ensembl_gene_lengths for examples.
• biomarker_values: (data frame) A data frame containing the true values of the biomarker in question.
• alpha: (numeric) Confidence level for error bounds.
• range_factor: (numeric) Value specifying how far beyond the range of max(biomarker) to plot confidence region.
• s: (numeric) If input predictions are for a range of panels, s chooses which panel (column in a pred_fit object) to produce predictions for.
• max_panel_length: (numeric) Select panel by maximum length.
• biomarker: (character) Which biomarker is being predicted.
• marker_mut_types: (character) If biomarker is not one of "TMB" or "TIB", then this is required to specify which mutation type groups constitute the biomarker.
• model: (character) The model (must be based on a linear estimator) for which prediction intervals are being generated.

Returns

A list with two entries:

• prediction_intervals:
• confidence_region:

Examples

example_intervals <- pred_intervals(predictions = get_predictions(example_refit_range,
               new_data = example_tables$val),
               pred_model = example_refit_range, biomarker_values = example_tmb_tables$val,
               gen_model = example_gen_model, training_matrix = example_tables$train$matrix,
               max_panel_length = 15000, gene_lengths = example_maf_data$gene_lengths)

example_confidence_plot <- ggplot2::ggplot() +
  ggplot2::geom_point(data = example_intervals$prediction_intervals,
             ggplot2::aes(x = true_value, y = estimated_value)) +
        ggplot2::geom_ribbon(data = example_intervals$confidence_region,
          ggplot2::aes(x = x, ymin = y_lower, ymax = y_upper),
                    fill = "red", alpha = 0.2) +
        ggplot2::geom_line(data = example_intervals$confidence_region,
          ggplot2::aes(x = x, y = y), linetype = 2) +
        ggplot2::scale_x_log10() + ggplot2::scale_y_log10()

plot(example_confidence_plot)