pydruglogics.utils.PlotUtil

Plotting utilities for ROC and PR curves, including ensemble and confidence interval visualizations.

static PlotUtil.plot_roc_and_pr_curve(predicted_synergy_scores, observed_synergy_scores, synergy_method, labels=None)

Plot the ROC and PR Curves for one or multiple sets of predicted synergy scores. Plots both ROC and PR for model predictions, with optional support for model ensembles.

Parameters:

• predicted_synergy_scores: List of predictions or single set.

• observed_synergy_scores: Observed synergy scores.

• synergy_method (str): Method used for scoring (plot titles).

• labels (list, optional): Optional labels for each set.

static PlotUtil.plot_pr_curve_with_ci(pr_df, auc_pr, boot_n, plot_discrete)

Plot a Precision-Recall (PR) curve with confidence intervals computed from bootstrap sampling. Accepts a DataFrame of PR/confidence data and supports both continuous confidence bands or discrete error bars to visualize model uncertainty.

Parameters

• pr_df (DataFrame): PR/confidence data.

• auc_pr (float): Area under PR.

• boot_n (int): Bootstrap samples.

• plot_discrete (bool): Discrete error bars or band.