valpas.visualization.ppv_comparison_heatmap.create_ppv_heatmap

valpas.visualization.ppv_comparison_heatmap.create_ppv_heatmap(data, title='Mean Positive Predictive Value for Top 1000 Predictions', figsize=(12, 8), colormap='custom', annotate=True, annotation_format='.3f', value_range=None, highlight_best=True, highlight_threshold=None, save_path=None, dpi=300, show_colorbar=True, font_sizes=None, border_color='white', border_width=0.5, round_corners=False, add_significance_markers=False, significance_data=None, custom_colors=None)

Create a publication-quality heatmap for PPV results

Parameters:

• data (DataFrame) – DataFrame with input types as rows, association metrics as columns

• title (str) – Title for the heatmap

• figsize (Tuple[int, int]) – Figure size (width, height)

• colormap (str) – Colormap to use (‘custom’, ‘viridis’, ‘plasma’, ‘RdYlBu_r’, etc.)

• annotate (bool) – Whether to show values in cells

• annotation_format (str) – Format string for cell annotations

• value_range (Tuple[float, float]) – Tuple of (min, max) for color scale

• highlight_best (bool) – Whether to highlight best performing cells

• highlight_threshold (float) – Threshold above which to highlight cells

• save_path (str | None) – Path to save the figure

• dpi (int) – Resolution for saved figure

• show_colorbar (bool) – Whether to show the colorbar

• font_sizes (Dict[str, int]) – Dictionary with font sizes for different elements

• border_color (str) – Color of cell borders

• border_width (float) – Width of cell borders

• round_corners (bool) – Whether to use rounded corners (experimental)

• add_significance_markers (bool) – Whether to add significance markers

• significance_data (DataFrame) – DataFrame with significance values (p-values)

• custom_colors (Dict[str, str]) – Dictionary for custom color scheme

Returns:

Tuple of (figure, axes) objects

Return type:

Tuple[Figure, Axes]