gridgene.overlay module#

gridgene.overlay.timeit(func)[source]
class gridgene.overlay.Overlay(mask_dict, segmentation, segmentation_type='auto', save_path=None, min_x=0, min_y=0, flip_masks=True, logger=None)[source]

Bases: object

Overlay segmentation with binary masks for comparison and visualization.

Supports polygon-based and label-mask segmentations.

Parameters:

  • mask_dict (dict[str, np.ndarray]) – Dictionary of masks, where keys are mask names and values are binary mask arrays.

  • segmentation (dict or np.ndarray or SpatialData) – Segmentation data, either GeoJSON-like dict, label mask ndarray, or SpatialData.

  • segmentation_type ({'auto', 'polygons', 'label_mask'}, optional) – Type of segmentation input. Default is ‘auto’ to detect automatically.

  • save_path (str, optional) – Optional path to save visualizations.

  • min_x (float, optional) – Minimum x coordinate to shift polygons. Default is 0.

  • min_y (float, optional) – Minimum y coordinate to shift polygons. Default is 0.

  • flip_masks (bool, optional) – Whether to flip masks vertically and rotate. Default is True.

  • logger (logging.Logger, optional) – Custom logger. If None, a default logger is used.

property mask_shape

Return the shape of the masks.

Returns:

Shape of the first mask in the dictionary.

Return type:

tuple

shift_polygons()[source]

Shift all polygon coordinates by (min_x, min_y).

Return type:

None

Modifies#

self.segmentationdict

Polygon coordinates are shifted in place.

compute_overlap()[source]

Compute overlap between masks and segmented regions.

Returns:

Nested dictionary of overlap counts {cell_id: {mask_name: pixel_overlap_count}}.

Return type:

dict

Raises:

NotImplementedError – If overlap for label masks is requested.

map_mask_cell_masks()[source]

Placeholder for label mask-based overlap computation.

Return type:

None

map_mask_cell_polygons()[source]

Compute overlap counts between each segmentation polygon and each mask.

Return type:

None

Stores results in self.results as:

{cell_id: {mask_name: pixel_overlap_count, …}, …}

plot_masks_overlay_segmentation(titles, colors, background='white', save_path=None, show=True, show_legend=True)[source]

Overlay binary masks and segmentation polygons for visualization.

Parameters:

  • titles (list of str) – Titles for each mask.

  • colors (list of str) – Colors corresponding to each mask.

  • background (str, optional) – Background color, by default ‘white’.

  • save_path (str, optional) – Path to save the overlay plot, by default None.

  • show (bool, optional) – Whether to display the plot, by default True.

  • show_legend (bool, optional) – Whether to show legend, by default True.

Return type:

None

plot_colored_by_mask_overlap(mask_to_color, color_map='Reds', show=True, save_path=None, figsize=(15, 15))[source]

Color segmented polygons based on overlap percentage with specified masks.

Parameters:

  • mask_to_color (list of str) – Mask names to base coloring on.

  • color_map (str, optional) – Matplotlib colormap, by default ‘Reds’.

  • show (bool, optional) – Whether to display the plot, by default True.

  • save_path (str, optional) – Path to save the plot, by default None.

  • figsize (tuple, optional) – Figure size, by default (15, 15).

Return type:

None

This Page