cellmap-analyze 0.4.2__tar.gz → 0.5.0__tar.gz

{cellmap_analyze-0.4.2/src/cellmap_analyze.egg-info → cellmap_analyze-0.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cellmap-analyze
-Version: 0.4.2
+Version: 0.5.0
 Summary: Code to perform analysis on segmentations like those produced by CellMap
 Author-email: David Ackerman <ackermand@janelia.hhmi.org>
 Maintainer-email: David Ackerman <ackermand@janelia.hhmi.org>
@@ -56,6 +56,15 @@ A suite of Dask-powered tools for processing and analyzing terabyte-scale 3D seg
 | **Label With Mask**      | `label-with-mask`            | Label one dataset with IDs from another.                    |
 | **Morphological Operations** | `morphological-operations` | Erosion and dilation of segmented datasets. Processing order across blocks is not guaranteed. |
 | **Skeletonize**          | `skeletonize`                | Generate skeletons from segmented objects with optional pruning and simplification. Automatically resamples to isotropic resolution before skeletonization. |
+| **Compute EDT**          | `compute-edt`                | Compute a Euclidean distance transform over a segmentation, for reuse across repeated `split-narrow-bridges` runs. |
+| **Split Narrow Bridges** | `split-narrow-bridges`       | Split accidentally-merged blob-like objects (nuclei, cells) at thin necks via EDT-watershed, preserving all original foreground voxels. |
+
+These tools chain together into a full predictions-to-segmentations pipeline:
+start from raw model output (`connected-components` thresholds prediction
+probability maps, or `mws` agglomerates affinities) to get an initial
+instance segmentation, optionally refine it (`split-narrow-bridges` to
+separate accidentally-merged instances, `fill-holes`, `morphological-operations`),
+then finalize with `clean-connected-components` and/or `filter-ids`.
 
 ### Analysis Tools
 

{cellmap_analyze-0.4.2 → cellmap_analyze-0.5.0}/README.md

@@ -21,6 +21,15 @@ A suite of Dask-powered tools for processing and analyzing terabyte-scale 3D seg
 | **Label With Mask**      | `label-with-mask`            | Label one dataset with IDs from another.                    |
 | **Morphological Operations** | `morphological-operations` | Erosion and dilation of segmented datasets. Processing order across blocks is not guaranteed. |
 | **Skeletonize**          | `skeletonize`                | Generate skeletons from segmented objects with optional pruning and simplification. Automatically resamples to isotropic resolution before skeletonization. |
+| **Compute EDT**          | `compute-edt`                | Compute a Euclidean distance transform over a segmentation, for reuse across repeated `split-narrow-bridges` runs. |
+| **Split Narrow Bridges** | `split-narrow-bridges`       | Split accidentally-merged blob-like objects (nuclei, cells) at thin necks via EDT-watershed, preserving all original foreground voxels. |
+
+These tools chain together into a full predictions-to-segmentations pipeline:
+start from raw model output (`connected-components` thresholds prediction
+probability maps, or `mws` agglomerates affinities) to get an initial
+instance segmentation, optionally refine it (`split-narrow-bridges` to
+separate accidentally-merged instances, `fill-holes`, `morphological-operations`),
+then finalize with `clean-connected-components` and/or `filter-ids`.
 
 ### Analysis Tools
 

{cellmap_analyze-0.4.2 → cellmap_analyze-0.5.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "cellmap-analyze"
-version = "0.4.2"
+version = "0.5.0"
 description = "Code to perform analysis on segmentations like those produced by CellMap"
 readme = "README.md"
 license-files = ["LICENSE"]
@@ -51,6 +51,8 @@ label-with-mask = "cli.cli:label_with_mask"
 measure = "cli.cli:measure"
 fit-lines-to-segmentations = "cli.cli:fit_lines_to_segmentations"
 assign-to-organelles = "cli.cli:assign_to_organelles"
+split-narrow-bridges = "cli.cli:split_narrow_bridges"
+compute-edt = "cli.cli:compute_edt"
 
 [tool.setuptools]
 package-dir = { "" = "src" }

{cellmap_analyze-0.4.2 → cellmap_analyze-0.5.0}/src/cellmap_analyze/analyze/measure.py

@@ -212,8 +212,15 @@ class Measure(ComputeConfigMixin):
 
         raw_idi = kwargs.get("raw_idi")
         if raw_idi is not None:
-            # Read raw at write_roi (no face-neighbor padding needed for intensity)
-            extra_kwargs["raw_data"] = raw_idi.to_ndarray_ts(block.write_roi)
+            # Read raw at write_roi (no face-neighbor padding needed for intensity).
+            # raw_valid marks voxels that came from padding/no-overlap fill
+            # rather than a real read, so they can be excluded from intensity
+            # stats instead of being silently counted as real intensity 0.
+            raw_data, raw_valid = raw_idi.to_ndarray_ts(
+                block.write_roi, return_valid_mask=True
+            )
+            extra_kwargs["raw_data"] = raw_data
+            extra_kwargs["raw_valid"] = raw_valid
 
         # get information only from actual block(not including padding)
         # Convert block offset from scaled coordinates back to true nm