Merge branch 'master' into lp-fix

.github/workflows/python-package.yml

@@ -5,9 +5,9 @@ name: Python Package Tests
 
 on:
   push:
-    branches: [ master, v2.0 ]
+    branches: [ master ]
   pull_request:
-    branches: [ master, v2.0 ]
+    branches: [ master ]
 
 jobs:
   build:
@@ -16,7 +16,7 @@ jobs:
     strategy:
       matrix:
         os : [ubuntu-latest, windows-latest, macos-latest]
-        python-version: ['3.9', '3.10', '3.11']
+        python-version: ['3.10', '3.11']
     steps:
     - uses: actions/[email protected]
     - name: Set up Python ${{ matrix.python-version }}
@@ -26,11 +26,11 @@ jobs:
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip
-        python -m pip install .[docs]
+        python -m pip install ".[docs]"
     - name: Lint & test coverage
       if: ${{ matrix.os == 'ubuntu-latest' && matrix.python-version == '3.10' }}
       run: |
-        pip install flake8
+        pip install flake8 pytest
         # stop the build if there are Python syntax errors or undefined names
         flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
         # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide

bento/geometry/_ops.py

@@ -61,7 +61,6 @@ def overlay(
     )
 
 
-
 @singledispatch
 def labels_to_shapes(labels: np.ndarray, attrs: dict, bg_value: int = 0):
     """
@@ -80,7 +79,7 @@ def labels_to_shapes(labels: np.ndarray, attrs: dict, bg_value: int = 0):
     -------
     GeoPandas DataFrame
         GeoPandas DataFrame containing the polygons extracted from the labeled image.
-    
+
     """
     import rasterio as rio
     import shapely.geometry
@@ -91,8 +90,8 @@ def labels_to_shapes(labels: np.ndarray, attrs: dict, bg_value: int = 0):
     shapes = gpd.GeoDataFrame(
         polygons[:, 1], geometry=gpd.GeoSeries(polygons[:, 0]).T, columns=["id"]
     )
-    shapes = shapes[shapes["id"] != bg_value] # Ignore background
-    
+    shapes = shapes[shapes["id"] != bg_value]  # Ignore background
+
     # Validate for SpatialData
     sd_shape = ShapesModel.parse(shapes)
     sd_shape.attrs = attrs
@@ -112,13 +111,13 @@ def _(labels: SpatialImage, attrs: dict, bg_value: int = 0):
         Dictionary of attributes to set for the SpatialData object.
     bg_value : int, optional
         Value of the background pixels, by default 0
-    
+
     Returns
     -------
     GeoPandas DataFrame
         GeoPandas DataFrame containing the polygons extracted from the labeled image.
     """
-        
+
     # Convert spatial_image.SpatialImage to np.ndarray
     labels = labels.values
-    return labels_to_shapes(labels, attrs, bg_value)
+    return labels_to_shapes(labels, attrs, bg_value)

bento/tools/_composition.py

@@ -73,7 +73,7 @@ def comp(sdata: SpatialData, points_key: str, shape_names: list):
     sdata : spatialdata.SpatialData
         Updates `sdata.table.uns` with average gene compositions for each shape.
     """
-    points = get_points(sdata,points_key=points_key, astype="pandas")
+    points = get_points(sdata, points_key=points_key, astype="pandas")
 
     instance_key = get_instance_key(sdata)
     feature_key = get_feature_key(sdata)
@@ -86,7 +86,9 @@ def comp(sdata: SpatialData, points_key: str, shape_names: list):
     sdata.table.uns["comp_stats"] = comp_stats
 
 
-def comp_diff(sdata: SpatialData, points_key: str, shape_names: list, groupby: str, ref_group: str):
+def comp_diff(
+    sdata: SpatialData, points_key: str, shape_names: list, groupby: str, ref_group: str
+):
     """Calculate the average difference in gene composition for shapes across batches of cells. Uses the Wasserstein distance.
 
     Parameters
@@ -109,7 +111,9 @@ def comp_diff(sdata: SpatialData, points_key: str, shape_names: list, groupby: s
     # Get average gene compositions for each batch
     comp_stats = dict()
     for group, pt_group in points.groupby(groupby):
-        comp_stats[group] = _get_compositions(pt_group, shape_names, instance_key=instance_key, feature_key=feature_key)
+        comp_stats[group] = _get_compositions(
+            pt_group, shape_names, instance_key=instance_key, feature_key=feature_key
+        )
 
     ref_comp = comp_stats[ref_group]
 

bento/tools/_flux.py

@@ -497,26 +497,23 @@ def fluxmap(
             columns=["fluxmap"],
         )
 
-        # Add back offsets
-        shapes["geometry"] = shapes["geometry"].translate(x_offset, y_offset)
-
         # Remove background shape
         shapes["fluxmap"] = shapes["fluxmap"].astype(int)
         shapes = shapes[shapes["fluxmap"] != 0]
 
         # Group same fields as MultiPolygons
         shapes = shapes.dissolve("fluxmap")["geometry"]
+
+        # Upscale to match original resolution, shift back to original coordinates
+        shapes = shapes.scale(xfact=1 / res, yfact=1 / res, origin=(0, 0)).translate(
+            x_offset, y_offset
+        )
+
         fluxmap_df[cell] = shapes
 
     fluxmap_df = pd.DataFrame.from_dict(fluxmap_df).T
     fluxmap_df.columns = "fluxmap" + fluxmap_df.columns.astype(str)
 
-    # Upscale to match original resolution
-    fluxmap_df = fluxmap_df.apply(
-        lambda col: gpd.GeoSeries(col).scale(
-            xfact=1 / res, yfact=1 / res, origin=(0, 0)
-        )
-    )
     pbar.update()
 
     pbar.set_description("Saving")

bento/tools/_flux_enrichment.py

@@ -99,13 +99,10 @@ def fe(
     cell_raster = get_points(
         sdata, points_key=f"{instance_key}_raster", astype="pandas", sync=False
     )[features]
-    cell_raster_matrix = np.mat(cell_raster)
-    mat = sparse.csr_matrix(cell_raster_matrix)  # sparse matrix in csr format
-
     samples = cell_raster.index.astype(str)
 
     enrichment = dc.run_wsum(
-        mat=[mat, samples, features],
+        mat=cell_raster,
         net=net,
         source=source,
         target=target,

bento/tools/_point_features.py

@@ -133,12 +133,12 @@ def analyze_points(
     for shape in list(
         set(obs_attrs).intersection(set([x for x in shape_keys if x != instance_key]))
     ):
-        points_df = (
-            points_df.join(
-                sdata.shapes[shape].set_index(instance_key), on=instance_key, lsuffix="", rsuffix=f"_{shape}"
-            )
-            .rename(columns={shape: f"{shape}_index", f"geometry_{shape}": shape})
-        )
+        points_df = points_df.join(
+            sdata.shapes[shape].set_index(instance_key),
+            on=instance_key,
+            lsuffix="",
+            rsuffix=f"_{shape}",
+        ).rename(columns={shape: f"{shape}_index", f"geometry_{shape}": shape})
 
     # Pull cell_boundaries shape features into the points dataframe
     points_df = (