v2.1.1

.github/workflows/python-package.yml

@@ -15,8 +15,8 @@ jobs:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os : [ubuntu-22.04, macos-11, macos-12, windows-2019]
-        python-version: ['3.9', '3.10']
+        os : [ubuntu-latest, windows-latest, macos-latest]
+        python-version: ['3.9', '3.10', '3.11']
     steps:
     - uses: actions/[email protected]
     - name: Set up Python ${{ matrix.python-version }}
@@ -28,7 +28,7 @@ jobs:
         python -m pip install --upgrade pip
         python -m pip install .[docs]
     - name: Lint & test coverage
-      if: ${{ matrix.os == 'ubuntu-22.04' && matrix.python-version == '3.8' }}
+      if: ${{ matrix.os == 'ubuntu-latest' && matrix.python-version == '3.10' }}
       run: |
         pip install flake8
         # stop the build if there are Python syntax errors or undefined names
@@ -38,7 +38,7 @@ jobs:
         pip install coverage
         coverage run -m unittest
     - name: Upload Coverage to Codecov
-      if: ${{ matrix.os == 'ubuntu-22.04' && matrix.python-version == '3.8' }}
+      if: ${{ matrix.os == 'ubuntu-latest' && matrix.python-version == '3.10' }}
       uses: codecov/codecov-action@v1
       with:
         fail_ci_if_error: true 

.gitignore

@@ -143,4 +143,6 @@ tests/data/processed/pre_transform.pt
 .DS_Store
 
 requirements.lock
-requirements-dev.lock
+requirements-dev.lock
+profile.html
+profile.json

bento/__init__.py

@@ -5,3 +5,4 @@
 from . import datasets as ds
 from . import io
 from .plotting import _colors as colors
+from ._constants import CosMx, Merscope, Xenium

bento/_constants.py

@@ -1,3 +1,26 @@
+from enum import Enum
+
 PATTERN_COLORS = ["#17becf", "#1f77b4", "#7f7f7f", "#ff7f0e", "#d62728"]
 PATTERN_NAMES = ["cell_edge", "cytoplasmic", "none", "nuclear", "nuclear_edge"]
 PATTERN_PROBS = [f"{p}_p" for p in PATTERN_NAMES]
+
+
+class CosMx(Enum):
+    """CosMx microscope constants"""
+
+    # https://nanostring-public-share.s3.us-west-2.amazonaws.com/SMI-Compressed/SMI-ReadMe.html
+    PIXEL_MICRONS = 0.18
+
+
+class Merscope(Enum):
+    """Merscope microscope constants"""
+
+    # https://vizgen.com/wp-content/uploads/2023/06/91600001_MERSCOPE-Instrument-User-Guide_Rev-G.pdf
+    PIXEL_MICRONS = 1
+
+
+class Xenium(Enum):
+    """Xenium microscope constants"""
+
+    # https://www.10xgenomics.com/instruments/xenium-analyzer
+    PIXEL_MICRONS = 0.2125

bento/_utils.py

@@ -1,13 +1,12 @@
 # Geometric operations for SpatialData ShapeElements wrapping GeoPandas GeoDataFrames.
-from typing import List, Optional
+from typing import List, Optional, Union
 
 import geopandas as gpd
 import numpy as np
 import pandas as pd
 from dask import dataframe as dd
 from spatialdata import SpatialData
 from spatialdata.models import PointsModel, ShapesModel, TableModel
-from spatialdata.transformations import get_transformation, set_transformation
 
 
 def filter_by_gene(
@@ -66,7 +65,7 @@ def get_points(
     points_key: str = "transcripts",
     astype: str = "pandas",
     sync: bool = True,
-) -> pd.DataFrame | dd.DataFrame | gpd.GeoDataFrame:
+) -> Union[pd.DataFrame, dd.DataFrame, gpd.GeoDataFrame]:
     """Get points DataFrame synced to AnnData object.
 
     Parameters
@@ -144,10 +143,10 @@ def get_shape(sdata: SpatialData, shape_key: str, sync: bool = True) -> gpd.GeoS
 
 def get_points_metadata(
     sdata: SpatialData,
-    metadata_keys: List[str] | str,
+    metadata_keys: Union[List[str], str],
     points_key: str,
-    astype="pandas",
-):
+    astype: str = "pandas",
+) -> Union[pd.DataFrame, dd.DataFrame]:
     """Get points metadata.
 
     Parameters
@@ -188,9 +187,9 @@ def get_points_metadata(
 
 def get_shape_metadata(
     sdata: SpatialData,
-    metadata_keys: List[str] | str,
+    metadata_keys: Union[List[str], str],
     shape_key: str,
-):
+) -> pd.DataFrame:
     """Get shape metadata.
 
     Parameters
@@ -204,7 +203,7 @@ def get_shape_metadata(
 
     Returns
     -------
-    pd.Dataframe
+    pd.DataFrame
         Returns `sdata.shapes[shape_key][metadata_keys]` as a `pd.DataFrame`
     """
     if shape_key not in sdata.shapes.keys():
@@ -223,9 +222,9 @@ def get_shape_metadata(
 def set_points_metadata(
     sdata: SpatialData,
     points_key: str,
-    metadata: List | pd.Series | pd.DataFrame | np.ndarray,
-    columns: List[str] | str,
-):
+    metadata: Union[List, pd.Series, pd.DataFrame, np.ndarray],
+    columns: Union[List[str], str],
+) -> None:
     """Write metadata in SpatialData points element as column(s). Aligns metadata index to shape index if present.
 
     Parameters
@@ -268,9 +267,9 @@ def set_points_metadata(
 def set_shape_metadata(
     sdata: SpatialData,
     shape_key: str,
-    metadata: List | pd.Series | pd.DataFrame | np.ndarray,
-    column_names: Optional[str | List[str]] = None,
-):
+    metadata: Union[List, pd.Series, pd.DataFrame, np.ndarray],
+    column_names: Union[List[str], str] = None,
+) -> None:
     """Write metadata in SpatialData shapes element as column(s). Aligns metadata index to shape index.
 
     Parameters
@@ -310,7 +309,9 @@ def set_shape_metadata(
         if "" not in metadata[col].cat.categories:
             metadata[col] = metadata[col].cat.add_categories([""]).fillna("")
 
-    sdata.shapes[shape_key].loc[:, metadata.columns] = metadata.reindex(shape_index)
+    sdata.shapes[shape_key].loc[:, metadata.columns] = metadata.reindex(
+        shape_index
+    ).fillna("")
 
 
 def _sync_points(sdata, points_key):
@@ -331,24 +332,20 @@ def _sync_points(sdata, points_key):
     """
     points = sdata.points[points_key].compute()
     instance_key = get_instance_key(sdata)
-    if instance_key not in points.columns:
-        raise ValueError(
-            f"Points {points_key} not synced to instance_key shape element. Run bento.io.prep() to setup SpatialData object for bento-tools."
-        )
-    else:
-        # Only keep points within instance_key shape
-        cells = set(sdata.shapes[instance_key].index)
-        transform = sdata.points[points_key].attrs
-        points_valid = points[
-            points[instance_key].isin(cells)
-        ]  # TODO why doesnt this grab the right cells
-        # Set points back to SpatialData object
-        points_valid = PointsModel.parse(
-            dd.from_pandas(points_valid, npartitions=1),
-            coordinates={"x": "x", "y": "y"},
-        )
-        points_valid.attrs = transform
-        sdata.points[points_key] = points_valid
+
+    # Only keep points within instance_key shape
+    cells = set(sdata.shapes[instance_key].index)
+    transform = sdata.points[points_key].attrs
+    points_valid = points[
+        points[instance_key].isin(cells)
+    ]  # TODO why doesnt this grab the right cells
+    # Set points back to SpatialData object
+    points_valid = PointsModel.parse(
+        dd.from_pandas(points_valid, npartitions=1),
+        coordinates={"x": "x", "y": "y"},
+    )
+    points_valid.attrs = transform
+    sdata.points[points_key] = points_valid
 
 
 def _sync_shapes(sdata, shape_key, instance_key):
@@ -371,22 +368,18 @@ def _sync_shapes(sdata, shape_key, instance_key):
     """
     shapes = sdata.shapes[shape_key]
     instance_shapes = sdata.shapes[instance_key]
-    if instance_key not in shapes.columns or shape_key not in instance_shapes.columns:
-        raise ValueError(
-            f"Shape {shape_key} not synced to instance_key shape element. Run bento.io.prep() to setup SpatialData object for bento-tools."
-        )
-    elif shape_key == instance_key:
+    if shape_key == instance_key:
         return
-    else:
-        # Only keep shapes within instance_key shape
-        cells = set(instance_shapes.index)
-        shapes = shapes[shapes[instance_key].isin(cells)]
 
-        # Set shapes back to SpatialData object
-        transform = sdata.shapes[shape_key].attrs
-        shapes_valid = ShapesModel.parse(shapes)
-        shapes_valid.attrs = transform
-        sdata.shapes[shape_key] = shapes_valid
+    # Only keep shapes within instance_key shape
+    cells = set(instance_shapes.index)
+    shapes = shapes[shapes[instance_key].isin(cells)]
+
+    # Set shapes back to SpatialData object
+    transform = sdata.shapes[shape_key].attrs
+    shapes_valid = ShapesModel.parse(shapes)
+    shapes_valid.attrs = transform
+    sdata.shapes[shape_key] = shapes_valid
 
 
 def get_instance_key(sdata: SpatialData):
@@ -406,7 +399,9 @@ def get_instance_key(sdata: SpatialData):
     try:
         return sdata.points["transcripts"].attrs["spatialdata_attrs"]["instance_key"]
     except KeyError:
-        raise KeyError("Instance key attribute not found in spatialdata object.")
+        raise KeyError(
+            "Instance key attribute not found in spatialdata object. Run bento.io.prep() to setup SpatialData object for bento-tools."
+        )
 
 
 def get_feature_key(sdata: SpatialData):

bento/datasets/merfish_sample.zarr.bak/points/.zgroup

@@ -0,0 +1,3 @@
+{
+    "zarr_format": 2
+}

bento/datasets/merfish_sample.zarr.bak/points/transcripts/.zattrs

@@ -0,0 +1,44 @@
+{
+    "axes": null,
+    "coordinateTransformations": [
+        {
+            "input": {
+                "axes": [
+                    {
+                        "name": "x",
+                        "type": "space",
+                        "unit": "unit"
+                    },
+                    {
+                        "name": "y",
+                        "type": "space",
+                        "unit": "unit"
+                    }
+                ],
+                "name": "xy"
+            },
+            "output": {
+                "axes": [
+                    {
+                        "name": "x",
+                        "type": "space",
+                        "unit": "unit"
+                    },
+                    {
+                        "name": "y",
+                        "type": "space",
+                        "unit": "unit"
+                    }
+                ],
+                "name": "global"
+            },
+            "type": "identity"
+        }
+    ],
+    "encoding-type": "ngff:points",
+    "spatialdata_attrs": {
+        "feature_key": "feature_name",
+        "instance_key": "cell_boundaries",
+        "version": "0.1"
+    }
+}

bento/datasets/merfish_sample.zarr.bak/points/transcripts/.zgroup

@@ -0,0 +1,3 @@
+{
+    "zarr_format": 2
+}

bento/datasets/merfish_sample.zarr.bak/shapes/.zgroup

@@ -0,0 +1,3 @@
+{
+    "zarr_format": 2
+}

bento/datasets/small_data.zarr/shapes/cell_circles/.zattrs → bento/datasets/merfish_sample.zarr.bak/shapes/cell_boundaries/.zattrs

@@ -32,18 +32,14 @@
                 ],
                 "name": "global"
             },
-            "scale": [
-                4.705882352941177,
-                4.705882352941177
-            ],
-            "type": "scale"
+            "type": "identity"
         }
     ],
     "encoding-type": "ngff:shapes",
     "spatialdata_attrs": {
         "geos": {
-            "name": "POINT",
-            "type": 0
+            "name": "POLYGON",
+            "type": 3
         },
         "version": "0.1"
     }

bento/datasets/merfish_sample.zarr.bak/shapes/cell_boundaries/.zgroup

@@ -0,0 +1,3 @@
+{
+    "zarr_format": 2
+}

bento/datasets/merfish_sample.zarr.bak/shapes/cell_boundaries/Index/.zarray

@@ -0,0 +1,25 @@
+{
+    "chunks": [
+        14
+    ],
+    "compressor": {
+        "blocksize": 0,
+        "clevel": 5,
+        "cname": "lz4",
+        "id": "blosc",
+        "shuffle": 1
+    },
+    "dimension_separator": "/",
+    "dtype": "|O",
+    "fill_value": 0,
+    "filters": [
+        {
+            "id": "vlen-utf8"
+        }
+    ],
+    "order": "C",
+    "shape": [
+        14
+    ],
+    "zarr_format": 2
+}

bento/datasets/merfish_sample.zarr.bak/shapes/cell_boundaries/coords/.zarray

@@ -0,0 +1,23 @@
+{
+    "chunks": [
+        1224,
+        2
+    ],
+    "compressor": {
+        "blocksize": 0,
+        "clevel": 5,
+        "cname": "lz4",
+        "id": "blosc",
+        "shuffle": 1
+    },
+    "dimension_separator": "/",
+    "dtype": "<f8",
+    "fill_value": 0.0,
+    "filters": null,
+    "order": "C",
+    "shape": [
+        1224,
+        2
+    ],
+    "zarr_format": 2
+}