spatialdata-io 0.1.3.post0__tar.gz → 0.1.4__tar.gz

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/.pre-commit-config.yaml

@@ -23,7 +23,7 @@ repos:
       hooks:
           - id: isort
     - repo: https://github.com/pre-commit/mirrors-mypy
-      rev: v1.10.1
+      rev: v1.11.0
       hooks:
           - id: mypy
             additional_dependencies: [numpy, types-PyYAML]
@@ -70,7 +70,7 @@ repos:
                 - flake8-bugbear
                 - flake8-blind-except
     - repo: https://github.com/asottile/pyupgrade
-      rev: v3.16.0
+      rev: v3.17.0
       hooks:
           - id: pyupgrade
             args: [--py3-plus, --py39-plus, --keep-runtime-typing]

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/CHANGELOG.md

@@ -10,6 +10,14 @@ and this project adheres to [Semantic Versioning][].
 
 ## [0.1.4] - xxxx-xx-xx
 
+## [0.1.4] - 2024-08-07
+
+### Changed
+
+-   (Xenium) changed default target of table to labels; radii of circles computed from cells, not nuclei #179
+-   (Visium HD) changed default geometry to squares from circles for the bins; added parameter to choose #183
+-   (CosMx) dropping points element with zero-length from the cosmx reader #191
+
 ## [0.1.3] - 2024-07-03
 
 ### Added

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: spatialdata-io
-Version: 0.1.3.post0
+Version: 0.1.4
 Summary: SpatialData IO for common techs
 Project-URL: Documentation, https://spatialdata-io.readthedocs.io/
 Project-URL: Source, https://github.com/scverse/spatialdata-io
@@ -49,6 +49,7 @@ Requires-Dist: readfcs
 Requires-Dist: scanpy
 Requires-Dist: scikit-image
 Requires-Dist: spatialdata>=0.2.1
+Requires-Dist: tifffile>=2023.8.12
 Provides-Extra: dev
 Requires-Dist: bump2version; extra == 'dev'
 Requires-Dist: pre-commit; extra == 'dev'
@@ -82,17 +83,20 @@ Description-Content-Type: text/markdown
 
 This package contains reader functions to load common spatial omics formats into SpatialData. Currently, we provide support for:
 
--   10x Genomics Visium
--   10x Genomics Visium HD
--   10x Genomics Xenium
--   Akoya PhenoCycler (formerly CODEX)
--   Curio Seeker
+-   10x Genomics Visium®
+-   10x Genomics Visium HD®
+-   10x Genomics Xenium®
+-   Akoya PhenoCycler® (formerly CODEX®)
+-   Curio Seeker®
 -   DBiT-seq
 -   MCMICRO (output data)
--   NanoString CosMx
+-   NanoString CosMx®
+-   Spatial Genomics GenePS® (seqFISH)
 -   Steinbock (output data)
--   STOmics Stereo-seq
--   Vizgen MERSCOPE (MERFISH)
+-   STOmics Stereo-seq®
+-   Vizgen MERSCOPE® (MERFISH)
+
+Note: all mentioned technologies are registered trademarks of their respective companies.
 
 ## Getting started
 
@@ -129,9 +133,13 @@ If you found a bug, please use the [issue tracker][issue-tracker].
 Technologies that can be read into `SpatialData` objects using third-party libraries:
 
 -   METASPACE (MALDI, ...): [metaspace-converter](https://github.com/metaspace2020/metaspace-converter)
--   PhenoCycler: [SOPA](https://github.com/gustaveroussy/sopa)
--   MACSima: [SOPA](https://github.com/gustaveroussy/sopa)
--   Hyperion (Imaging Mass Cytometry): [SOPA](https://github.com/gustaveroussy/sopa)
+-   PhenoCycler®: [SOPA](https://github.com/gustaveroussy/sopa)
+-   MACSima®: [SOPA](https://github.com/gustaveroussy/sopa)
+-   Hyperion® (Imaging Mass Cytometry): [SOPA](https://github.com/gustaveroussy/sopa)
+
+## Disclaimer
+
+This library is community maintained and is not officially endorsed by the aforementioned spatial technology companies. As such, we cannot offer any warranty of the correctness of the representation. Furthermore, we cannot ensure the correctness of the readers for every data version as the technologies evolve and update their formats. If you find a bug or notice a misrepresentation of the data please report it via our [Bug Tracking System](https://github.com/scverse/spatialdata-io/issues?q=sort%3Aupdated-desc+is%3Aissue+is%3Aopen) so that it can be addressed either by the maintainers of this library or by the community.
 
 ## Citation
 

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/README.md

@@ -12,17 +12,20 @@
 
 This package contains reader functions to load common spatial omics formats into SpatialData. Currently, we provide support for:
 
--   10x Genomics Visium
--   10x Genomics Visium HD
--   10x Genomics Xenium
--   Akoya PhenoCycler (formerly CODEX)
--   Curio Seeker
+-   10x Genomics Visium®
+-   10x Genomics Visium HD®
+-   10x Genomics Xenium®
+-   Akoya PhenoCycler® (formerly CODEX®)
+-   Curio Seeker®
 -   DBiT-seq
 -   MCMICRO (output data)
--   NanoString CosMx
+-   NanoString CosMx®
+-   Spatial Genomics GenePS® (seqFISH)
 -   Steinbock (output data)
--   STOmics Stereo-seq
--   Vizgen MERSCOPE (MERFISH)
+-   STOmics Stereo-seq®
+-   Vizgen MERSCOPE® (MERFISH)
+
+Note: all mentioned technologies are registered trademarks of their respective companies.
 
 ## Getting started
 
@@ -59,9 +62,13 @@ If you found a bug, please use the [issue tracker][issue-tracker].
 Technologies that can be read into `SpatialData` objects using third-party libraries:
 
 -   METASPACE (MALDI, ...): [metaspace-converter](https://github.com/metaspace2020/metaspace-converter)
--   PhenoCycler: [SOPA](https://github.com/gustaveroussy/sopa)
--   MACSima: [SOPA](https://github.com/gustaveroussy/sopa)
--   Hyperion (Imaging Mass Cytometry): [SOPA](https://github.com/gustaveroussy/sopa)
+-   PhenoCycler®: [SOPA](https://github.com/gustaveroussy/sopa)
+-   MACSima®: [SOPA](https://github.com/gustaveroussy/sopa)
+-   Hyperion® (Imaging Mass Cytometry): [SOPA](https://github.com/gustaveroussy/sopa)
+
+## Disclaimer
+
+This library is community maintained and is not officially endorsed by the aforementioned spatial technology companies. As such, we cannot offer any warranty of the correctness of the representation. Furthermore, we cannot ensure the correctness of the readers for every data version as the technologies evolve and update their formats. If you find a bug or notice a misrepresentation of the data please report it via our [Bug Tracking System](https://github.com/scverse/spatialdata-io/issues?q=sort%3Aupdated-desc+is%3Aissue+is%3Aopen) so that it can be addressed either by the maintainers of this library or by the community.
 
 ## Citation
 

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '0.1.3.post0'
-__version_tuple__ = version_tuple = (0, 1, 3)
+__version__ = version = '0.1.4'
+__version_tuple__ = version_tuple = (0, 1, 4)

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/pyproject.toml

@@ -33,6 +33,7 @@ dependencies = [
     "dask-image",
     "pyarrow",
     "readfcs",
+    "tifffile>=2023.8.12",
 ]
 
 [project.optional-dependencies]

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/src/spatialdata_io/converters/legacy_anndata.py

@@ -268,6 +268,8 @@ def from_legacy_anndata(adata: AnnData) -> SpatialData:
             keys = set(adata.uns[SPATIAL][dataset_id].keys())
             tissue_hires_scalef = None
             tissue_lowres_scalef = None
+            hires = None
+            lowres = None
             if SCALEFACTORS in keys:
                 scalefactors = adata.uns[SPATIAL][dataset_id][SCALEFACTORS]
                 if TISSUE_HIRES_SCALEF in scalefactors:

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/src/spatialdata_io/readers/cosmx.py

@@ -271,17 +271,18 @@ def cosmx(
                 sub_table[CosmxKeys.INSTANCE_KEY] = sub_table[CosmxKeys.INSTANCE_KEY].astype("category")
                 # we rename z because we want to treat the data as 2d
                 sub_table.rename(columns={"z": "z_raw"}, inplace=True)
-                points[f"{fov}_points"] = PointsModel.parse(
-                    sub_table,
-                    coordinates={"x": CosmxKeys.X_LOCAL_TRANSCRIPT, "y": CosmxKeys.Y_LOCAL_TRANSCRIPT},
-                    feature_key=CosmxKeys.TARGET_OF_TRANSCRIPT,
-                    instance_key=CosmxKeys.INSTANCE_KEY,
-                    transformations={
-                        fov: Identity(),
-                        "global": aff,
-                        "global_only_labels": aff,
-                    },
-                )
+                if len(sub_table) > 0:
+                    points[f"{fov}_points"] = PointsModel.parse(
+                        sub_table,
+                        coordinates={"x": CosmxKeys.X_LOCAL_TRANSCRIPT, "y": CosmxKeys.Y_LOCAL_TRANSCRIPT},
+                        feature_key=CosmxKeys.TARGET_OF_TRANSCRIPT,
+                        instance_key=CosmxKeys.INSTANCE_KEY,
+                        transformations={
+                            fov: Identity(),
+                            "global": aff,
+                            "global_only_labels": aff,
+                        },
+                    )
 
     # TODO: what to do with fov file?
     # if fov_file is not None:

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/src/spatialdata_io/readers/visium_hd.py

@@ -14,6 +14,7 @@ import numpy as np
 import pandas as pd
 import scanpy as sc
 from dask_image.imread import imread
+from geopandas import GeoDataFrame
 from imageio import imread as imread2
 from multiscale_spatial_image import MultiscaleSpatialImage
 from spatial_image import SpatialImage
@@ -38,6 +39,7 @@ def visium_hd(
     dataset_id: str | None = None,
     filtered_counts_file: bool = True,
     bin_size: int | list[int] | None = None,
+    bins_as_squares: bool = True,
     fullres_image_file: str | Path | None = None,
     load_all_images: bool = False,
     imread_kwargs: Mapping[str, Any] = MappingProxyType({}),
@@ -63,6 +65,9 @@ def visium_hd(
     bin_size
         When specified, load the data of a specific bin size, or a list of bin sizes. By default, it loads all the
         available bin sizes.
+    bins_as_squares
+        If `True`, the bins are represented as squares. If `False`, the bins are represented as circles. For a correct
+        visualization one should use squares.
     fullres_image_file
         Path to the full-resolution image. By default the image is searched in the ``{vx.MICROSCOPE_IMAGE!r}``
         directory.
@@ -219,19 +224,27 @@ def visium_hd(
             axes=("x", "y"),
         )
         # parse shapes
+        shapes_name = dataset_id + "_" + bin_size_str
+        radius = scalefactors[VisiumHDKeys.SCALEFACTORS_SPOT_DIAMETER_FULLRES] / 2.0
+        transformations = {
+            "global": transform_original,
+            "downscaled_hires": transform_hires,
+            "downscaled_lowres": transform_lowres,
+        }
         circles = ShapesModel.parse(
             adata.obsm["spatial"],
             geometry=0,
-            radius=scalefactors[VisiumHDKeys.SCALEFACTORS_SPOT_DIAMETER_FULLRES] / 2.0,
+            radius=radius,
             index=adata.obs[VisiumHDKeys.INSTANCE_KEY].copy(),
-            transformations={
-                "global": transform_original,
-                "downscaled_hires": transform_hires,
-                "downscaled_lowres": transform_lowres,
-            },
+            transformations=transformations,
         )
-        shapes_name = dataset_id + "_" + bin_size_str
-        shapes[shapes_name] = circles
+        if not bins_as_squares:
+            shapes[shapes_name] = circles
+        else:
+            squares_series = circles.buffer(radius, cap_style=3)
+            shapes[shapes_name] = ShapesModel.parse(
+                GeoDataFrame(geometry=squares_series), transformations=transformations
+            )
 
         # parse table
         adata.obs[VisiumHDKeys.REGION_KEY] = shapes_name

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/src/spatialdata_io/readers/xenium.py

@@ -56,7 +56,7 @@ def xenium(
     *,
     cells_boundaries: bool = True,
     nucleus_boundaries: bool = True,
-    cells_as_circles: bool = True,
+    cells_as_circles: bool | None = None,
     cells_labels: bool = True,
     nucleus_labels: bool = True,
     transcripts: bool = True,
@@ -96,10 +96,8 @@ def xenium(
     nucleus_boundaries
         Whether to read nucleus boundaries (polygons).
     cells_as_circles
-        Whether to read cells also as circles. Useful for performant visualization. The radii of the nuclei,
-        not the ones of cells, will be used; using the radii of cells would make the visualization too cluttered
-        (the cell boundaries are computed as a maximum expansion of the nuclei location and therefore the
-        corresponding circles would show considerable overlap).
+        Whether to read cells also as circles (the center and the radius of each circle is computed from the
+        corresponding labels cell).
     cells_labels
         Whether to read cell labels (raster). The polygonal version of the cell labels are simplified
         for visualization purposes, and using the raster version is recommended for analysis.
@@ -128,7 +126,34 @@ def xenium(
     Returns
     -------
     :class:`spatialdata.SpatialData`
+
+    Notes
+    -----
+    Old versions. Until spatialdata-io v0.1.3post0: previously, `cells_as_circles` was `True` by default; the table was associated to the
+    circles when `cells_as_circles` was `True`, and the table was associated to the polygons when `cells_as_circles`
+    was `False`; the radii of the circles were computed form the nuclei instead of the cells.
+
+    Performance. You can improve visualization performance (at the cost of accuracy) by setting `cells_as_circles` to `True`.
+
+    Examples
+    --------
+    This code shows how to change the annotation target of the table from the cell circles to the cell labels.
+    >>> from spatialdata_io import xenium
+    >>> sdata = xenium("path/to/raw/data", cells_as_circles=True)
+    >>> sdata["table"].obs["region"] = "cell_labels"
+    >>> sdata.set_table_annotates_spatialelement(
+    ...     table_name="table", region="cell_labels", region_key="region", instance_key="cell_labels"
+    ... )
+    >>> sdata.write("path/to/data.zarr")
     """
+    if cells_as_circles is None:
+        cells_as_circles = True
+        warnings.warn(
+            "The default value of `cells_as_circles` will change to `False` in the next release. "
+            "Please pass `True` explicitly to maintain the current behavior.",
+            DeprecationWarning,
+            stacklevel=3,
+        )
     image_models_kwargs, labels_models_kwargs = _initialize_raster_models_kwargs(
         image_models_kwargs, labels_models_kwargs
     )
@@ -138,7 +163,7 @@ def xenium(
     # to trigger the warning if the version cannot be parsed
     version = _parse_version_of_xenium_analyzer(specs, hide_warning=False)
 
-    specs["region"] = "cell_circles" if cells_as_circles else "cell_boundaries"
+    specs["region"] = "cell_circles" if cells_as_circles else "cell_labels"
 
     # the table is required in some cases
     if not cells_table:
@@ -210,14 +235,16 @@ def xenium(
         if cell_labels_indices_mapping is not None and table is not None:
             if not pd.DataFrame.equals(cell_labels_indices_mapping["cell_id"], table.obs[str(XeniumKeys.CELL_ID)]):
                 warnings.warn(
-                    "The cell_id column in the cell_labels_table does not match the cell_id column derived from the cell "
-                    "labels data. This could be due to trying to read a new version that is not supported yet. Please "
-                    "report this issue.",
+                    "The cell_id column in the cell_labels_table does not match the cell_id column derived from the "
+                    "cell labels data. This could be due to trying to read a new version that is not supported yet. "
+                    "Please report this issue.",
                     UserWarning,
                     stacklevel=2,
                 )
             else:
                 table.obs["cell_labels"] = cell_labels_indices_mapping["label_index"]
+                if not cells_as_circles:
+                    table.uns[TableModel.ATTRS_KEY][TableModel.INSTANCE_KEY] = "cell_labels"
 
     if nucleus_boundaries:
         polygons["nucleus_boundaries"] = _get_polygons(
@@ -506,7 +533,7 @@ def _get_tables_and_circles(
     table = TableModel.parse(adata, region=specs["region"], region_key="region", instance_key=str(XeniumKeys.CELL_ID))
     if cells_as_circles:
         transform = Scale([1.0 / specs["pixel_size"], 1.0 / specs["pixel_size"]], axes=("x", "y"))
-        radii = np.sqrt(adata.obs[XeniumKeys.CELL_NUCLEUS_AREA].to_numpy() / np.pi)
+        radii = np.sqrt(adata.obs[XeniumKeys.CELL_AREA].to_numpy() / np.pi)
         circles = ShapesModel.parse(
             circ,
             geometry=0,

{spatialdata_io-0.1.3.post0 → spatialdata_io-0.1.4}/tests/test_xenium.py

@@ -51,13 +51,13 @@ def test_roundtrip_with_data_limits() -> None:
     "dataset,expected",
     [
         ("Xenium_V1_human_Breast_2fov_outs", "{'y': (0, 3529), 'x': (0, 5792), 'z': (10, 25)}"),
-        ("Xenium_V1_human_Lung_2fov_outs", "{'y': (-1, 3556), 'x': (0, 5794), 'z': (7, 32)}"),
+        ("Xenium_V1_human_Lung_2fov_outs", "{'y': (0, 3553), 'x': (0, 5793), 'z': (7, 32)}"),
     ],
 )
 def test_example_data(dataset: str, expected: str) -> None:
     f = Path("./data") / dataset
     assert f.is_dir()
-    sdata = xenium(f)
+    sdata = xenium(f, cells_as_circles=False)
     from spatialdata import get_extent
 
     extent = get_extent(sdata, exact=False)