Version bump (#490)

* Version bump

* Bump js version and update blobs notebook

* Update script

* Lint

* Update

Author: keller-mark

.coveragerc_omit

@@ -14,4 +14,5 @@ omit =
     src/vitessce/data_utils/multivec.py
     src/vitessce/data_utils/spatialdata_points_zorder.py
     src/vitessce/widget_plugins/demo_plugin.py
-    src/vitessce/widget_plugins/spatial_query.py
+    src/vitessce/widget_plugins/spatial_query.py
+    tests/*.py

docs/notebooks/spatial_data_blobs.ipynb

@@ -30,7 +30,9 @@
    "outputs": [],
    "source": [
     "import spatialdata\n",
-    "from os.path import join"
+    "from os.path import join\n",
+    "import pandas as pd\n",
+    "from anndata import AnnData"
    ]
   },
   {
@@ -39,7 +41,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "sdata = spatialdata.datasets.blobs()"
+    "sdata = spatialdata.datasets.blobs()\n",
+    "sdata"
    ]
   },
   {
@@ -48,7 +51,13 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "spatialdata_filepath = join(\"data\", \"blobs.spatialdata.zarr\")"
+    "# This blobs dataset only contains a table with a var.index corresponding to the shapes features.\n",
+    "# We need to construct a table with a var.index corresponding to the points features:\n",
+    "ddf = sdata.points[\"blobs_points\"]\n",
+    "unique_gene_ids = ddf[\"genes\"].unique().compute().tolist()\n",
+    "points_var_df = pd.DataFrame(index=unique_gene_ids, data=[], columns=[])\n",
+    "points_table = AnnData(var=points_var_df, obs=None, X=None)\n",
+    "sdata.tables['table_points'] = points_table"
    ]
   },
   {
@@ -57,7 +66,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "sdata.write(spatialdata_filepath, overwrite=True)"
+    "spatialdata_filepath = join(\"data\", \"blobs.spatialdata.zarr\")"
    ]
   },
   {
@@ -66,13 +75,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "# Change data type of x and y columns from int64 to int32\n",
-    "# Temporary workaround, will be resolved by https://github.com/vitessce/vitessce/issues/2292\n",
-    "ddf = sdata[\"blobs_points\"]\n",
-    "ddf[\"x\"] = ddf[\"x\"].astype('int32')\n",
-    "ddf[\"y\"] = ddf[\"y\"].astype('int32')\n",
-    "sdata[\"blobs_points_2\"] = ddf\n",
-    "sdata.write_element(\"blobs_points_2\")"
+    "sdata.write(spatialdata_filepath, overwrite=True)"
    ]
   },
   {
@@ -112,23 +115,28 @@
     ")\n",
     "# Add data to the configuration:\n",
     "wrapper = SpatialDataWrapper(\n",
-    "    sdata_store=spatialdata_filepath,\n",
+    "    sdata_path=spatialdata_filepath,\n",
     "    # The following paths are relative to the root of the SpatialData zarr store on-disk.\n",
     "    image_path=\"images/blobs_image\",\n",
     "    obs_segmentations_path=\"labels/blobs_labels\",\n",
+    "    obs_embedding_paths=[\"tables/table/obsm/X_umap\"],\n",
+    "    obs_feature_matrix_path=\"tables/table/X\",\n",
     "    coordinate_system=\"global\",\n",
     "    coordination_values={\n",
     "        \"obsType\": \"blob\",\n",
+    "        \"featureType\": \"channel\",\n",
     "        \"fileUid\": \"my_unique_id\"\n",
     "    }\n",
     ")\n",
     "points_wrapper = SpatialDataWrapper(\n",
-    "    sdata_store=spatialdata_filepath,\n",
+    "    sdata_path=spatialdata_filepath,\n",
     "    # The following paths are relative to the root of the SpatialData zarr store on-disk.\n",
-    "    obs_points_path=\"points/blobs_points_2\",\n",
+    "    obs_points_path=\"points/blobs_points\",\n",
+    "    obs_feature_matrix_path=\"tables/table_points/X\", # TODO\n",
     "    coordinate_system=\"global\",\n",
     "    coordination_values={\n",
     "        \"obsType\": \"point\",\n",
+    "        \"featureType\": \"gene\",\n",
     "        \"fileUid\": \"other_unique_id\"\n",
     "    }\n",
     ")\n",
@@ -169,6 +177,7 @@
     "        'segmentationChannel': CL([{\n",
     "            'spatialChannelVisible': True,\n",
     "            'obsType': 'blob',\n",
+    "            \"featureType\": \"channel\",\n",
     "        }]),\n",
     "    }]),\n",
     "}, scope_prefix=get_initial_coordination_scope_prefix(\"A\", \"obsSegmentations\"))\n",
@@ -177,7 +186,10 @@
     "    'pointLayer': CL([{\n",
     "        \"fileUid\": \"other_unique_id\",\n",
     "        \"obsType\": \"point\",\n",
+    "        \"featureType\": \"gene\",\n",
     "        \"obsHighlight\": None,\n",
+    "        \"obsColorEncoding\": \"randomByFeature\",\n",
+    "        \"spatialLayerOpacity\": 1.0,\n",
     "    }]),\n",
     "}, scope_prefix=get_initial_coordination_scope_prefix(\"A\", \"obsPoints\"))\n",
     "\n",

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "vitessce"
-version = "3.8.0"
+version = "3.8.1"
 authors = [
   { name="Mark Keller", email="mark_keller@hms.harvard.edu" },
 ]

src/vitessce/file_def_utils.py

@@ -147,7 +147,7 @@ def gen_sdata_obs_segmentations_schema(options, path: str, table_path: str = "ta
     return options
 
 
-def gen_sdata_obs_points_schema(options, path: str, table_path: str = "tables/table", coordinate_system: Optional[str] = None) -> dict:
+def gen_sdata_obs_points_schema(options, path: str, table_path: str = "tables/table", coordinate_system: Optional[str] = None, feature_index_column: Optional[str] = None, morton_code_column: Optional[str] = None) -> dict:
     if path is not None:
         options["obsPoints"] = {
             "path": path
@@ -156,6 +156,10 @@ def gen_sdata_obs_points_schema(options, path: str, table_path: str = "tables/ta
             options["obsPoints"]['tablePath'] = table_path
         if coordinate_system is not None:
             options["obsPoints"]['coordinateSystem'] = coordinate_system
+        if feature_index_column is not None:
+            options["obsPoints"]['featureIndexColumn'] = feature_index_column
+        if morton_code_column is not None:
+            options["obsPoints"]['mortonCodeColumn'] = morton_code_column
     return options
 
 

src/vitessce/widget.py

@@ -729,7 +729,7 @@ class VitessceWidget(anywidget.AnyWidget):
 
     next_port = DEFAULT_PORT
 
-    js_package_version = Unicode('3.9.0').tag(sync=True)
+    js_package_version = Unicode('3.9.2').tag(sync=True)
     js_dev_mode = Bool(False).tag(sync=True)
     custom_js_url = Unicode('').tag(sync=True)
     plugin_esm = List(trait=Unicode(''), default_value=[]).tag(sync=True)
@@ -742,7 +742,7 @@ class VitessceWidget(anywidget.AnyWidget):
 
     store_urls = List(trait=Unicode(''), default_value=[]).tag(sync=True)
 
-    def __init__(self, config, height=600, theme='auto', uid=None, port=None, proxy=False, js_package_version='3.9.0', js_dev_mode=False, custom_js_url='', plugins=None, remount_on_uid_change=True, prefer_local=True, invoke_timeout=300000, invoke_batched=True, page_mode=False, page_esm=None, prevent_scroll=True, server_host=None):
+    def __init__(self, config, height=600, theme='auto', uid=None, port=None, proxy=False, js_package_version='3.9.2', js_dev_mode=False, custom_js_url='', plugins=None, remount_on_uid_change=True, prefer_local=True, invoke_timeout=300000, invoke_batched=True, page_mode=False, page_esm=None, prevent_scroll=True, server_host=None):
         """
         Construct a new Vitessce widget. Not intended to be instantiated directly; instead, use ``VitessceConfig.widget``.
 
@@ -876,7 +876,7 @@ def _plugin_command(self, params, buffers):
 # Launch Vitessce using plain HTML representation (no ipywidgets)
 
 
-def ipython_display(config, height=600, theme='auto', base_url=None, host_name=None, uid=None, port=None, proxy=False, js_package_version='3.9.0', js_dev_mode=False, custom_js_url='', plugins=None, remount_on_uid_change=True, page_mode=False, page_esm=None, server_host=None):
+def ipython_display(config, height=600, theme='auto', base_url=None, host_name=None, uid=None, port=None, proxy=False, js_package_version='3.9.2', js_dev_mode=False, custom_js_url='', plugins=None, remount_on_uid_change=True, page_mode=False, page_esm=None, server_host=None):
     from IPython.display import display, HTML
     uid_str = "vitessce" + get_uid_str(uid)
 

src/vitessce/wrappers.py

@@ -1407,7 +1407,7 @@ def auto_view_config(self, vc):
 
 class SpatialDataWrapper(AnnDataWrapper):
 
-    def __init__(self, sdata_path: Optional[str] = None, sdata_url: Optional[str] = None, sdata_store: Optional[Union[str, zarr.storage.StoreLike]] = None, sdata_artifact: Optional[ln.Artifact] = None, image_path: Optional[str] = None, region: Optional[str] = None, coordinate_system: Optional[str] = None, obs_spots_path: Optional[str] = None, obs_segmentations_path: Optional[str] = None, obs_points_path: Optional[str] = None, table_path: str = "tables/table", is_zip=None, coordination_values=None, **kwargs):
+    def __init__(self, sdata_path: Optional[str] = None, sdata_url: Optional[str] = None, sdata_store: Optional[Union[str, zarr.storage.StoreLike]] = None, sdata_artifact: Optional[ln.Artifact] = None, image_path: Optional[str] = None, region: Optional[str] = None, coordinate_system: Optional[str] = None, obs_spots_path: Optional[str] = None, obs_segmentations_path: Optional[str] = None, obs_points_path: Optional[str] = None, obs_points_feature_index_column: Optional[str] = None, obs_points_morton_code_column: Optional[str] = None, table_path: str = "tables/table", is_zip=None, coordination_values=None, **kwargs):
         """
         Wrap a SpatialData object.
 
@@ -1432,6 +1432,10 @@ def __init__(self, sdata_path: Optional[str] = None, sdata_url: Optional[str] =
         :type obs_segmentations_path: Optional[str]
         :param obs_points_path: Path to a points element, by default None
         :type obs_points_path: Optional[str]
+        :param obs_points_feature_index_column: Column in the points dataframe that contains a feature index value (i.e., index into table.var.index to specify a gene) for each point, by default None
+        :type obs_points_feature_index_column: Optional[str]
+        :param obs_points_morton_code_column: Column in the points dataframe that contains a morton code for each point, by default None
+        :type obs_points_morton_code_column: Optional[str]
         :param str feature_labels_path: Path to a table var column containing feature labels (e.g., alternate gene symbols), instead of the default index column of the `var` dataframe.
         :param list[str] obs_embedding_paths: Column names like `['obsm/X_umap', 'obsm/X_pca']` for showing scatterplots
         :param list[str] obs_embedding_names: Overriding names like `['UMAP', 'PCA']` for displaying above scatterplots
@@ -1466,6 +1470,8 @@ def __init__(self, sdata_path: Optional[str] = None, sdata_url: Optional[str] =
         self._obs_spots_path = obs_spots_path
         self._obs_segmentations_path = obs_segmentations_path
         self._obs_points_path = obs_points_path
+        self._obs_points_feature_index_column = obs_points_feature_index_column
+        self._obs_points_morton_code_column = obs_points_morton_code_column
         if self._adata_path is not None:
             self.zarr_folder = 'spatialdata.zarr'
         self.obs_type_label = None
@@ -1553,7 +1559,10 @@ def generator(base_url):
             options = gen_sdata_obs_spots_schema(options, self._obs_spots_path, self._table_path, self._region, self._coordinate_system)
             options = gen_sdata_image_schema(options, self._image_path, self._coordinate_system)
             options = gen_sdata_obs_segmentations_schema(options, self._obs_segmentations_path, self._table_path, self._coordinate_system)
-            options = gen_sdata_obs_points_schema(options, self._obs_points_path, self._table_path, self._coordinate_system)
+            options = gen_sdata_obs_points_schema(
+                options, self._obs_points_path, self._table_path, self._coordinate_system,
+                self._obs_points_feature_index_column, self._obs_points_morton_code_column
+            )
             options = gen_feature_labels_schema(self._feature_labels, options)
             options = gen_obs_embedding_schema(options, self._mappings_obsm, self._mappings_obsm_names, self._mappings_obsm_dims)
             if len(options.keys()) > 0:

tests/create_xenium_filtered_points.py

@@ -8,44 +8,48 @@
 from spatialdata import read_zarr, SpatialData
 
 
-data_dir = "data"
-zip_filepath = join(data_dir, "xenium_rep1_io.spatialdata.zarr.zip")
-spatialdata_filepath = join(data_dir, "xenium_rep1_io.spatialdata.zarr")
+def create_xenium_filtered_points():
+    # 1. Download and extract the Xenium dataset if not already present
+    data_dir = "data"
+    zip_filepath = join(data_dir, "xenium_rep1_io.spatialdata.zarr.zip")
+    spatialdata_filepath = join(data_dir, "xenium_rep1_io.spatialdata.zarr")
 
+    if not isdir(spatialdata_filepath):
+        if not isfile(zip_filepath):
+            os.makedirs(data_dir, exist_ok=True)
+            # zip_url = 'https://s3.embl.de/spatialdata/spatialdata-sandbox/xenium_rep1_io.zip'
+            zip_url = 'https://s3.embl.de/spatialdata/spatialdata-sandbox/xenium_rep1_io_spatialdata_0.7.1.zip'
+            urlretrieve(zip_url, zip_filepath)
+        with zipfile.ZipFile(zip_filepath, "r") as zip_ref:
+            zip_ref.extractall(data_dir)
+            os.rename(join(data_dir, "data.zarr"), spatialdata_filepath)
 
-if not isdir(spatialdata_filepath):
-    if not isfile(zip_filepath):
-        os.makedirs(data_dir, exist_ok=True)
-        urlretrieve('https://s3.embl.de/spatialdata/spatialdata-sandbox/xenium_rep1_io.zip', zip_filepath)
-    with zipfile.ZipFile(zip_filepath, "r") as zip_ref:
-        zip_ref.extractall(data_dir)
-        os.rename(join(data_dir, "data.zarr"), spatialdata_filepath)
+            # This Xenium dataset has an AnnData "raw" element.
+            # Reference: https://github.com/giovp/spatialdata-sandbox/issues/55
+            raw_dir = join(spatialdata_filepath, "tables", "table", "raw")
+            if isdir(raw_dir):
+                shutil.rmtree(raw_dir)
 
-        # This Xenium dataset has an AnnData "raw" element.
-        # Reference: https://github.com/giovp/spatialdata-sandbox/issues/55
-        raw_dir = join(spatialdata_filepath, "tables", "table", "raw")
-        if isdir(raw_dir):
-            shutil.rmtree(raw_dir)
+    sdata = read_zarr(spatialdata_filepath)
 
-sdata = read_zarr(spatialdata_filepath)
+    ddf = sdata.points["transcripts"]
 
-ddf = sdata.points["transcripts"]
+    # 2. Define a function to take every 100th row from a partition
 
-# 2. Define a function to take every 100th row from a partition
+    def select_every_200th(partition):
+        # Each 'partition' is a Pandas DataFrame
+        # .iloc[::100] is the efficient pandas way to get every 100th row
+        return partition.iloc[::200]
 
+    # 3. Apply this function to every partition in the Dask DataFrame
+    result = ddf.map_partitions(select_every_200th)
 
-def select_every_200th(partition):
-    # Each 'partition' is a Pandas DataFrame
-    # .iloc[::100] is the efficient pandas way to get every 100th row
-    return partition.iloc[::200]
+    # 4. Compute the result to see it
+    filtered_ddf = result[["x", "y", "z", "feature_name", "cell_id"]]
 
+    small_sdata = SpatialData(points={"transcripts": filtered_ddf})
 
-# 3. Apply this function to every partition in the Dask DataFrame
-result = ddf.map_partitions(select_every_200th)
+    small_sdata.write("xenium_rep1_io.points_only.spatialdata.zarr", overwrite=True)
 
-# 4. Compute the result to see it
-filtered_ddf = result[["x", "y", "z", "feature_name", "cell_id"]]
-
-small_sdata = SpatialData(points={"transcripts": filtered_ddf})
-
-small_sdata.write("xenium_rep1_io.points_only.spatialdata.zarr", overwrite=True)
+# Uncomment to run.
+# create_xenium_filtered_points()