[pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

notebooks/examples/technology_spacem.ipynb

@@ -21,10 +21,10 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "import json\n",
     "from pathlib import Path\n",
     "\n",
     "import anndata as ad\n",
-    "import json\n",
     "import matplotlib\n",
     "import numpy as np\n",
     "import pandas as pd\n",
@@ -601,7 +601,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "well_shapes_instance = adata.obs.loc[(adata.obs.region==layout_name) & (adata.obs.slide_id==slide_id) & (adata.obs.well_id==well_id)].iloc[0].instance_id\n",
+    "well_shapes_instance = (\n",
+    "    adata.obs.loc[(adata.obs.region == layout_name) & (adata.obs.slide_id == slide_id) & (adata.obs.well_id == well_id)]\n",
+    "    .iloc[0]\n",
+    "    .instance_id\n",
+    ")\n",
     "well_polygon = spacem_sdata.shapes[layout_name].geometry.iloc[well_shapes_instance]\n",
     "# Note that Shapely has XY order, but we use YX. Bounds: min_x, min_y, max_x, max_y\n",
     "# Add some offset to visualize a bit more of the image\n",
@@ -639,21 +643,19 @@
     "%matplotlib inline\n",
     "\n",
     "# FIXME: wrong image region is plotted with bounding box query\n",
-    "#sdata_cropped = spacem_sdata.query.bounding_box(\n",
+    "# sdata_cropped = spacem_sdata.query.bounding_box(\n",
     "#    axes=YX,\n",
     "#    min_coordinate=min_coordinate_yx,\n",
     "#    max_coordinate=max_coordinate_yx,\n",
     "#    target_coordinate_system=COORD_SYS_GLOBAL,\n",
-    "#)\n",
+    "# )\n",
     "sdata_cropped = spacem_sdata\n",
     "fig = (\n",
     "    sdata_cropped.pl.render_images(elements=ablation_marks_image_name, palette=[\"gray\", \"blue\"])\n",
-    "    .pl.render_labels(\n",
-    "        elements=ablation_marks_labels_name, color=INSTANCE_KEY, cmap=\"viridis\", fill_alpha=1.0\n",
-    "    )\n",
-    "    #.pl.render_shapes(\n",
+    "    .pl.render_labels(elements=ablation_marks_labels_name, color=INSTANCE_KEY, cmap=\"viridis\", fill_alpha=1.0)\n",
+    "    # .pl.render_shapes(\n",
     "    #    elements=maldi_regions_name, fill_alpha=0.0, outline=True, outline_color=\"mediumvioletred\"\n",
-    "    #)\n",
+    "    # )\n",
     "    .pl.show(\n",
     "        coordinate_systems=COORD_SYS_GLOBAL,\n",
     "        title=\"Ablation marks\",\n",
@@ -697,9 +699,10 @@
     "%matplotlib inline\n",
     "\n",
     "(\n",
-    "    sdata_cropped.pl.render_images(elements=cells_image_name, palette=[\"gray\", \"lime\"])\n",
-    "    .pl.render_labels(elements=cells_labels_name, fill_alpha=1.0)\n",
-    "    #.pl.render_shapes(elements=maldi_regions_name, fill_alpha=0.0, outline=True, outline_color=\"mediumvioletred\")\n",
+    "    sdata_cropped.pl.render_images(elements=cells_image_name, palette=[\"gray\", \"lime\"]).pl.render_labels(\n",
+    "        elements=cells_labels_name, fill_alpha=1.0\n",
+    "    )\n",
+    "    # .pl.render_shapes(elements=maldi_regions_name, fill_alpha=0.0, outline=True, outline_color=\"mediumvioletred\")\n",
     "    .pl.show(coordinate_systems=COORD_SYS_GLOBAL, title=f\"Cell segmentation (pre_maldi)\")\n",
     ")\n",
     "None"
@@ -741,7 +744,7 @@
    "source": [
     "# Here, we automatically select the first non-zero ion with highest variance,\n",
     "# but any other ion from var_names can be selected manually.\n",
-    "_adata_roi = adata[adata.obs.region==ablation_marks_labels_name]\n",
+    "_adata_roi = adata[adata.obs.region == ablation_marks_labels_name]\n",
     "selected_ion = _adata_roi.var_names[np.nanargmax(np.nanvar(_adata_roi.X, axis=0))]\n",
     "selected_ion"
    ]
@@ -805,7 +808,7 @@
    "source": [
     "# Here, we automatically select the first non-zero ion with highest variance,\n",
     "# but any other ion from var_names can be selected manually.\n",
-    "_adata_roi = adata[adata.obs.region==cells_labels_name]\n",
+    "_adata_roi = adata[adata.obs.region == cells_labels_name]\n",
     "selected_ion = _adata_roi.var_names[np.nanargmax(np.nanvar(_adata_roi.X, axis=0))]\n",
     "selected_ion"
    ]
@@ -877,7 +880,7 @@
    ],
    "source": [
     "adata = spacem_sdata.tables[\"table\"]\n",
-    "cells_adata = adata[adata.obs.object_type==\"cells\", :]\n",
+    "cells_adata = adata[adata.obs.object_type == \"cells\", :]\n",
     "not_nan_ions = np.all(np.isfinite(cells_adata.X), axis=0)\n",
     "cells_adata = cells_adata[:, not_nan_ions]\n",
     "cells_adata.shape"
@@ -983,13 +986,22 @@
     "\n",
     "# Re-index batches before data integration to ensure matching batch numbers exist over all biological conditions.\n",
     "# (Otherwise, biological differences can get lost.)\n",
+    "\n",
+    "\n",
     "def reindex_batch(df: pd.DataFrame, batch_key: str) -> pd.DataFrame:\n",
     "    batches = df[batch_key].unique()\n",
     "    indexed = range(len(batches))\n",
     "    mapping = dict(zip(batches, indexed))\n",
     "    df[batch_key] = df[batch_key].replace(mapping)\n",
     "    return df\n",
-    "cells_adata.obs = cells_adata.obs.groupby(\"condition\", observed=True).apply(lambda group: reindex_batch(group, \"replicate\"), include_groups=True).reset_index(level=\"condition\", drop=True).sort_index()\n",
+    "\n",
+    "\n",
+    "cells_adata.obs = (\n",
+    "    cells_adata.obs.groupby(\"condition\", observed=True)\n",
+    "    .apply(lambda group: reindex_batch(group, \"replicate\"), include_groups=True)\n",
+    "    .reset_index(level=\"condition\", drop=True)\n",
+    "    .sort_index()\n",
+    ")\n",
     "\n",
     "sc.external.pp.bbknn(cells_adata, batch_key=\"replicate\")\n",
     "sc.tl.leiden(cells_adata)\n",
@@ -1031,7 +1043,9 @@
     "\n",
     "# Since ion names are not very expressive, label the plot with the first molecule candidate.\n",
     "# (The ion is not necessarily this candidate molecule.)\n",
-    "cells_adata.var[\"first_molecule_name\"] = cells_adata.var[\"moleculeNames\"].map(lambda names_json: next(iter(json.loads(names_json)), \"\")[:50])\n",
+    "cells_adata.var[\"first_molecule_name\"] = cells_adata.var[\"moleculeNames\"].map(\n",
+    "    lambda names_json: next(iter(json.loads(names_json)), \"\")[:50]\n",
+    ")\n",
     "sc.pl.rank_genes_groups(cells_adata, sharey=False, gene_symbols=\"first_molecule_name\")"
    ]
   },
@@ -1075,7 +1089,9 @@
     "best_ion_per_group = cells_adata.uns[\"rank_genes_groups\"][\"names\"][0]\n",
     "# Extend obs for plotting single conditions\n",
     "for group in groups:\n",
-    "    cells_adata.obs[group] = pd.Categorical([group if t==group else np.nan for t in cells_adata.obs[CONDITION_COL].values], categories=groups)\n",
+    "    cells_adata.obs[group] = pd.Categorical(\n",
+    "        [group if t == group else np.nan for t in cells_adata.obs[CONDITION_COL].values], categories=groups\n",
+    "    )\n",
     "sc.pl.umap(cells_adata, color=groups)\n",
     "sc.pl.umap(cells_adata, color=best_ion_per_group, cmap=\"viridis_r\")"
    ]