{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "nbsphinx": "hidden"
   },
   "source": [
    "# Vitessce Widget Tutorial"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualization of a SpatialData object"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import dependencies\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from os.path import join, isfile, isdir\n",
    "from urllib.request import urlretrieve\n",
    "import zipfile\n",
    "import shutil\n",
    "\n",
    "from vitessce import (\n",
    "    VitessceConfig,\n",
    "    ViewType as vt,\n",
    "    CoordinationType as ct,\n",
    "    CoordinationLevel as CL,\n",
    "    SpatialDataWrapper,\n",
    "    get_initial_coordination_scope_prefix\n",
    ")\n",
    "\n",
    "from vitessce.data_utils import (\n",
    "    sdata_morton_sort_points,\n",
    "    sdata_points_process_columns,\n",
    "    sdata_points_write_bounding_box_attrs,\n",
    "    sdata_points_modify_row_group_size,\n",
    "    sdata_morton_query_rect,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from spatialdata import read_zarr\n",
    "\n",
    "import anndata as ad\n",
    "\n",
    "ad.settings.zarr_write_format = 3\n",
    "print(ad.settings.zarr_write_format)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import dask\n",
    "import tempfile\n",
    "\n",
    "# Point Dask temp dir to your external drive\n",
    "dask.config.set({'temporary_directory': '/Volumes/T7 Shield/tmp'})\n",
    "\n",
    "# Create the dir if it doesn't exist\n",
    "import os\n",
    "os.makedirs('/Volumes/T7 Shield/tmp', exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "from spatialdata_io import xenium\n",
    "import spatialdata as sd\n",
    "import pandas as pd\n",
    "\n",
    "xenium_dir = Path(\"../data/instrument_data/Xenium_V1_hPancreas_Cancer_Add_on_FFPE_outs/\")   # folder containing experiment.xenium\n",
    "out_zarr = Path(\"../data/processed_data/vitessce/Xenium_V1_hPancreas_Cancer_Add_on_FFPE_outs.zarr\")\n",
    "\n",
    "sdata = xenium(\n",
    "    xenium_dir,\n",
    "    cells_boundaries=True,\n",
    "    nucleus_boundaries=True,\n",
    "    cells_labels=True,\n",
    "    nucleus_labels=True,\n",
    "    transcripts=True,\n",
    "    morphology_focus=True,\n",
    "    aligned_images=True,\n",
    "    cells_table=True,\n",
    "    gex_only=True,\n",
    ")\n",
    "\n",
    "print(sdata)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add cluster labels from Xenium instrument output\n",
    "clusters = pd.read_csv(\n",
    "    xenium_dir / \"analysis/clustering/gene_expression_graphclust/clusters.csv\"\n",
    ")\n",
    "sdata.tables[\"table\"].obs = sdata.tables[\"table\"].obs.merge(\n",
    "    clusters.set_index(\"Barcode\")[[\"Cluster\"]].rename(columns={\"Cluster\": \"leiden\"}),\n",
    "    left_index=True,\n",
    "    right_index=True,\n",
    "    how=\"left\",\n",
    ")\n",
    "sdata.tables[\"table\"].obs[\"leiden\"] = sdata.tables[\"table\"].obs[\"leiden\"].astype(str)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save as a SpatialData Zarr store\n",
    "sdata.write(out_zarr)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdata[\"transcripts\"].shape[0].compute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdata.tables[\"table\"].X = sdata.tables[\"table\"].X.toarray()\n",
    "sdata.tables[\"dense_table\"] = sdata.tables[\"table\"]\n",
    "sdata.write_element(\"dense_table\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: store the two separate images as a single image with two channels.\n",
    "# Similar to https://github.com/EricMoerthVis/tissue-map-tools/pull/12"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sdata.tables['table'].obs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sdata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sdata.points['transcripts'].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sorting Points and creating a new Points element in the SpatialData object"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 1. Sort rows with `sdata_morton_sort_points`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import importlib.metadata\n",
    "print(importlib.metadata.version(\"vitessce\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sdata = sdata_morton_sort_points(sdata, \"transcripts\")\n",
    "from vitessce.data_utils.spatialdata_points_zorder import norm_ddf_to_uint, morton_interleave\n",
    "\n",
    "element = \"transcripts\"\n",
    "ddf = sdata.points[element]\n",
    "attrs = ddf.attrs.copy()\n",
    "\n",
    "ddf = norm_ddf_to_uint(ddf)\n",
    "ddf[\"morton_code_2d\"] = morton_interleave(ddf)\n",
    "sorted_ddf = ddf.sort_values(by=\"morton_code_2d\", ascending=True)\n",
    "sorted_ddf.attrs.update(attrs)\n",
    "sdata.points[element] = sorted_ddf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 2. Clean up columns with `sdata_points_process_columns`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add feature_index column to dataframe, and reorder columns so that feature_name (dict column) is the rightmost column.\n",
    "ddf = sdata_points_process_columns(sdata, \"transcripts\", var_name_col=\"feature_name\", table_name=\"table\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ddf.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3. Save sorted dataframe to new Points element"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sdata[\"transcripts_with_morton_codes\"] = ddf\n",
    "# sdata.write_element(\"transcripts_with_morton_codes\")\n",
    "\n",
    "from spatialdata.models import PointsModel\n",
    "\n",
    "transformations = sdata[\"transcripts\"].attrs[\"transform\"]\n",
    "del ddf.attrs[\"transform\"]\n",
    "\n",
    "sdata[\"transcripts_with_morton_codes\"] = PointsModel.parse(\n",
    "    ddf, feature_key=\"feature_name\", instance_key=\"cell_id\", transformations=transformations\n",
    ")\n",
    "sdata.write_element(\"transcripts_with_morton_codes\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 4. Write bounding box metadata with `sdata_points_write_bounding_box_attrs`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import shutil\n",
    "import os\n",
    "\n",
    "tmp_dir = '/Volumes/T7 Shield/tmp'\n",
    "shutil.rmtree(tmp_dir)\n",
    "os.makedirs(tmp_dir)\n",
    "print(\"Done\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdata_points_write_bounding_box_attrs(sdata, \"transcripts_with_morton_codes\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 5. Modify the row group sizes of the Parquet files with `sdata_points_modify_row_group_size`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import shutil\n",
    "import os\n",
    "\n",
    "tmp_dir = '/Volumes/T7 Shield/tmp'\n",
    "shutil.rmtree(tmp_dir)\n",
    "os.makedirs(tmp_dir)\n",
    "print(\"Done\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdata_points_modify_row_group_size(sdata, \"transcripts_with_morton_codes\", row_group_size=25_000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Done"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Optionally, check the number of row groups in one of the parquet file parts.\n",
    "import pyarrow.parquet as pq\n",
    "from os.path import join\n",
    "\n",
    "parquet_file = pq.ParquetFile(join(sdata.path, \"points\", \"transcripts_with_morton_codes\", \"points.parquet\", \"part.0.parquet\"))\n",
    "\n",
    "# Get the number of row groups in this part-0 file.\n",
    "num_groups = parquet_file.num_row_groups\n",
    "num_groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.1"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "undefined": {
      "model_module": "anywidget",
      "model_module_version": "2.0.0",
      "model_name": "AnyModel",
      "state": {
       "_view_name": "ErrorWidgetView",
       "error": {},
       "msg": "Model class 'AnyModel' from module 'anywidget' is loaded but can not be instantiated"
      }
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}