walkthrough.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pandasaurus CxG Extension Walkthrough "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Overview\n",
    "Welcome to this Jupyter notebook walkthrough for pandasaurus_cxg! This library provides powerful tools for analyzing and enriching AnnData objects, enabling you to gain deeper insights into your single-cell RNA sequencing (scRNA-seq) data.\n",
    "\n",
    "In this notebook, we will explore two main classes: `AnndataEnricher` and `AnndataAnalyzer`. Let's dive in and see how these classes can help us in our scRNA-seq analysis.\n",
    "\n",
    "Now, let's get started with an example workflow that demonstrates the capabilities of these classes. We'll load an example dataset, perform enrichment, analysis, and visualization steps to gain a better understanding of our scRNA-seq data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test Data\n",
    "The following files are used in the walkthrough. Please download them manually to a folder of your choice. Ensure that you adjust the file paths used in the examples to match your local file paths.\n",
    "- [Time-resolved Systems Immunology Reveals a Late Juncture Linked to Fatal COVID-19: Adaptive Cells](https://cellxgene.cziscience.com/collections/db14ce52-5dd6-4649-a9e9-7fb2572d0605)\n",
    "- [Integrated Single-nucleus and Single-cell RNA-seq of the Adult Human Kidney](https://cellxgene.cziscience.com/collections/36b8480d-114e-42fe-b6a9-bdf79a7eb1fc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## AnndataEnricher Walkthrough"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Initialization\n",
    "Let's import the necessary modules and initialize our AnndataEnricher"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pandasaurus_cxg.anndata_enricher import AnndataEnricher"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Using Time-resolved Systems Immunology Reveals a Late Juncture Linked to Fatal COVID-19: Adaptive Cells dataset\n",
    "ade = AnndataEnricher.from_file_path(\"./immune_example.h5ad\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "AnnData object loaded into AnnDataEnricher"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tissue_ontology_term_id</th>\n",
       "      <th>author_cell_type</th>\n",
       "      <th>disease_ontology_term_id</th>\n",
       "      <th>age</th>\n",
       "      <th>days_since_hospitalized</th>\n",
       "      <th>donor_id</th>\n",
       "      <th>severity</th>\n",
       "      <th>dsm_severity_score</th>\n",
       "      <th>ever_admitted_to_icu</th>\n",
       "      <th>days_since_onset</th>\n",
       "      <th>...</th>\n",
       "      <th>organism_ontology_term_id</th>\n",
       "      <th>suspension_type</th>\n",
       "      <th>cell_type</th>\n",
       "      <th>assay</th>\n",
       "      <th>disease</th>\n",
       "      <th>organism</th>\n",
       "      <th>sex</th>\n",
       "      <th>tissue</th>\n",
       "      <th>self_reported_ethnicity</th>\n",
       "      <th>development_stage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AAACCTGAGACGACGT-1_1</th>\n",
       "      <td>UBERON:0000178</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>PATO:0000461</td>\n",
       "      <td>32.0</td>\n",
       "      <td></td>\n",
       "      <td>AA220907</td>\n",
       "      <td></td>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>...</td>\n",
       "      <td>NCBITaxon:9606</td>\n",
       "      <td>cell</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>10x 5' v1</td>\n",
       "      <td>normal</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>female</td>\n",
       "      <td>blood</td>\n",
       "      <td>unknown</td>\n",
       "      <td>32-year-old human stage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAACCTGAGCTAGTTC-1_1</th>\n",
       "      <td>UBERON:0000178</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>MONDO:0100096</td>\n",
       "      <td>37.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>HGR0000079</td>\n",
       "      <td>Critical</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>30.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NCBITaxon:9606</td>\n",
       "      <td>cell</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>10x 5' v1</td>\n",
       "      <td>COVID-19</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>female</td>\n",
       "      <td>blood</td>\n",
       "      <td>European</td>\n",
       "      <td>37-year-old human stage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAACCTGCATAGACTC-1_1</th>\n",
       "      <td>UBERON:0000178</td>\n",
       "      <td>memory B cell</td>\n",
       "      <td>PATO:0000461</td>\n",
       "      <td>32.0</td>\n",
       "      <td></td>\n",
       "      <td>AA220907</td>\n",
       "      <td></td>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>...</td>\n",
       "      <td>NCBITaxon:9606</td>\n",
       "      <td>cell</td>\n",
       "      <td>memory B cell</td>\n",
       "      <td>10x 5' v1</td>\n",
       "      <td>normal</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>female</td>\n",
       "      <td>blood</td>\n",
       "      <td>unknown</td>\n",
       "      <td>32-year-old human stage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAACCTGCATTAACCG-1_1</th>\n",
       "      <td>UBERON:0000178</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>MONDO:0100096</td>\n",
       "      <td>54.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>HGR0000143</td>\n",
       "      <td>Critical</td>\n",
       "      <td>1.674056</td>\n",
       "      <td>True</td>\n",
       "      <td>8.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NCBITaxon:9606</td>\n",
       "      <td>cell</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>10x 5' v1</td>\n",
       "      <td>COVID-19</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>female</td>\n",
       "      <td>blood</td>\n",
       "      <td>East Asian</td>\n",
       "      <td>54-year-old human stage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAACCTGGTCATCGGC-1_1</th>\n",
       "      <td>UBERON:0000178</td>\n",
       "      <td>gamma-delta T cell</td>\n",
       "      <td>MONDO:0100096</td>\n",
       "      <td>55.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>HGR0000083</td>\n",
       "      <td>Moderate</td>\n",
       "      <td>-1.950858</td>\n",
       "      <td>False</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NCBITaxon:9606</td>\n",
       "      <td>cell</td>\n",
       "      <td>gamma-delta T cell</td>\n",
       "      <td>10x 5' v1</td>\n",
       "      <td>COVID-19</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>European</td>\n",
       "      <td>55-year-old human stage</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     tissue_ontology_term_id    author_cell_type  \\\n",
       "AAACCTGAGACGACGT-1_1          UBERON:0000178        naive B cell   \n",
       "AAACCTGAGCTAGTTC-1_1          UBERON:0000178        naive B cell   \n",
       "AAACCTGCATAGACTC-1_1          UBERON:0000178       memory B cell   \n",
       "AAACCTGCATTAACCG-1_1          UBERON:0000178        naive B cell   \n",
       "AAACCTGGTCATCGGC-1_1          UBERON:0000178  gamma-delta T cell   \n",
       "\n",
       "                     disease_ontology_term_id   age days_since_hospitalized  \\\n",
       "AAACCTGAGACGACGT-1_1             PATO:0000461  32.0                           \n",
       "AAACCTGAGCTAGTTC-1_1            MONDO:0100096  37.0                     6.0   \n",
       "AAACCTGCATAGACTC-1_1             PATO:0000461  32.0                           \n",
       "AAACCTGCATTAACCG-1_1            MONDO:0100096  54.0                     1.0   \n",
       "AAACCTGGTCATCGGC-1_1            MONDO:0100096  55.0                     1.0   \n",
       "\n",
       "                        donor_id  severity  dsm_severity_score  \\\n",
       "AAACCTGAGACGACGT-1_1    AA220907                           NaN   \n",
       "AAACCTGAGCTAGTTC-1_1  HGR0000079  Critical                 NaN   \n",
       "AAACCTGCATAGACTC-1_1    AA220907                           NaN   \n",
       "AAACCTGCATTAACCG-1_1  HGR0000143  Critical            1.674056   \n",
       "AAACCTGGTCATCGGC-1_1  HGR0000083  Moderate           -1.950858   \n",
       "\n",
       "                     ever_admitted_to_icu days_since_onset  ...  \\\n",
       "AAACCTGAGACGACGT-1_1                                        ...   \n",
       "AAACCTGAGCTAGTTC-1_1                False             30.0  ...   \n",
       "AAACCTGCATAGACTC-1_1                                        ...   \n",
       "AAACCTGCATTAACCG-1_1                 True              8.0  ...   \n",
       "AAACCTGGTCATCGGC-1_1                False              6.0  ...   \n",
       "\n",
       "                     organism_ontology_term_id suspension_type  \\\n",
       "AAACCTGAGACGACGT-1_1            NCBITaxon:9606            cell   \n",
       "AAACCTGAGCTAGTTC-1_1            NCBITaxon:9606            cell   \n",
       "AAACCTGCATAGACTC-1_1            NCBITaxon:9606            cell   \n",
       "AAACCTGCATTAACCG-1_1            NCBITaxon:9606            cell   \n",
       "AAACCTGGTCATCGGC-1_1            NCBITaxon:9606            cell   \n",
       "\n",
       "                               cell_type      assay   disease      organism  \\\n",
       "AAACCTGAGACGACGT-1_1        naive B cell  10x 5' v1    normal  Homo sapiens   \n",
       "AAACCTGAGCTAGTTC-1_1        naive B cell  10x 5' v1  COVID-19  Homo sapiens   \n",
       "AAACCTGCATAGACTC-1_1       memory B cell  10x 5' v1    normal  Homo sapiens   \n",
       "AAACCTGCATTAACCG-1_1        naive B cell  10x 5' v1  COVID-19  Homo sapiens   \n",
       "AAACCTGGTCATCGGC-1_1  gamma-delta T cell  10x 5' v1  COVID-19  Homo sapiens   \n",
       "\n",
       "                         sex tissue self_reported_ethnicity  \\\n",
       "AAACCTGAGACGACGT-1_1  female  blood                 unknown   \n",
       "AAACCTGAGCTAGTTC-1_1  female  blood                European   \n",
       "AAACCTGCATAGACTC-1_1  female  blood                 unknown   \n",
       "AAACCTGCATTAACCG-1_1  female  blood              East Asian   \n",
       "AAACCTGGTCATCGGC-1_1    male  blood                European   \n",
       "\n",
       "                            development_stage  \n",
       "AAACCTGAGACGACGT-1_1  32-year-old human stage  \n",
       "AAACCTGAGCTAGTTC-1_1  37-year-old human stage  \n",
       "AAACCTGCATAGACTC-1_1  32-year-old human stage  \n",
       "AAACCTGCATTAACCG-1_1  54-year-old human stage  \n",
       "AAACCTGGTCATCGGC-1_1  55-year-old human stage  \n",
       "\n",
       "[5 rows x 30 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ade.anndata.obs[0:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Enrichment\n",
    "\n",
    "Enrichment methods use the structure of the ontology to provide relevant semantic enrichment to cell type ontology annotation in the CxG Anndata file.  Results are returned as a simple pandas dataframe.\n",
    "\n",
    "Users can choose from a range of techniques to limit enrichment to some relevant context.\n",
    "\n",
    "#### Simple enrichment\n",
    "\n",
    "In simple enichment the context is just the terms CL terms used in annotation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>s</th>\n",
       "      <th>s_label</th>\n",
       "      <th>p</th>\n",
       "      <th>o</th>\n",
       "      <th>o_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CL:0000798</td>\n",
       "      <td>gamma-delta T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CL:0000809</td>\n",
       "      <td>double-positive, alpha-beta thymocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CL:0000813</td>\n",
       "      <td>memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CL:0000815</td>\n",
       "      <td>regulatory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CL:0000895</td>\n",
       "      <td>naive thymus-derived CD4-positive, alpha-beta ...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>CL:0000897</td>\n",
       "      <td>CD4-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>CL:0000897</td>\n",
       "      <td>CD4-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000813</td>\n",
       "      <td>memory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>CL:0000900</td>\n",
       "      <td>naive thymus-derived CD8-positive, alpha-beta ...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>CL:0000909</td>\n",
       "      <td>CD8-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>CL:0000909</td>\n",
       "      <td>CD8-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000813</td>\n",
       "      <td>memory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>CL:0000940</td>\n",
       "      <td>mucosal invariant T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>CL:0002489</td>\n",
       "      <td>double negative thymocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             s                                            s_label  \\\n",
       "0   CL:0000798                                 gamma-delta T cell   \n",
       "1   CL:0000809              double-positive, alpha-beta thymocyte   \n",
       "2   CL:0000813                                      memory T cell   \n",
       "3   CL:0000815                                  regulatory T cell   \n",
       "4   CL:0000895  naive thymus-derived CD4-positive, alpha-beta ...   \n",
       "5   CL:0000897             CD4-positive, alpha-beta memory T cell   \n",
       "6   CL:0000897             CD4-positive, alpha-beta memory T cell   \n",
       "7   CL:0000900  naive thymus-derived CD8-positive, alpha-beta ...   \n",
       "8   CL:0000909             CD8-positive, alpha-beta memory T cell   \n",
       "9   CL:0000909             CD8-positive, alpha-beta memory T cell   \n",
       "10  CL:0000940                           mucosal invariant T cell   \n",
       "11  CL:0002489                          double negative thymocyte   \n",
       "\n",
       "                  p           o        o_label  \n",
       "0   rdfs:subClassOf  CL:0000084         T cell  \n",
       "1   rdfs:subClassOf  CL:0000084         T cell  \n",
       "2   rdfs:subClassOf  CL:0000084         T cell  \n",
       "3   rdfs:subClassOf  CL:0000084         T cell  \n",
       "4   rdfs:subClassOf  CL:0000084         T cell  \n",
       "5   rdfs:subClassOf  CL:0000084         T cell  \n",
       "6   rdfs:subClassOf  CL:0000813  memory T cell  \n",
       "7   rdfs:subClassOf  CL:0000084         T cell  \n",
       "8   rdfs:subClassOf  CL:0000084         T cell  \n",
       "9   rdfs:subClassOf  CL:0000813  memory T cell  \n",
       "10  rdfs:subClassOf  CL:0000084         T cell  \n",
       "11  rdfs:subClassOf  CL:0000084         T cell  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ade.simple_enrichment()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the returned dataframe, terms in the `obs.cell_type` field are listed in columns `s` and `s_label`. Columns `o` an `o_label` show which of these terms classify other terms used in `obs.cell_type`.\n",
    "\n",
    "We can use this get a list of all T cells in the anndata file.\n",
    "\n",
    "(TBA)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Using Slim context for enrichment\n",
    "\n",
    "The cell ontology contains a number of upper slims - sets of high level terms that can be used to summarise annotations with CL terms.  To see which slims are available, we can run:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'name': 'blood_and_immune_upper_slim',\n",
       "  'description': 'a subset of general classes related to blood and the immune system, primarily of hematopoietic origin'},\n",
       " {'name': 'eye_upper_slim',\n",
       "  'description': 'a subset of general classes related to specific cell types in the eye.'},\n",
       " {'name': 'general_cell_types_upper_slim',\n",
       "  'description': 'a subset of general classes of cell types in the cell ontology.'}]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ade.slim_list"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Minimal slim enrichment\n",
    "\n",
    "As the dataset we are working with consists of immune cells, we will choose the blood_and_immune_upper_slim as context. (The method takes a list of slims, allowing them to be combined.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>s</th>\n",
       "      <th>s_label</th>\n",
       "      <th>p</th>\n",
       "      <th>o</th>\n",
       "      <th>o_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CL:0000787</td>\n",
       "      <td>memory B cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000145</td>\n",
       "      <td>professional antigen presenting cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CL:0000787</td>\n",
       "      <td>memory B cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000236</td>\n",
       "      <td>B cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CL:0000787</td>\n",
       "      <td>memory B cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CL:0000788</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>CL:0000788</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000145</td>\n",
       "      <td>professional antigen presenting cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>CL:0000788</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000236</td>\n",
       "      <td>B cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>CL:0000798</td>\n",
       "      <td>gamma-delta T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>CL:0000798</td>\n",
       "      <td>gamma-delta T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>CL:0000809</td>\n",
       "      <td>double-positive, alpha-beta thymocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>CL:0000809</td>\n",
       "      <td>double-positive, alpha-beta thymocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000789</td>\n",
       "      <td>alpha-beta T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>CL:0000809</td>\n",
       "      <td>double-positive, alpha-beta thymocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>CL:0000809</td>\n",
       "      <td>double-positive, alpha-beta thymocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002420</td>\n",
       "      <td>immature T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>CL:0000813</td>\n",
       "      <td>memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>CL:0000813</td>\n",
       "      <td>memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>CL:0000815</td>\n",
       "      <td>regulatory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>CL:0000815</td>\n",
       "      <td>regulatory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>CL:0000895</td>\n",
       "      <td>naive thymus-derived CD4-positive, alpha-beta ...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>CL:0000895</td>\n",
       "      <td>naive thymus-derived CD4-positive, alpha-beta ...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>CL:0000895</td>\n",
       "      <td>naive thymus-derived CD4-positive, alpha-beta ...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000789</td>\n",
       "      <td>alpha-beta T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>CL:0000897</td>\n",
       "      <td>CD4-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>CL:0000897</td>\n",
       "      <td>CD4-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000789</td>\n",
       "      <td>alpha-beta T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>CL:0000897</td>\n",
       "      <td>CD4-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000813</td>\n",
       "      <td>memory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>CL:0000897</td>\n",
       "      <td>CD4-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>CL:0000900</td>\n",
       "      <td>naive thymus-derived CD8-positive, alpha-beta ...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>CL:0000900</td>\n",
       "      <td>naive thymus-derived CD8-positive, alpha-beta ...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000789</td>\n",
       "      <td>alpha-beta T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>CL:0000900</td>\n",
       "      <td>naive thymus-derived CD8-positive, alpha-beta ...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>CL:0000909</td>\n",
       "      <td>CD8-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>CL:0000909</td>\n",
       "      <td>CD8-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000789</td>\n",
       "      <td>alpha-beta T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>CL:0000909</td>\n",
       "      <td>CD8-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>CL:0000909</td>\n",
       "      <td>CD8-positive, alpha-beta memory T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000813</td>\n",
       "      <td>memory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>CL:0000940</td>\n",
       "      <td>mucosal invariant T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000789</td>\n",
       "      <td>alpha-beta T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>CL:0000940</td>\n",
       "      <td>mucosal invariant T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>CL:0000940</td>\n",
       "      <td>mucosal invariant T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>CL:0000980</td>\n",
       "      <td>plasmablast</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000145</td>\n",
       "      <td>professional antigen presenting cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>CL:0000980</td>\n",
       "      <td>plasmablast</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000236</td>\n",
       "      <td>B cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>CL:0000980</td>\n",
       "      <td>plasmablast</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>CL:0002489</td>\n",
       "      <td>double negative thymocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>CL:0002489</td>\n",
       "      <td>double negative thymocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>CL:0002489</td>\n",
       "      <td>double negative thymocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002420</td>\n",
       "      <td>immature T cell</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             s                                            s_label  \\\n",
       "0   CL:0000084                                             T cell   \n",
       "1   CL:0000787                                      memory B cell   \n",
       "2   CL:0000787                                      memory B cell   \n",
       "3   CL:0000787                                      memory B cell   \n",
       "4   CL:0000788                                       naive B cell   \n",
       "5   CL:0000788                                       naive B cell   \n",
       "6   CL:0000788                                       naive B cell   \n",
       "7   CL:0000798                                 gamma-delta T cell   \n",
       "8   CL:0000798                                 gamma-delta T cell   \n",
       "9   CL:0000809              double-positive, alpha-beta thymocyte   \n",
       "10  CL:0000809              double-positive, alpha-beta thymocyte   \n",
       "11  CL:0000809              double-positive, alpha-beta thymocyte   \n",
       "12  CL:0000809              double-positive, alpha-beta thymocyte   \n",
       "13  CL:0000813                                      memory T cell   \n",
       "14  CL:0000813                                      memory T cell   \n",
       "15  CL:0000815                                  regulatory T cell   \n",
       "16  CL:0000815                                  regulatory T cell   \n",
       "17  CL:0000895  naive thymus-derived CD4-positive, alpha-beta ...   \n",
       "18  CL:0000895  naive thymus-derived CD4-positive, alpha-beta ...   \n",
       "19  CL:0000895  naive thymus-derived CD4-positive, alpha-beta ...   \n",
       "20  CL:0000897             CD4-positive, alpha-beta memory T cell   \n",
       "21  CL:0000897             CD4-positive, alpha-beta memory T cell   \n",
       "22  CL:0000897             CD4-positive, alpha-beta memory T cell   \n",
       "23  CL:0000897             CD4-positive, alpha-beta memory T cell   \n",
       "24  CL:0000900  naive thymus-derived CD8-positive, alpha-beta ...   \n",
       "25  CL:0000900  naive thymus-derived CD8-positive, alpha-beta ...   \n",
       "26  CL:0000900  naive thymus-derived CD8-positive, alpha-beta ...   \n",
       "27  CL:0000909             CD8-positive, alpha-beta memory T cell   \n",
       "28  CL:0000909             CD8-positive, alpha-beta memory T cell   \n",
       "29  CL:0000909             CD8-positive, alpha-beta memory T cell   \n",
       "30  CL:0000909             CD8-positive, alpha-beta memory T cell   \n",
       "31  CL:0000940                           mucosal invariant T cell   \n",
       "32  CL:0000940                           mucosal invariant T cell   \n",
       "33  CL:0000940                           mucosal invariant T cell   \n",
       "34  CL:0000980                                        plasmablast   \n",
       "35  CL:0000980                                        plasmablast   \n",
       "36  CL:0000980                                        plasmablast   \n",
       "37  CL:0002489                          double negative thymocyte   \n",
       "38  CL:0002489                          double negative thymocyte   \n",
       "39  CL:0002489                          double negative thymocyte   \n",
       "\n",
       "                  p           o                               o_label  \n",
       "0   rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "1   rdfs:subClassOf  CL:0000145  professional antigen presenting cell  \n",
       "2   rdfs:subClassOf  CL:0000236                                B cell  \n",
       "3   rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "4   rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "5   rdfs:subClassOf  CL:0000145  professional antigen presenting cell  \n",
       "6   rdfs:subClassOf  CL:0000236                                B cell  \n",
       "7   rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "8   rdfs:subClassOf  CL:0000084                                T cell  \n",
       "9   rdfs:subClassOf  CL:0000084                                T cell  \n",
       "10  rdfs:subClassOf  CL:0000789                     alpha-beta T cell  \n",
       "11  rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "12  rdfs:subClassOf  CL:0002420                       immature T cell  \n",
       "13  rdfs:subClassOf  CL:0000084                                T cell  \n",
       "14  rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "15  rdfs:subClassOf  CL:0000084                                T cell  \n",
       "16  rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "17  rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "18  rdfs:subClassOf  CL:0000084                                T cell  \n",
       "19  rdfs:subClassOf  CL:0000789                     alpha-beta T cell  \n",
       "20  rdfs:subClassOf  CL:0000084                                T cell  \n",
       "21  rdfs:subClassOf  CL:0000789                     alpha-beta T cell  \n",
       "22  rdfs:subClassOf  CL:0000813                         memory T cell  \n",
       "23  rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "24  rdfs:subClassOf  CL:0000084                                T cell  \n",
       "25  rdfs:subClassOf  CL:0000789                     alpha-beta T cell  \n",
       "26  rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "27  rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "28  rdfs:subClassOf  CL:0000789                     alpha-beta T cell  \n",
       "29  rdfs:subClassOf  CL:0000084                                T cell  \n",
       "30  rdfs:subClassOf  CL:0000813                         memory T cell  \n",
       "31  rdfs:subClassOf  CL:0000789                     alpha-beta T cell  \n",
       "32  rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "33  rdfs:subClassOf  CL:0000084                                T cell  \n",
       "34  rdfs:subClassOf  CL:0000145  professional antigen presenting cell  \n",
       "35  rdfs:subClassOf  CL:0000236                                B cell  \n",
       "36  rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "37  rdfs:subClassOf  CL:0000084                                T cell  \n",
       "38  rdfs:subClassOf  CL:0000842                      mononuclear cell  \n",
       "39  rdfs:subClassOf  CL:0002420                       immature T cell  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ade.minimal_slim_enrichment([\"blood_and_immune_upper_slim\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `s` column now includes terms from the slim as well as terms used in `obs.cell_type`, so we get mappings up to terms like 'B cell'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Full slim enrichment\n",
    "\n",
    "We can get a more extensive enrichment by using the `full_slim_enrichment` method. In this case we get mappings to all terms on the classification path up to the term in the slim."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "164"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_enrichment =  ade.full_slim_enrichment([\"blood_and_immune_upper_slim\"])\n",
    "len(full_enrichment)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>s</th>\n",
       "      <th>s_label</th>\n",
       "      <th>p</th>\n",
       "      <th>o</th>\n",
       "      <th>o_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000542</td>\n",
       "      <td>lymphocyte</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CL:0000084</td>\n",
       "      <td>T cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000842</td>\n",
       "      <td>mononuclear cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CL:0000787</td>\n",
       "      <td>memory B cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0000145</td>\n",
       "      <td>professional antigen presenting cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CL:0000787</td>\n",
       "      <td>memory B cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0001200</td>\n",
       "      <td>lymphocyte of B lineage, CD19-positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CL:0000787</td>\n",
       "      <td>memory B cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0001201</td>\n",
       "      <td>B cell, CD19-positive</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            s        s_label                p           o  \\\n",
       "0  CL:0000084         T cell  rdfs:subClassOf  CL:0000542   \n",
       "1  CL:0000084         T cell  rdfs:subClassOf  CL:0000842   \n",
       "2  CL:0000787  memory B cell  rdfs:subClassOf  CL:0000145   \n",
       "3  CL:0000787  memory B cell  rdfs:subClassOf  CL:0001200   \n",
       "4  CL:0000787  memory B cell  rdfs:subClassOf  CL:0001201   \n",
       "\n",
       "                                  o_label  \n",
       "0                              lymphocyte  \n",
       "1                        mononuclear cell  \n",
       "2    professional antigen presenting cell  \n",
       "3  lymphocyte of B lineage, CD19-positive  \n",
       "4                   B cell, CD19-positive  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_enrichment[0:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Contextual enrichment\n",
    "\n",
    "The contextual enrichment method uses the anatomical context (obs.tissue) in the anndata file to define s context in CL.\n",
    "\n",
    "To demonstrate this we will use a second example in which the anatomical context (contents of the tissuer field) is kidney."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Using Integrated Single-nucleus and Single-cell RNA-seq of the Adult Human Kidney dataset\n",
    "ade = AnndataEnricher.from_file_path(\"./blue_lake_kidney.h5ad\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ade._AnndataEnricher__context_list  # Future versions will look up the name of the anatomical context term(s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>s</th>\n",
       "      <th>s_label</th>\n",
       "      <th>p</th>\n",
       "      <th>o</th>\n",
       "      <th>o_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002584</td>\n",
       "      <td>renal cortical epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000449</td>\n",
       "      <td>epithelial cell of nephron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000450</td>\n",
       "      <td>epithelial cell of glomerular capsule</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000510</td>\n",
       "      <td>kidney glomerular epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000612</td>\n",
       "      <td>kidney corpuscule cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002681</td>\n",
       "      <td>kidney cortical cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000497</td>\n",
       "      <td>kidney cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002518</td>\n",
       "      <td>kidney epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002584</td>\n",
       "      <td>renal cortical epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000746</td>\n",
       "      <td>glomerular cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002681</td>\n",
       "      <td>kidney cortical cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000746</td>\n",
       "      <td>glomerular cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000612</td>\n",
       "      <td>kidney corpuscule cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000510</td>\n",
       "      <td>kidney glomerular epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000450</td>\n",
       "      <td>epithelial cell of glomerular capsule</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>CL:0002306</td>\n",
       "      <td>epithelial cell of proximal tubule</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000507</td>\n",
       "      <td>kidney tubule cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>CL:0002306</td>\n",
       "      <td>epithelial cell of proximal tubule</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000494</td>\n",
       "      <td>nephron tubule epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>CL:0002306</td>\n",
       "      <td>epithelial cell of proximal tubule</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000449</td>\n",
       "      <td>epithelial cell of nephron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>CL:0002306</td>\n",
       "      <td>epithelial cell of proximal tubule</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002584</td>\n",
       "      <td>renal cortical epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>CL:0002306</td>\n",
       "      <td>epithelial cell of proximal tubule</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002681</td>\n",
       "      <td>kidney cortical cell</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             s                             s_label                p  \\\n",
       "0   CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "1   CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "2   CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "3   CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "4   CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "5   CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "6   CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "7   CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "8   CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "9   CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "10  CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "11  CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "12  CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "13  CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "14  CL:0000653                            podocyte  rdfs:subClassOf   \n",
       "15  CL:0002306  epithelial cell of proximal tubule  rdfs:subClassOf   \n",
       "16  CL:0002306  epithelial cell of proximal tubule  rdfs:subClassOf   \n",
       "17  CL:0002306  epithelial cell of proximal tubule  rdfs:subClassOf   \n",
       "18  CL:0002306  epithelial cell of proximal tubule  rdfs:subClassOf   \n",
       "19  CL:0002306  epithelial cell of proximal tubule  rdfs:subClassOf   \n",
       "\n",
       "             o                                o_label  \n",
       "0   CL:0002584         renal cortical epithelial cell  \n",
       "1   CL:1000449             epithelial cell of nephron  \n",
       "2   CL:1000450  epithelial cell of glomerular capsule  \n",
       "3   CL:1000510      kidney glomerular epithelial cell  \n",
       "4   CL:1000612                 kidney corpuscule cell  \n",
       "5   CL:0002681                   kidney cortical cell  \n",
       "6   CL:1000497                            kidney cell  \n",
       "7   CL:0002518                 kidney epithelial cell  \n",
       "8   CL:0002584         renal cortical epithelial cell  \n",
       "9   CL:1000746                        glomerular cell  \n",
       "10  CL:0002681                   kidney cortical cell  \n",
       "11  CL:1000746                        glomerular cell  \n",
       "12  CL:1000612                 kidney corpuscule cell  \n",
       "13  CL:1000510      kidney glomerular epithelial cell  \n",
       "14  CL:1000450  epithelial cell of glomerular capsule  \n",
       "15  CL:1000507                     kidney tubule cell  \n",
       "16  CL:1000494         nephron tubule epithelial cell  \n",
       "17  CL:1000449             epithelial cell of nephron  \n",
       "18  CL:0002584         renal cortical epithelial cell  \n",
       "19  CL:0002681                   kidney cortical cell  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ade.contextual_slim_enrichment()[0:20]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "contextual_slim_enrichment enriches using all cell type terms from CL that are defined as part_of the anatomical context.  For example, podocyte is classified as an \"epitheliual cell of the glomerular capsule\"."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## AnndataAnalyzer walkthrough"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Initialization\n",
    "Let's import the necessary modules and initialize our AnndataAnalyzer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pandasaurus_cxg.anndata_analyzer import AnndataAnalyzer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# temporarily using a placeholder schema for free text cell types \n",
    "ada = AnndataAnalyzer.from_file_path(\"./immune_example.h5ad\", author_cell_type_list = ['subclass.full', 'subclass.l3', 'subclass.l2', 'subclass.l1', 'class', 'author_cell_type'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Analyzer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Co-annotation report\n",
    "Generates a co-annotation report based on the provided schema.  Each annotation (obs key:value pair) defines a set of cells in the matrix. If the cell sets defined by two annotations match (contain exactly the same cells), we record this as cluster_matches.  If one cell set containts the other, we record this using a subcluster_of relationship.  If the two sets partially overlap (intersect) we record this with a cluster_overlaps relationship."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>field_name1</th>\n",
       "      <th>value1</th>\n",
       "      <th>predicate</th>\n",
       "      <th>field_name2</th>\n",
       "      <th>value2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>naive B cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>memory B cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>memory B cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>gamma-delta T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>gamma-delta T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>plasmablast</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>plasmablast</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>regulatory T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>regulatory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>CD4-positive, alpha-beta memory T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>CD4-positive, alpha-beta memory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>CD8-positive, alpha-beta memory T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>CD8-positive, alpha-beta memory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>naive thymus-derived CD8-positive, alpha-beta ...</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>naive CD8+ T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>naive thymus-derived CD4-positive, alpha-beta ...</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>naive CD4+ T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>mucosal invariant T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>mucosal invariant T cell (MAIT)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>memory T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>TissueResMemT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>double-positive, alpha-beta thymocyte</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>double-positive T cell (DPT)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>double negative thymocyte</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>double negative T cell (DNT)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>cell_type</td>\n",
       "      <td>T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>TCRVbeta13.1pos</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>naive B cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>naive B cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>memory B cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>memory B cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>gamma-delta T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>gamma-delta T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>plasmablast</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>plasmablast</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>regulatory T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>regulatory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>CD4-positive, alpha-beta memory T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>CD4-positive, alpha-beta memory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>CD8-positive, alpha-beta memory T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>CD8-positive, alpha-beta memory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>naive CD8+ T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>naive thymus-derived CD8-positive, alpha-beta ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>naive CD4+ T cell</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>naive thymus-derived CD4-positive, alpha-beta ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>mucosal invariant T cell (MAIT)</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>mucosal invariant T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>TissueResMemT</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>memory T cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>double-positive T cell (DPT)</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>double-positive, alpha-beta thymocyte</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>double negative T cell (DNT)</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>double negative thymocyte</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>author_cell_type</td>\n",
       "      <td>TCRVbeta13.1pos</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>T cell</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         field_name1                                             value1  \\\n",
       "0          cell_type                                       naive B cell   \n",
       "1          cell_type                                      memory B cell   \n",
       "2          cell_type                                 gamma-delta T cell   \n",
       "3          cell_type                                        plasmablast   \n",
       "4          cell_type                                  regulatory T cell   \n",
       "5          cell_type             CD4-positive, alpha-beta memory T cell   \n",
       "6          cell_type             CD8-positive, alpha-beta memory T cell   \n",
       "7          cell_type  naive thymus-derived CD8-positive, alpha-beta ...   \n",
       "8          cell_type  naive thymus-derived CD4-positive, alpha-beta ...   \n",
       "9          cell_type                           mucosal invariant T cell   \n",
       "10         cell_type                                      memory T cell   \n",
       "11         cell_type              double-positive, alpha-beta thymocyte   \n",
       "12         cell_type                          double negative thymocyte   \n",
       "13         cell_type                                             T cell   \n",
       "14  author_cell_type                                       naive B cell   \n",
       "15  author_cell_type                                      memory B cell   \n",
       "16  author_cell_type                                 gamma-delta T cell   \n",
       "17  author_cell_type                                        plasmablast   \n",
       "18  author_cell_type                                  regulatory T cell   \n",
       "19  author_cell_type             CD4-positive, alpha-beta memory T cell   \n",
       "20  author_cell_type             CD8-positive, alpha-beta memory T cell   \n",
       "21  author_cell_type                                  naive CD8+ T cell   \n",
       "22  author_cell_type                                  naive CD4+ T cell   \n",
       "23  author_cell_type                    mucosal invariant T cell (MAIT)   \n",
       "24  author_cell_type                                      TissueResMemT   \n",
       "25  author_cell_type                       double-positive T cell (DPT)   \n",
       "26  author_cell_type                       double negative T cell (DNT)   \n",
       "27  author_cell_type                                    TCRVbeta13.1pos   \n",
       "\n",
       "          predicate       field_name2  \\\n",
       "0   cluster_matches  author_cell_type   \n",
       "1   cluster_matches  author_cell_type   \n",
       "2   cluster_matches  author_cell_type   \n",
       "3   cluster_matches  author_cell_type   \n",
       "4   cluster_matches  author_cell_type   \n",
       "5   cluster_matches  author_cell_type   \n",
       "6   cluster_matches  author_cell_type   \n",
       "7   cluster_matches  author_cell_type   \n",
       "8   cluster_matches  author_cell_type   \n",
       "9   cluster_matches  author_cell_type   \n",
       "10  cluster_matches  author_cell_type   \n",
       "11  cluster_matches  author_cell_type   \n",
       "12  cluster_matches  author_cell_type   \n",
       "13  cluster_matches  author_cell_type   \n",
       "14  cluster_matches         cell_type   \n",
       "15  cluster_matches         cell_type   \n",
       "16  cluster_matches         cell_type   \n",
       "17  cluster_matches         cell_type   \n",
       "18  cluster_matches         cell_type   \n",
       "19  cluster_matches         cell_type   \n",
       "20  cluster_matches         cell_type   \n",
       "21  cluster_matches         cell_type   \n",
       "22  cluster_matches         cell_type   \n",
       "23  cluster_matches         cell_type   \n",
       "24  cluster_matches         cell_type   \n",
       "25  cluster_matches         cell_type   \n",
       "26  cluster_matches         cell_type   \n",
       "27  cluster_matches         cell_type   \n",
       "\n",
       "                                               value2  \n",
       "0                                        naive B cell  \n",
       "1                                       memory B cell  \n",
       "2                                  gamma-delta T cell  \n",
       "3                                         plasmablast  \n",
       "4                                   regulatory T cell  \n",
       "5              CD4-positive, alpha-beta memory T cell  \n",
       "6              CD8-positive, alpha-beta memory T cell  \n",
       "7                                   naive CD8+ T cell  \n",
       "8                                   naive CD4+ T cell  \n",
       "9                     mucosal invariant T cell (MAIT)  \n",
       "10                                      TissueResMemT  \n",
       "11                       double-positive T cell (DPT)  \n",
       "12                       double negative T cell (DNT)  \n",
       "13                                    TCRVbeta13.1pos  \n",
       "14                                       naive B cell  \n",
       "15                                      memory B cell  \n",
       "16                                 gamma-delta T cell  \n",
       "17                                        plasmablast  \n",
       "18                                  regulatory T cell  \n",
       "19             CD4-positive, alpha-beta memory T cell  \n",
       "20             CD8-positive, alpha-beta memory T cell  \n",
       "21  naive thymus-derived CD8-positive, alpha-beta ...  \n",
       "22  naive thymus-derived CD4-positive, alpha-beta ...  \n",
       "23                           mucosal invariant T cell  \n",
       "24                                      memory T cell  \n",
       "25              double-positive, alpha-beta thymocyte  \n",
       "26                          double negative thymocyte  \n",
       "27                                             T cell  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ada.co_annotation_report()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this example, each cell set defined by an 'author_cell_type' value has a matching cell set defined by the cell_type field (which contains a cell ontology term)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Secondary Example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>field_name1</th>\n",
       "      <th>value1</th>\n",
       "      <th>predicate</th>\n",
       "      <th>field_name2</th>\n",
       "      <th>value2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>subclass.l3</td>\n",
       "      <td>dPT</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>subclass.full</td>\n",
       "      <td>Degenerative Proximal Tubule Epithelial Cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>subclass.l3</td>\n",
       "      <td>aPT</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>subclass.full</td>\n",
       "      <td>Adaptive / Maladaptive / Repairing Proximal Tu...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>subclass.l3</td>\n",
       "      <td>M-FIB</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>subclass.full</td>\n",
       "      <td>Medullary Fibroblast</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>subclass.l3</td>\n",
       "      <td>MD</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>subclass.full</td>\n",
       "      <td>Macula Densa Cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>subclass.l3</td>\n",
       "      <td>NKC/T</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>subclass.full</td>\n",
       "      <td>Natural Killer Cell / Natural Killer T Cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1198</th>\n",
       "      <td>subclass.l1</td>\n",
       "      <td>PEC</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>parietal epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1199</th>\n",
       "      <td>subclass.l1</td>\n",
       "      <td>PapE</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>papillary tips cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1200</th>\n",
       "      <td>class</td>\n",
       "      <td>immune cells</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>leukocyte</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1201</th>\n",
       "      <td>class</td>\n",
       "      <td>endothelial cells</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>endothelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1202</th>\n",
       "      <td>class</td>\n",
       "      <td>neural cells</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>neural cell</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1203 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      field_name1             value1        predicate    field_name2  \\\n",
       "0     subclass.l3                dPT  cluster_matches  subclass.full   \n",
       "1     subclass.l3                aPT  cluster_matches  subclass.full   \n",
       "2     subclass.l3              M-FIB  cluster_matches  subclass.full   \n",
       "3     subclass.l3                 MD  cluster_matches  subclass.full   \n",
       "4     subclass.l3              NKC/T  cluster_matches  subclass.full   \n",
       "...           ...                ...              ...            ...   \n",
       "1198  subclass.l1                PEC  cluster_matches      cell_type   \n",
       "1199  subclass.l1               PapE  cluster_matches      cell_type   \n",
       "1200        class       immune cells  cluster_matches      cell_type   \n",
       "1201        class  endothelial cells  cluster_matches      cell_type   \n",
       "1202        class       neural cells  cluster_matches      cell_type   \n",
       "\n",
       "                                                 value2  \n",
       "0          Degenerative Proximal Tubule Epithelial Cell  \n",
       "1     Adaptive / Maladaptive / Repairing Proximal Tu...  \n",
       "2                                  Medullary Fibroblast  \n",
       "3                                     Macula Densa Cell  \n",
       "4           Natural Killer Cell / Natural Killer T Cell  \n",
       "...                                                 ...  \n",
       "1198                           parietal epithelial cell  \n",
       "1199                                papillary tips cell  \n",
       "1200                                          leukocyte  \n",
       "1201                                   endothelial cell  \n",
       "1202                                        neural cell  \n",
       "\n",
       "[1203 rows x 5 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ada = AnndataAnalyzer.from_file_path(\"./blue_lake_kidney.h5ad\", author_cell_type_list = ['subclass.full', 'subclass.l3', 'subclass.l2', 'subclass.l1', 'class', 'author_cell_type'])\n",
    "ada.co_annotation_report()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this case, we have many subcluster_of relationships.  In many cases these define more granular cell type/state annotations than those recorded using a cell ontology term in the cell_type field, e.g.\n",
    "\n",
    "| field_name1\t| value1 |\tpredicate| field_name2\t| value2\n",
    "| -- | -- | -- | -- | -- |\n",
    "| Cycling Natural Killer Cell / Natural Killer T... |\tsubcluster_of |\tcell_type | leukocyte`|"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualising co-annotation and enrichment as a graph\n",
    "\n",
    "The subcluster_of relationships in a co-annotation define a hierarchy from more general set sets (epithelial cells, lymphocyte) to more granular ones (podocyte, T cell).  These in turn nest under the cell type heirarchy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## AnndataEnrichmentAnalyzer Walkthrough"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A wrapper for AnndataEnricher and AnndataAnalyzer."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Initialization\n",
    "Let's import the necessary modules and initialize our AnndataAnalyzer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pandasaurus_cxg.enrichment_analysis import AnndataEnrichmentAnalyzer\n",
    "from pandasaurus_cxg.graph_generator.graph_generator import GraphGenerator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "cell_type_list=['subclass.full', 'subclass.l3', 'subclass.l2', 'subclass.l1', 'class', 'author_cell_type']\n",
    "aea = AnndataEnrichmentAnalyzer(\"test/data/human_kidney.h5ad\", author_cell_type_list=cell_type_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Contextual enrichment\n",
    "\n",
    "The contextual enrichment method uses the anatomical context (obs.tissue) in the anndata file to define s context in CL.\n",
    "\n",
    "To demonstrate this we will use a second example in which the anatomical context (contents of the tissuer field) is kidney."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>s</th>\n",
       "      <th>s_label</th>\n",
       "      <th>p</th>\n",
       "      <th>o</th>\n",
       "      <th>o_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002518</td>\n",
       "      <td>kidney epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000746</td>\n",
       "      <td>glomerular cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000612</td>\n",
       "      <td>kidney corpuscule cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000510</td>\n",
       "      <td>kidney glomerular epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CL:0000653</td>\n",
       "      <td>podocyte</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000450</td>\n",
       "      <td>epithelial cell of glomerular capsule</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132</th>\n",
       "      <td>CL:1001432</td>\n",
       "      <td>kidney collecting duct intercalated cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000454</td>\n",
       "      <td>kidney collecting duct epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>CL:1001432</td>\n",
       "      <td>kidney collecting duct intercalated cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1000497</td>\n",
       "      <td>kidney cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>134</th>\n",
       "      <td>CL:1001432</td>\n",
       "      <td>kidney collecting duct intercalated cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0002518</td>\n",
       "      <td>kidney epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>CL:1001432</td>\n",
       "      <td>kidney collecting duct intercalated cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:1001225</td>\n",
       "      <td>kidney collecting duct cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>CL:1001432</td>\n",
       "      <td>kidney collecting duct intercalated cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>CL:0005010</td>\n",
       "      <td>renal intercalated cell</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>137 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              s                                   s_label                p  \\\n",
       "0    CL:0000653                                  podocyte  rdfs:subClassOf   \n",
       "1    CL:0000653                                  podocyte  rdfs:subClassOf   \n",
       "2    CL:0000653                                  podocyte  rdfs:subClassOf   \n",
       "3    CL:0000653                                  podocyte  rdfs:subClassOf   \n",
       "4    CL:0000653                                  podocyte  rdfs:subClassOf   \n",
       "..          ...                                       ...              ...   \n",
       "132  CL:1001432  kidney collecting duct intercalated cell  rdfs:subClassOf   \n",
       "133  CL:1001432  kidney collecting duct intercalated cell  rdfs:subClassOf   \n",
       "134  CL:1001432  kidney collecting duct intercalated cell  rdfs:subClassOf   \n",
       "135  CL:1001432  kidney collecting duct intercalated cell  rdfs:subClassOf   \n",
       "136  CL:1001432  kidney collecting duct intercalated cell  rdfs:subClassOf   \n",
       "\n",
       "              o                                 o_label  \n",
       "0    CL:0002518                  kidney epithelial cell  \n",
       "1    CL:1000746                         glomerular cell  \n",
       "2    CL:1000612                  kidney corpuscule cell  \n",
       "3    CL:1000510       kidney glomerular epithelial cell  \n",
       "4    CL:1000450   epithelial cell of glomerular capsule  \n",
       "..          ...                                     ...  \n",
       "132  CL:1000454  kidney collecting duct epithelial cell  \n",
       "133  CL:1000497                             kidney cell  \n",
       "134  CL:0002518                  kidney epithelial cell  \n",
       "135  CL:1001225             kidney collecting duct cell  \n",
       "136  CL:0005010                 renal intercalated cell  \n",
       "\n",
       "[137 rows x 5 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aea.contextual_slim_enrichment()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Co-annotation report\n",
    "Generates a co-annotation report based on the provided schema.  Each annotation (obs key:value pair) defines a set of cells in the matrix. If the cell sets defined by two annotations match (contain exactly the same cells), we record this as cluster_matches.  If one cell set containts the other, we record this using a subcluster_of relationship.  If the two sets partially overlap (intersect) we record this with a cluster_overlaps relationship."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>field_name1</th>\n",
       "      <th>value1</th>\n",
       "      <th>predicate</th>\n",
       "      <th>field_name2</th>\n",
       "      <th>value2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>subclass.l3</td>\n",
       "      <td>dPT</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>subclass.full</td>\n",
       "      <td>Degenerative Proximal Tubule Epithelial Cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>subclass.l3</td>\n",
       "      <td>aPT</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>subclass.full</td>\n",
       "      <td>Adaptive / Maladaptive / Repairing Proximal Tu...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>subclass.l3</td>\n",
       "      <td>M-FIB</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>subclass.full</td>\n",
       "      <td>Medullary Fibroblast</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>subclass.l3</td>\n",
       "      <td>MD</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>subclass.full</td>\n",
       "      <td>Macula Densa Cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>subclass.l3</td>\n",
       "      <td>NKC/T</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>subclass.full</td>\n",
       "      <td>Natural Killer Cell / Natural Killer T Cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1198</th>\n",
       "      <td>subclass.l1</td>\n",
       "      <td>PEC</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>parietal epithelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1199</th>\n",
       "      <td>subclass.l1</td>\n",
       "      <td>PapE</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>papillary tips cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1200</th>\n",
       "      <td>class</td>\n",
       "      <td>immune cells</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>leukocyte</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1201</th>\n",
       "      <td>class</td>\n",
       "      <td>endothelial cells</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>endothelial cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1202</th>\n",
       "      <td>class</td>\n",
       "      <td>neural cells</td>\n",
       "      <td>cluster_matches</td>\n",
       "      <td>cell_type</td>\n",
       "      <td>neural cell</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1203 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      field_name1             value1        predicate    field_name2  \\\n",
       "0     subclass.l3                dPT  cluster_matches  subclass.full   \n",
       "1     subclass.l3                aPT  cluster_matches  subclass.full   \n",
       "2     subclass.l3              M-FIB  cluster_matches  subclass.full   \n",
       "3     subclass.l3                 MD  cluster_matches  subclass.full   \n",
       "4     subclass.l3              NKC/T  cluster_matches  subclass.full   \n",
       "...           ...                ...              ...            ...   \n",
       "1198  subclass.l1                PEC  cluster_matches      cell_type   \n",
       "1199  subclass.l1               PapE  cluster_matches      cell_type   \n",
       "1200        class       immune cells  cluster_matches      cell_type   \n",
       "1201        class  endothelial cells  cluster_matches      cell_type   \n",
       "1202        class       neural cells  cluster_matches      cell_type   \n",
       "\n",
       "                                                 value2  \n",
       "0          Degenerative Proximal Tubule Epithelial Cell  \n",
       "1     Adaptive / Maladaptive / Repairing Proximal Tu...  \n",
       "2                                  Medullary Fibroblast  \n",
       "3                                     Macula Densa Cell  \n",
       "4           Natural Killer Cell / Natural Killer T Cell  \n",
       "...                                                 ...  \n",
       "1198                           parietal epithelial cell  \n",
       "1199                                papillary tips cell  \n",
       "1200                                          leukocyte  \n",
       "1201                                   endothelial cell  \n",
       "1202                                        neural cell  \n",
       "\n",
       "[1203 rows x 5 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aea.co_annotation_report()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Graph Generator\n",
    "We'll explore the features of the GraphGenerator class, a powerful tool for creating, enriching, saving, and visualizing RDF graphs. This class is designed to make working with RDF data and Cell Ontology terms intuitive and efficient."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To begin, we'll create an RDF graph using a sample DataFrame containing relevant data. The GraphGenerator class takes care of converting co_annotation DataFrame into RDF triples.\n",
    "We can set label adding priorty to assign a label to each cell set node, and then add labels to the nodes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "gg = GraphGenerator(aea)\n",
    "gg.generate_rdf_graph()\n",
    "gg.set_label_adding_priority([\"class\", \"cell_type\", \"subclass.l1\", \"subclass.l1\", \"subclass.full\", \"subclass.l2\", \"subclass.l3\"])\n",
    "gg.add_label_to_terms()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we'll enrich our RDF graph using Cell Ontology terms from an enrichment DataFrame. The enrich_rdf_graph method of the GraphGenerator class simplifies this process."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "gg.enrich_rdf_graph()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The GraphGenerator class also provides methods to save the RDF graph to a file and visualize it using popular graph visualization tools."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# gg.save_rdf_graph(file_name=\"kidney_new\", _format=\"ttl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "gg.visualize_rdf_graph(\n",
    "    node_selector={\"property\": \"label\", \"value\": \"Cortical Collecting Duct Intercalated Cell Type A\"},\n",
    "    predicate=None,\n",
    "    file_path=\"kidney_new.ttl\",\n",
    ")"
   ]
  },
  {
   "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": "3.9.17"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}