{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "beb3c9fb",
   "metadata": {},
   "source": [
    "# OAK Developers Tutorial\n",
    "\n",
    "This tutorial is primarily for **Python Developers** who wish to use OAK in their applications. These include\n",
    "applications such as:\n",
    "\n",
    "- building ontology-driven data portals\n",
    "- creating curation tools\n",
    "- data science and machine learning applications\n",
    "- web services\n",
    "\n",
    "Some basic knowledge of the overall architecture and capabilities of OAK is assumed.\n",
    "\n",
    "You may want to start with the slides on the command line here: https://doi.org/10.5281/zenodo.7708963\n",
    "\n",
    "Or part 1 of the tutorial here: https://incatools.github.io/ontology-access-kit/intro/tutorial01.html\n",
    "\n",
    "There is a video of the walkthrough of this tutorial: https://www.youtube.com/watch?v=nVTWazO_Gu0\n",
    "\n",
    "\n",
    "## How to follow this tutorial\n",
    "\n",
    "The easiest way to run this tutorial is to clone the repo and run locally:\n",
    "\n",
    "1. clone the repo here https://github.com/INCATools/ontology-access-kit/\n",
    "2. cd ontology-access-kit\n",
    "3. poetry install\n",
    "4. poetry run jupyter notebook\n",
    "\n",
    "Alternatively, everything here should work on a fresh install of oak from pypi. You will need to make sure the test files from [tests/input](https://github.com/INCATools/ontology-access-kit/tree/main/tests/input) are accessible.\n",
    "\n",
    "Some of the examples work with these test files - others will work with versions of ontologies on the web."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c66dd5c",
   "metadata": {},
   "source": [
    "### Change directory so that test files are directly accessible\n",
    "\n",
    "Note: this is necessary if you are running from a checkout of the OAK repo, since this notebook is in a subfolder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "45adb248",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/cjm/repos/ontology-access-kit\n"
     ]
    }
   ],
   "source": [
    "%cd .."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aad26aed",
   "metadata": {},
   "source": [
    "The OAK documentation makes heavy use of some of the unit test files in the [tests/input](https://github.com/INCATools/ontology-access-kit/tree/main/tests/input) folder.\n",
    "\n",
    "This include a little mini test subset of GO, available in different formats for the purposes of testing different adapters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e2438fb1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tests/input/go-nucleus.cx      tests/input/go-nucleus.ofn\r\n",
      "tests/input/go-nucleus.db      tests/input/go-nucleus.owl\r\n",
      "tests/input/go-nucleus.json    tests/input/go-nucleus.owl.ttl\r\n",
      "tests/input/go-nucleus.obo\r\n"
     ]
    }
   ],
   "source": [
    "!ls tests/input/go-nucleus.*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8c0c62a0",
   "metadata": {},
   "source": [
    "If you are an OAK core developer it helps to be aware of these files, as you will likely be writing new unit tests.\n",
    "\n",
    "If one the other hand you just want to use OAK in your own code you don't need to know anything about these except that they are handy for quick testing."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dae5de64",
   "metadata": {},
   "source": [
    "## Running Examples from the OAK sphinx docs\n",
    "\n",
    "The sphinx docs include code examples, these are visible from the `>>>`s\n",
    "\n",
    "For example, in:\n",
    "\n",
    "https://incatools.github.io/ontology-access-kit/packages/interfaces/basic\n",
    "\n",
    "You can see sections like this:\n",
    "\n",
    "![images/oak-docs-code.png](images/oak-docs-code.png)\n",
    "\n",
    "If you click the \"copy\" button it will copy the code only (no `>>>`s, and no output) such that you\n",
    "can paste directly into a Python REPL or a notebook (provided the paths are preserved).\n",
    "\n",
    "E.g. try copying this section from the docs\n",
    "\n",
    "```python\n",
    ">>> from oaklib import get_adapter\n",
    ">>> adapter = get_adapter('tests/input/go-nucleus.db')\n",
    ">>> print(adapter.label(\"GO:0005634\"))\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2ec55470",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "nucleus\n"
     ]
    }
   ],
   "source": [
    "from oaklib import get_adapter\n",
    "adapter = get_adapter('tests/input/go-nucleus.db')\n",
    "print(adapter.label(\"GO:0005634\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1fa26abb",
   "metadata": {},
   "source": [
    "Hurray! All examples throughout the OAK docs should work"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1733bae5",
   "metadata": {},
   "source": [
    "Note that you can play with using different input formats - this should give the same results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "34e9a5a1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "nucleus\n"
     ]
    }
   ],
   "source": [
    "from oaklib import get_adapter\n",
    "adapter = get_adapter('tests/input/go-nucleus.obo')\n",
    "print(adapter.label(\"GO:0005634\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b85e1771",
   "metadata": {},
   "source": [
    "## Working with whole ontologies\n",
    "\n",
    "Almost all the examples in this tutorial make use of pre-made sqlite versions of ontologies.\n",
    "\n",
    "These are specified using [selector](https://incatools.github.io/ontology-access-kit/packages/selectors.html) syntax:\n",
    "\n",
    "```\n",
    "sqlite:obo:ONTID\n",
    "```\n",
    "\n",
    "E.g.\n",
    "\n",
    "```\n",
    "sqlite:obo:cl\n",
    "```\n",
    "\n",
    "When you use this for the first time it will download and cache the file (using pystow), so there may be an initial lag"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1f324799",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "neuron\n"
     ]
    }
   ],
   "source": [
    "from oaklib import get_adapter\n",
    "adapter = get_adapter(\"sqlite:obo:cl\")\n",
    "print(adapter.label(\"CL:0000540\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a53f8987",
   "metadata": {},
   "source": [
    "Hurray! We successfully fetched the label (name) for a class ID in the cell ontology!\n",
    "\n",
    "Note that the `label` method is part of the BasicOntologyInterface"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd705733",
   "metadata": {},
   "source": [
    "### BasicOntologyInterface\n",
    "\n",
    "The [BasicOntologyInterface](https://incatools.github.io/ontology-access-kit/packages/interfaces/basic) provides basic methods that encompass the majority of what most people need to do when\n",
    "working with ontologies - lookups of various kinds as well as simple graph operations.\n",
    "\n",
    "OAK has the architectural concept of separating interfaces from implementations. It helps to read about this concept, but for now you don't need to worry about it. We are using the sql adapter which fully *implements* almost all the existing OAK interfaces"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e84e5425",
   "metadata": {},
   "source": [
    "### Fetching ancestors\n",
    "\n",
    "Next we are going to fetch ancestors. \n",
    "\n",
    "Note: it would help to review [tutorial part 1](https://incatools.github.io/ontology-access-kit/intro/tutorial01.html) to understand basic concepts of edges, ancestors, and predicates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "67273969",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CL:0000540 neuron\n",
      "BFO:0000002 continuant\n",
      "BFO:0000004 independent continuant\n",
      "BFO:0000040 material entity\n",
      "CL:0000000 cell\n",
      "CL:0000003 native cell\n",
      "CL:0000211 electrically active cell\n",
      "CL:0000255 eukaryotic cell\n",
      "CL:0000393 electrically responsive cell\n",
      "CL:0000404 electrically signaling cell\n",
      "CL:0000540 neuron\n",
      "CL:0000548 animal cell\n",
      "CL:0002319 neural cell\n",
      "CL:0002371 somatic cell\n"
     ]
    }
   ],
   "source": [
    "from oaklib.datamodels.vocabulary import IS_A\n",
    "for anc in adapter.ancestors(\"CL:0000540\", predicates=[IS_A]):\n",
    "    print(anc, adapter.label(anc))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91474cd7",
   "metadata": {},
   "source": [
    "### Fetching descendants\n",
    "\n",
    "Let's try working with descendants.\n",
    "\n",
    "This time we are going to demonstrate how OAK deals with collections.\n",
    "\n",
    "__note__ _we expect the following code **not** to work_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "954b22b5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PROBLEM: object of type 'generator' has no len()\n"
     ]
    }
   ],
   "source": [
    "neurons = adapter.descendants(\"CL:0000540\", predicates=[IS_A])\n",
    "try:\n",
    "    print(len(neurons))\n",
    "except(Exception) as e:\n",
    "    print(f\"PROBLEM: {e}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "265d0cd9",
   "metadata": {},
   "source": [
    "Why didn't this work? What does `object of type 'generator' has no len()` mean?\n",
    "\n",
    "To understand why we will mention a key concept in OAK, that of the iterator"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae574c65",
   "metadata": {},
   "source": [
    "### Iterators\n",
    "\n",
    "OAK methods rarely return lists - instead they return iterators. This means that code is better adaptable to use cases\n",
    "where you want to work with potentially very large lists or you want to *stream* results. See:\n",
    "\n",
    "See [best practice](https://incatools.github.io/ontology-access-kit/packages/best-practice.html#iterators)\n",
    "\n",
    "However, if you don't care about this you can simple use `list(...)` to get a list:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "4239552a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "454\n"
     ]
    }
   ],
   "source": [
    "neurons = list(adapter.descendants(\"CL:0000540\", predicates=[IS_A]))\n",
    "print(len(neurons))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "818c83f9",
   "metadata": {},
   "source": [
    "You can also cast to `set()` to use set operations like intersections.\n",
    "\n",
    "For example, let's say we want to compose our neuron query above with a query to fetch all things in the forebrain,\n",
    "to get all neurons in the forebrain:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "695eef7c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "219\n"
     ]
    }
   ],
   "source": [
    "from oaklib.datamodels.vocabulary import IS_A, PART_OF\n",
    "parts_of_forebrain = set(adapter.descendants(\"UBERON:0001890\", predicates=[IS_A, PART_OF]))\n",
    "print(len(parts_of_forebrain))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "942af1e8",
   "metadata": {},
   "source": [
    "You may be wondering what Uberon terms are doing here given that we requested the cell ontology in `get_adapter`\n",
    "\n",
    "One under-appreciated fact of OBO is that many ontologies are in fact mini \"knowledge graphs\", linking out to nodes in other ontologies. See \n",
    "\n",
    " - [extracting using robot](https://oboacademy.github.io/obook/tutorial/robot-tutorial-1/) (OBO Academy)\n",
    " - [owl format variants](https://oboacademy.github.io/obook/explanation/owl-format-variants/) (OBO Academy)\n",
    " - [OAK basics](https://incatools.github.io/ontology-access-kit/guide/basics.html) (OAK Guide)\n",
    " \n",
    " \n",
    " OK next lets do an **intersection** between the two lists"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "96a1a814",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CL:1001435 periglomerular cell\n",
      "CL:4023040 L2/3-6 intratelencephalic projecting glutamatergic cortical neuron\n",
      "CL:1001571 hippocampal pyramidal neuron\n",
      "CL:1001502 mitral cell\n",
      "CL:1001434 olfactory bulb interneuron\n",
      "CL:4023048 L4/5 intratelencephalic projecting glutamatergic neuron of the primary motor cortex\n",
      "CL:1001505 parvocellular neurosecretory cell\n",
      "CL:4023008 intratelencephalic-projecting glutamatergic cortical neuron\n",
      "CL:4023049 L5 intratelencephalic projecting glutamatergic neuron of the primary motor cortex\n",
      "CL:4023047 L2/3 intratelencephalic projecting glutamatergic neuron of the primary motor cortex\n",
      "CL:4023081 inverted L6 intratelencephalic projecting glutamatergic neuron of the primary motor cortex (Mmus)\n",
      "CL:4023050 L6 intratelencephalic projecting glutamatergic neuron of the primary motor cortex\n",
      "CL:1001503 olfactory bulb tufted cell\n",
      "CL:4023080 stellate L6 intratelencephalic projecting glutamatergic neuron of the primary motor cortex (Mmus)\n"
     ]
    }
   ],
   "source": [
    "for cell in parts_of_forebrain.intersection(neurons):\n",
    "    print(cell, adapter.label(cell))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "562170d3",
   "metadata": {},
   "source": [
    "This is **all the neurons that are part of the forebrain**.\n",
    "\n",
    "Readers familiar with OWL and Protege may like to think of this as similar to a DL query for the expression `neuron and part-of some forebrain` -- there are some theoretical differences we won't get into here but for practical purposes the results should be the same."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "615f4deb",
   "metadata": {},
   "source": [
    "### Relationships\n",
    "\n",
    "The above example uses the ancestors and descendants query in [BasicOntologyInterface](https://incatools.github.io/ontology-access-kit/packages/interfaces/basic).\n",
    "\n",
    "We can also get the relationships using the `relationships` method:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "b61c3832",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('CL:0000540', 'RO:0002215', 'GO:0019226')\n",
      "('CL:0000540', 'rdfs:subClassOf', 'BFO:0000040')\n",
      "('CL:0000540', 'rdfs:subClassOf', 'CL:0000393')\n",
      "('CL:0000540', 'rdfs:subClassOf', 'CL:0000404')\n",
      "('CL:0000540', 'rdfs:subClassOf', 'CL:0002319')\n"
     ]
    }
   ],
   "source": [
    "for rel in adapter.relationships([\"CL:0000540\"]):\n",
    "    print(rel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "eb1e98a2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " RO:0002215 capable of GO:0019226 transmission of nerve impulse\n",
      " rdfs:subClassOf None BFO:0000040 material entity\n",
      " rdfs:subClassOf None CL:0000393 electrically responsive cell\n",
      " rdfs:subClassOf None CL:0000404 electrically signaling cell\n",
      " rdfs:subClassOf None CL:0002319 neural cell\n"
     ]
    }
   ],
   "source": [
    "for _s, p, o in adapter.relationships([\"CL:0000540\"]):\n",
    "    print(f\" {p} {adapter.label(p)} {o} {adapter.label(o)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "e13fd38c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " rdfs:subClassOf FROM: CL:0000028 CNS neuron (sensu Nematoda and Protostomia)\n",
      " rdfs:subClassOf FROM: CL:0000029 neural crest derived neuron\n",
      " rdfs:subClassOf FROM: CL:0000099 interneuron\n",
      " rdfs:subClassOf FROM: CL:0000102 polymodal neuron\n",
      " rdfs:subClassOf FROM: CL:0000104 multipolar neuron\n",
      " rdfs:subClassOf FROM: CL:0000105 pseudounipolar neuron\n",
      " rdfs:subClassOf FROM: CL:0000106 unipolar neuron\n",
      " rdfs:subClassOf FROM: CL:0000108 cholinergic neuron\n",
      " rdfs:subClassOf FROM: CL:0000109 adrenergic neuron\n",
      " rdfs:subClassOf FROM: CL:0000110 peptidergic neuron\n"
     ]
    }
   ],
   "source": [
    "for s, p, _o in list(adapter.relationships(objects=[\"CL:0000540\"]))[0:10]:\n",
    "    print(f\" {p} FROM: {s} {adapter.label(s)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "049cfd10",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BFO:0000002 continuant\n",
      "BFO:0000004 independent continuant\n",
      "BFO:0000040 material entity\n",
      "CARO:0000000 anatomical entity\n",
      "CARO:0000006 material anatomical entity\n",
      "CARO:0030000 biological entity\n",
      "UBERON:0000061 anatomical structure\n",
      "UBERON:0000465 material anatomical entity\n",
      "UBERON:0000467 anatomical system\n",
      "UBERON:0000468 multicellular organism\n",
      "UBERON:0001016 nervous system\n",
      "UBERON:0001062 anatomical entity\n",
      "UBERON:0010000 multicellular anatomical structure\n"
     ]
    }
   ],
   "source": [
    "for _s, _p, o in adapter.relationships([\"CL:0000540\"], predicates=[PART_OF], include_entailed=True):\n",
    "    print(o, adapter.label(o))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d478e8fb",
   "metadata": {},
   "source": [
    "This includes all the entailed part-of relationships from neuron, including trivial ones (\"every neuron is part of a material entity\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "b52e0519",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CL:0000705 R6 photoreceptor cell\n",
      "CL:4023108 oxytocin-secreting magnocellular cell\n",
      "CL:0004240 WF1 amacrine cell\n",
      "CL:0004242 WF3-1 amacrine cell\n",
      "CL:1000380 type 1 vestibular sensory cell of epithelium of macula of saccule of membranous labyrinth\n",
      "CL:1001582 lateral ventricle neuron\n",
      "CL:4023128 rostral periventricular region of the third ventricle KDNy neuron\n",
      "CL:0003020 retinal ganglion cell C outer\n",
      "CL:4023094 tufted pyramidal neuron\n",
      "CL:4023057 cerebellar inhibitory GABAergic interneuron\n"
     ]
    }
   ],
   "source": [
    "for s, _p, o in list(adapter.relationships(objects=[\"CL:0000540\"], predicates=[IS_A], include_entailed=True))[0:10]:\n",
    "    print(s, adapter.label(s))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d07c30cc",
   "metadata": {},
   "source": [
    "### Creating a Data Frame for Relationships\n",
    "\n",
    "Next we will see how to create a small data frame for relationships for forebrain neurons:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f3afbee7",
   "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>p_label</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:1001434</td>\n",
       "      <td>olfactory bulb interneuron</td>\n",
       "      <td>BFO:0000050</td>\n",
       "      <td>part of</td>\n",
       "      <td>UBERON:0002264</td>\n",
       "      <td>olfactory bulb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CL:1001434</td>\n",
       "      <td>olfactory bulb interneuron</td>\n",
       "      <td>RO:0002100</td>\n",
       "      <td>has soma location</td>\n",
       "      <td>UBERON:0002264</td>\n",
       "      <td>olfactory bulb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CL:1001434</td>\n",
       "      <td>olfactory bulb interneuron</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:0000101</td>\n",
       "      <td>sensory neuron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CL:1001434</td>\n",
       "      <td>olfactory bulb interneuron</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:0000402</td>\n",
       "      <td>CNS interneuron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CL:1001434</td>\n",
       "      <td>olfactory bulb interneuron</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:0012001</td>\n",
       "      <td>neuron of the forebrain</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>CL:1001435</td>\n",
       "      <td>periglomerular cell</td>\n",
       "      <td>RO:0002100</td>\n",
       "      <td>has soma location</td>\n",
       "      <td>UBERON:0005377</td>\n",
       "      <td>olfactory bulb glomerular layer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>CL:1001435</td>\n",
       "      <td>periglomerular cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:1001434</td>\n",
       "      <td>olfactory bulb interneuron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>CL:1001502</td>\n",
       "      <td>mitral cell</td>\n",
       "      <td>RO:0002100</td>\n",
       "      <td>has soma location</td>\n",
       "      <td>UBERON:0004186</td>\n",
       "      <td>olfactory bulb mitral cell layer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>CL:1001502</td>\n",
       "      <td>mitral cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:1001434</td>\n",
       "      <td>olfactory bulb interneuron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>CL:1001503</td>\n",
       "      <td>olfactory bulb tufted cell</td>\n",
       "      <td>BFO:0000050</td>\n",
       "      <td>part of</td>\n",
       "      <td>UBERON:0005376</td>\n",
       "      <td>olfactory bulb external plexiform layer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>CL:1001503</td>\n",
       "      <td>olfactory bulb tufted cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CARO:0000000</td>\n",
       "      <td>anatomical entity</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>CL:1001503</td>\n",
       "      <td>olfactory bulb tufted cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:0000540</td>\n",
       "      <td>neuron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>CL:1001505</td>\n",
       "      <td>parvocellular neurosecretory cell</td>\n",
       "      <td>BFO:0000050</td>\n",
       "      <td>part of</td>\n",
       "      <td>UBERON:0001930</td>\n",
       "      <td>paraventricular nucleus of hypothalamus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>CL:1001505</td>\n",
       "      <td>parvocellular neurosecretory cell</td>\n",
       "      <td>RO:0002215</td>\n",
       "      <td>capable of</td>\n",
       "      <td>GO:0030103</td>\n",
       "      <td>vasopressin secretion</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>CL:1001505</td>\n",
       "      <td>parvocellular neurosecretory cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CARO:0000000</td>\n",
       "      <td>anatomical entity</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>CL:1001505</td>\n",
       "      <td>parvocellular neurosecretory cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:0000167</td>\n",
       "      <td>peptide hormone secreting cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>CL:1001505</td>\n",
       "      <td>parvocellular neurosecretory cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:0000381</td>\n",
       "      <td>neurosecretory neuron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>CL:1001505</td>\n",
       "      <td>parvocellular neurosecretory cell</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:2000030</td>\n",
       "      <td>hypothalamus cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>CL:1001571</td>\n",
       "      <td>hippocampal pyramidal neuron</td>\n",
       "      <td>BFO:0000050</td>\n",
       "      <td>part of</td>\n",
       "      <td>UBERON:0002313</td>\n",
       "      <td>hippocampus pyramidal layer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>CL:1001571</td>\n",
       "      <td>hippocampal pyramidal neuron</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:0002608</td>\n",
       "      <td>hippocampal neuron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>CL:1001571</td>\n",
       "      <td>hippocampal pyramidal neuron</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:4023111</td>\n",
       "      <td>cerebral cortex pyramidal neuron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>CL:4023008</td>\n",
       "      <td>intratelencephalic-projecting glutamatergic co...</td>\n",
       "      <td>RO:0000053</td>\n",
       "      <td>bearer of</td>\n",
       "      <td>PATO:0070034</td>\n",
       "      <td>intratelencephalic projecting</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>CL:4023008</td>\n",
       "      <td>intratelencephalic-projecting glutamatergic co...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:0000679</td>\n",
       "      <td>glutamatergic neuron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>CL:4023008</td>\n",
       "      <td>intratelencephalic-projecting glutamatergic co...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:0010012</td>\n",
       "      <td>cerebral cortex neuron</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>CL:4023040</td>\n",
       "      <td>L2/3-6 intratelencephalic projecting glutamate...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:4023008</td>\n",
       "      <td>intratelencephalic-projecting glutamatergic co...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>CL:4023047</td>\n",
       "      <td>L2/3 intratelencephalic projecting glutamaterg...</td>\n",
       "      <td>RO:0002100</td>\n",
       "      <td>has soma location</td>\n",
       "      <td>UBERON:0001384</td>\n",
       "      <td>primary motor cortex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>CL:4023047</td>\n",
       "      <td>L2/3 intratelencephalic projecting glutamaterg...</td>\n",
       "      <td>RO:0002100</td>\n",
       "      <td>has soma location</td>\n",
       "      <td>UBERON:8440000</td>\n",
       "      <td>cortical layer II/III</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>CL:4023047</td>\n",
       "      <td>L2/3 intratelencephalic projecting glutamaterg...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:4023040</td>\n",
       "      <td>L2/3-6 intratelencephalic projecting glutamate...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>CL:4023048</td>\n",
       "      <td>L4/5 intratelencephalic projecting glutamaterg...</td>\n",
       "      <td>RO:0002100</td>\n",
       "      <td>has soma location</td>\n",
       "      <td>UBERON:0001384</td>\n",
       "      <td>primary motor cortex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>CL:4023048</td>\n",
       "      <td>L4/5 intratelencephalic projecting glutamaterg...</td>\n",
       "      <td>RO:0002100</td>\n",
       "      <td>has soma location</td>\n",
       "      <td>UBERON:8440001</td>\n",
       "      <td>cortical layer IV/V</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>CL:4023048</td>\n",
       "      <td>L4/5 intratelencephalic projecting glutamaterg...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:4023040</td>\n",
       "      <td>L2/3-6 intratelencephalic projecting glutamate...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>CL:4023049</td>\n",
       "      <td>L5 intratelencephalic projecting glutamatergic...</td>\n",
       "      <td>RO:0002100</td>\n",
       "      <td>has soma location</td>\n",
       "      <td>UBERON:0001384</td>\n",
       "      <td>primary motor cortex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>CL:4023049</td>\n",
       "      <td>L5 intratelencephalic projecting glutamatergic...</td>\n",
       "      <td>RO:0002100</td>\n",
       "      <td>has soma location</td>\n",
       "      <td>UBERON:0005394</td>\n",
       "      <td>cortical layer V</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>CL:4023049</td>\n",
       "      <td>L5 intratelencephalic projecting glutamatergic...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:4023040</td>\n",
       "      <td>L2/3-6 intratelencephalic projecting glutamate...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>CL:4023050</td>\n",
       "      <td>L6 intratelencephalic projecting glutamatergic...</td>\n",
       "      <td>RO:0000053</td>\n",
       "      <td>bearer of</td>\n",
       "      <td>PATO:0070019</td>\n",
       "      <td>untufted pyramidal morphology</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>CL:4023050</td>\n",
       "      <td>L6 intratelencephalic projecting glutamatergic...</td>\n",
       "      <td>RO:0002100</td>\n",
       "      <td>has soma location</td>\n",
       "      <td>UBERON:0005395</td>\n",
       "      <td>cortical layer VI</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>CL:4023050</td>\n",
       "      <td>L6 intratelencephalic projecting glutamatergic...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:2000049</td>\n",
       "      <td>primary motor cortex pyramidal cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>CL:4023050</td>\n",
       "      <td>L6 intratelencephalic projecting glutamatergic...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:4023040</td>\n",
       "      <td>L2/3-6 intratelencephalic projecting glutamate...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>CL:4023080</td>\n",
       "      <td>stellate L6 intratelencephalic projecting glut...</td>\n",
       "      <td>RO:0000053</td>\n",
       "      <td>bearer of</td>\n",
       "      <td>PATO:0070020</td>\n",
       "      <td>stellate pyramidal morphology</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>CL:4023080</td>\n",
       "      <td>stellate L6 intratelencephalic projecting glut...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:4023050</td>\n",
       "      <td>L6 intratelencephalic projecting glutamatergic...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>CL:4023081</td>\n",
       "      <td>inverted L6 intratelencephalic projecting glut...</td>\n",
       "      <td>RO:0000053</td>\n",
       "      <td>bearer of</td>\n",
       "      <td>PATO:0070021</td>\n",
       "      <td>inverted pyramidal morphology</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>CL:4023081</td>\n",
       "      <td>inverted L6 intratelencephalic projecting glut...</td>\n",
       "      <td>rdfs:subClassOf</td>\n",
       "      <td>None</td>\n",
       "      <td>CL:4023050</td>\n",
       "      <td>L6 intratelencephalic projecting glutamatergic...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             s                                            s_label  \\\n",
       "0   CL:1001434                         olfactory bulb interneuron   \n",
       "1   CL:1001434                         olfactory bulb interneuron   \n",
       "2   CL:1001434                         olfactory bulb interneuron   \n",
       "3   CL:1001434                         olfactory bulb interneuron   \n",
       "4   CL:1001434                         olfactory bulb interneuron   \n",
       "5   CL:1001435                                periglomerular cell   \n",
       "6   CL:1001435                                periglomerular cell   \n",
       "7   CL:1001502                                        mitral cell   \n",
       "8   CL:1001502                                        mitral cell   \n",
       "9   CL:1001503                         olfactory bulb tufted cell   \n",
       "10  CL:1001503                         olfactory bulb tufted cell   \n",
       "11  CL:1001503                         olfactory bulb tufted cell   \n",
       "12  CL:1001505                  parvocellular neurosecretory cell   \n",
       "13  CL:1001505                  parvocellular neurosecretory cell   \n",
       "14  CL:1001505                  parvocellular neurosecretory cell   \n",
       "15  CL:1001505                  parvocellular neurosecretory cell   \n",
       "16  CL:1001505                  parvocellular neurosecretory cell   \n",
       "17  CL:1001505                  parvocellular neurosecretory cell   \n",
       "18  CL:1001571                       hippocampal pyramidal neuron   \n",
       "19  CL:1001571                       hippocampal pyramidal neuron   \n",
       "20  CL:1001571                       hippocampal pyramidal neuron   \n",
       "21  CL:4023008  intratelencephalic-projecting glutamatergic co...   \n",
       "22  CL:4023008  intratelencephalic-projecting glutamatergic co...   \n",
       "23  CL:4023008  intratelencephalic-projecting glutamatergic co...   \n",
       "24  CL:4023040  L2/3-6 intratelencephalic projecting glutamate...   \n",
       "25  CL:4023047  L2/3 intratelencephalic projecting glutamaterg...   \n",
       "26  CL:4023047  L2/3 intratelencephalic projecting glutamaterg...   \n",
       "27  CL:4023047  L2/3 intratelencephalic projecting glutamaterg...   \n",
       "28  CL:4023048  L4/5 intratelencephalic projecting glutamaterg...   \n",
       "29  CL:4023048  L4/5 intratelencephalic projecting glutamaterg...   \n",
       "30  CL:4023048  L4/5 intratelencephalic projecting glutamaterg...   \n",
       "31  CL:4023049  L5 intratelencephalic projecting glutamatergic...   \n",
       "32  CL:4023049  L5 intratelencephalic projecting glutamatergic...   \n",
       "33  CL:4023049  L5 intratelencephalic projecting glutamatergic...   \n",
       "34  CL:4023050  L6 intratelencephalic projecting glutamatergic...   \n",
       "35  CL:4023050  L6 intratelencephalic projecting glutamatergic...   \n",
       "36  CL:4023050  L6 intratelencephalic projecting glutamatergic...   \n",
       "37  CL:4023050  L6 intratelencephalic projecting glutamatergic...   \n",
       "38  CL:4023080  stellate L6 intratelencephalic projecting glut...   \n",
       "39  CL:4023080  stellate L6 intratelencephalic projecting glut...   \n",
       "40  CL:4023081  inverted L6 intratelencephalic projecting glut...   \n",
       "41  CL:4023081  inverted L6 intratelencephalic projecting glut...   \n",
       "\n",
       "                  p            p_label               o  \\\n",
       "0       BFO:0000050            part of  UBERON:0002264   \n",
       "1        RO:0002100  has soma location  UBERON:0002264   \n",
       "2   rdfs:subClassOf               None      CL:0000101   \n",
       "3   rdfs:subClassOf               None      CL:0000402   \n",
       "4   rdfs:subClassOf               None      CL:0012001   \n",
       "5        RO:0002100  has soma location  UBERON:0005377   \n",
       "6   rdfs:subClassOf               None      CL:1001434   \n",
       "7        RO:0002100  has soma location  UBERON:0004186   \n",
       "8   rdfs:subClassOf               None      CL:1001434   \n",
       "9       BFO:0000050            part of  UBERON:0005376   \n",
       "10  rdfs:subClassOf               None    CARO:0000000   \n",
       "11  rdfs:subClassOf               None      CL:0000540   \n",
       "12      BFO:0000050            part of  UBERON:0001930   \n",
       "13       RO:0002215         capable of      GO:0030103   \n",
       "14  rdfs:subClassOf               None    CARO:0000000   \n",
       "15  rdfs:subClassOf               None      CL:0000167   \n",
       "16  rdfs:subClassOf               None      CL:0000381   \n",
       "17  rdfs:subClassOf               None      CL:2000030   \n",
       "18      BFO:0000050            part of  UBERON:0002313   \n",
       "19  rdfs:subClassOf               None      CL:0002608   \n",
       "20  rdfs:subClassOf               None      CL:4023111   \n",
       "21       RO:0000053          bearer of    PATO:0070034   \n",
       "22  rdfs:subClassOf               None      CL:0000679   \n",
       "23  rdfs:subClassOf               None      CL:0010012   \n",
       "24  rdfs:subClassOf               None      CL:4023008   \n",
       "25       RO:0002100  has soma location  UBERON:0001384   \n",
       "26       RO:0002100  has soma location  UBERON:8440000   \n",
       "27  rdfs:subClassOf               None      CL:4023040   \n",
       "28       RO:0002100  has soma location  UBERON:0001384   \n",
       "29       RO:0002100  has soma location  UBERON:8440001   \n",
       "30  rdfs:subClassOf               None      CL:4023040   \n",
       "31       RO:0002100  has soma location  UBERON:0001384   \n",
       "32       RO:0002100  has soma location  UBERON:0005394   \n",
       "33  rdfs:subClassOf               None      CL:4023040   \n",
       "34       RO:0000053          bearer of    PATO:0070019   \n",
       "35       RO:0002100  has soma location  UBERON:0005395   \n",
       "36  rdfs:subClassOf               None      CL:2000049   \n",
       "37  rdfs:subClassOf               None      CL:4023040   \n",
       "38       RO:0000053          bearer of    PATO:0070020   \n",
       "39  rdfs:subClassOf               None      CL:4023050   \n",
       "40       RO:0000053          bearer of    PATO:0070021   \n",
       "41  rdfs:subClassOf               None      CL:4023050   \n",
       "\n",
       "                                              o_label  \n",
       "0                                      olfactory bulb  \n",
       "1                                      olfactory bulb  \n",
       "2                                      sensory neuron  \n",
       "3                                     CNS interneuron  \n",
       "4                             neuron of the forebrain  \n",
       "5                     olfactory bulb glomerular layer  \n",
       "6                          olfactory bulb interneuron  \n",
       "7                    olfactory bulb mitral cell layer  \n",
       "8                          olfactory bulb interneuron  \n",
       "9             olfactory bulb external plexiform layer  \n",
       "10                                  anatomical entity  \n",
       "11                                             neuron  \n",
       "12            paraventricular nucleus of hypothalamus  \n",
       "13                              vasopressin secretion  \n",
       "14                                  anatomical entity  \n",
       "15                     peptide hormone secreting cell  \n",
       "16                              neurosecretory neuron  \n",
       "17                                  hypothalamus cell  \n",
       "18                        hippocampus pyramidal layer  \n",
       "19                                 hippocampal neuron  \n",
       "20                   cerebral cortex pyramidal neuron  \n",
       "21                      intratelencephalic projecting  \n",
       "22                               glutamatergic neuron  \n",
       "23                             cerebral cortex neuron  \n",
       "24  intratelencephalic-projecting glutamatergic co...  \n",
       "25                               primary motor cortex  \n",
       "26                              cortical layer II/III  \n",
       "27  L2/3-6 intratelencephalic projecting glutamate...  \n",
       "28                               primary motor cortex  \n",
       "29                                cortical layer IV/V  \n",
       "30  L2/3-6 intratelencephalic projecting glutamate...  \n",
       "31                               primary motor cortex  \n",
       "32                                   cortical layer V  \n",
       "33  L2/3-6 intratelencephalic projecting glutamate...  \n",
       "34                      untufted pyramidal morphology  \n",
       "35                                  cortical layer VI  \n",
       "36                primary motor cortex pyramidal cell  \n",
       "37  L2/3-6 intratelencephalic projecting glutamate...  \n",
       "38                      stellate pyramidal morphology  \n",
       "39  L6 intratelencephalic projecting glutamatergic...  \n",
       "40                      inverted pyramidal morphology  \n",
       "41  L6 intratelencephalic projecting glutamatergic...  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "forebrain_neurons = parts_of_forebrain.intersection(neurons)\n",
    "\n",
    "objs = []\n",
    "for s, p, o in adapter.relationships(forebrain_neurons):\n",
    "    objs.append({\"s\": s, \"s_label\": adapter.label(s), \n",
    "                 \"p\": p, \"p_label\": adapter.label(p),\n",
    "                 \"o\": o, \"o_label\": adapter.label(o)})\n",
    "\n",
    "df = pd.DataFrame(objs)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7802cfcc",
   "metadata": {},
   "source": [
    "### Aliases\n",
    "\n",
    "The BasicOntologyInterface has a deliberately simple datamodel for aliases that can be expressed by returning\n",
    "simple strings and tuples. Later on we can see how to leverage the more advanced OBO Graphs data model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "567b4685",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['nerve cell', 'neuron']"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adapter.entity_aliases(\"CL:0000540\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "75ad933b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rdfs:label neuron\n",
      "oio:hasExactSynonym nerve cell\n"
     ]
    }
   ],
   "source": [
    "for pred, alias in adapter.alias_relationships(\"CL:0000540\"):\n",
    "    print(pred, alias)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cce14612",
   "metadata": {},
   "source": [
    "### Mappings\n",
    "\n",
    "Similar to aliases, the BasicOntologyInterface has a very simple model of mappings. Later on we will see how we\n",
    "can use the MappingProviderInterface to get more granular information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "a3af89a5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "oio:hasDbXref FMA:62364\n"
     ]
    }
   ],
   "source": [
    "for pred, xref in adapter.simple_mappings_by_curie(\"CL:0000202\"):\n",
    "    print(pred, xref)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "236fd724",
   "metadata": {},
   "source": [
    "## Subsets\n",
    "\n",
    "See [Subsets](https://incatools.github.io/ontology-access-kit/glossary.html#term-Subset) in the OAK Glossary.\n",
    "\n",
    "Subsets allow terms to be placed into groups outside the hierarchy for different purposes.\n",
    "\n",
    "To illustrate we will switch our example to use GO which has a rich variety of subsets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "2d23990d",
   "metadata": {},
   "outputs": [],
   "source": [
    "go_adapter = get_adapter(\"sqlite:obo:go\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "b32ece7e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "chebi_ph7_3\n",
      "3_STAR\n",
      "1_STAR\n",
      "goslim_plant\n",
      "goslim_pir\n",
      "goslim_flybase_ribbon\n",
      "goslim_chembl\n",
      "goslim_agr\n",
      "goslim_metagenomics\n",
      "goslim_yeast\n",
      "goslim_pombe\n",
      "gocheck_do_not_annotate\n",
      "goslim_generic\n",
      "goslim_drosophila\n",
      "goslim_candida\n",
      "prokaryote_subset\n",
      "gocheck_do_not_manually_annotate\n",
      "goslim_synapse\n",
      "goslim_mouse\n",
      "SOFA\n",
      "Alliance_of_Genome_Resources\n",
      "biosapiens\n"
     ]
    }
   ],
   "source": [
    "for subset in go_adapter.subsets():\n",
    "    print(subset)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a0941b7",
   "metadata": {},
   "source": [
    "note this includes subsets for ontologies that have been merged in"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "f3c814cc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GO:0000228 nuclear chromosome\n",
      "GO:0000278 mitotic cell cycle\n",
      "GO:0000910 cytokinesis\n",
      "GO:0001618 virus receptor activity\n",
      "GO:0002181 cytoplasmic translation\n",
      "GO:0002376 immune system process\n",
      "GO:0003012 muscle system process\n",
      "GO:0003013 circulatory system process\n",
      "GO:0003014 renal system process\n",
      "GO:0003016 respiratory system process\n"
     ]
    }
   ],
   "source": [
    "for e in list(go_adapter.subset_members(\"goslim_generic\"))[0:10]:\n",
    "    print(e, go_adapter.label(e))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "930cd6df",
   "metadata": {},
   "source": [
    "### Plotting how GO subsets inter-relate\n",
    "\n",
    "Now we are ready for a simple mini application - showing commonalities between "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "072dce81",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "sets = []\n",
    "\n",
    "for subset in go_adapter.subsets():\n",
    "    members = set([x for x in go_adapter.subset_members(subset) if x.startswith(\"GO:\")])\n",
    "    if members:\n",
    "        sets.append((subset, members))\n",
    "\n",
    "\n",
    "# Number of sets\n",
    "N = len(sets)\n",
    "\n",
    "# Initialize an empty matrix to store the number of members in common\n",
    "intersection_matrix = np.zeros((N, N))\n",
    "\n",
    "# Calculate the intersections between each pair of sets\n",
    "for i in range(N):\n",
    "    for j in range(N):\n",
    "        intersection_matrix[i, j] = len(sets[i][1].intersection(sets[j][1]))\n",
    "\n",
    "# Get the set names\n",
    "set_names = [s[0] for s in sets]\n",
    "\n",
    "# Create a pandas DataFrame with the intersection matrix and set names as index and columns\n",
    "intersection_df = pd.DataFrame(intersection_matrix, index=set_names, columns=set_names)\n",
    "\n",
    "# Plot the clustermap with dendrograms\n",
    "sns.clustermap(intersection_df, annot=True, cmap='viridis', fmt='g', figsize=(8, 6))\n",
    "plt.title('Clustermap of Common Members Between Sets with Dendrograms', y=1.03)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "beb3469c",
   "metadata": {},
   "source": [
    "## Search Interface\n",
    "\n",
    "So far we have been doing basic lookup information, assuming we know the ID in advance.\n",
    "\n",
    "What if we don't know the ID but just have a label, or if we don't even have a particular concept in mind, and just want to search?\n",
    "\n",
    "If so, the [SearchInterface](https://incatools.github.io/ontology-access-kit/packages/interfaces/search.html) is your friend!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f9349be3",
   "metadata": {},
   "source": [
    "### Lookup by label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "58e1d5a5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CL:0000540\n"
     ]
    }
   ],
   "source": [
    "for result in adapter.basic_search(\"neuron\"):\n",
    "    print(result)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f3b5459",
   "metadata": {},
   "source": [
    "now let's try searching for the capitalized form:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "708d0258",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(list(adapter.basic_search(\"Neuron\")))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e63227b",
   "metadata": {},
   "source": [
    "uh oh!\n",
    "\n",
    "By design, the default is case sensitive. But we can pass a SearchConfiguration to make search more customizable.\n",
    "\n",
    "You can read more about the SearchConfiguration datamodel here:\n",
    "\n",
    "- https://w3id.org/oak/search\n",
    "\n",
    "__a note on data models__ the BasicOntologyInterface is designed to work without any particular data model, returning only simple lists and tuples. Other interfaces typically need to work with more sophisticated structures, so we use data models here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "795b40af",
   "metadata": {},
   "outputs": [],
   "source": [
    "from oaklib.datamodels.search import SearchConfiguration, SearchTermSyntax, SearchProperty"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "c2ee9f4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "config = SearchConfiguration(force_case_insensitive=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "492d2ab6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(list(adapter.basic_search(\"Neuron\", config)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "cbcb7dd3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(list(adapter.basic_search(\"NeUrOn\", config)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "784a1c78",
   "metadata": {},
   "source": [
    "We can also do regexes, starts-with, ends with etc (but see below for caveat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "3104e5cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "config = SearchConfiguration(syntax=SearchTermSyntax.STARTS_WITH)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "4bb14054",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CARO:0001001 neuron projection bundle\n",
      "CL:0000006 neuronal receptor cell\n",
      "CL:0000095 neuron associated cell\n",
      "CL:0000123 neuron associated cell (sensu Vertebrata)\n",
      "CL:0000130 neuron associated cell (sensu Nematoda and Protostomia)\n",
      "CL:0000540 neuron\n",
      "CL:0000555 neuronal brush cell\n",
      "CL:0002611 neuron of the dorsal spinal cord\n",
      "CL:0002612 neuron of the ventral spinal cord\n",
      "CL:0002614 neuron of the substantia nigra\n",
      "CL:0012001 neuron of the forebrain\n",
      "GO:0001764 neuron migration\n",
      "GO:0019228 neuronal action potential\n",
      "GO:0030182 neuron differentiation\n",
      "GO:0031175 neuron projection development\n",
      "GO:0032589 neuron projection membrane\n",
      "GO:0042551 neuron maturation\n",
      "GO:0043005 neuron projection\n",
      "GO:0043025 neuronal cell body\n",
      "GO:0044306 neuron projection terminus\n",
      "GO:0048666 neuron development\n",
      "GO:0048812 neuron projection morphogenesis\n",
      "GO:0051402 neuron apoptotic process\n",
      "GO:0060705 neuron differentiation involved in salivary gland development\n",
      "GO:0070050 neuron cellular homeostasis\n",
      "GO:0070997 neuron death\n",
      "GO:0106027 neuron projection organization\n",
      "GO:0120111 neuron projection cytoplasm\n",
      "GO:0150099 neuron-glial cell signaling\n",
      "PATO:0070033 neuron projection quality\n",
      "PR:000005460 neuronal acetylcholine receptor subunit alpha-7\n",
      "PR:000044062 neuronal acetylcholine receptor subunit alpha-7, signal peptide removed form\n",
      "PR:000044063 neuronal acetylcholine receptor subunit alpha-7, signal peptide removed form (human)\n",
      "PR:P36544 neuronal acetylcholine receptor subunit alpha-7 (human)\n",
      "PR:P49582 neuronal acetylcholine receptor subunit alpha-7 (mouse)\n",
      "UBERON:0000122 neuron projection bundle\n",
      "UBERON:0004904 neuron projection bundle connecting eye with brain\n"
     ]
    }
   ],
   "source": [
    "for result in adapter.basic_search(\"neuron\", config):\n",
    "    print(result, adapter.label(result))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1fc847e6",
   "metadata": {},
   "source": [
    "now we can try a regex:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "8f36650c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CARO:0001001 neuron projection bundle\n",
      "CL:0000006 neuronal receptor cell\n",
      "CL:0000095 neuron associated cell\n",
      "CL:0000123 neuron associated cell (sensu Vertebrata)\n",
      "CL:0000130 neuron associated cell (sensu Nematoda and Protostomia)\n",
      "CL:0000540 neuron\n",
      "CL:0000555 neuronal brush cell\n",
      "CL:0002611 neuron of the dorsal spinal cord\n",
      "CL:0002612 neuron of the ventral spinal cord\n",
      "CL:0002614 neuron of the substantia nigra\n",
      "CL:0012001 neuron of the forebrain\n",
      "GO:0001764 neuron migration\n",
      "GO:0019228 neuronal action potential\n",
      "GO:0030182 neuron differentiation\n",
      "GO:0031175 neuron projection development\n",
      "GO:0032589 neuron projection membrane\n",
      "GO:0042551 neuron maturation\n",
      "GO:0043005 neuron projection\n",
      "GO:0043025 neuronal cell body\n",
      "GO:0044306 neuron projection terminus\n",
      "GO:0048666 neuron development\n",
      "GO:0048812 neuron projection morphogenesis\n",
      "GO:0051402 neuron apoptotic process\n",
      "GO:0060705 neuron differentiation involved in salivary gland development\n",
      "GO:0070050 neuron cellular homeostasis\n",
      "GO:0070997 neuron death\n",
      "GO:0106027 neuron projection organization\n",
      "GO:0120111 neuron projection cytoplasm\n",
      "GO:0150099 neuron-glial cell signaling\n",
      "PATO:0070033 neuron projection quality\n",
      "PR:000005460 neuronal acetylcholine receptor subunit alpha-7\n",
      "PR:000044062 neuronal acetylcholine receptor subunit alpha-7, signal peptide removed form\n",
      "PR:000044063 neuronal acetylcholine receptor subunit alpha-7, signal peptide removed form (human)\n",
      "PR:P36544 neuronal acetylcholine receptor subunit alpha-7 (human)\n",
      "PR:P49582 neuronal acetylcholine receptor subunit alpha-7 (mouse)\n",
      "UBERON:0000122 neuron projection bundle\n",
      "UBERON:0004904 neuron projection bundle connecting eye with brain\n"
     ]
    }
   ],
   "source": [
    "config = SearchConfiguration(syntax=SearchTermSyntax.REGULAR_EXPRESSION)\n",
    "for result in adapter.basic_search(\"^neuron\", config):\n",
    "    print(result, adapter.label(result))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f668d66e",
   "metadata": {},
   "source": [
    "### Caveat on regexes\n",
    "\n",
    "If your adapter is talking to sqlite, then the regex must be of a form that can be translated to a LIKE query\n",
    "\n",
    "(OAK takes care of this translation - as a developer you should only care about the interface, not implementation)\n",
    "\n",
    "In future we may have strategies to allow more powerful lexical search with sqlite..."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0b4bbb16",
   "metadata": {},
   "source": [
    "### Searching on mapped identifiers\n",
    "\n",
    "You can search on arbitrary properties, such as synonyms or even mapped identifiers (`object_id` in SSSOM lingo)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "221c3e6a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CL:0000202 auditory hair cell\n",
      "CL:4023120 cochlea auditory hair cell\n"
     ]
    }
   ],
   "source": [
    "config = SearchConfiguration(properties=[SearchProperty.MAPPED_IDENTIFIER])\n",
    "for result in adapter.basic_search(\"FMA:62364\", config):\n",
    "    print(result, adapter.label(result))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a47980d5",
   "metadata": {},
   "source": [
    "### SSSOM Mappings\n",
    "\n",
    "Up above we saw that the default datamodel for mappings in OAK is simple. For more advanced operations, you can use:\n",
    "\n",
    "[MappingProviderInterface](https://incatools.github.io/ontology-access-kit/packages/interfaces/mapping-provider.html)\n",
    "\n",
    "This makes use of the https://w3id.org/sssom data model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "26eccc9f",
   "metadata": {},
   "outputs": [],
   "source": [
    "neurons = list(adapter.descendants(\"CL:0000540\", predicates=[IS_A]))\n",
    "mappings = list(adapter.sssom_mappings(neurons))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "eb41b4f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "186"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(mappings)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "5da17c85",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mapping(subject_id='CL:0000099', predicate_id='oio:hasDbXref', object_id='BTO:0003811', mapping_justification='semapv:UnspecifiedMatching', subject_label=None, subject_category=None, predicate_label=None, predicate_modifier=None, object_label=None, object_category=None, author_id=[], author_label=[], reviewer_id=[], reviewer_label=[], creator_id=[], creator_label=[], license=None, subject_type=None, subject_source='CL', subject_source_version=None, object_type=None, object_source='BTO', object_source_version=None, mapping_provider=None, mapping_source=None, mapping_cardinality=None, mapping_tool=None, mapping_tool_version=None, mapping_date=None, confidence=None, curation_rule=[], curation_rule_text=[], subject_match_field=[], object_match_field=[], match_string=[], subject_preprocessing=[], object_preprocessing=[], semantic_similarity_score=None, semantic_similarity_measure=None, see_also=[], other=None, comment=None)\n"
     ]
    }
   ],
   "source": [
    "print(mappings[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "d0196ae7",
   "metadata": {},
   "outputs": [],
   "source": [
    "from linkml_runtime.dumpers import yaml_dumper"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "d916d6d9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- subject_id: CL:0000099\n",
      "  predicate_id: oio:hasDbXref\n",
      "  object_id: BTO:0003811\n",
      "  mapping_justification: semapv:UnspecifiedMatching\n",
      "  subject_source: CL\n",
      "  object_source: BTO\n",
      "- subject_id: CL:0000099\n",
      "  predicate_id: oio:hasDbXref\n",
      "  object_id: FBbt:00005125\n",
      "  mapping_justification: semapv:UnspecifiedMatching\n",
      "  subject_source: CL\n",
      "  object_source: FBbt\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(yaml_dumper.dumps(mappings[0:2]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "55b8c04c",
   "metadata": {},
   "source": [
    "## Text Annotation\n",
    "\n",
    "Interface: [TextAnnotatorInterface](https://incatools.github.io/ontology-access-kit/packages/interfaces/text-annotator.html)\n",
    "\n",
    "The text annotator uses the https://w3id.org/linkml/text-annotator data model. This models each annotation as an TextAnnotation object with fields such as subject_start and subject_end (marking the span in the text) and object_id and object_label (the matched concept):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "b61ee543",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18 21 CARO:0000013 cell\n",
      "18 21 CL:0000000 cell\n",
      "9 9 CHEBI:15339 A\n",
      "11 11 CHEBI:15428 G\n",
      "18 18 CHEBI:27594 C\n",
      "11 21 CL:0000160 goblet cell\n",
      "1 2 PR:000016301 TH\n",
      "1 2 PR:P07101 TH\n",
      "1 2 PR:P24529 Th\n",
      "1 2 UBERON:0001897 Th\n",
      "43 52 UBERON:0000483 epithelium\n",
      "32 52 UBERON:0001277 intestinal epithelium\n"
     ]
    }
   ],
   "source": [
    "for ann in adapter.annotate_text(\"this is a goblet cell from the intestinal epithelium\"):\n",
    "    print(ann.subject_start, ann.subject_end, ann.object_id, ann.object_label)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b38acc0",
   "metadata": {},
   "source": [
    "### OBO Graph Interface\n",
    "\n",
    "[OboGraphInterface](https://incatools.github.io/ontology-access-kit/packages/interfaces/obograph.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "dc4ba581",
   "metadata": {},
   "outputs": [],
   "source": [
    "graph = adapter.ancestor_graph([\"CL:0000540\"], predicates=[IS_A, PART_OF])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "36a599dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(graph.nodes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "09979d95",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "29"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(graph.edges)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e43d9cb1",
   "metadata": {},
   "source": [
    "### Exporting subgraphs to GraphViz\n",
    "\n",
    "See also [part 5](https://incatools.github.io/ontology-access-kit/intro/tutorial05.html) of the tutorial"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "62c2e672",
   "metadata": {},
   "outputs": [],
   "source": [
    "from oaklib.utilities.obograph_utils import graph_to_image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "a3c605ec",
   "metadata": {},
   "outputs": [],
   "source": "graph_to_image(graph, seeds=[\"CL:0000540\"], imgfile=\"examples/output/neuron-v1.png\")"
  },
  {
   "cell_type": "markdown",
   "id": "3200472b",
   "metadata": {},
   "source": [
    "![img](output/neuron-v1.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0aa822e",
   "metadata": {},
   "source": [
    "### Adding a stylesheet\n",
    "\n",
    "The graph above is a little plain and boring looking. We can spice it up using a StyleMap.\n",
    "\n",
    "For now we will use the standard stylemap in [src/oaklib/conf/obograph-style.json](https://github.com/INCATools/ontology-access-kit/blob/main/src/oaklib/conf/obograph-style.json):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "deda20f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "from oaklib.utilities.obograph_utils import default_stylemap_path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "edae4b80",
   "metadata": {},
   "outputs": [],
   "source": "graph_to_image(graph, seeds=[\"CL:0000540\"], imgfile=\"examples/output/neuron-v2.png\", stylemap=default_stylemap_path())"
  },
  {
   "cell_type": "markdown",
   "id": "09c581a6",
   "metadata": {},
   "source": [
    "![img](output/neuron-v2.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f53d9efb",
   "metadata": {},
   "source": [
    "## Working with annotations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "41dabf06",
   "metadata": {},
   "outputs": [],
   "source": [
    "hp = get_adapter(\"src/oaklib/conf/hpoa-g2p-input-spec.yaml\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "3ea4ace2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "238269"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(list(hp.associations()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "8c929abf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Association(subject='NCBIGene:ncbi_gene_id', predicate=None, object='hpo_id', property_values=[]) None\n",
      "Association(subject='NCBIGene:10', predicate=None, object='HP:0000007', property_values=[]) Autosomal recessive inheritance\n",
      "Association(subject='NCBIGene:10', predicate=None, object='HP:0001939', property_values=[]) Abnormality of metabolism/homeostasis\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0002460', property_values=[]) Distal muscle weakness\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0002451', property_values=[]) Limb dystonia\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0010871', property_values=[]) Sensory ataxia\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0009886', property_values=[]) Trichorrhexis nodosa\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0002421', property_values=[]) Poor head control\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0001298', property_values=[]) Encephalopathy\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0001290', property_values=[]) Generalized hypotonia\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0001273', property_values=[]) Abnormal corpus callosum morphology\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0001268', property_values=[]) Mental deterioration\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0002599', property_values=[]) Head titubation\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0001284', property_values=[]) Areflexia\n",
      "Association(subject='NCBIGene:16', predicate=None, object='HP:0001250', property_values=[]) Seizure\n"
     ]
    }
   ],
   "source": [
    "for assoc in list(hp.associations())[0:15]:\n",
    "    print(assoc, hp.label(assoc.object))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "ae02da0c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "15"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Fetch sensory ataxia genes (including those annotated to is-a descendants of the term)\n",
    "ataxia_assocs = list(hp.associations(objects=[\"HP:0010871\"], object_closure_predicates=[IS_A]))\n",
    "len(ataxia_assocs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "17f238ef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "15"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "genes = list(set([assoc.subject for assoc in ataxia_assocs]))\n",
    "len(genes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "8961221f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['NCBIGene:56652',\n",
       " 'NCBIGene:5428',\n",
       " 'NCBIGene:16',\n",
       " 'NCBIGene:1959',\n",
       " 'NCBIGene:57716']"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "genes[0:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "d7773ee5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'UniProtKB:A0A024RDV7',\n",
       " 'UniProtKB:A0A140VJE4',\n",
       " 'UniProtKB:A0A2R8Y4V4',\n",
       " 'UniProtKB:A0A2R8Y746',\n",
       " 'UniProtKB:A0A2U3TZU2',\n",
       " 'UniProtKB:A0A514TP98',\n",
       " 'UniProtKB:A0A5F9ZI26',\n",
       " 'UniProtKB:A0A8I5KYI5',\n",
       " 'UniProtKB:A8KA82',\n",
       " 'UniProtKB:A8MU75',\n",
       " 'UniProtKB:B2RB38',\n",
       " 'UniProtKB:B4DE36',\n",
       " 'UniProtKB:E5KNU5',\n",
       " 'UniProtKB:E5KSY5',\n",
       " 'UniProtKB:O00505',\n",
       " 'UniProtKB:P06744',\n",
       " 'UniProtKB:P11161',\n",
       " 'UniProtKB:P25189',\n",
       " 'UniProtKB:P49588',\n",
       " 'UniProtKB:P54098',\n",
       " 'UniProtKB:P54802',\n",
       " 'UniProtKB:Q01453',\n",
       " 'UniProtKB:Q13217',\n",
       " 'UniProtKB:Q6FH25',\n",
       " 'UniProtKB:Q8TF17',\n",
       " 'UniProtKB:Q96K19',\n",
       " 'UniProtKB:Q96RR1',\n",
       " 'UniProtKB:Q9BXM0',\n",
       " 'UniProtKB:Q9H5I5',\n",
       " 'UniProtKB:Q9H6V3',\n",
       " 'UniProtKB:Q9Y5Y0'}"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "node_normalizer = get_adapter(\"translator:\")\n",
    "uniprot_ids = set()\n",
    "for gene in genes:\n",
    "    for m in node_normalizer.sssom_mappings([gene], source=\"UniProtKB\"):\n",
    "        uniprot_ids.add(m.object_id)\n",
    "uniprot_ids"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "b14e0520",
   "metadata": {},
   "outputs": [],
   "source": [
    "go = get_adapter(\"src/oaklib/conf/go-human-input-spec.yaml\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "10a687cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "results = list(go.enriched_classes(uniprot_ids, object_closure_predicates=[IS_A, PART_OF], autolabel=True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "b22cb405",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GO:0042552 'myelination' 9.72e-04\n",
      "GO:0008366 'axon ensheathment' 1.06e-03\n",
      "GO:0007272 'ensheathment of neurons' 1.06e-03\n",
      "GO:0007422 'peripheral nervous system development' 6.38e-03\n",
      "GO:0014037 'Schwann cell differentiation' 2.75e-02\n"
     ]
    }
   ],
   "source": [
    "for result in results:\n",
    "    print(f\"{result.class_id} '{result.class_label}' {result.p_value_adjusted:0.2e}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "6d382ca5",
   "metadata": {},
   "outputs": [],
   "source": [
    "terms = [r.class_id for r in results]\n",
    "graph = go.ancestor_graph(terms, predicates=[IS_A, PART_OF])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "a656d404",
   "metadata": {},
   "outputs": [],
   "source": "graph_to_image(graph, seeds=terms, imgfile=\"examples/output/go-enrichment-from-hp.png\", stylemap=default_stylemap_path())"
  },
  {
   "cell_type": "markdown",
   "id": "0e7d6fce",
   "metadata": {},
   "source": [
    "![img](output/go-enrichment-from-hp.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "117ca65c",
   "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": "3.9.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}