{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "pursuant-separate",
   "metadata": {
    "papermill": {
     "duration": 0.026869,
     "end_time": "2021-03-22T10:46:16.607493",
     "exception": false,
     "start_time": "2021-03-22T10:46:16.580624",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Meta functionalities of the EpiGraphDB platform"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "written-decade",
   "metadata": {
    "papermill": {
     "duration": 0.025289,
     "end_time": "2021-03-22T10:46:16.659074",
     "exception": false,
     "start_time": "2021-03-22T10:46:16.633785",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "In this notebook we show the following aspects of the EpiGraphDB platform, and how to use the API to get the information:\n",
    "\n",
    "1. Metadata: meta nodes and meta edges, and the overall schema.\n",
    "2. Search for a specific node under the meta node.\n",
    "3. Cypher: how to query the database directly using Neo4j Cypher\n",
    "\n",
    "For detailed documentation on the API endpoints please visit:\n",
    "\n",
    "- The Swagger interface: http://api.epigraphdb.org\n",
    "- The sections regarding API endpoints on the documentation site: http://docs.epigraphdb.org/api/api-endpoints/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "united-valve",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:16.692381Z",
     "iopub.status.busy": "2021-03-22T10:46:16.692014Z",
     "iopub.status.idle": "2021-03-22T10:46:17.093963Z",
     "shell.execute_reply": "2021-03-22T10:46:17.094608Z"
    },
    "papermill": {
     "duration": 0.421274,
     "end_time": "2021-03-22T10:46:17.094891",
     "exception": false,
     "start_time": "2021-03-22T10:46:16.673617",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from pprint import pformat\n",
    "\n",
    "import networkx as nx\n",
    "import pandas as pd\n",
    "import requests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "narrative-smell",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:17.128485Z",
     "iopub.status.busy": "2021-03-22T10:46:17.128115Z",
     "iopub.status.idle": "2021-03-22T10:46:17.130122Z",
     "shell.execute_reply": "2021-03-22T10:46:17.129777Z"
    },
    "papermill": {
     "duration": 0.019947,
     "end_time": "2021-03-22T10:46:17.130200",
     "exception": false,
     "start_time": "2021-03-22T10:46:17.110253",
     "status": "completed"
    },
    "tags": [
     "parameters"
    ]
   },
   "outputs": [],
   "source": [
    "# default parameters\n",
    "API_URL = \"https://api.epigraphdb.org\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "sporting-blind",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:17.160280Z",
     "iopub.status.busy": "2021-03-22T10:46:17.159938Z",
     "iopub.status.idle": "2021-03-22T10:46:17.162153Z",
     "shell.execute_reply": "2021-03-22T10:46:17.161625Z"
    },
    "papermill": {
     "duration": 0.018866,
     "end_time": "2021-03-22T10:46:17.162285",
     "exception": false,
     "start_time": "2021-03-22T10:46:17.143419",
     "status": "completed"
    },
    "tags": [
     "injected-parameters"
    ]
   },
   "outputs": [],
   "source": [
    "# Parameters\n",
    "API_URL = \"https://api.epigraphdb.org\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "rocky-spider",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:17.202936Z",
     "iopub.status.busy": "2021-03-22T10:46:17.202394Z",
     "iopub.status.idle": "2021-03-22T10:46:17.347014Z",
     "shell.execute_reply": "2021-03-22T10:46:17.347389Z"
    },
    "papermill": {
     "duration": 0.161663,
     "end_time": "2021-03-22T10:46:17.347511",
     "exception": false,
     "start_time": "2021-03-22T10:46:17.185848",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "https://api.epigraphdb.org\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(API_URL)\n",
    "requests.get(f\"{API_URL}/ping\").json()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "verbal-missile",
   "metadata": {
    "papermill": {
     "duration": 0.018825,
     "end_time": "2021-03-22T10:46:17.385164",
     "exception": false,
     "start_time": "2021-03-22T10:46:17.366339",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Metadata"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "subsequent-helping",
   "metadata": {
    "papermill": {
     "duration": 0.018396,
     "end_time": "2021-03-22T10:46:17.423403",
     "exception": false,
     "start_time": "2021-03-22T10:46:17.405007",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Here we query for the metadata information using the endpoint `GET /meta/schema`, which will be used for downstream processing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "royal-mason",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:17.459168Z",
     "iopub.status.busy": "2021-03-22T10:46:17.458820Z",
     "iopub.status.idle": "2021-03-22T10:46:17.623210Z",
     "shell.execute_reply": "2021-03-22T10:46:17.623835Z"
    },
    "papermill": {
     "duration": 0.187001,
     "end_time": "2021-03-22T10:46:17.624010",
     "exception": false,
     "start_time": "2021-03-22T10:46:17.437009",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['nodes', 'edges', 'connections']) \n",
      "\n",
      "# nodes:\n",
      "{'Disease': {'count': 38960,\n",
      "             'properties': {'_id': {'indexed': True,\n",
      "                                    'type': 'STRING',\n",
      "                                    'unique': False},\n",
      "                            '_name': {'indexed': True,\n",
      "                                      'type': 'STRING',\n",
      "                                      'unique': False},\n",
      "                            '_source': {'indexed': False,\n",
      "                                        'type': 'LIST',\n",
      "                                        'unique': False},\n",
      "                            'definition': {'indexed': False,\n",
      "                                           'type': 'STRING',\n",
      "                                           'unique': False},\n",
      "                            'doid': {'indexed': True,\n",
      "                                     'type': 'LIST',\n",
      "                                     'unique': False},\n",
      "                            'efo': {'indexed': False,\n",
      "                                    'type': 'LIST',\n",
      "                     \n",
      "\n",
      "# edges:\n",
      "{'BIORXIV_OBJ': {'count': 32651,\n",
      "                 'properties': {'_source': {'array': True, 'type': 'LIST'}}},\n",
      " 'BIORXIV_PREDICATE': {'count': 32648,\n",
      "                       'properties': {'_source': {'array': True,\n",
      "                                                  'type': 'LIST'},\n",
      "                                      'count': {'array': False,\n",
      "                                                'type': 'INTEGER'},\n",
      "                                      'predicate': {'array': False,\n",
      "                                                    'type': 'STRING'}}},\n",
      " 'BIORXIV_SUB': {'count': 32657,\n",
      "                 'properties': {'_source': {'array': True, 'type': 'LIST'}}},\n",
      " 'BIORXIV_TO_LIT': {'count': 35211,\n",
      "                    'properties': {'_source': {'array': True, 'type': 'LIST'}}},\n",
      " 'CPIC': {'count': 375,\n",
      "          'properties': {'_source': {'array': True, 'type': 'LIST'},\n",
      "                         'cpic_level': {'array': False, 'type': 'STRING'},\n",
      "                         'guideline': {'array': F \n",
      "\n",
      "# connections:\n",
      "[{'count': 2461,\n",
      "  'from_node': 'Drug',\n",
      "  'rel': 'OPENTARGETS_DRUG_TO_DISEASE',\n",
      "  'to_node': 'Disease'},\n",
      " {'count': 5763,\n",
      "  'from_node': 'Gene',\n",
      "  'rel': 'GENE_TO_DISEASE',\n",
      "  'to_node': 'Disease'},\n",
      " {'count': 8247,\n",
      "  'from_node': 'Disease',\n",
      "  'rel': 'MONDO_MAP_UMLS',\n",
      "  'to_node': 'LiteratureTerm'},\n",
      " {'count': 2819,\n",
      "  'from_node': 'Disease',\n",
      "  'rel': 'MONDO_MAP_EFO',\n",
      "  'to_node': 'Efo'},\n",
      " {'count': 2463,\n",
      "  'from_node': 'Pathway',\n",
      "  'rel': 'PATHWAY_CHILD_OF',\n",
      "  'to_node': 'Pathway'},\n",
      " {'count': 121873,\n",
      "  'from_node': 'Protein',\n",
      "  'rel': 'PROTEIN_IN_PATHWAY',\n",
      "  'to_node': 'Pathway'},\n",
      " {'count': 1969,\n",
      "  'from_node': 'LiteratureTerm',\n",
      "  'rel': 'MEDRXIV_PREDICATE',\n",
      "  'to_node': 'LiteratureTerm'},\n",
      " {'count': 16435,\n",
      "  'from_node': 'LiteratureTerm',\n",
      "  'rel': 'TERM_TO_GENE',\n",
      "  'to_node': 'Gene'},\n",
      " {'count': 32651,\n",
      "  'from_node': 'LiteratureTriple',\n",
      "  'rel': 'BIORXIV_OBJ',\n",
      "  'to_node': 'LiteratureTerm'},\n",
      " {'count': 32657,\n",
      "  'from_node': 'LiteratureTriple',\n",
      "  'rel': 'BIORXIV_SUB',\n",
      "  'to_node': 'Li \n",
      "\n"
     ]
    }
   ],
   "source": [
    "endpoint = \"/meta/schema\"\n",
    "params = {\"graphviz\": False, \"plot\": False}\n",
    "r = requests.get(f\"{API_URL}{endpoint}\", params=params)\n",
    "r.raise_for_status()\n",
    "metadata = r.json()\n",
    "\n",
    "# Preview of metadata information\n",
    "keys = metadata.keys()\n",
    "print(pformat(keys), \"\\n\")\n",
    "for key in list(keys):\n",
    "    print(f\"# {key}:\")\n",
    "    print(pformat(metadata[key])[:1000], \"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "floral-extension",
   "metadata": {
    "papermill": {
     "duration": 0.026417,
     "end_time": "2021-03-22T10:46:17.676760",
     "exception": false,
     "start_time": "2021-03-22T10:46:17.650343",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Meta nodes\n",
    "\n",
    "We can extract the specific meta node information as a pandas dataframe from the metadata."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "filled-blood",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:17.759212Z",
     "iopub.status.busy": "2021-03-22T10:46:17.744039Z",
     "iopub.status.idle": "2021-03-22T10:46:17.795459Z",
     "shell.execute_reply": "2021-03-22T10:46:17.794912Z"
    },
    "papermill": {
     "duration": 0.093095,
     "end_time": "2021-03-22T10:46:17.795612",
     "exception": false,
     "start_time": "2021-03-22T10:46:17.702517",
     "status": "completed"
    },
    "tags": []
   },
   "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>count</th>\n",
       "      <th>properties</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Disease</th>\n",
       "      <td>38,960</td>\n",
       "      <td>{'_name': {'type': 'STRING', 'indexed': True, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Drug</th>\n",
       "      <td>2,697</td>\n",
       "      <td>{'molecule_type': {'type': 'STRING', 'indexed'...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Efo</th>\n",
       "      <td>25,390</td>\n",
       "      <td>{'_name': {'type': 'STRING', 'indexed': True, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gene</th>\n",
       "      <td>57,737</td>\n",
       "      <td>{'druggability_tier': {'type': 'STRING', 'inde...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gwas</th>\n",
       "      <td>34,494</td>\n",
       "      <td>{'note': {'type': 'STRING', 'indexed': False, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Literature</th>\n",
       "      <td>3,995,672</td>\n",
       "      <td>{'issn': {'type': 'STRING', 'indexed': False, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LiteratureTerm</th>\n",
       "      <td>108,905</td>\n",
       "      <td>{'_name': {'type': 'STRING', 'indexed': True, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LiteratureTriple</th>\n",
       "      <td>5,609,945</td>\n",
       "      <td>{'subject_id': {'type': 'STRING', 'indexed': T...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pathway</th>\n",
       "      <td>2,441</td>\n",
       "      <td>{'_name': {'type': 'STRING', 'indexed': True, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Protein</th>\n",
       "      <td>20,280</td>\n",
       "      <td>{'name': {'type': 'STRING', 'indexed': True, '...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tissue</th>\n",
       "      <td>54</td>\n",
       "      <td>{'name': {'type': 'STRING', 'indexed': True, '...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Variant</th>\n",
       "      <td>99,005</td>\n",
       "      <td>{'ref': {'type': 'STRING', 'indexed': False, '...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      count                                         properties\n",
       "Disease              38,960  {'_name': {'type': 'STRING', 'indexed': True, ...\n",
       "Drug                  2,697  {'molecule_type': {'type': 'STRING', 'indexed'...\n",
       "Efo                  25,390  {'_name': {'type': 'STRING', 'indexed': True, ...\n",
       "Gene                 57,737  {'druggability_tier': {'type': 'STRING', 'inde...\n",
       "Gwas                 34,494  {'note': {'type': 'STRING', 'indexed': False, ...\n",
       "Literature        3,995,672  {'issn': {'type': 'STRING', 'indexed': False, ...\n",
       "LiteratureTerm      108,905  {'_name': {'type': 'STRING', 'indexed': True, ...\n",
       "LiteratureTriple  5,609,945  {'subject_id': {'type': 'STRING', 'indexed': T...\n",
       "Pathway               2,441  {'_name': {'type': 'STRING', 'indexed': True, ...\n",
       "Protein              20,280  {'name': {'type': 'STRING', 'indexed': True, '...\n",
       "Tissue                   54  {'name': {'type': 'STRING', 'indexed': True, '...\n",
       "Variant              99,005  {'ref': {'type': 'STRING', 'indexed': False, '..."
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "meta_node_df = pd.DataFrame.from_dict(metadata[\"nodes\"], orient=\"index\")\n",
    "\n",
    "(\n",
    "    meta_node_df.sort_index().assign(\n",
    "        count=lambda df: df[\"count\"].apply(lambda x: f\"{x:,}\")\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "casual-demonstration",
   "metadata": {
    "papermill": {
     "duration": 0.029886,
     "end_time": "2021-03-22T10:46:17.853291",
     "exception": false,
     "start_time": "2021-03-22T10:46:17.823405",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Meta relationships and connections\n",
    "\n",
    "We can also extract the meta relationship (edge) information, and the connections."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "separate-reflection",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:17.902561Z",
     "iopub.status.busy": "2021-03-22T10:46:17.901677Z",
     "iopub.status.idle": "2021-03-22T10:46:17.968732Z",
     "shell.execute_reply": "2021-03-22T10:46:17.969440Z"
    },
    "papermill": {
     "duration": 0.096284,
     "end_time": "2021-03-22T10:46:17.969693",
     "exception": false,
     "start_time": "2021-03-22T10:46:17.873409",
     "status": "completed"
    },
    "tags": []
   },
   "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>count</th>\n",
       "      <th>properties</th>\n",
       "      <th>from_node</th>\n",
       "      <th>to_node</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MONDO_MAP_EFO</th>\n",
       "      <td>2,819</td>\n",
       "      <td>{'_source': {'array': False, 'type': 'STRING'}}</td>\n",
       "      <td>Disease</td>\n",
       "      <td>Efo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MONDO_MAP_UMLS</th>\n",
       "      <td>8,247</td>\n",
       "      <td>{'_source': {'array': False, 'type': 'STRING'}}</td>\n",
       "      <td>Disease</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OPENTARGETS_DRUG_TO_DISEASE</th>\n",
       "      <td>2,461</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>Drug</td>\n",
       "      <td>Disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CPIC</th>\n",
       "      <td>375</td>\n",
       "      <td>{'pharmgkb_level_of_evidence': {'array': False...</td>\n",
       "      <td>Drug</td>\n",
       "      <td>Gene</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OPENTARGETS_DRUG_TO_TARGET</th>\n",
       "      <td>6,534</td>\n",
       "      <td>{'phase': {'array': False, 'type': 'STRING'}, ...</td>\n",
       "      <td>Drug</td>\n",
       "      <td>Gene</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EFO_CHILD_OF</th>\n",
       "      <td>43,132</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>Efo</td>\n",
       "      <td>Efo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GENE_TO_DISEASE</th>\n",
       "      <td>5,763</td>\n",
       "      <td>{'last_updated': {'array': False, 'type': 'STR...</td>\n",
       "      <td>Gene</td>\n",
       "      <td>Disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XQTL_MULTI_SNP_MR</th>\n",
       "      <td>3,015,233</td>\n",
       "      <td>{'p': {'array': False, 'type': 'FLOAT'}, 'se':...</td>\n",
       "      <td>Gene</td>\n",
       "      <td>Gwas</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XQTL_SINGLE_SNP_MR_GENE_GWAS</th>\n",
       "      <td>8,449,779</td>\n",
       "      <td>{'p': {'array': False, 'type': 'FLOAT'}, 'se':...</td>\n",
       "      <td>Gene</td>\n",
       "      <td>Gwas</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GENE_TO_PROTEIN</th>\n",
       "      <td>19,142</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>Gene</td>\n",
       "      <td>Protein</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EXPRESSED_IN</th>\n",
       "      <td>2,918,240</td>\n",
       "      <td>{'tpm': {'array': False, 'type': 'FLOAT'}, '_s...</td>\n",
       "      <td>Gene</td>\n",
       "      <td>Tissue</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GWAS_NLP_EFO</th>\n",
       "      <td>12,302</td>\n",
       "      <td>{'score': {'array': False, 'type': 'FLOAT'}, '...</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>Efo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GWAS_EFO_EBI</th>\n",
       "      <td>281</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>Efo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PRS</th>\n",
       "      <td>118,124</td>\n",
       "      <td>{'p': {'array': False, 'type': 'FLOAT'}, 'r2':...</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>Gwas</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MR_EVE_MR</th>\n",
       "      <td>25,804,945</td>\n",
       "      <td>{'b': {'array': False, 'type': 'FLOAT'}, 'se':...</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>Gwas</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GEN_COR</th>\n",
       "      <td>840,960</td>\n",
       "      <td>{'h2_intercept_SE': {'array': False, 'type': '...</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>Gwas</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OBS_COR</th>\n",
       "      <td>17,932</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}, '...</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>Gwas</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GWAS_NLP</th>\n",
       "      <td>89,239,773</td>\n",
       "      <td>{'score': {'array': False, 'type': 'FLOAT'}, '...</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>Gwas</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GWAS_TO_LITERATURE</th>\n",
       "      <td>28,111,669</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>Literature</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>METAMAP_LITE</th>\n",
       "      <td>5,556</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}, '...</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GWAS_TO_LITERATURE_TRIPLE</th>\n",
       "      <td>17,531,153</td>\n",
       "      <td>{'pval': {'array': False, 'type': 'FLOAT'}, 'g...</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>LiteratureTriple</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OPENGWAS_TOPHITS</th>\n",
       "      <td>160,283</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}, '...</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>Variant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GWAS_TO_VARIANT</th>\n",
       "      <td>26,436</td>\n",
       "      <td>{'se': {'array': False, 'type': 'FLOAT'}, 'nca...</td>\n",
       "      <td>Gwas</td>\n",
       "      <td>Variant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TERM_TO_GENE</th>\n",
       "      <td>16,435</td>\n",
       "      <td>{'_source': {'array': False, 'type': 'STRING'}}</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "      <td>Gene</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SEMMEDDB_PREDICATE</th>\n",
       "      <td>5,584,547</td>\n",
       "      <td>{'count': {'array': False, 'type': 'INTEGER'},...</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BIORXIV_PREDICATE</th>\n",
       "      <td>32,648</td>\n",
       "      <td>{'count': {'array': False, 'type': 'INTEGER'},...</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDRXIV_PREDICATE</th>\n",
       "      <td>1,969</td>\n",
       "      <td>{'count': {'array': False, 'type': 'INTEGER'},...</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BIORXIV_TO_LIT</th>\n",
       "      <td>35,211</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>LiteratureTriple</td>\n",
       "      <td>Literature</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SEMMEDDB_TO_LIT</th>\n",
       "      <td>10,589,785</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>LiteratureTriple</td>\n",
       "      <td>Literature</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SEMMEDDB_SUB</th>\n",
       "      <td>5,584,547</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>LiteratureTriple</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BIORXIV_OBJ</th>\n",
       "      <td>32,651</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>LiteratureTriple</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BIORXIV_SUB</th>\n",
       "      <td>32,657</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>LiteratureTriple</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SEMMEDDB_OBJ</th>\n",
       "      <td>5,584,547</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>LiteratureTriple</td>\n",
       "      <td>LiteratureTerm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PATHWAY_CHILD_OF</th>\n",
       "      <td>2,463</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>Pathway</td>\n",
       "      <td>Pathway</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PROTEIN_IN_PATHWAY</th>\n",
       "      <td>121,873</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>Protein</td>\n",
       "      <td>Pathway</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STRING_INTERACT_WITH</th>\n",
       "      <td>827,184</td>\n",
       "      <td>{'score': {'array': False, 'type': 'FLOAT'}, '...</td>\n",
       "      <td>Protein</td>\n",
       "      <td>Protein</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VARIANT_TO_GENE</th>\n",
       "      <td>108,561</td>\n",
       "      <td>{'amino_acids': {'array': False, 'type': 'STRI...</td>\n",
       "      <td>Variant</td>\n",
       "      <td>Gene</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XQTL_SINGLE_SNP_MR_SNP_GENE</th>\n",
       "      <td>41,564</td>\n",
       "      <td>{'_source': {'array': True, 'type': 'LIST'}}</td>\n",
       "      <td>Variant</td>\n",
       "      <td>Gene</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   count  \\\n",
       "MONDO_MAP_EFO                      2,819   \n",
       "MONDO_MAP_UMLS                     8,247   \n",
       "OPENTARGETS_DRUG_TO_DISEASE        2,461   \n",
       "CPIC                                 375   \n",
       "OPENTARGETS_DRUG_TO_TARGET         6,534   \n",
       "EFO_CHILD_OF                      43,132   \n",
       "GENE_TO_DISEASE                    5,763   \n",
       "XQTL_MULTI_SNP_MR              3,015,233   \n",
       "XQTL_SINGLE_SNP_MR_GENE_GWAS   8,449,779   \n",
       "GENE_TO_PROTEIN                   19,142   \n",
       "EXPRESSED_IN                   2,918,240   \n",
       "GWAS_NLP_EFO                      12,302   \n",
       "GWAS_EFO_EBI                         281   \n",
       "PRS                              118,124   \n",
       "MR_EVE_MR                     25,804,945   \n",
       "GEN_COR                          840,960   \n",
       "OBS_COR                           17,932   \n",
       "GWAS_NLP                      89,239,773   \n",
       "GWAS_TO_LITERATURE            28,111,669   \n",
       "METAMAP_LITE                       5,556   \n",
       "GWAS_TO_LITERATURE_TRIPLE     17,531,153   \n",
       "OPENGWAS_TOPHITS                 160,283   \n",
       "GWAS_TO_VARIANT                   26,436   \n",
       "TERM_TO_GENE                      16,435   \n",
       "SEMMEDDB_PREDICATE             5,584,547   \n",
       "BIORXIV_PREDICATE                 32,648   \n",
       "MEDRXIV_PREDICATE                  1,969   \n",
       "BIORXIV_TO_LIT                    35,211   \n",
       "SEMMEDDB_TO_LIT               10,589,785   \n",
       "SEMMEDDB_SUB                   5,584,547   \n",
       "BIORXIV_OBJ                       32,651   \n",
       "BIORXIV_SUB                       32,657   \n",
       "SEMMEDDB_OBJ                   5,584,547   \n",
       "PATHWAY_CHILD_OF                   2,463   \n",
       "PROTEIN_IN_PATHWAY               121,873   \n",
       "STRING_INTERACT_WITH             827,184   \n",
       "VARIANT_TO_GENE                  108,561   \n",
       "XQTL_SINGLE_SNP_MR_SNP_GENE       41,564   \n",
       "\n",
       "                                                                     properties  \\\n",
       "MONDO_MAP_EFO                   {'_source': {'array': False, 'type': 'STRING'}}   \n",
       "MONDO_MAP_UMLS                  {'_source': {'array': False, 'type': 'STRING'}}   \n",
       "OPENTARGETS_DRUG_TO_DISEASE        {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "CPIC                          {'pharmgkb_level_of_evidence': {'array': False...   \n",
       "OPENTARGETS_DRUG_TO_TARGET    {'phase': {'array': False, 'type': 'STRING'}, ...   \n",
       "EFO_CHILD_OF                       {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "GENE_TO_DISEASE               {'last_updated': {'array': False, 'type': 'STR...   \n",
       "XQTL_MULTI_SNP_MR             {'p': {'array': False, 'type': 'FLOAT'}, 'se':...   \n",
       "XQTL_SINGLE_SNP_MR_GENE_GWAS  {'p': {'array': False, 'type': 'FLOAT'}, 'se':...   \n",
       "GENE_TO_PROTEIN                    {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "EXPRESSED_IN                  {'tpm': {'array': False, 'type': 'FLOAT'}, '_s...   \n",
       "GWAS_NLP_EFO                  {'score': {'array': False, 'type': 'FLOAT'}, '...   \n",
       "GWAS_EFO_EBI                       {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "PRS                           {'p': {'array': False, 'type': 'FLOAT'}, 'r2':...   \n",
       "MR_EVE_MR                     {'b': {'array': False, 'type': 'FLOAT'}, 'se':...   \n",
       "GEN_COR                       {'h2_intercept_SE': {'array': False, 'type': '...   \n",
       "OBS_COR                       {'_source': {'array': True, 'type': 'LIST'}, '...   \n",
       "GWAS_NLP                      {'score': {'array': False, 'type': 'FLOAT'}, '...   \n",
       "GWAS_TO_LITERATURE                 {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "METAMAP_LITE                  {'_source': {'array': True, 'type': 'LIST'}, '...   \n",
       "GWAS_TO_LITERATURE_TRIPLE     {'pval': {'array': False, 'type': 'FLOAT'}, 'g...   \n",
       "OPENGWAS_TOPHITS              {'_source': {'array': True, 'type': 'LIST'}, '...   \n",
       "GWAS_TO_VARIANT               {'se': {'array': False, 'type': 'FLOAT'}, 'nca...   \n",
       "TERM_TO_GENE                    {'_source': {'array': False, 'type': 'STRING'}}   \n",
       "SEMMEDDB_PREDICATE            {'count': {'array': False, 'type': 'INTEGER'},...   \n",
       "BIORXIV_PREDICATE             {'count': {'array': False, 'type': 'INTEGER'},...   \n",
       "MEDRXIV_PREDICATE             {'count': {'array': False, 'type': 'INTEGER'},...   \n",
       "BIORXIV_TO_LIT                     {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "SEMMEDDB_TO_LIT                    {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "SEMMEDDB_SUB                       {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "BIORXIV_OBJ                        {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "BIORXIV_SUB                        {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "SEMMEDDB_OBJ                       {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "PATHWAY_CHILD_OF                   {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "PROTEIN_IN_PATHWAY                 {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "STRING_INTERACT_WITH          {'score': {'array': False, 'type': 'FLOAT'}, '...   \n",
       "VARIANT_TO_GENE               {'amino_acids': {'array': False, 'type': 'STRI...   \n",
       "XQTL_SINGLE_SNP_MR_SNP_GENE        {'_source': {'array': True, 'type': 'LIST'}}   \n",
       "\n",
       "                                     from_node           to_node  \n",
       "MONDO_MAP_EFO                          Disease               Efo  \n",
       "MONDO_MAP_UMLS                         Disease    LiteratureTerm  \n",
       "OPENTARGETS_DRUG_TO_DISEASE               Drug           Disease  \n",
       "CPIC                                      Drug              Gene  \n",
       "OPENTARGETS_DRUG_TO_TARGET                Drug              Gene  \n",
       "EFO_CHILD_OF                               Efo               Efo  \n",
       "GENE_TO_DISEASE                           Gene           Disease  \n",
       "XQTL_MULTI_SNP_MR                         Gene              Gwas  \n",
       "XQTL_SINGLE_SNP_MR_GENE_GWAS              Gene              Gwas  \n",
       "GENE_TO_PROTEIN                           Gene           Protein  \n",
       "EXPRESSED_IN                              Gene            Tissue  \n",
       "GWAS_NLP_EFO                              Gwas               Efo  \n",
       "GWAS_EFO_EBI                              Gwas               Efo  \n",
       "PRS                                       Gwas              Gwas  \n",
       "MR_EVE_MR                                 Gwas              Gwas  \n",
       "GEN_COR                                   Gwas              Gwas  \n",
       "OBS_COR                                   Gwas              Gwas  \n",
       "GWAS_NLP                                  Gwas              Gwas  \n",
       "GWAS_TO_LITERATURE                        Gwas        Literature  \n",
       "METAMAP_LITE                              Gwas    LiteratureTerm  \n",
       "GWAS_TO_LITERATURE_TRIPLE                 Gwas  LiteratureTriple  \n",
       "OPENGWAS_TOPHITS                          Gwas           Variant  \n",
       "GWAS_TO_VARIANT                           Gwas           Variant  \n",
       "TERM_TO_GENE                    LiteratureTerm              Gene  \n",
       "SEMMEDDB_PREDICATE              LiteratureTerm    LiteratureTerm  \n",
       "BIORXIV_PREDICATE               LiteratureTerm    LiteratureTerm  \n",
       "MEDRXIV_PREDICATE               LiteratureTerm    LiteratureTerm  \n",
       "BIORXIV_TO_LIT                LiteratureTriple        Literature  \n",
       "SEMMEDDB_TO_LIT               LiteratureTriple        Literature  \n",
       "SEMMEDDB_SUB                  LiteratureTriple    LiteratureTerm  \n",
       "BIORXIV_OBJ                   LiteratureTriple    LiteratureTerm  \n",
       "BIORXIV_SUB                   LiteratureTriple    LiteratureTerm  \n",
       "SEMMEDDB_OBJ                  LiteratureTriple    LiteratureTerm  \n",
       "PATHWAY_CHILD_OF                       Pathway           Pathway  \n",
       "PROTEIN_IN_PATHWAY                     Protein           Pathway  \n",
       "STRING_INTERACT_WITH                   Protein           Protein  \n",
       "VARIANT_TO_GENE                        Variant              Gene  \n",
       "XQTL_SINGLE_SNP_MR_SNP_GENE            Variant              Gene  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "meta_rel_df = pd.DataFrame.from_dict(metadata[\"edges\"], orient=\"index\").merge(\n",
    "    pd.DataFrame.from_dict(\n",
    "        {_[\"rel\"]: _ for _ in metadata[\"connections\"]}, orient=\"index\"\n",
    "    )[[\"from_node\", \"to_node\"]],\n",
    "    left_index=True,\n",
    "    right_index=True,\n",
    ")\n",
    "\n",
    "(\n",
    "    meta_rel_df.sort_values(by=[\"from_node\", \"to_node\"]).assign(\n",
    "        count=lambda df: df[\"count\"].apply(lambda x: f\"{x:,}\")\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "informational-indie",
   "metadata": {
    "papermill": {
     "duration": 0.028925,
     "end_time": "2021-03-22T10:46:18.030025",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.001100",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Schema plot"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "electric-monitor",
   "metadata": {
    "papermill": {
     "duration": 0.014955,
     "end_time": "2021-03-22T10:46:18.064021",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.049066",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "We can generate a network diagram of the graph db schema using `networkx`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "grateful-rapid",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:18.098560Z",
     "iopub.status.busy": "2021-03-22T10:46:18.098217Z",
     "iopub.status.idle": "2021-03-22T10:46:18.100621Z",
     "shell.execute_reply": "2021-03-22T10:46:18.100281Z"
    },
    "papermill": {
     "duration": 0.020885,
     "end_time": "2021-03-22T10:46:18.100694",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.079809",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "graph = nx.from_pandas_edgelist(meta_rel_df, source=\"from_node\", target=\"to_node\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "corresponding-recorder",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:18.135351Z",
     "iopub.status.busy": "2021-03-22T10:46:18.134965Z",
     "iopub.status.idle": "2021-03-22T10:46:18.551217Z",
     "shell.execute_reply": "2021-03-22T10:46:18.551675Z"
    },
    "papermill": {
     "duration": 0.436452,
     "end_time": "2021-03-22T10:46:18.551837",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.115385",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "nx.draw(\n",
    "    G=graph,\n",
    "    pos=nx.kamada_kawai_layout(graph),\n",
    "    with_labels=True,\n",
    "    node_color=\"white\",\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "standard-invitation",
   "metadata": {
    "papermill": {
     "duration": 0.016735,
     "end_time": "2021-03-22T10:46:18.591504",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.574769",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "A detailed version of the shema plot can be obtained from the API:\n",
    "\n",
    "![schema_plot](https://api.epigraphdb.org/meta/schema?graphviz=true&plot=true)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fiscal-polyester",
   "metadata": {
    "papermill": {
     "duration": 0.016705,
     "end_time": "2021-03-22T10:46:18.623618",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.606913",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Search for specific node"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "annual-exhibit",
   "metadata": {
    "papermill": {
     "duration": 0.016179,
     "end_time": "2021-03-22T10:46:18.655534",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.639355",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Users can use [the explorer on the Web UI](http://dev.epigraphdb.org/explore) to search for a specific node by:\n",
    "\n",
    "- fuzzy matching by \"name\" field.\n",
    "- exact matching by \"ID\" field if you know the its ID (e.g. the ID to a GWAS from IEU GWAS Database).\n",
    "\n",
    "Here we show how these are done at the API level using `Gwas` nodes as an example."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "floating-acquisition",
   "metadata": {
    "papermill": {
     "duration": 0.016488,
     "end_time": "2021-03-22T10:46:18.688480",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.671992",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "First we need to know what the \"ID\" and \"name\" fields are for the meta nodes using `GET /meta/nodes/id-name-schema`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "thirty-syntax",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:18.726381Z",
     "iopub.status.busy": "2021-03-22T10:46:18.726030Z",
     "iopub.status.idle": "2021-03-22T10:46:18.863583Z",
     "shell.execute_reply": "2021-03-22T10:46:18.864101Z"
    },
    "papermill": {
     "duration": 0.159271,
     "end_time": "2021-03-22T10:46:18.864224",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.704953",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Disease': {'id': 'id', 'name': 'label'},\n",
       " 'Drug': {'id': 'label', 'name': 'label'},\n",
       " 'Efo': {'id': 'id', 'name': 'value'},\n",
       " 'Gene': {'id': 'ensembl_id', 'name': 'name'},\n",
       " 'Gwas': {'id': 'id', 'name': 'trait'},\n",
       " 'Literature': {'id': 'id', 'name': 'id'},\n",
       " 'LiteratureTerm': {'id': 'id', 'name': 'name'},\n",
       " 'LiteratureTriple': {'id': 'id', 'name': 'name'},\n",
       " 'Pathway': {'id': 'id', 'name': 'name'},\n",
       " 'Protein': {'id': 'uniprot_id', 'name': 'uniprot_id'},\n",
       " 'Tissue': {'id': 'id', 'name': 'name'},\n",
       " 'Variant': {'id': 'name', 'name': 'name'}}"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r = requests.get(f\"{API_URL}/meta/nodes/id-name-schema\")\n",
    "r.raise_for_status()\n",
    "\n",
    "meta_node_fields = r.json()\n",
    "meta_node_fields"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "increasing-yield",
   "metadata": {
    "papermill": {
     "duration": 0.021819,
     "end_time": "2021-03-22T10:46:18.906403",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.884584",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Fuzzy matching"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "statutory-feedback",
   "metadata": {
    "papermill": {
     "duration": 0.017427,
     "end_time": "2021-03-22T10:46:18.946237",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.928810",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Here we search for nodes can contain \"body mass index\" in their traits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "tested-bottle",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:18.984611Z",
     "iopub.status.busy": "2021-03-22T10:46:18.981919Z",
     "iopub.status.idle": "2021-03-22T10:46:19.143190Z",
     "shell.execute_reply": "2021-03-22T10:46:19.142620Z"
    },
    "papermill": {
     "duration": 0.180714,
     "end_time": "2021-03-22T10:46:19.143294",
     "exception": false,
     "start_time": "2021-03-22T10:46:18.962580",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'metadata': {'empty_results': False,\n",
      "              'query': 'MATCH (node: Gwas) WHERE node.trait =~ \"(?i).*body '\n",
      "                       'mass index.*\" RETURN node LIMIT 10;',\n",
      "              'total_seconds': 0.021328},\n",
      " 'results': [{'node': {'_id': 'ieu-a-1089',\n",
      "                       '_name': 'Body mass index',\n",
      "                       '_source': ['OpenGWAS-2020-10-13'],\n",
      "                       'author': 'Wood',\n",
      "                       'build': 'HG19/GRCh37',\n",
      "                       'category': 'Risk factor',\n",
      "                       'id': 'ieu-a-1089',\n",
      "                       'mr': '0',\n",
      "                       'note': 'Dominance model? If so then not necessarily of '\n",
      "                               'value for MR; Results from interim Biobank '\n",
      "                               'release enriched for smokers; could lead to '\n",
      "                               'bias through collider issues in MR',\n",
      "                       'nsnp': '8654252',\n",
      "                       'pmid': '26961502.0',\n",
      "                       'population': 'European',\n",
      "                       'sample_size': '120286.0',\n",
      "                       'sex': 'Males and Females',\n",
      "                       'subcategory': 'Anthropometric',\n",
      "                       'trait': 'Body mass index',\n",
      "                       'year': '2016.0'}},\n",
      "             {'node': {'_id': 'ieu-a-974',\n",
      "                       '_name': 'Body mass index',\n",
      "                       '_source': ['OpenGWAS-2020-10-13'],\n",
      "                       'author': 'Locke AE',\n",
      "                       'build': 'HG19/GRCh37',\n",
      "                       'category': 'Risk factor',\n",
      "                       'id': 'ieu-a-974',\n",
      "                       'mr': '1',\n",
      "                       'nsnp': '2494613',\n",
      "                       'pmid': '25673413.0',\n",
      "                       'population': 'European',\n",
      "                       'sample_size': '171977.0',\n",
      "                       'sd': '4.77',\n",
      "                       'sex': 'Females',\n",
      "                       'subcategory': 'Anthropometric',\n",
      "                       'trait': 'Body mass index',\n",
      "                       'year': '2015.0'}},\n",
      "             {'node': {'_id': 'ieu-a-95',\n",
      "                       '_name': 'Body mass index',\n",
      "                       '_source': ['OpenGWAS-2020-10-13'],\n",
      "                       'author': 'Randall JC',\n",
      "                       'build': 'HG19/GRCh37',\n",
      "                       'category': 'Risk factor',\n",
      "                       'id': 'ieu-a-95',\n",
      "                       'mr': '1',\n",
      "                       'nsnp': '2736876',\n",
      "                       'pmid': '23754948.0',\n",
      "                       'population': 'European',\n",
      "                       'sample_size': '73137.0',\n",
      "                       'sd': '4.77',\n",
      "                       'sex': 'Females',\n",
      "                       'subcategory': 'Anthropometric',\n",
      "                       'trait': 'Body mass index',\n",
      "                       'year': '2013.0'}},\n",
      "             {'node': {'_id': 'ebi-a-GCST004904',\n",
      "                       '_name': 'Body mass index',\n",
      "                       '_source': ['OpenGWAS-2020-\n"
     ]
    }
   ],
   "source": [
    "name = \"body mass index\"\n",
    "\n",
    "r = requests.get(f\"{API_URL}/meta/nodes/Gwas/search\", params={\"name\": name})\n",
    "r.raise_for_status()\n",
    "\n",
    "print(pformat(r.json())[:3000])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "common-draft",
   "metadata": {
    "papermill": {
     "duration": 0.022528,
     "end_time": "2021-03-22T10:46:19.187856",
     "exception": false,
     "start_time": "2021-03-22T10:46:19.165328",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Exact matching"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "thousand-listing",
   "metadata": {
    "papermill": {
     "duration": 0.020625,
     "end_time": "2021-03-22T10:46:19.231203",
     "exception": false,
     "start_time": "2021-03-22T10:46:19.210578",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Similarly, we can exact match a specific node by its ID."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "challenging-offering",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:19.272770Z",
     "iopub.status.busy": "2021-03-22T10:46:19.272432Z",
     "iopub.status.idle": "2021-03-22T10:46:19.413948Z",
     "shell.execute_reply": "2021-03-22T10:46:19.413550Z"
    },
    "papermill": {
     "duration": 0.164685,
     "end_time": "2021-03-22T10:46:19.414047",
     "exception": false,
     "start_time": "2021-03-22T10:46:19.249362",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'metadata': {'empty_results': False,\n",
      "              'query': 'MATCH (node: Gwas {id: \"ieu-a-2\"}) RETURN node LIMIT '\n",
      "                       '10;',\n",
      "              'total_seconds': 0.012118},\n",
      " 'results': [{'node': {'_id': 'ieu-a-2',\n",
      "                       '_name': 'Body mass index',\n",
      "                       '_source': ['OpenGWAS-2020-10-13'],\n",
      "                       'author': 'Locke AE',\n",
      "                       'build': 'HG19/GRCh37',\n",
      "                       'category': 'Risk factor',\n",
      "                       'id': 'ieu-a-2',\n",
      "                       'mr': '1',\n",
      "                       'nsnp': '2555511',\n",
      "                       'pmid': '25673413.0',\n",
      "                       'population': 'Mixed',\n",
      "                       'sample_size': '339224.0',\n",
      "                       'sd': '4.77',\n",
      "                       'sex': 'Males and Females',\n",
      "                       'subcategory': 'Anthropometric',\n",
      "                       'trait': 'Body mass index',\n",
      "                       'year': '2015.0'}}]}\n"
     ]
    }
   ],
   "source": [
    "id = \"ieu-a-2\"\n",
    "\n",
    "r = requests.get(f\"{API_URL}/meta/nodes/Gwas/search\", params={\"id\": id})\n",
    "r.raise_for_status()\n",
    "\n",
    "print(pformat(r.json())[:3000])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "vietnamese-companion",
   "metadata": {
    "papermill": {
     "duration": 0.030923,
     "end_time": "2021-03-22T10:46:19.474984",
     "exception": false,
     "start_time": "2021-03-22T10:46:19.444061",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Cypher (advanced)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "proper-fantasy",
   "metadata": {
    "papermill": {
     "duration": 0.017481,
     "end_time": "2021-03-22T10:46:19.516610",
     "exception": false,
     "start_time": "2021-03-22T10:46:19.499129",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Advanced users that are familiar with Neo4j Cypher can query the database using Cypher directly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "incorporated-salem",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:19.558119Z",
     "iopub.status.busy": "2021-03-22T10:46:19.557752Z",
     "iopub.status.idle": "2021-03-22T10:46:19.705581Z",
     "shell.execute_reply": "2021-03-22T10:46:19.704903Z"
    },
    "papermill": {
     "duration": 0.171252,
     "end_time": "2021-03-22T10:46:19.705741",
     "exception": false,
     "start_time": "2021-03-22T10:46:19.534489",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'metadata': {'empty_results': False,\n",
      "              'query': 'MATCH (exposure:Gwas)-[mr:MR_EVE_MR]->(outcome:Gwas) '\n",
      "                       'WHERE exposure.trait = \"Body mass index\" RETURN '\n",
      "                       'exposure, outcome, mr LIMIT 2',\n",
      "              'total_seconds': 0.011951},\n",
      " 'results': [{'exposure': {'_id': 'ieu-a-974',\n",
      "                           '_name': 'Body mass index',\n",
      "                           '_source': ['OpenGWAS-2020-10-13'],\n",
      "                           'author': 'Locke AE',\n",
      "                           'build': 'HG19/GRCh37',\n",
      "                           'category': 'Risk factor',\n",
      "                           'id': 'ieu-a-974',\n",
      "                           'mr': '1',\n",
      "                           'nsnp': '2494613',\n",
      "                           'pmid': '25673413.0',\n",
      "                           'population': 'European',\n",
      "                           'sample_size': '171977.0',\n",
      "                           'sd': '4.77',\n",
      "                           'sex': 'Females',\n",
      "                           'subcategory': 'Anthropometric',\n",
      "                           'trait': 'Body mass index',\n",
      "                           'year': '2015.0'},\n",
      "              'mr': {'_source': ['MR-EvE-2021-03-10'],\n",
      "                     'b': 0.0289678834636601,\n",
      "                     'ci_low': -0.165265753333202,\n",
      "                     'ci_upp': 0.22320152026052198,\n",
      "                     'method': 'Simple mean',\n",
      "                     'moescore': 0.77,\n",
      "                     'nsnp': 37,\n",
      "                     'pval': 0.771725286643359,\n",
      "                     'se': 0.0990987942841133,\n",
      "                     'selection': 'HF'},\n",
      "              'outcome': {'_id': 'prot-a-1729',\n",
      "                          '_name': 'Galectin-7',\n",
      "                          '_source': ['OpenGWAS-2020-10-13'],\n",
      "                          'author': 'Sun BB',\n",
      "                          'build': 'HG19/GRCh37',\n",
      "                          'category': 'Immune system',\n",
      "                          'id': 'prot-a-1729',\n",
      "                          'mr': '1',\n",
      "                          'nsnp': '10534735',\n",
      "                          'pmid': '29875488.0',\n",
      "                          'population': 'European',\n",
      "                          'sample_size': '3301.0',\n",
      "                          'sex': 'Males and Females',\n",
      "                          'subcategory': 'Protein',\n",
      "                          'trait': 'Galectin-7',\n",
      "                          'year': '2018.0'}},\n",
      "             {'exposure': {'_id': 'ieu-a-974',\n",
      "                           '_name': 'Body mass index',\n",
      "                           '_source': ['OpenGWAS-2020-10-13'],\n",
      "                           'author': 'Locke AE',\n",
      "                           'build': 'HG19/GRCh37',\n",
      "                           'category': 'Risk factor',\n",
      "                           'id': 'ieu-a-974',\n",
      "                           'mr': '1',\n",
      "                           'nsnp': '2494613',\n",
      "                           'pmid': '25673413.0',\n",
      "                           'population': 'European',\n",
      "                           'sample_size': '171977.0',\n",
      "    \n"
     ]
    }
   ],
   "source": [
    "query = \"\"\"\n",
    "    MATCH (exposure:Gwas)-[mr:MR_EVE_MR]->(outcome:Gwas) \n",
    "    WHERE exposure.trait = \"Body mass index\"\n",
    "    RETURN exposure, outcome, mr LIMIT 2\n",
    "\"\"\"\n",
    "\n",
    "r = requests.post(f\"{API_URL}/cypher\", json={\"query\": query})\n",
    "r.raise_for_status()\n",
    "\n",
    "print(pformat(r.json())[:3000])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "defined-guinea",
   "metadata": {
    "papermill": {
     "duration": 0.031733,
     "end_time": "2021-03-22T10:46:19.770472",
     "exception": false,
     "start_time": "2021-03-22T10:46:19.738739",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Again for the detailed documentation on the API endpoints please visit:\n",
    "\n",
    "- The Swagger interface: http://api.epigraphdb.org\n",
    "- The sections regarding API endpoints on the documentation site: http://docs.epigraphdb.org/api/api-endpoints/"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.7"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 5.210277,
   "end_time": "2021-03-22T10:46:21.001532",
   "environment_variables": {},
   "exception": null,
   "input_path": "general-examples/platform-meta-functionalities.ipynb",
   "output_path": "general-examples/platform-meta-functionalities.ipynb",
   "parameters": {
    "API_URL": "https://api.epigraphdb.org"
   },
   "start_time": "2021-03-22T10:46:15.791255",
   "version": "2.3.3"
  },
  "toc-autonumbering": false
 },
 "nbformat": 4,
 "nbformat_minor": 5
}