paper-case-studies/case-3-literature-triangulation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "impressed-chinese",
   "metadata": {
    "papermill": {
     "duration": 0.037923,
     "end_time": "2021-03-22T10:46:57.656922",
     "exception": false,
     "start_time": "2021-03-22T10:46:57.618999",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Case study 3 - Triangulating causal estimates with literature evidence\n",
    "\n",
    "The biomedical literature contains a wealth of information than far exceeds our capacity for systematic manual extraction. For this reason, there are many existing literature mining methods to extract the key concepts and content. Here we use data from [SemMedDB](https://skr3.nlm.nih.gov/SemMedDB/), a well established database that provides subject-predicate-object triples from all PubMed titles and abstracts. Using a subset of this data we created MELODI-presto (https://melodi-presto.mrcieu.ac.uk/), a method to assign triples to any given biomedical query via a PubMed search and some basic enrichment, and have applied this systematically to traits represented in EpiGraphDB. This allows us to identify overlapping terms connecting any set of GWAS traits, e.g. exposure and disease outcome. From here we can attempt to triangulate causal estimates, and conversely, check the mechanisms identified from the literature against the causal evidence. \n",
    "\n",
    "This case study goes as follows:\n",
    "\n",
    "- For an exposure trait we identify its causal evidence against other outcome traits,\n",
    "  then link the outcomes to diseases\n",
    "- Then for an exposure-outcome pair we look for its literature evidence,\n",
    "  and the mechanisms (as SemMed triples) identified from the literature evidence\n",
    "- Finally we look into detail the mechanisms specifically related to an overlapping term and visualise these mechanisms."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "proof-scholarship",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:57.713040Z",
     "iopub.status.busy": "2021-03-22T10:46:57.712648Z",
     "iopub.status.idle": "2021-03-22T10:46:58.473299Z",
     "shell.execute_reply": "2021-03-22T10:46:58.473695Z"
    },
    "papermill": {
     "duration": 0.786198,
     "end_time": "2021-03-22T10:46:58.473957",
     "exception": false,
     "start_time": "2021-03-22T10:46:57.687759",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import json\n",
    "import networkx as nx\n",
    "from networkx.drawing.nx_pydot import graphviz_layout\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import requests\n",
    "import math\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "turned-beatles",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:58.528947Z",
     "iopub.status.busy": "2021-03-22T10:46:58.528314Z",
     "iopub.status.idle": "2021-03-22T10:46:58.530576Z",
     "shell.execute_reply": "2021-03-22T10:46:58.529907Z"
    },
    "papermill": {
     "duration": 0.031928,
     "end_time": "2021-03-22T10:46:58.530719",
     "exception": false,
     "start_time": "2021-03-22T10:46:58.498791",
     "status": "completed"
    },
    "tags": [
     "parameters"
    ]
   },
   "outputs": [],
   "source": [
    "# Default parameters\n",
    "API_URL = \"https://api.epigraphdb.org\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "identical-quantum",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:58.579647Z",
     "iopub.status.busy": "2021-03-22T10:46:58.579033Z",
     "iopub.status.idle": "2021-03-22T10:46:58.582207Z",
     "shell.execute_reply": "2021-03-22T10:46:58.581766Z"
    },
    "papermill": {
     "duration": 0.02988,
     "end_time": "2021-03-22T10:46:58.582333",
     "exception": false,
     "start_time": "2021-03-22T10:46:58.552453",
     "status": "completed"
    },
    "tags": [
     "injected-parameters"
    ]
   },
   "outputs": [],
   "source": [
    "# Parameters\n",
    "API_URL = \"https://api.epigraphdb.org\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "chicken-saying",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:58.633877Z",
     "iopub.status.busy": "2021-03-22T10:46:58.633484Z",
     "iopub.status.idle": "2021-03-22T10:46:58.772275Z",
     "shell.execute_reply": "2021-03-22T10:46:58.772560Z"
    },
    "papermill": {
     "duration": 0.16833,
     "end_time": "2021-03-22T10:46:58.772665",
     "exception": false,
     "start_time": "2021-03-22T10:46:58.604335",
     "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": "rolled-original",
   "metadata": {
    "papermill": {
     "duration": 0.032379,
     "end_time": "2021-03-22T10:46:58.830640",
     "exception": false,
     "start_time": "2021-03-22T10:46:58.798261",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Parameters\n",
    "\n",
    "Here we set the starting trait, which we will use to explore associated disease traits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "connected-length",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:58.886262Z",
     "iopub.status.busy": "2021-03-22T10:46:58.885776Z",
     "iopub.status.idle": "2021-03-22T10:46:58.887342Z",
     "shell.execute_reply": "2021-03-22T10:46:58.887812Z"
    },
    "papermill": {
     "duration": 0.023624,
     "end_time": "2021-03-22T10:46:58.887967",
     "exception": false,
     "start_time": "2021-03-22T10:46:58.864343",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "STARTING_TRAIT = \"Sleep duration\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "other-cemetery",
   "metadata": {
    "papermill": {
     "duration": 0.017356,
     "end_time": "2021-03-22T10:46:58.922614",
     "exception": false,
     "start_time": "2021-03-22T10:46:58.905258",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Get traits MR association \n",
    "\n",
    "Given an exposure trait, find all traits with causal evidence. This method searches the causal evidence data for cases where our exposure trait has a potential casual effect on an outcome trait. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "different-shipping",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:58.967272Z",
     "iopub.status.busy": "2021-03-22T10:46:58.966875Z",
     "iopub.status.idle": "2021-03-22T10:46:59.801562Z",
     "shell.execute_reply": "2021-03-22T10:46:59.802051Z"
    },
    "papermill": {
     "duration": 0.861698,
     "end_time": "2021-03-22T10:46:59.802192",
     "exception": false,
     "start_time": "2021-03-22T10:46:58.940494",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1178 entries, 0 to 1177\n",
      "Data columns (total 10 columns):\n",
      " #   Column          Non-Null Count  Dtype  \n",
      "---  ------          --------------  -----  \n",
      " 0   exposure.id     1178 non-null   object \n",
      " 1   exposure.trait  1178 non-null   object \n",
      " 2   outcome.id      1178 non-null   object \n",
      " 3   outcome.trait   1178 non-null   object \n",
      " 4   mr.b            1178 non-null   float64\n",
      " 5   mr.se           1178 non-null   float64\n",
      " 6   mr.pval         1178 non-null   float64\n",
      " 7   mr.method       1178 non-null   object \n",
      " 8   mr.selection    1178 non-null   object \n",
      " 9   mr.moescore     1178 non-null   float64\n",
      "dtypes: float64(4), object(6)\n",
      "memory usage: 92.2+ KB\n",
      "None\n"
     ]
    },
    {
     "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>exposure.id</th>\n",
       "      <th>exposure.trait</th>\n",
       "      <th>outcome.id</th>\n",
       "      <th>outcome.trait</th>\n",
       "      <th>mr.b</th>\n",
       "      <th>mr.se</th>\n",
       "      <th>mr.pval</th>\n",
       "      <th>mr.method</th>\n",
       "      <th>mr.selection</th>\n",
       "      <th>mr.moescore</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>met-a-353</td>\n",
       "      <td>Cortisol</td>\n",
       "      <td>0.059177</td>\n",
       "      <td>0.001386</td>\n",
       "      <td>0.0</td>\n",
       "      <td>FE IVW</td>\n",
       "      <td>DF</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>met-a-307</td>\n",
       "      <td>Cholesterol</td>\n",
       "      <td>-0.024282</td>\n",
       "      <td>0.000054</td>\n",
       "      <td>0.0</td>\n",
       "      <td>FE IVW</td>\n",
       "      <td>DF</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>ieu-a-99</td>\n",
       "      <td>Hip circumference</td>\n",
       "      <td>-0.288662</td>\n",
       "      <td>0.005928</td>\n",
       "      <td>0.0</td>\n",
       "      <td>FE IVW</td>\n",
       "      <td>DF</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>ieu-a-59</td>\n",
       "      <td>Hip circumference</td>\n",
       "      <td>0.072874</td>\n",
       "      <td>0.000162</td>\n",
       "      <td>0.0</td>\n",
       "      <td>FE IVW</td>\n",
       "      <td>DF</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>ieu-a-58</td>\n",
       "      <td>Hip circumference</td>\n",
       "      <td>0.072874</td>\n",
       "      <td>0.000162</td>\n",
       "      <td>0.0</td>\n",
       "      <td>FE IVW</td>\n",
       "      <td>DF</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  exposure.id  exposure.trait outcome.id      outcome.trait      mr.b  \\\n",
       "0  ieu-a-1088  Sleep duration  met-a-353           Cortisol  0.059177   \n",
       "1  ieu-a-1088  Sleep duration  met-a-307        Cholesterol -0.024282   \n",
       "2  ieu-a-1088  Sleep duration   ieu-a-99  Hip circumference -0.288662   \n",
       "3  ieu-a-1088  Sleep duration   ieu-a-59  Hip circumference  0.072874   \n",
       "4  ieu-a-1088  Sleep duration   ieu-a-58  Hip circumference  0.072874   \n",
       "\n",
       "      mr.se  mr.pval mr.method mr.selection  mr.moescore  \n",
       "0  0.001386      0.0    FE IVW           DF          1.0  \n",
       "1  0.000054      0.0    FE IVW           DF          1.0  \n",
       "2  0.005928      0.0    FE IVW           DF          1.0  \n",
       "3  0.000162      0.0    FE IVW           DF          1.0  \n",
       "4  0.000162      0.0    FE IVW           DF          1.0  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def get_mr(trait):\n",
    "    endpoint = \"/mr\"\n",
    "    url = f\"{API_URL}{endpoint}\"\n",
    "    params = {\n",
    "        \"exposure_trait\": trait,\n",
    "        \"pval_threshold\": 1e-10,\n",
    "    }\n",
    "    r = requests.get(url, params=params)\n",
    "    r.raise_for_status()\n",
    "    mr_df = pd.json_normalize(r.json()[\"results\"])\n",
    "    return mr_df\n",
    "\n",
    "\n",
    "mr_df = get_mr(STARTING_TRAIT)\n",
    "print(mr_df.info())\n",
    "mr_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "spectacular-priest",
   "metadata": {
    "papermill": {
     "duration": 0.020097,
     "end_time": "2021-03-22T10:46:59.838826",
     "exception": false,
     "start_time": "2021-03-22T10:46:59.818729",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Map outcome traits to disease\n",
    "\n",
    "For this example, we are interested in traits mapped to a disease node. To do this we utilise the mapping from GWAS trait to Disease via EFO term. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "becoming-haiti",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:46:59.882437Z",
     "iopub.status.busy": "2021-03-22T10:46:59.882016Z",
     "iopub.status.idle": "2021-03-22T10:49:15.982950Z",
     "shell.execute_reply": "2021-03-22T10:49:15.983507Z"
    },
    "papermill": {
     "duration": 136.127175,
     "end_time": "2021-03-22T10:49:15.983685",
     "exception": false,
     "start_time": "2021-03-22T10:46:59.856510",
     "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>outcome.trait</th>\n",
       "      <th>disease</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Alzheimer's disease</td>\n",
       "      <td>[Alzheimer disease]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Non-cancer illness code, self-reported: mening...</td>\n",
       "      <td>[bacterial meningitis, infectious meningitis, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Diagnoses - secondary ICD10: K30 Dyspepsia</td>\n",
       "      <td>[dyspepsia]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Anorexia Nervosa</td>\n",
       "      <td>[anorexia nervosa]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Lymphomas</td>\n",
       "      <td>[Burkitt lymphoma, lymphoma]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Parkinson's disease</td>\n",
       "      <td>[Parkinson disease]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Cardioembolic stroke</td>\n",
       "      <td>[stroke disorder]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Non-cancer illness code  self-reported: gout</td>\n",
       "      <td>[gout]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Insomnia complaints</td>\n",
       "      <td>[insomnia (disease)]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Cancer code, self-reported: cervical cancer</td>\n",
       "      <td>[cancer]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Non-cancer illness code, self-reported: allerg...</td>\n",
       "      <td>[allergic disease]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Sleeplessness / insomnia</td>\n",
       "      <td>[insomnia (disease)]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Coronary heart disease</td>\n",
       "      <td>[coronary artery disease]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Non-cancer illness code, self-reported: bronch...</td>\n",
       "      <td>[chronic bronchitis, bronchitis]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Diagnoses - main ICD10: K35 Acute appendicitis</td>\n",
       "      <td>[appendicitis]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Cancer code  self-reported: lung cancer</td>\n",
       "      <td>[cancer, lung carcinoma]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Non-cancer illness code, self-reported: multip...</td>\n",
       "      <td>[relapsing-remitting multiple sclerosis, multi...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        outcome.trait  \\\n",
       "0                                 Alzheimer's disease   \n",
       "1   Non-cancer illness code, self-reported: mening...   \n",
       "2          Diagnoses - secondary ICD10: K30 Dyspepsia   \n",
       "3                                    Anorexia Nervosa   \n",
       "4                                           Lymphomas   \n",
       "5                                 Parkinson's disease   \n",
       "6                                Cardioembolic stroke   \n",
       "7        Non-cancer illness code  self-reported: gout   \n",
       "8                                 Insomnia complaints   \n",
       "9         Cancer code, self-reported: cervical cancer   \n",
       "10  Non-cancer illness code, self-reported: allerg...   \n",
       "11                           Sleeplessness / insomnia   \n",
       "12                             Coronary heart disease   \n",
       "13  Non-cancer illness code, self-reported: bronch...   \n",
       "14     Diagnoses - main ICD10: K35 Acute appendicitis   \n",
       "15            Cancer code  self-reported: lung cancer   \n",
       "16  Non-cancer illness code, self-reported: multip...   \n",
       "\n",
       "                                              disease  \n",
       "0                                 [Alzheimer disease]  \n",
       "1   [bacterial meningitis, infectious meningitis, ...  \n",
       "2                                         [dyspepsia]  \n",
       "3                                  [anorexia nervosa]  \n",
       "4                        [Burkitt lymphoma, lymphoma]  \n",
       "5                                 [Parkinson disease]  \n",
       "6                                   [stroke disorder]  \n",
       "7                                              [gout]  \n",
       "8                                [insomnia (disease)]  \n",
       "9                                            [cancer]  \n",
       "10                                 [allergic disease]  \n",
       "11                               [insomnia (disease)]  \n",
       "12                          [coronary artery disease]  \n",
       "13                   [chronic bronchitis, bronchitis]  \n",
       "14                                     [appendicitis]  \n",
       "15                           [cancer, lung carcinoma]  \n",
       "16  [relapsing-remitting multiple sclerosis, multi...  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def trait_to_disease(row):\n",
    "    endpoint = \"/ontology/gwas-efo-disease\"\n",
    "    url = f\"{API_URL}{endpoint}\"\n",
    "    params = {\n",
    "        \"trait\": row[\"outcome.trait\"],\n",
    "    }\n",
    "    r = requests.get(url, params=params)\n",
    "    r.raise_for_status()\n",
    "    disease_df = pd.json_normalize(r.json()[\"results\"])\n",
    "    if \"disease.label\" in disease_df:\n",
    "        return disease_df[\"disease.label\"].drop_duplicates()\n",
    "\n",
    "\n",
    "outcome_disease_df = (\n",
    "    mr_df[[\"outcome.trait\"]]\n",
    "    .drop_duplicates()\n",
    "    .assign(disease=lambda df: df.apply(trait_to_disease, axis=1))\n",
    "    .dropna()\n",
    "    .reset_index(drop=True)\n",
    ")\n",
    "outcome_disease_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "stable-haven",
   "metadata": {
    "papermill": {
     "duration": 0.022731,
     "end_time": "2021-03-22T10:49:16.039603",
     "exception": false,
     "start_time": "2021-03-22T10:49:16.016872",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Take one example to look for literature evidence\n",
    "\n",
    "For the multiple `exposure -> outcome` relationships as reported from the table above,\n",
    "here we look at the literature evidence for one pair in detail:\n",
    "\n",
    "- Trait X: \"Sleep duration\" \n",
    "  ([ieu-a-1088](https://gwas.mrcieu.ac.uk/datasets/ieu-a-1088/)) \n",
    "- Trait Y: \"Coronary heart disease\" \n",
    "  ([ieu-a-6](https://gwas.mrcieu.ac.uk/datasets/ieu-a-6/)). \n",
    "\n",
    "The following looks for enriched triples of information (Subject-Predicate-Object) associated with our two traits. These have been derived via PubMed searches and corresponding [SemMedDB](https://skr3.nlm.nih.gov/SemMedDB/) data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "christian-liver",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:16.083420Z",
     "iopub.status.busy": "2021-03-22T10:49:16.083048Z",
     "iopub.status.idle": "2021-03-22T10:49:18.193134Z",
     "shell.execute_reply": "2021-03-22T10:49:18.192463Z"
    },
    "papermill": {
     "duration": 2.137116,
     "end_time": "2021-03-22T10:49:18.193278",
     "exception": false,
     "start_time": "2021-03-22T10:49:16.056162",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INTERACTS_WITH        719\n",
      "STIMULATES             96\n",
      "INHIBITS               93\n",
      "NEG_INTERACTS_WITH     51\n",
      "higher_than            23\n",
      "COEXISTS_WITH          15\n",
      "NEG_INHIBITS            3\n",
      "Name: s1.predicate, dtype: int64\n",
      "TREATS                 191\n",
      "STIMULATES             132\n",
      "CAUSES                 120\n",
      "PREVENTS               117\n",
      "ASSOCIATED_WITH        103\n",
      "PREDISPOSES             83\n",
      "AFFECTS                 71\n",
      "INTERACTS_WITH          62\n",
      "INHIBITS                47\n",
      "COEXISTS_WITH           34\n",
      "DISRUPTS                19\n",
      "NEG_PREVENTS            13\n",
      "higher_than              4\n",
      "AUGMENTS                 2\n",
      "NEG_ASSOCIATED_WITH      1\n",
      "same_as                  1\n",
      "Name: s2.predicate, dtype: int64\n"
     ]
    },
    {
     "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>gwas.id</th>\n",
       "      <th>gwas.trait</th>\n",
       "      <th>gs1.pval</th>\n",
       "      <th>gs1.localCount</th>\n",
       "      <th>st1.name</th>\n",
       "      <th>s1.id</th>\n",
       "      <th>s1.subject_id</th>\n",
       "      <th>s1.object_id</th>\n",
       "      <th>s1.predicate</th>\n",
       "      <th>st.name</th>\n",
       "      <th>st.type</th>\n",
       "      <th>s2.id</th>\n",
       "      <th>s2.subject_id</th>\n",
       "      <th>s2.object_id</th>\n",
       "      <th>s2.predicate</th>\n",
       "      <th>gs2.pval</th>\n",
       "      <th>gs2.localCount</th>\n",
       "      <th>st2.name</th>\n",
       "      <th>assoc_gwas.id</th>\n",
       "      <th>assoc_gwas.trait</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000095</td>\n",
       "      <td>2</td>\n",
       "      <td>Sulpiride</td>\n",
       "      <td>C0038803:INTERACTS_WITH:C0003596</td>\n",
       "      <td>C0038803</td>\n",
       "      <td>C0003596</td>\n",
       "      <td>INTERACTS_WITH</td>\n",
       "      <td>Apomorphine</td>\n",
       "      <td>[orch, phsu]</td>\n",
       "      <td>C0003596:TREATS:C0242350</td>\n",
       "      <td>C0003596</td>\n",
       "      <td>C0242350</td>\n",
       "      <td>TREATS</td>\n",
       "      <td>0.045878</td>\n",
       "      <td>3</td>\n",
       "      <td>Erectile dysfunction</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Glycine</td>\n",
       "      <td>C0017890:STIMULATES:C0033487</td>\n",
       "      <td>C0017890</td>\n",
       "      <td>C0033487</td>\n",
       "      <td>STIMULATES</td>\n",
       "      <td>propofol</td>\n",
       "      <td>[orch, phsu]</td>\n",
       "      <td>C0033487:AUGMENTS:C0151744</td>\n",
       "      <td>C0033487</td>\n",
       "      <td>C0151744</td>\n",
       "      <td>AUGMENTS</td>\n",
       "      <td>0.002682</td>\n",
       "      <td>2</td>\n",
       "      <td>Myocardial Ischemia</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Glycine</td>\n",
       "      <td>C0017890:STIMULATES:C0033487</td>\n",
       "      <td>C0017890</td>\n",
       "      <td>C0033487</td>\n",
       "      <td>STIMULATES</td>\n",
       "      <td>propofol</td>\n",
       "      <td>[orch, phsu]</td>\n",
       "      <td>C0033487:TREATS:C0010054</td>\n",
       "      <td>C0033487</td>\n",
       "      <td>C0010054</td>\n",
       "      <td>TREATS</td>\n",
       "      <td>0.001627</td>\n",
       "      <td>2</td>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>2</td>\n",
       "      <td>gamma-aminobutyric acid</td>\n",
       "      <td>C0016904:STIMULATES:C0033487</td>\n",
       "      <td>C0016904</td>\n",
       "      <td>C0033487</td>\n",
       "      <td>STIMULATES</td>\n",
       "      <td>propofol</td>\n",
       "      <td>[orch, phsu]</td>\n",
       "      <td>C0033487:AUGMENTS:C0151744</td>\n",
       "      <td>C0033487</td>\n",
       "      <td>C0151744</td>\n",
       "      <td>AUGMENTS</td>\n",
       "      <td>0.002682</td>\n",
       "      <td>2</td>\n",
       "      <td>Myocardial Ischemia</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>2</td>\n",
       "      <td>gamma-aminobutyric acid</td>\n",
       "      <td>C0016904:STIMULATES:C0033487</td>\n",
       "      <td>C0016904</td>\n",
       "      <td>C0033487</td>\n",
       "      <td>STIMULATES</td>\n",
       "      <td>propofol</td>\n",
       "      <td>[orch, phsu]</td>\n",
       "      <td>C0033487:TREATS:C0010054</td>\n",
       "      <td>C0033487</td>\n",
       "      <td>C0010054</td>\n",
       "      <td>TREATS</td>\n",
       "      <td>0.001627</td>\n",
       "      <td>2</td>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      gwas.id      gwas.trait  gs1.pval  gs1.localCount  \\\n",
       "0  ieu-a-1088  Sleep duration  0.000095               2   \n",
       "1  ieu-a-1088  Sleep duration  0.000002               2   \n",
       "2  ieu-a-1088  Sleep duration  0.000002               2   \n",
       "3  ieu-a-1088  Sleep duration  0.000019               2   \n",
       "4  ieu-a-1088  Sleep duration  0.000019               2   \n",
       "\n",
       "                  st1.name                             s1.id s1.subject_id  \\\n",
       "0                Sulpiride  C0038803:INTERACTS_WITH:C0003596      C0038803   \n",
       "1                  Glycine      C0017890:STIMULATES:C0033487      C0017890   \n",
       "2                  Glycine      C0017890:STIMULATES:C0033487      C0017890   \n",
       "3  gamma-aminobutyric acid      C0016904:STIMULATES:C0033487      C0016904   \n",
       "4  gamma-aminobutyric acid      C0016904:STIMULATES:C0033487      C0016904   \n",
       "\n",
       "  s1.object_id    s1.predicate      st.name       st.type  \\\n",
       "0     C0003596  INTERACTS_WITH  Apomorphine  [orch, phsu]   \n",
       "1     C0033487      STIMULATES     propofol  [orch, phsu]   \n",
       "2     C0033487      STIMULATES     propofol  [orch, phsu]   \n",
       "3     C0033487      STIMULATES     propofol  [orch, phsu]   \n",
       "4     C0033487      STIMULATES     propofol  [orch, phsu]   \n",
       "\n",
       "                        s2.id s2.subject_id s2.object_id s2.predicate  \\\n",
       "0    C0003596:TREATS:C0242350      C0003596     C0242350       TREATS   \n",
       "1  C0033487:AUGMENTS:C0151744      C0033487     C0151744     AUGMENTS   \n",
       "2    C0033487:TREATS:C0010054      C0033487     C0010054       TREATS   \n",
       "3  C0033487:AUGMENTS:C0151744      C0033487     C0151744     AUGMENTS   \n",
       "4    C0033487:TREATS:C0010054      C0033487     C0010054       TREATS   \n",
       "\n",
       "   gs2.pval  gs2.localCount                   st2.name assoc_gwas.id  \\\n",
       "0  0.045878               3       Erectile dysfunction       ieu-a-6   \n",
       "1  0.002682               2        Myocardial Ischemia       ieu-a-6   \n",
       "2  0.001627               2  Coronary Arteriosclerosis       ieu-a-6   \n",
       "3  0.002682               2        Myocardial Ischemia       ieu-a-6   \n",
       "4  0.001627               2  Coronary Arteriosclerosis       ieu-a-6   \n",
       "\n",
       "         assoc_gwas.trait  \n",
       "0  Coronary heart disease  \n",
       "1  Coronary heart disease  \n",
       "2  Coronary heart disease  \n",
       "3  Coronary heart disease  \n",
       "4  Coronary heart disease  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def get_gwas_pair_literature(gwas_id, assoc_gwas_id):\n",
    "    endpoint = \"/literature/gwas/pairwise\"\n",
    "    url = f\"{API_URL}{endpoint}\"\n",
    "    params = {\n",
    "        \"gwas_id\": gwas_id,\n",
    "        \"assoc_gwas_id\": assoc_gwas_id,\n",
    "        \"by_gwas_id\": \"true\",\n",
    "        \"pval_threshold\": 1e-1,\n",
    "        \"semmantic_types\": [\"nusq\", \"dsyn\"],\n",
    "        \"blacklist\": \"True\",\n",
    "        \"limit\": 1000,\n",
    "    }\n",
    "    r = requests.get(url, params=params)\n",
    "    r.raise_for_status()\n",
    "    lit_df = pd.json_normalize(r.json()[\"results\"])\n",
    "    return lit_df\n",
    "\n",
    "\n",
    "GWAS_ID_X = \"ieu-a-1088\"\n",
    "GWAS_ID_Y = \"ieu-a-6\"\n",
    "lit_df = get_gwas_pair_literature(GWAS_ID_X, GWAS_ID_Y)\n",
    "\n",
    "# Predicate counts for SemMed triples for trait X\n",
    "print(lit_df[\"s1.predicate\"].value_counts())\n",
    "# Predicate counts for SemMed triples for trait Y\n",
    "print(lit_df[\"s2.predicate\"].value_counts())\n",
    "\n",
    "lit_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "unexpected-albania",
   "metadata": {
    "papermill": {
     "duration": 0.033576,
     "end_time": "2021-03-22T10:49:18.261462",
     "exception": false,
     "start_time": "2021-03-22T10:49:18.227886",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Filter the data by predicates\n",
    "\n",
    "Sometimes it is preferable to filter the SemMedDB data, e.g. to remove less informative Predicates, such as `COEXISTS_WITH` and `ASSOCIATED_WITH`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "expected-finding",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:18.308793Z",
     "iopub.status.busy": "2021-03-22T10:49:18.308356Z",
     "iopub.status.idle": "2021-03-22T10:49:18.323677Z",
     "shell.execute_reply": "2021-03-22T10:49:18.323373Z"
    },
    "papermill": {
     "duration": 0.03912,
     "end_time": "2021-03-22T10:49:18.323759",
     "exception": false,
     "start_time": "2021-03-22T10:49:18.284639",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INTERACTS_WITH        649\n",
      "INHIBITS               87\n",
      "STIMULATES             66\n",
      "NEG_INTERACTS_WITH     26\n",
      "higher_than            22\n",
      "Name: s1.predicate, dtype: int64\n",
      "TREATS                 180\n",
      "STIMULATES             132\n",
      "CAUSES                 120\n",
      "PREVENTS               117\n",
      "PREDISPOSES             83\n",
      "AFFECTS                 70\n",
      "INTERACTS_WITH          62\n",
      "INHIBITS                46\n",
      "DISRUPTS                19\n",
      "NEG_PREVENTS            13\n",
      "higher_than              4\n",
      "AUGMENTS                 2\n",
      "NEG_ASSOCIATED_WITH      1\n",
      "same_as                  1\n",
      "Name: s2.predicate, dtype: int64\n"
     ]
    },
    {
     "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>st1.name</th>\n",
       "      <th>s1.predicate</th>\n",
       "      <th>st.name</th>\n",
       "      <th>s2.predicate</th>\n",
       "      <th>st2.name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>felbamate</td>\n",
       "      <td>INTERACTS_WITH</td>\n",
       "      <td>ethanol</td>\n",
       "      <td>PREVENTS</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>TREATS</td>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>felbamate</td>\n",
       "      <td>INTERACTS_WITH</td>\n",
       "      <td>ethanol</td>\n",
       "      <td>AFFECTS</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>aspirin</td>\n",
       "      <td>INHIBITS</td>\n",
       "      <td>Oral anticoagulants</td>\n",
       "      <td>TREATS</td>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>PREDISPOSES</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>felbamate</td>\n",
       "      <td>INTERACTS_WITH</td>\n",
       "      <td>ethanol</td>\n",
       "      <td>TREATS</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>TREATS</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>felbamate</td>\n",
       "      <td>INTERACTS_WITH</td>\n",
       "      <td>ethanol</td>\n",
       "      <td>PREVENTS</td>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>felbamate</td>\n",
       "      <td>INTERACTS_WITH</td>\n",
       "      <td>ethanol</td>\n",
       "      <td>PREDISPOSES</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>PREDISPOSES</td>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         st1.name        s1.predicate              st.name s2.predicate  \\\n",
       "0       felbamate      INTERACTS_WITH              ethanol     PREVENTS   \n",
       "1  Corticosterone  NEG_INTERACTS_WITH               leptin       TREATS   \n",
       "2       felbamate      INTERACTS_WITH              ethanol      AFFECTS   \n",
       "3         aspirin            INHIBITS  Oral anticoagulants       TREATS   \n",
       "4  Corticosterone  NEG_INTERACTS_WITH               leptin  PREDISPOSES   \n",
       "5       felbamate      INTERACTS_WITH              ethanol       TREATS   \n",
       "6  Corticosterone  NEG_INTERACTS_WITH               leptin       TREATS   \n",
       "7       felbamate      INTERACTS_WITH              ethanol     PREVENTS   \n",
       "8       felbamate      INTERACTS_WITH              ethanol  PREDISPOSES   \n",
       "9  Corticosterone  NEG_INTERACTS_WITH               leptin  PREDISPOSES   \n",
       "\n",
       "                    st2.name  \n",
       "0     Coronary heart disease  \n",
       "1  Coronary Arteriosclerosis  \n",
       "2     Coronary heart disease  \n",
       "3  Coronary Arteriosclerosis  \n",
       "4     Coronary heart disease  \n",
       "5     Coronary heart disease  \n",
       "6     Coronary heart disease  \n",
       "7  Coronary Arteriosclerosis  \n",
       "8     Coronary heart disease  \n",
       "9  Coronary Arteriosclerosis  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# filter out some predicates that are not informative\n",
    "pred_filter = [\"COEXISTS_WITH\", \"ASSOCIATED_WITH\"]\n",
    "\n",
    "lit_df_filter = lit_df[\n",
    "    ~lit_df[\"s1.predicate\"].isin(pred_filter)\n",
    "    & ~lit_df[\"s2.predicate\"].isin(pred_filter)\n",
    "]\n",
    "print(lit_df_filter[\"s1.predicate\"].value_counts())\n",
    "print(lit_df_filter[\"s2.predicate\"].value_counts())\n",
    "\n",
    "keep_columns = [\"st1.name\", \"s1.predicate\", \"st.name\", \"s2.predicate\", \"st2.name\"]\n",
    "(\n",
    "    lit_df_filter.sort_values([\"gs1.pval\", \"gs2.pval\"], ascending=True)\n",
    "    .head(n=10)\n",
    "    .reset_index(drop=True)[keep_columns]\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "outside-entity",
   "metadata": {
    "papermill": {
     "duration": 0.01941,
     "end_time": "2021-03-22T10:49:18.368156",
     "exception": false,
     "start_time": "2021-03-22T10:49:18.348746",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Literature results\n",
    "\n",
    "If we explore the full table in ```lit_df_filter```, we can see lots of links between the two traits, pinned on specific overlapping terms.\n",
    "\n",
    "We can summarise the [SemMedDB semantic type](https://mmtx.nlm.nih.gov/MMTx/semanticTypes.shtml) and number of overlapping terms:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "functioning-valve",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:18.411830Z",
     "iopub.status.busy": "2021-03-22T10:49:18.411433Z",
     "iopub.status.idle": "2021-03-22T10:49:18.414135Z",
     "shell.execute_reply": "2021-03-22T10:49:18.414423Z"
    },
    "papermill": {
     "duration": 0.026858,
     "end_time": "2021-03-22T10:49:18.414541",
     "exception": false,
     "start_time": "2021-03-22T10:49:18.387683",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ethanol                                 571\n",
       "melatonin                                88\n",
       "agonists                                 31\n",
       "leptin                                   26\n",
       "Oral anticoagulants                      24\n",
       "Metoprolol                               20\n",
       "Antidepressive Agents                    12\n",
       "Anesthetics                               8\n",
       "morphine                                  7\n",
       "Cytochrome P450                           6\n",
       "Somatotropin                              6\n",
       "Medroxyprogesterone 17-Acetate            5\n",
       "Hormones                                  4\n",
       "mitogen-activated protein kinase p38      4\n",
       "propofol                                  4\n",
       "Benzodiazepines                           4\n",
       "Glycoproteins                             3\n",
       "Thiopental                                2\n",
       "AHSA1                                     2\n",
       "MAPK14                                    2\n",
       "AIMP2                                     2\n",
       "POLDIP2                                   2\n",
       "Diazepam                                  2\n",
       "MAPK1                                     2\n",
       "Protein Kinase C                          2\n",
       "Midazolam                                 2\n",
       "CRK                                       2\n",
       "GRAP2                                     2\n",
       "glutathione                               1\n",
       "Apomorphine                               1\n",
       "gamma hydroxybutyrate                     1\n",
       "Neostigmine                               1\n",
       "Luteolin                                  1\n",
       "Name: st.name, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lit_counts = lit_df_filter[\"st.name\"].value_counts()\n",
    "lit_counts"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "tested-highland",
   "metadata": {
    "papermill": {
     "duration": 0.017101,
     "end_time": "2021-03-22T10:49:18.449305",
     "exception": false,
     "start_time": "2021-03-22T10:49:18.432204",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Note, the SemMedDB semantic types have been pre-filtered to only include a subset of possibilities. \n",
    "\n",
    "Further examples of these term IDs and descriptions can be found here - [https://metamap.nlm.nih.gov/Docs/SemanticTypes_2018AB.txt](https://metamap.nlm.nih.gov/Docs/SemanticTypes_2018AB.txt)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "nervous-regression",
   "metadata": {
    "papermill": {
     "duration": 0.017442,
     "end_time": "2021-03-22T10:49:18.484130",
     "exception": false,
     "start_time": "2021-03-22T10:49:18.466688",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "#### Plotting the overlapping term frequencies\n",
    "\n",
    "We can also visualise the above table as a bar chart. In this case we will remove Ethanol as it is an outlier."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "logical-cowboy",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:18.526726Z",
     "iopub.status.busy": "2021-03-22T10:49:18.524923Z",
     "iopub.status.idle": "2021-03-22T10:49:21.637327Z",
     "shell.execute_reply": "2021-03-22T10:49:21.637865Z"
    },
    "papermill": {
     "duration": 3.136734,
     "end_time": "2021-03-22T10:49:21.638049",
     "exception": false,
     "start_time": "2021-03-22T10:49:18.501315",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# remove Ethanol\n",
    "lit_counts = lit_counts[lit_counts < 100]\n",
    "\n",
    "# convert to df\n",
    "lit_counts = lit_counts.reset_index(name=\"counts\").rename(columns={\"index\": \"st.name\"})\n",
    "\n",
    "g = sns.catplot(\n",
    "    y=\"st.name\",\n",
    "    x=\"counts\",\n",
    "    data=lit_counts,\n",
    "    height=5,\n",
    "    kind=\"bar\",\n",
    "    palette=\"muted\",\n",
    "    orient=\"horizontal\",\n",
    "    dodge=False,\n",
    ")\n",
    "# add numbers to the bars\n",
    "for i, v in enumerate(lit_counts[\"counts\"]):\n",
    "    plt.text(v + 1, i + 0.25, str(v))\n",
    "g.set(ylabel=\"Overlapping term\", xlabel=\"Term frequency\")\n",
    "plt.savefig(\"overlaps.png\", dpi=1000)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "divine-harris",
   "metadata": {
    "papermill": {
     "duration": 0.036386,
     "end_time": "2021-03-22T10:49:21.712269",
     "exception": false,
     "start_time": "2021-03-22T10:49:21.675883",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Look in detail at one overlapping term\n",
    "\n",
    "Here we look at cases where `leptin` is the central overlapping term."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "instant-julian",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:21.774605Z",
     "iopub.status.busy": "2021-03-22T10:49:21.773876Z",
     "iopub.status.idle": "2021-03-22T10:49:21.776545Z",
     "shell.execute_reply": "2021-03-22T10:49:21.776184Z"
    },
    "papermill": {
     "duration": 0.044082,
     "end_time": "2021-03-22T10:49:21.776628",
     "exception": false,
     "start_time": "2021-03-22T10:49:21.732546",
     "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>gwas.id</th>\n",
       "      <th>gwas.trait</th>\n",
       "      <th>gs1.pval</th>\n",
       "      <th>gs1.localCount</th>\n",
       "      <th>st1.name</th>\n",
       "      <th>s1.id</th>\n",
       "      <th>s1.subject_id</th>\n",
       "      <th>s1.object_id</th>\n",
       "      <th>s1.predicate</th>\n",
       "      <th>st.name</th>\n",
       "      <th>st.type</th>\n",
       "      <th>s2.id</th>\n",
       "      <th>s2.subject_id</th>\n",
       "      <th>s2.object_id</th>\n",
       "      <th>s2.predicate</th>\n",
       "      <th>gs2.pval</th>\n",
       "      <th>gs2.localCount</th>\n",
       "      <th>st2.name</th>\n",
       "      <th>assoc_gwas.id</th>\n",
       "      <th>assoc_gwas.trait</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>746</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>C0010124</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>[aapp, gngm, horm]</td>\n",
       "      <td>C0299583:TREATS:C0520679</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>C0520679</td>\n",
       "      <td>TREATS</td>\n",
       "      <td>0.055690</td>\n",
       "      <td>2</td>\n",
       "      <td>Sleep Apnea, Obstructive</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>748</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>C0010124</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>[aapp, gngm, horm]</td>\n",
       "      <td>C0299583:STIMULATES:C0031727</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>C0031727</td>\n",
       "      <td>STIMULATES</td>\n",
       "      <td>0.009237</td>\n",
       "      <td>2</td>\n",
       "      <td>Phosphotransferases</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>749</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>C0010124</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>[aapp, gngm, horm]</td>\n",
       "      <td>C0299583:STIMULATES:3717</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>3717</td>\n",
       "      <td>STIMULATES</td>\n",
       "      <td>0.031956</td>\n",
       "      <td>2</td>\n",
       "      <td>JAK2</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>750</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>C0010124</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>[aapp, gngm, horm]</td>\n",
       "      <td>C0299583:STIMULATES:C0169661</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>C0169661</td>\n",
       "      <td>STIMULATES</td>\n",
       "      <td>0.031956</td>\n",
       "      <td>2</td>\n",
       "      <td>Janus kinase 2</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>751</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>C0010124</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>[aapp, gngm, horm]</td>\n",
       "      <td>C0299583:INTERACTS_WITH:C0021641</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>C0021641</td>\n",
       "      <td>INTERACTS_WITH</td>\n",
       "      <td>0.078833</td>\n",
       "      <td>2</td>\n",
       "      <td>Insulin</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>754</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>C0010124</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>[aapp, gngm, horm]</td>\n",
       "      <td>C0299583:STIMULATES:C0389071</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>C0389071</td>\n",
       "      <td>STIMULATES</td>\n",
       "      <td>0.073994</td>\n",
       "      <td>2</td>\n",
       "      <td>Adiponectin</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>758</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>C0010124</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>[aapp, gngm, horm]</td>\n",
       "      <td>C0299583:PREDISPOSES:C0010068</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>C0010068</td>\n",
       "      <td>PREDISPOSES</td>\n",
       "      <td>0.001627</td>\n",
       "      <td>2</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>760</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>C0010124</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>[aapp, gngm, horm]</td>\n",
       "      <td>C0299583:INTERACTS_WITH:C0751973</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>C0751973</td>\n",
       "      <td>INTERACTS_WITH</td>\n",
       "      <td>0.001627</td>\n",
       "      <td>2</td>\n",
       "      <td>Proteome</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>761</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>C0010124</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>[aapp, gngm, horm]</td>\n",
       "      <td>C0299583:CAUSES:C0856169</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>C0856169</td>\n",
       "      <td>CAUSES</td>\n",
       "      <td>0.016383</td>\n",
       "      <td>2</td>\n",
       "      <td>Endothelial dysfunction</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>763</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>Sleep duration</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>C0010124</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>NEG_INTERACTS_WITH</td>\n",
       "      <td>leptin</td>\n",
       "      <td>[aapp, gngm, horm]</td>\n",
       "      <td>C0299583:PREDISPOSES:C0007222</td>\n",
       "      <td>C0299583</td>\n",
       "      <td>C0007222</td>\n",
       "      <td>PREDISPOSES</td>\n",
       "      <td>0.099101</td>\n",
       "      <td>2</td>\n",
       "      <td>Cardiovascular Diseases</td>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        gwas.id      gwas.trait  gs1.pval  gs1.localCount        st1.name  \\\n",
       "746  ieu-a-1088  Sleep duration  0.000002               2  Corticosterone   \n",
       "748  ieu-a-1088  Sleep duration  0.000002               2  Corticosterone   \n",
       "749  ieu-a-1088  Sleep duration  0.000002               2  Corticosterone   \n",
       "750  ieu-a-1088  Sleep duration  0.000002               2  Corticosterone   \n",
       "751  ieu-a-1088  Sleep duration  0.000002               2  Corticosterone   \n",
       "754  ieu-a-1088  Sleep duration  0.000002               2  Corticosterone   \n",
       "758  ieu-a-1088  Sleep duration  0.000002               2  Corticosterone   \n",
       "760  ieu-a-1088  Sleep duration  0.000002               2  Corticosterone   \n",
       "761  ieu-a-1088  Sleep duration  0.000002               2  Corticosterone   \n",
       "763  ieu-a-1088  Sleep duration  0.000002               2  Corticosterone   \n",
       "\n",
       "                                    s1.id s1.subject_id s1.object_id  \\\n",
       "746  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
       "748  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
       "749  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
       "750  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
       "751  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
       "754  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
       "758  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
       "760  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
       "761  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
       "763  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
       "\n",
       "           s1.predicate st.name             st.type  \\\n",
       "746  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
       "748  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
       "749  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
       "750  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
       "751  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
       "754  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
       "758  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
       "760  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
       "761  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
       "763  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
       "\n",
       "                                s2.id s2.subject_id s2.object_id  \\\n",
       "746          C0299583:TREATS:C0520679      C0299583     C0520679   \n",
       "748      C0299583:STIMULATES:C0031727      C0299583     C0031727   \n",
       "749          C0299583:STIMULATES:3717      C0299583         3717   \n",
       "750      C0299583:STIMULATES:C0169661      C0299583     C0169661   \n",
       "751  C0299583:INTERACTS_WITH:C0021641      C0299583     C0021641   \n",
       "754      C0299583:STIMULATES:C0389071      C0299583     C0389071   \n",
       "758     C0299583:PREDISPOSES:C0010068      C0299583     C0010068   \n",
       "760  C0299583:INTERACTS_WITH:C0751973      C0299583     C0751973   \n",
       "761          C0299583:CAUSES:C0856169      C0299583     C0856169   \n",
       "763     C0299583:PREDISPOSES:C0007222      C0299583     C0007222   \n",
       "\n",
       "       s2.predicate  gs2.pval  gs2.localCount                  st2.name  \\\n",
       "746          TREATS  0.055690               2  Sleep Apnea, Obstructive   \n",
       "748      STIMULATES  0.009237               2       Phosphotransferases   \n",
       "749      STIMULATES  0.031956               2                      JAK2   \n",
       "750      STIMULATES  0.031956               2            Janus kinase 2   \n",
       "751  INTERACTS_WITH  0.078833               2                   Insulin   \n",
       "754      STIMULATES  0.073994               2               Adiponectin   \n",
       "758     PREDISPOSES  0.001627               2    Coronary heart disease   \n",
       "760  INTERACTS_WITH  0.001627               2                  Proteome   \n",
       "761          CAUSES  0.016383               2   Endothelial dysfunction   \n",
       "763     PREDISPOSES  0.099101               2   Cardiovascular Diseases   \n",
       "\n",
       "    assoc_gwas.id        assoc_gwas.trait  \n",
       "746       ieu-a-6  Coronary heart disease  \n",
       "748       ieu-a-6  Coronary heart disease  \n",
       "749       ieu-a-6  Coronary heart disease  \n",
       "750       ieu-a-6  Coronary heart disease  \n",
       "751       ieu-a-6  Coronary heart disease  \n",
       "754       ieu-a-6  Coronary heart disease  \n",
       "758       ieu-a-6  Coronary heart disease  \n",
       "760       ieu-a-6  Coronary heart disease  \n",
       "761       ieu-a-6  Coronary heart disease  \n",
       "763       ieu-a-6  Coronary heart disease  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "focus_term = \"leptin\"\n",
    "lit_detail = lit_df_filter[lit_df_filter[\"st.name\"] == focus_term]\n",
    "lit_detail.head(n=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "killing-stylus",
   "metadata": {
    "papermill": {
     "duration": 0.019612,
     "end_time": "2021-03-22T10:49:21.815789",
     "exception": false,
     "start_time": "2021-03-22T10:49:21.796177",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "We can create a network diagram to visualise these relationships."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "tribal-rwanda",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:21.861724Z",
     "iopub.status.busy": "2021-03-22T10:49:21.861376Z",
     "iopub.status.idle": "2021-03-22T10:49:21.878051Z",
     "shell.execute_reply": "2021-03-22T10:49:21.878373Z"
    },
    "papermill": {
     "duration": 0.042337,
     "end_time": "2021-03-22T10:49:21.878473",
     "exception": false,
     "start_time": "2021-03-22T10:49:21.836136",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        gwas.id      gwas.trait  gs1.pval  gs1.localCount        st1.name  \\\n",
      "746  ieu-a-1088  SLEEP DURATION  0.000002               2  Corticosterone   \n",
      "748  ieu-a-1088  SLEEP DURATION  0.000002               2  Corticosterone   \n",
      "749  ieu-a-1088  SLEEP DURATION  0.000002               2  Corticosterone   \n",
      "750  ieu-a-1088  SLEEP DURATION  0.000002               2  Corticosterone   \n",
      "751  ieu-a-1088  SLEEP DURATION  0.000002               2  Corticosterone   \n",
      "\n",
      "                                    s1.id s1.subject_id s1.object_id  \\\n",
      "746  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
      "748  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
      "749  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
      "750  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
      "751  C0010124:NEG_INTERACTS_WITH:C0299583      C0010124     C0299583   \n",
      "\n",
      "           s1.predicate st.name             st.type  \\\n",
      "746  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
      "748  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
      "749  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
      "750  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
      "751  NEG_INTERACTS_WITH  leptin  [aapp, gngm, horm]   \n",
      "\n",
      "                                s2.id s2.subject_id s2.object_id  \\\n",
      "746          C0299583:TREATS:C0520679      C0299583     C0520679   \n",
      "748      C0299583:STIMULATES:C0031727      C0299583     C0031727   \n",
      "749          C0299583:STIMULATES:3717      C0299583         3717   \n",
      "750      C0299583:STIMULATES:C0169661      C0299583     C0169661   \n",
      "751  C0299583:INTERACTS_WITH:C0021641      C0299583     C0021641   \n",
      "\n",
      "       s2.predicate  gs2.pval  gs2.localCount                  st2.name  \\\n",
      "746          TREATS  0.055690               2  Sleep Apnea, Obstructive   \n",
      "748      STIMULATES  0.009237               2       Phosphotransferases   \n",
      "749      STIMULATES  0.031956               2                      JAK2   \n",
      "750      STIMULATES  0.031956               2            Janus kinase 2   \n",
      "751  INTERACTS_WITH  0.078833               2                   Insulin   \n",
      "\n",
      "    assoc_gwas.id        assoc_gwas.trait  \n",
      "746       ieu-a-6  CORONARY HEART DISEASE  \n",
      "748       ieu-a-6  CORONARY HEART DISEASE  \n",
      "749       ieu-a-6  CORONARY HEART DISEASE  \n",
      "750       ieu-a-6  CORONARY HEART DISEASE  \n",
      "751       ieu-a-6  CORONARY HEART DISEASE  \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/yiliu/.local/share/virtualenvs/epigraphdb-notebooks/lib/python3.7/site-packages/pandas/core/indexing.py:966: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self.obj[item] = s\n"
     ]
    }
   ],
   "source": [
    "# set things up\n",
    "exposure = \"Sleep duration\".upper()\n",
    "outcome = \"Coronary heart disease\".upper()\n",
    "lit_detail.loc[:, \"gwas.trait\"] = lit_detail[\"gwas.trait\"].apply(lambda x: x.upper())\n",
    "lit_detail.loc[:, \"assoc_gwas.trait\"] = lit_detail[\"assoc_gwas.trait\"].apply(\n",
    "    lambda x: x.upper()\n",
    ")\n",
    "print(lit_detail.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "governmental-tunisia",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:21.951902Z",
     "iopub.status.busy": "2021-03-22T10:49:21.941132Z",
     "iopub.status.idle": "2021-03-22T10:49:21.955414Z",
     "shell.execute_reply": "2021-03-22T10:49:21.955057Z"
    },
    "papermill": {
     "duration": 0.057699,
     "end_time": "2021-03-22T10:49:21.955496",
     "exception": false,
     "start_time": "2021-03-22T10:49:21.897797",
     "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>node</th>\n",
       "      <th>node_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SLEEP DURATION</td>\n",
       "      <td>trait</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td>trait</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Sleep Apnea, Obstructive</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Phosphotransferases</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>JAK2</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Janus kinase 2</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Insulin</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Adiponectin</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Coronary heart disease</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Proteome</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Endothelial dysfunction</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Cardiovascular Diseases</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Congestive heart failure</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Acute Coronary Syndrome</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Stable angina</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Metabolic syndrome</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Acute myocardial infarction</td>\n",
       "      <td>literature_term</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>leptin</td>\n",
       "      <td>overlapping_term</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           node         node_type\n",
       "0                SLEEP DURATION             trait\n",
       "1        CORONARY HEART DISEASE             trait\n",
       "2                Corticosterone   literature_term\n",
       "3      Sleep Apnea, Obstructive   literature_term\n",
       "4           Phosphotransferases   literature_term\n",
       "5                          JAK2   literature_term\n",
       "6                Janus kinase 2   literature_term\n",
       "7                       Insulin   literature_term\n",
       "8                   Adiponectin   literature_term\n",
       "9        Coronary heart disease   literature_term\n",
       "10                     Proteome   literature_term\n",
       "11      Endothelial dysfunction   literature_term\n",
       "12      Cardiovascular Diseases   literature_term\n",
       "13     Congestive heart failure   literature_term\n",
       "14      Acute Coronary Syndrome   literature_term\n",
       "15    Coronary Arteriosclerosis   literature_term\n",
       "16                Stable angina   literature_term\n",
       "17           Metabolic syndrome   literature_term\n",
       "18  Acute myocardial infarction   literature_term\n",
       "19                       leptin  overlapping_term"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# nodes\n",
    "nodes = (\n",
    "    pd.concat(\n",
    "        [\n",
    "            # exposure trait\n",
    "            lit_detail[[\"gwas.trait\"]]\n",
    "            .rename(columns={\"gwas.trait\": \"node\"})\n",
    "            .assign(node_type=\"trait\"),\n",
    "            # outcome trait\n",
    "            lit_detail[[\"assoc_gwas.trait\"]]\n",
    "            .rename(columns={\"assoc_gwas.trait\": \"node\"})\n",
    "            .assign(node_type=\"trait\"),\n",
    "            # subject nodes\n",
    "            lit_detail[[\"st1.name\"]]\n",
    "            .rename(columns={\"st1.name\": \"node\"})\n",
    "            .assign(node_type=\"literature_term\"),\n",
    "            # object nodes\n",
    "            lit_detail[[\"st2.name\"]]\n",
    "            .rename(columns={\"st2.name\": \"node\"})\n",
    "            .assign(node_type=\"literature_term\"),\n",
    "            # overlapping term nodes\n",
    "            lit_detail[[\"st.name\"]]\n",
    "            .rename(columns={\"st.name\": \"node\"})\n",
    "            .assign(node_type=\"overlapping_term\"),\n",
    "        ]\n",
    "    )\n",
    "    .drop_duplicates()\n",
    "    .reset_index(drop=True)\n",
    ")\n",
    "nodes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "mediterranean-diploma",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:22.026787Z",
     "iopub.status.busy": "2021-03-22T10:49:22.007091Z",
     "iopub.status.idle": "2021-03-22T10:49:22.066283Z",
     "shell.execute_reply": "2021-03-22T10:49:22.066714Z"
    },
    "papermill": {
     "duration": 0.089621,
     "end_time": "2021-03-22T10:49:22.066856",
     "exception": false,
     "start_time": "2021-03-22T10:49:21.977235",
     "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>source</th>\n",
       "      <th>target</th>\n",
       "      <th>label</th>\n",
       "      <th>weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SLEEP DURATION</td>\n",
       "      <td>Corticosterone</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Sleep Apnea, Obstructive</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Phosphotransferases</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>JAK2</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Janus kinase 2</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Insulin</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Adiponectin</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Coronary heart disease</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Proteome</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Endothelial dysfunction</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Cardiovascular Diseases</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Congestive heart failure</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Acute Coronary Syndrome</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Stable angina</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Metabolic syndrome</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Acute myocardial infarction</td>\n",
       "      <td>CORONARY HEART DISEASE</td>\n",
       "      <td></td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Corticosterone</td>\n",
       "      <td>leptin</td>\n",
       "      <td>NEG_INTERACTS_WITH(2, 2.48e-06)</td>\n",
       "      <td>5.605808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Sleep Apnea, Obstructive</td>\n",
       "      <td>TREATS(2, 0.0557)</td>\n",
       "      <td>1.254224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Phosphotransferases</td>\n",
       "      <td>STIMULATES(2, 0.00924)</td>\n",
       "      <td>2.034473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>leptin</td>\n",
       "      <td>JAK2</td>\n",
       "      <td>STIMULATES(2, 0.032)</td>\n",
       "      <td>1.495445</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Janus kinase 2</td>\n",
       "      <td>STIMULATES(2, 0.032)</td>\n",
       "      <td>1.495445</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Insulin</td>\n",
       "      <td>INTERACTS_WITH(2, 0.0788)</td>\n",
       "      <td>1.103291</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Adiponectin</td>\n",
       "      <td>STIMULATES(2, 0.074)</td>\n",
       "      <td>1.130802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "      <td>PREDISPOSES(2, 0.00163)</td>\n",
       "      <td>2.788519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Proteome</td>\n",
       "      <td>INTERACTS_WITH(2, 0.00163)</td>\n",
       "      <td>2.788519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Endothelial dysfunction</td>\n",
       "      <td>CAUSES(2, 0.0164)</td>\n",
       "      <td>1.785612</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Cardiovascular Diseases</td>\n",
       "      <td>PREDISPOSES(2, 0.0991)</td>\n",
       "      <td>1.003922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "      <td>NEG_ASSOCIATED_WITH(2, 0.00163)</td>\n",
       "      <td>2.788519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Congestive heart failure</td>\n",
       "      <td>PREDISPOSES(2, 0.00163)</td>\n",
       "      <td>2.788519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Acute Coronary Syndrome</td>\n",
       "      <td>TREATS(3, 0.000243)</td>\n",
       "      <td>3.614532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "      <td>TREATS(3, 8.9e-05)</td>\n",
       "      <td>4.050805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Sleep Apnea, Obstructive</td>\n",
       "      <td>TREATS(3, 0.000508)</td>\n",
       "      <td>3.294404</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Cardiovascular Diseases</td>\n",
       "      <td>TREATS(3, 0.000154)</td>\n",
       "      <td>3.813212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Stable angina</td>\n",
       "      <td>TREATS(3, 8.9e-05)</td>\n",
       "      <td>4.050805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Metabolic syndrome</td>\n",
       "      <td>TREATS(4, 0.00149)</td>\n",
       "      <td>2.826490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Acute myocardial infarction</td>\n",
       "      <td>TREATS(4, 9.06e-05)</td>\n",
       "      <td>4.042728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Endothelial dysfunction</td>\n",
       "      <td>AFFECTS(4, 1.49e-05)</td>\n",
       "      <td>4.826651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Metabolic syndrome</td>\n",
       "      <td>PREDISPOSES(4, 0.00361)</td>\n",
       "      <td>2.441962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "      <td>TREATS(5, 3.01e-07)</td>\n",
       "      <td>6.521391</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "      <td>PREDISPOSES(7, 1.1e-09)</td>\n",
       "      <td>8.958524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "      <td>TREATS(10, 2.6e-13)</td>\n",
       "      <td>12.584627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Coronary heart disease</td>\n",
       "      <td>PREDISPOSES(12, 1.02e-15)</td>\n",
       "      <td>14.990710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>leptin</td>\n",
       "      <td>Coronary Arteriosclerosis</td>\n",
       "      <td>TREATS(28, 7.72e-35)</td>\n",
       "      <td>34.112222</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         source                       target  \\\n",
       "0                SLEEP DURATION               Corticosterone   \n",
       "1      Sleep Apnea, Obstructive       CORONARY HEART DISEASE   \n",
       "2           Phosphotransferases       CORONARY HEART DISEASE   \n",
       "3                          JAK2       CORONARY HEART DISEASE   \n",
       "4                Janus kinase 2       CORONARY HEART DISEASE   \n",
       "5                       Insulin       CORONARY HEART DISEASE   \n",
       "6                   Adiponectin       CORONARY HEART DISEASE   \n",
       "7        Coronary heart disease       CORONARY HEART DISEASE   \n",
       "8                      Proteome       CORONARY HEART DISEASE   \n",
       "9       Endothelial dysfunction       CORONARY HEART DISEASE   \n",
       "10      Cardiovascular Diseases       CORONARY HEART DISEASE   \n",
       "11     Congestive heart failure       CORONARY HEART DISEASE   \n",
       "12      Acute Coronary Syndrome       CORONARY HEART DISEASE   \n",
       "13    Coronary Arteriosclerosis       CORONARY HEART DISEASE   \n",
       "14                Stable angina       CORONARY HEART DISEASE   \n",
       "15           Metabolic syndrome       CORONARY HEART DISEASE   \n",
       "16  Acute myocardial infarction       CORONARY HEART DISEASE   \n",
       "17               Corticosterone                       leptin   \n",
       "18                       leptin     Sleep Apnea, Obstructive   \n",
       "19                       leptin          Phosphotransferases   \n",
       "20                       leptin                         JAK2   \n",
       "21                       leptin               Janus kinase 2   \n",
       "22                       leptin                      Insulin   \n",
       "23                       leptin                  Adiponectin   \n",
       "24                       leptin       Coronary heart disease   \n",
       "25                       leptin                     Proteome   \n",
       "26                       leptin      Endothelial dysfunction   \n",
       "27                       leptin      Cardiovascular Diseases   \n",
       "28                       leptin       Coronary heart disease   \n",
       "29                       leptin     Congestive heart failure   \n",
       "30                       leptin      Acute Coronary Syndrome   \n",
       "31                       leptin    Coronary Arteriosclerosis   \n",
       "32                       leptin     Sleep Apnea, Obstructive   \n",
       "33                       leptin      Cardiovascular Diseases   \n",
       "34                       leptin                Stable angina   \n",
       "35                       leptin           Metabolic syndrome   \n",
       "36                       leptin  Acute myocardial infarction   \n",
       "37                       leptin      Endothelial dysfunction   \n",
       "38                       leptin           Metabolic syndrome   \n",
       "39                       leptin    Coronary Arteriosclerosis   \n",
       "40                       leptin    Coronary Arteriosclerosis   \n",
       "41                       leptin       Coronary heart disease   \n",
       "42                       leptin       Coronary heart disease   \n",
       "43                       leptin    Coronary Arteriosclerosis   \n",
       "\n",
       "                              label     weight  \n",
       "0                                     1.000000  \n",
       "1                                     1.000000  \n",
       "2                                     1.000000  \n",
       "3                                     1.000000  \n",
       "4                                     1.000000  \n",
       "5                                     1.000000  \n",
       "6                                     1.000000  \n",
       "7                                     1.000000  \n",
       "8                                     1.000000  \n",
       "9                                     1.000000  \n",
       "10                                    1.000000  \n",
       "11                                    1.000000  \n",
       "12                                    1.000000  \n",
       "13                                    1.000000  \n",
       "14                                    1.000000  \n",
       "15                                    1.000000  \n",
       "16                                    1.000000  \n",
       "17  NEG_INTERACTS_WITH(2, 2.48e-06)   5.605808  \n",
       "18                TREATS(2, 0.0557)   1.254224  \n",
       "19           STIMULATES(2, 0.00924)   2.034473  \n",
       "20             STIMULATES(2, 0.032)   1.495445  \n",
       "21             STIMULATES(2, 0.032)   1.495445  \n",
       "22        INTERACTS_WITH(2, 0.0788)   1.103291  \n",
       "23             STIMULATES(2, 0.074)   1.130802  \n",
       "24          PREDISPOSES(2, 0.00163)   2.788519  \n",
       "25       INTERACTS_WITH(2, 0.00163)   2.788519  \n",
       "26                CAUSES(2, 0.0164)   1.785612  \n",
       "27           PREDISPOSES(2, 0.0991)   1.003922  \n",
       "28  NEG_ASSOCIATED_WITH(2, 0.00163)   2.788519  \n",
       "29          PREDISPOSES(2, 0.00163)   2.788519  \n",
       "30              TREATS(3, 0.000243)   3.614532  \n",
       "31               TREATS(3, 8.9e-05)   4.050805  \n",
       "32              TREATS(3, 0.000508)   3.294404  \n",
       "33              TREATS(3, 0.000154)   3.813212  \n",
       "34               TREATS(3, 8.9e-05)   4.050805  \n",
       "35               TREATS(4, 0.00149)   2.826490  \n",
       "36              TREATS(4, 9.06e-05)   4.042728  \n",
       "37             AFFECTS(4, 1.49e-05)   4.826651  \n",
       "38          PREDISPOSES(4, 0.00361)   2.441962  \n",
       "39              TREATS(5, 3.01e-07)   6.521391  \n",
       "40          PREDISPOSES(7, 1.1e-09)   8.958524  \n",
       "41              TREATS(10, 2.6e-13)  12.584627  \n",
       "42        PREDISPOSES(12, 1.02e-15)  14.990710  \n",
       "43             TREATS(28, 7.72e-35)  34.112222  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "edges = (\n",
    "    pd.concat(\n",
    "        [\n",
    "            # exposure -> s1 subject\n",
    "            lit_detail.rename(\n",
    "                columns={\"gwas.trait\": \"source\", \"st1.name\": \"target\"}\n",
    "            ).assign(label=\"\", weight=1.0)[[\"source\", \"target\", \"label\", \"weight\"]],\n",
    "            # s2 object -> outcome\n",
    "            lit_detail.rename(\n",
    "                columns={\"st2.name\": \"source\", \"assoc_gwas.trait\": \"target\"}\n",
    "            ).assign(label=\"\", weight=1.0)[[\"source\", \"target\", \"label\", \"weight\"]],\n",
    "            # s1 subject - s1 predicate -> st\n",
    "            lit_detail.rename(\n",
    "                columns={\"st1.name\": \"source\", \"st.name\": \"target\"}\n",
    "            ).assign(\n",
    "                label=lambda df: df.apply(\n",
    "                    lambda row: f\"{row['s1.predicate']}({row['gs1.localCount']}, {row['gs1.pval']:.3g})\",\n",
    "                    axis=1,\n",
    "                ),\n",
    "                weight=lambda df: df[\"gs1.pval\"].apply(lambda x: math.log10(x) * -1),\n",
    "            )[\n",
    "                [\"source\", \"target\", \"label\", \"weight\"]\n",
    "            ],\n",
    "            # st - s2 predicate -> s2 name\n",
    "            lit_detail.rename(\n",
    "                columns={\"st.name\": \"source\", \"st2.name\": \"target\"}\n",
    "            ).assign(\n",
    "                label=lambda df: df.apply(\n",
    "                    lambda row: f\"{row['s2.predicate']}({row['gs2.localCount']}, {row['gs2.pval']:.3g})\",\n",
    "                    axis=1,\n",
    "                ),\n",
    "                weight=lambda df: df[\"gs2.pval\"].apply(lambda x: math.log10(x) * -1),\n",
    "            )[\n",
    "                [\"source\", \"target\", \"label\", \"weight\"]\n",
    "            ],\n",
    "        ]\n",
    "    )\n",
    "    .drop_duplicates()\n",
    "    .reset_index(drop=True)\n",
    ")\n",
    "edges"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "liked-piece",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:22.136061Z",
     "iopub.status.busy": "2021-03-22T10:49:22.135270Z",
     "iopub.status.idle": "2021-03-22T10:49:22.138222Z",
     "shell.execute_reply": "2021-03-22T10:49:22.137597Z"
    },
    "papermill": {
     "duration": 0.050099,
     "end_time": "2021-03-22T10:49:22.138371",
     "exception": false,
     "start_time": "2021-03-22T10:49:22.088272",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "G = nx.from_pandas_edgelist(edges)\n",
    "\n",
    "edgelist = [(_[\"source\"], _[\"target\"]) for idx, _ in edges.iterrows()]\n",
    "edge_labels = {}\n",
    "for idx, row in edges.iterrows():\n",
    "    edge_labels[(row[\"source\"], row[\"target\"])] = row[\"label\"]\n",
    "node_labels = {}\n",
    "for idx, row in nodes.iterrows():\n",
    "    node_labels[row[\"node\"]] = row[\"node\"].replace(\" \", \"\\n\")\n",
    "color_map = {\"trait\": \"red\", \"literature_term\": \"lightblue\", \"overlapping_term\": \"gold\"}\n",
    "node_colors = [color_map[_] for _ in nodes[\"node_type\"].tolist()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "pregnant-headline",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:22.209622Z",
     "iopub.status.busy": "2021-03-22T10:49:22.209236Z",
     "iopub.status.idle": "2021-03-22T10:49:23.195037Z",
     "shell.execute_reply": "2021-03-22T10:49:23.195600Z"
    },
    "papermill": {
     "duration": 1.017134,
     "end_time": "2021-03-22T10:49:23.195778",
     "exception": false,
     "start_time": "2021-03-22T10:49:22.178644",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pos = graphviz_layout(G)\n",
    "plt.figure(figsize=(15, 15))\n",
    "nx.draw_networkx_edges(G, pos, edgelist=edgelist, width=edges[\"weight\"].tolist())\n",
    "nx.draw_networkx_nodes(\n",
    "    G, pos, nodelist=nodes[\"node\"].tolist(), node_size=8000, node_color=node_colors\n",
    ")\n",
    "_ = nx.draw_networkx_edge_labels(G, pos, edge_labels)\n",
    "_ = nx.draw_networkx_labels(G, pos, node_labels)\n",
    "plt.savefig(f\"{focus_term}.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "serial-plant",
   "metadata": {
    "papermill": {
     "duration": 0.038005,
     "end_time": "2021-03-22T10:49:23.279881",
     "exception": false,
     "start_time": "2021-03-22T10:49:23.241876",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Checking the source literature\n",
    "\n",
    "We can refer back to the articles to check the text that was used to derive the SemMedDB data. This is important due to the imperfect nature of the SemRep annotation process (https://semrep.nlm.nih.gov/). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "australian-canyon",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:23.337366Z",
     "iopub.status.busy": "2021-03-22T10:49:23.336611Z",
     "iopub.status.idle": "2021-03-22T10:49:26.992812Z",
     "shell.execute_reply": "2021-03-22T10:49:26.992181Z"
    },
    "papermill": {
     "duration": 3.690579,
     "end_time": "2021-03-22T10:49:26.992972",
     "exception": false,
     "start_time": "2021-03-22T10:49:23.302393",
     "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>gwas_id</th>\n",
       "      <th>triple_id</th>\n",
       "      <th>literature_epigraphdb</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ieu-a-1088</td>\n",
       "      <td>C0010124:NEG_INTERACTS_WITH:C0299583</td>\n",
       "      <td>{'triple_name': ['Corticosterone NEG_INTERACTS...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:TREATS:C0520679</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP TREATS Sleep Apne...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:STIMULATES:C0031727</td>\n",
       "      <td>{'triple_name': ['Leptin STIMULATES Phosphotra...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:STIMULATES:3717</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP STIMULATES Janus ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:STIMULATES:C0169661</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP STIMULATES Janus ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:INTERACTS_WITH:C0021641</td>\n",
       "      <td>{'triple_name': ['Leptin INTERACTS_WITH Insuli...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:STIMULATES:C0389071</td>\n",
       "      <td>{'triple_name': ['Leptin STIMULATES Adiponecti...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:PREDISPOSES:C0010068</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP PREDISPOSES Coron...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:INTERACTS_WITH:C0751973</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP INTERACTS_WITH Pr...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:CAUSES:C0856169</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP CAUSES Endothelia...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:PREDISPOSES:C0007222</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP PREDISPOSES Cardi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:NEG_ASSOCIATED_WITH:C0010068</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP NEG_ASSOCIATED_WI...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:PREDISPOSES:C0018802</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP PREDISPOSES Conge...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:TREATS:C0948089</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP TREATS Acute coro...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:TREATS:C0010054</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP TREATS Coronary A...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:TREATS:C0007222</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP TREATS Cardiovasc...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:TREATS:C0340288</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP TREATS Stable ang...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:TREATS:C0948265</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP TREATS Metabolic ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:TREATS:C0155626</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP TREATS Acute myoc...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:AFFECTS:C0856169</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP AFFECTS Endotheli...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:PREDISPOSES:C0948265</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP PREDISPOSES Metab...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:PREDISPOSES:C0010054</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP PREDISPOSES Coron...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>ieu-a-6</td>\n",
       "      <td>C0299583:TREATS:C0010068</td>\n",
       "      <td>{'triple_name': ['Leptin|LEP TREATS Coronary h...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       gwas_id                              triple_id  \\\n",
       "0   ieu-a-1088   C0010124:NEG_INTERACTS_WITH:C0299583   \n",
       "1      ieu-a-6               C0299583:TREATS:C0520679   \n",
       "2      ieu-a-6           C0299583:STIMULATES:C0031727   \n",
       "3      ieu-a-6               C0299583:STIMULATES:3717   \n",
       "4      ieu-a-6           C0299583:STIMULATES:C0169661   \n",
       "5      ieu-a-6       C0299583:INTERACTS_WITH:C0021641   \n",
       "6      ieu-a-6           C0299583:STIMULATES:C0389071   \n",
       "7      ieu-a-6          C0299583:PREDISPOSES:C0010068   \n",
       "8      ieu-a-6       C0299583:INTERACTS_WITH:C0751973   \n",
       "9      ieu-a-6               C0299583:CAUSES:C0856169   \n",
       "10     ieu-a-6          C0299583:PREDISPOSES:C0007222   \n",
       "11     ieu-a-6  C0299583:NEG_ASSOCIATED_WITH:C0010068   \n",
       "12     ieu-a-6          C0299583:PREDISPOSES:C0018802   \n",
       "13     ieu-a-6               C0299583:TREATS:C0948089   \n",
       "14     ieu-a-6               C0299583:TREATS:C0010054   \n",
       "15     ieu-a-6               C0299583:TREATS:C0007222   \n",
       "16     ieu-a-6               C0299583:TREATS:C0340288   \n",
       "17     ieu-a-6               C0299583:TREATS:C0948265   \n",
       "18     ieu-a-6               C0299583:TREATS:C0155626   \n",
       "19     ieu-a-6              C0299583:AFFECTS:C0856169   \n",
       "20     ieu-a-6          C0299583:PREDISPOSES:C0948265   \n",
       "21     ieu-a-6          C0299583:PREDISPOSES:C0010054   \n",
       "22     ieu-a-6               C0299583:TREATS:C0010068   \n",
       "\n",
       "                                literature_epigraphdb  \n",
       "0   {'triple_name': ['Corticosterone NEG_INTERACTS...  \n",
       "1   {'triple_name': ['Leptin|LEP TREATS Sleep Apne...  \n",
       "2   {'triple_name': ['Leptin STIMULATES Phosphotra...  \n",
       "3   {'triple_name': ['Leptin|LEP STIMULATES Janus ...  \n",
       "4   {'triple_name': ['Leptin|LEP STIMULATES Janus ...  \n",
       "5   {'triple_name': ['Leptin INTERACTS_WITH Insuli...  \n",
       "6   {'triple_name': ['Leptin STIMULATES Adiponecti...  \n",
       "7   {'triple_name': ['Leptin|LEP PREDISPOSES Coron...  \n",
       "8   {'triple_name': ['Leptin|LEP INTERACTS_WITH Pr...  \n",
       "9   {'triple_name': ['Leptin|LEP CAUSES Endothelia...  \n",
       "10  {'triple_name': ['Leptin|LEP PREDISPOSES Cardi...  \n",
       "11  {'triple_name': ['Leptin|LEP NEG_ASSOCIATED_WI...  \n",
       "12  {'triple_name': ['Leptin|LEP PREDISPOSES Conge...  \n",
       "13  {'triple_name': ['Leptin|LEP TREATS Acute coro...  \n",
       "14  {'triple_name': ['Leptin|LEP TREATS Coronary A...  \n",
       "15  {'triple_name': ['Leptin|LEP TREATS Cardiovasc...  \n",
       "16  {'triple_name': ['Leptin|LEP TREATS Stable ang...  \n",
       "17  {'triple_name': ['Leptin|LEP TREATS Metabolic ...  \n",
       "18  {'triple_name': ['Leptin|LEP TREATS Acute myoc...  \n",
       "19  {'triple_name': ['Leptin|LEP AFFECTS Endotheli...  \n",
       "20  {'triple_name': ['Leptin|LEP PREDISPOSES Metab...  \n",
       "21  {'triple_name': ['Leptin|LEP PREDISPOSES Coron...  \n",
       "22  {'triple_name': ['Leptin|LEP TREATS Coronary h...  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def get_literature(gwas_id, triple_id):\n",
    "    endpoint = \"/literature/gwas\"\n",
    "    url = f\"{API_URL}{endpoint}\"\n",
    "    params = {\n",
    "        \"gwas_id\": gwas_id,\n",
    "        \"semmed_triple_id\": triple_id,\n",
    "        \"by_gwas_id\": \"true\",\n",
    "        \"pval_threshold\": 1e-1,\n",
    "    }\n",
    "    r = requests.get(url, params=params)\n",
    "    r.raise_for_status()\n",
    "    df = pd.json_normalize(r.json()[\"results\"])\n",
    "    if len(df) > 0:\n",
    "        res = {\n",
    "            \"triple_name\": df[\"triple.name\"].drop_duplicates().tolist(),\n",
    "            \"pubmed_id\": df[\"lit.id\"].tolist(),\n",
    "        }\n",
    "        return res\n",
    "\n",
    "\n",
    "gwas_pub_df = pd.concat(\n",
    "    [\n",
    "        lit_detail.rename(columns={\"gwas.id\": \"gwas_id\", \"s1.id\": \"triple_id\"})[\n",
    "            [\"gwas_id\", \"triple_id\"]\n",
    "        ]\n",
    "        .drop_duplicates()\n",
    "        .assign(\n",
    "            literature_epigraphdb=lambda df: df.apply(\n",
    "                lambda row: get_literature(row[\"gwas_id\"], row[\"triple_id\"]), axis=1\n",
    "            )\n",
    "        ),\n",
    "        lit_detail.rename(columns={\"assoc_gwas.id\": \"gwas_id\", \"s2.id\": \"triple_id\"})[\n",
    "            [\"gwas_id\", \"triple_id\"]\n",
    "        ]\n",
    "        .drop_duplicates()\n",
    "        .assign(\n",
    "            literature_epigraphdb=lambda df: df.apply(\n",
    "                lambda row: get_literature(row[\"gwas_id\"], row[\"triple_id\"]), axis=1\n",
    "            )\n",
    "        ),\n",
    "    ]\n",
    ").reset_index(drop=True)\n",
    "gwas_pub_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "lesbian-cabinet",
   "metadata": {
    "papermill": {
     "duration": 0.039889,
     "end_time": "2021-03-22T10:49:27.081193",
     "exception": false,
     "start_time": "2021-03-22T10:49:27.041304",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Check the text\n",
    "\n",
    "Do the data from SemMedDB match the literature origins? Here we can use the MELODI Presto API (http://melodi-presto.mrcieu.ac.uk/docs/) to examine the SemMedDB data in more detail."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "adopted-serbia",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-03-22T10:49:27.137245Z",
     "iopub.status.busy": "2021-03-22T10:49:27.131312Z",
     "iopub.status.idle": "2021-03-22T10:49:31.906450Z",
     "shell.execute_reply": "2021-03-22T10:49:31.906725Z"
    },
    "papermill": {
     "duration": 4.803021,
     "end_time": "2021-03-22T10:49:31.906836",
     "exception": false,
     "start_time": "2021-03-22T10:49:27.103815",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "### ['Corticosterone NEG_INTERACTS_WITH Leptin|LEP']  ###\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Interestingly, such changes were associated with increased serum levels of glucagon, corticosterone, and norepinephrine, but no effects on leptin, insulin, or ghrelin were observed. \n",
      "\n",
      "SD failed to change leptin concentrations, but it promptly stimulated plasma ghrelin and induced eating. \n",
      "\n",
      "SD increased plasma corticosterone, but corticosterone did not seem to influence either leptin or ghrelin. \n",
      "\n",
      "\n",
      "### ['Leptin|LEP TREATS Sleep Apnea, Obstructive']  ###\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BACKGROUND: Obstructive sleep apnea syndrome (OSAS) is a common disorder in obese subjects. \n",
      "\n",
      "Changes in intra-abdominal visceral fat and serum leptin levels in patients with obstructive sleep apnea syndrome following nasal continuous positive airway pressure therapy. \n",
      "\n",
      "Serum leptin levels of another 21 OSAS patients were measured before and after 3 to 4 days of NCPAP to gain insight into the mechanism by which NCPAP affects fat distribution. \n",
      "\n",
      "The effect of nasal continuous positive airway pressure (NCPAP) treatment on VFA and serum leptin levels in OSAS patients has not been known. \n",
      "\n",
      "BACKGROUND: Obstructive sleep apnea syndrome (OSAS) is a common disorder in obese subjects. \n",
      "\n",
      "Changes in intra-abdominal visceral fat and serum leptin levels in patients with obstructive sleep apnea syndrome following nasal continuous positive airway pressure therapy. \n",
      "\n",
      "Serum leptin levels of another 21 OSAS patients were measured before and after 3 to 4 days of NCPAP to gain insight into the mechanism by which NCPAP affects fat distribution. \n",
      "\n",
      "The effect of nasal continuous positive airway pressure (NCPAP) treatment on VFA and serum leptin levels in OSAS patients has not been known. \n",
      "\n",
      "\n",
      "### ['Leptin STIMULATES Phosphotransferases']  ###\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "These effects of leptin on [Ca(2+)](i) were inhibited by blockers of JAK2 and tyrosine kinases. \n",
      "\n",
      "It has been reported that leptin is an independent risk factor for the coronary artery disease in obese patients and that leptin is involved in the pathogenesis of cardiovascular diseases. \n",
      "\n",
      "Leptin potentiates ADP-induced [Ca(2+)](i) increase via JAK2 and tyrosine kinases in a megakaryoblast cell line. \n",
      "\n",
      "The results indicate that leptin enhances ADP-induced [Ca(2+)](i) increases via JAK2 and tyrosine kinases in a megakaryoblast cell line. \n",
      "\n",
      "Plasma leptin levels are elevated in most of obese individuals, and obesity is associated with high incidence of cardiovascular diseases. \n",
      "\n",
      "\n",
      "### ['Leptin|LEP STIMULATES Janus kinase 2|JAK2']  ###\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "These effects of leptin on [Ca(2+)](i) were inhibited by blockers of JAK2 and tyrosine kinases. \n",
      "\n",
      "It has been reported that leptin is an independent risk factor for the coronary artery disease in obese patients and that leptin is involved in the pathogenesis of cardiovascular diseases. \n",
      "\n",
      "Leptin potentiates ADP-induced [Ca(2+)](i) increase via JAK2 and tyrosine kinases in a megakaryoblast cell line. \n",
      "\n",
      "The results indicate that leptin enhances ADP-induced [Ca(2+)](i) increases via JAK2 and tyrosine kinases in a megakaryoblast cell line. \n",
      "\n",
      "Plasma leptin levels are elevated in most of obese individuals, and obesity is associated with high incidence of cardiovascular diseases. \n",
      "\n",
      "\n",
      "### ['Leptin|LEP STIMULATES Janus kinase 2|JAK2']  ###\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "These effects of leptin on [Ca(2+)](i) were inhibited by blockers of JAK2 and tyrosine kinases. \n",
      "\n",
      "It has been reported that leptin is an independent risk factor for the coronary artery disease in obese patients and that leptin is involved in the pathogenesis of cardiovascular diseases. \n",
      "\n",
      "Leptin potentiates ADP-induced [Ca(2+)](i) increase via JAK2 and tyrosine kinases in a megakaryoblast cell line. \n",
      "\n",
      "The results indicate that leptin enhances ADP-induced [Ca(2+)](i) increases via JAK2 and tyrosine kinases in a megakaryoblast cell line. \n",
      "\n",
      "Plasma leptin levels are elevated in most of obese individuals, and obesity is associated with high incidence of cardiovascular diseases. \n",
      "\n"
     ]
    }
   ],
   "source": [
    "def get_sentence_data(pmid):\n",
    "    url = \"https://melodi-presto.mrcieu.ac.uk/api/sentence/\"\n",
    "    params = {\n",
    "        \"pmid\": pmid,\n",
    "    }\n",
    "    response = requests.post(url, data=json.dumps(params))\n",
    "    response.raise_for_status()\n",
    "    res = response.json()[\"data\"]\n",
    "    df = pd.json_normalize(res)\n",
    "    if len(df) > 0:\n",
    "        return df[\"SENTENCE\"].drop_duplicates().tolist()\n",
    "\n",
    "\n",
    "max_items = 5\n",
    "for idx, data in gwas_pub_df.head(max_items).iterrows():\n",
    "    print(\"\\n###\", data[\"literature_epigraphdb\"][\"triple_name\"], \" ###\\n\")\n",
    "    sentence_results = [\n",
    "        get_sentence_data(_) for _ in data[\"literature_epigraphdb\"][\"pubmed_id\"]\n",
    "    ]\n",
    "    # flatten\n",
    "    sentence_results = [_ for sublist in sentence_results for _ in sublist]\n",
    "    for _ in sentence_results:\n",
    "        print(_, \"\\n\")"
   ]
  }
 ],
 "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": 155.513784,
   "end_time": "2021-03-22T10:49:32.340772",
   "environment_variables": {},
   "exception": null,
   "input_path": "paper-case-studies/case-3-literature-triangulation.ipynb",
   "output_path": "paper-case-studies/case-3-literature-triangulation.ipynb",
   "parameters": {
    "API_URL": "https://api.epigraphdb.org"
   },
   "start_time": "2021-03-22T10:46:56.826988",
   "version": "2.3.3"
  },
  "toc-autonumbering": false
 },
 "nbformat": 4,
 "nbformat_minor": 5
}