notebooks/5_p_metapath_generation.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Metagraph and Metapath Generation\n",
    "\n",
    "Here, we use the same syntax as [HetNetPy](https://github.com/hetio/hetnetpy/tree/e1ff1f8296b766dd5a63e5870a8a1a7d82427a80) to create a metagraph and extract metapaths from the KG.\n",
    "\n",
    "Metagraph creation is based on [this notebook](https://github.com/dhimmel/integrate/blob/93feba1765fbcd76fd79e22f25121f5399629148/integrate.ipynb)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import os\n",
    "import os.path as osp\n",
    "\n",
    "import hetnetpy.hetnet\n",
    "import hetnetpy.readwrite\n",
    "import hetnetpy.stats\n",
    "\n",
    "from hetnetpy.abbreviation import metaedges_from_metapath\n",
    "from hetnetpy.pathtools import DWPC, paths_between\n",
    "\n",
    "import re\n",
    "import json\n",
    "import tqdm\n",
    "from collections import Counter\n",
    "\n",
    "from itertools import chain\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Using the protein subclasses\n",
    "\n",
    "In this version of the KG, we subclassed some of the proteins to UniProt keywords. Since we need letters representing these subclasses, we'll assign that here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "MAPPING_DIR = '../data/mappings/'\n",
    "\n",
    "KG_DIR = '../data/kg/'\n",
    "SPLITS_DIR = osp.join(KG_DIR, 'splits')\n",
    "MOA_NET_PROTCLASSED = osp.join(SPLITS_DIR, 'MoA-net-protclass')\n",
    "MARS_DIR = osp.join(MOA_NET_PROTCLASSED, 'MARS')\n",
    "METAPATHS_DIR = osp.join(MOA_NET_PROTCLASSED, 'metapaths')\n",
    "\n",
    "for dir in [SPLITS_DIR, MARS_DIR, MOA_NET_PROTCLASSED, METAPATHS_DIR]:\n",
    "    if not osp.exists(dir):\n",
    "        os.makedirs(dir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Read in the json file containing the classes:\n",
    "with open(osp.join(MAPPING_DIR, 'node2class.json'), 'r') as f:\n",
    "    prot_classes = json.load(f)\n",
    "\n",
    "prot_classes = {kw for keyword in prot_classes.values() for kw in keyword}\n",
    "\n",
    "prot_classes = sorted(list(prot_classes))  # sort so we get consistent results each time code is run:"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Be sure to exclude C and G because we've already assigned them to Compound and Gene."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Antimicrobial',\n",
       " 'Chaperone',\n",
       " 'Cytokine',\n",
       " 'Developmental protein',\n",
       " 'Growth factor',\n",
       " 'Heparin-binding',\n",
       " 'Hormone',\n",
       " 'Hydrolase',\n",
       " 'Ion channel',\n",
       " 'Isomerase',\n",
       " 'Protease inhibitor',\n",
       " 'Translocase']"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prot_classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Antimicrobial': 'A',\n",
       " 'Chaperone': 'B',\n",
       " 'Cytokine': 'D',\n",
       " 'Developmental protein': 'E',\n",
       " 'Growth factor': 'F',\n",
       " 'Heparin-binding': 'H',\n",
       " 'Hormone': 'I',\n",
       " 'Hydrolase': 'J',\n",
       " 'Ion channel': 'K',\n",
       " 'Isomerase': 'L',\n",
       " 'Protease inhibitor': 'M',\n",
       " 'Translocase': 'N'}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# for each item in keep_kws, assign a letter of the alphabet in a dictionary, excluding C or G\n",
    "class2letter = dict() \n",
    "i = 0\n",
    "for kw in prot_classes:\n",
    "    if kw == 'Gene':\n",
    "        continue\n",
    "    if chr(i + 65) in {'C', 'G'}:\n",
    "        i += 1\n",
    "    class2letter[kw] = chr(i + 65)\n",
    "    i += 1\n",
    "\n",
    "class2letter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "kg = pd.read_csv(osp.join(MOA_NET_PROTCLASSED, 'kg_no_cmp_bp.tsv'), sep='\\t')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>source</th>\n",
       "      <th>source_node_type</th>\n",
       "      <th>target</th>\n",
       "      <th>target_node_type</th>\n",
       "      <th>edge_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [source, source_node_type, target, target_node_type, edge_type]\n",
       "Index: []"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# show me the duplicated rows in the dataframe\n",
    "kg[kg.duplicated(keep=False)]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define and Instantiate Metagraph:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "kind_to_abbev = {\n",
    "    \n",
    "    # metanodes\n",
    "    'Compound': 'C',\n",
    "    'Gene': 'G',\n",
    "    'Biological Process': 'BP',\n",
    "    \n",
    "    # metaedges\n",
    "    'upregulates': 'u',\n",
    "    'downregulates': 'd',\n",
    "    'interacts': 'i',\n",
    "    'participates': 'p',\n",
    "    'induces': 't',\n",
    "}\n",
    "\n",
    "# add the new protein subclasses to the abbreviations:\n",
    "kind_to_abbev.update(class2letter)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "metaedge_tuples = list({(row['source_node_type'], row['target_node_type'], row['edge_type'], 'forward') for i, row in kg.iterrows()})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# now make the metagraph\n",
    "metagraph = hetnetpy.hetnet.MetaGraph.from_edge_tuples(metaedge_tuples, kind_to_abbev)\n",
    "graph = hetnetpy.hetnet.Graph(metagraph)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "seen = set()\n",
    "\n",
    "for i, row in kg.iterrows():\n",
    "    if not (row['source'], row['source_node_type']) in seen:\n",
    "        graph.add_node(kind=row['source_node_type'], identifier=row['source'])\n",
    "        seen.add((row['source'], row['source_node_type']))\n",
    "    if not (row['target'], row['target_node_type']) in seen:\n",
    "        graph.add_node(kind=row['target_node_type'], identifier=row['target'])\n",
    "        seen.add((row['target'], row['target_node_type']))\n",
    "\n",
    "    src_id = row['source_node_type'], row['source']\n",
    "    trgt_id = row['target_node_type'], row['target']\n",
    "\n",
    "    graph.add_edge(source_id=src_id, target_id=trgt_id, kind=row['edge_type'], direction='forward')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Network Visualization and Stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>metanode</th>\n",
       "      <th>abbreviation</th>\n",
       "      <th>metaedges</th>\n",
       "      <th>nodes</th>\n",
       "      <th>unconnected_nodes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Antimicrobial</td>\n",
       "      <td>A</td>\n",
       "      <td>10</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Biological Process</td>\n",
       "      <td>BP</td>\n",
       "      <td>12</td>\n",
       "      <td>61</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Chaperone</td>\n",
       "      <td>B</td>\n",
       "      <td>19</td>\n",
       "      <td>95</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Compound</td>\n",
       "      <td>C</td>\n",
       "      <td>22</td>\n",
       "      <td>205</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Cytokine</td>\n",
       "      <td>D</td>\n",
       "      <td>25</td>\n",
       "      <td>164</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Developmental protein</td>\n",
       "      <td>E</td>\n",
       "      <td>26</td>\n",
       "      <td>422</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Growth factor</td>\n",
       "      <td>F</td>\n",
       "      <td>23</td>\n",
       "      <td>39</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Heparin-binding</td>\n",
       "      <td>H</td>\n",
       "      <td>24</td>\n",
       "      <td>65</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Hormone</td>\n",
       "      <td>I</td>\n",
       "      <td>24</td>\n",
       "      <td>59</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Hydrolase</td>\n",
       "      <td>J</td>\n",
       "      <td>26</td>\n",
       "      <td>815</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Ion channel</td>\n",
       "      <td>K</td>\n",
       "      <td>20</td>\n",
       "      <td>215</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Isomerase</td>\n",
       "      <td>L</td>\n",
       "      <td>19</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Protease inhibitor</td>\n",
       "      <td>M</td>\n",
       "      <td>21</td>\n",
       "      <td>41</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Translocase</td>\n",
       "      <td>N</td>\n",
       "      <td>22</td>\n",
       "      <td>59</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 metanode abbreviation  metaedges  nodes  unconnected_nodes\n",
       "12          Antimicrobial            A         10     14                  0\n",
       "13     Biological Process           BP         12     61                  0\n",
       "9               Chaperone            B         19     95                  0\n",
       "0                Compound            C         22    205                  0\n",
       "1                Cytokine            D         25    164                  0\n",
       "8   Developmental protein            E         26    422                  0\n",
       "11          Growth factor            F         23     39                  0\n",
       "10        Heparin-binding            H         24     65                  0\n",
       "6                 Hormone            I         24     59                  0\n",
       "3               Hydrolase            J         26    815                  0\n",
       "5             Ion channel            K         20    215                  0\n",
       "2               Isomerase            L         19     58                  0\n",
       "7      Protease inhibitor            M         21     41                  0\n",
       "4             Translocase            N         22     59                  0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metanode_df = hetnetpy.stats.get_metanode_df(graph)\n",
    "metanode_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "rel_mapping = {str(row['abbreviation']): str(row['metanode']) for i, row in metanode_df.iterrows()}\n",
    "# write to json file\n",
    "with open(osp.join(MAPPING_DIR, 'rel_mapping.json'), 'w') as f:\n",
    "    json.dump(rel_mapping, f, indent=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.barplot(x='metanode', y='nodes', data=metanode_df.sort_values('nodes'))\n",
    "for tick in ax.get_xticklabels():\n",
    "    tick.set_rotation(90)\n",
    "ax.set_xlabel(''); ax.set_ylabel('nodes');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2312"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Total number of nodes\n",
    "metanode_df.nodes.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Summary of metaedges and cooresponding edges\n",
    "metaedge_df = hetnetpy.stats.get_metaedge_df(graph)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>metaedge</th>\n",
       "      <th>abbreviation</th>\n",
       "      <th>edges</th>\n",
       "      <th>source_nodes</th>\n",
       "      <th>target_nodes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Developmental protein &gt; interacts &gt; Developmen...</td>\n",
       "      <td>Ei&gt;E</td>\n",
       "      <td>1227</td>\n",
       "      <td>202</td>\n",
       "      <td>271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74</th>\n",
       "      <td>Developmental protein &gt; interacts &gt; Hydrolase</td>\n",
       "      <td>Ei&gt;J</td>\n",
       "      <td>654</td>\n",
       "      <td>141</td>\n",
       "      <td>330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Hydrolase &gt; participates &gt; Biological Process</td>\n",
       "      <td>Jp&gt;BP</td>\n",
       "      <td>583</td>\n",
       "      <td>396</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Hydrolase &gt; interacts &gt; Hydrolase</td>\n",
       "      <td>Ji&gt;J</td>\n",
       "      <td>514</td>\n",
       "      <td>190</td>\n",
       "      <td>224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>Developmental protein &gt; participates &gt; Biologi...</td>\n",
       "      <td>Ep&gt;BP</td>\n",
       "      <td>312</td>\n",
       "      <td>221</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Cytokine &gt; interacts &gt; Cytokine</td>\n",
       "      <td>Di&gt;D</td>\n",
       "      <td>264</td>\n",
       "      <td>63</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>150</th>\n",
       "      <td>Cytokine &gt; interacts &gt; Developmental protein</td>\n",
       "      <td>Di&gt;E</td>\n",
       "      <td>263</td>\n",
       "      <td>96</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>Hydrolase &gt; interacts &gt; Developmental protein</td>\n",
       "      <td>Ji&gt;E</td>\n",
       "      <td>252</td>\n",
       "      <td>91</td>\n",
       "      <td>116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82</th>\n",
       "      <td>Compound &gt; downregulates &gt; Ion channel</td>\n",
       "      <td>Cd&gt;K</td>\n",
       "      <td>246</td>\n",
       "      <td>74</td>\n",
       "      <td>89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Compound &gt; upregulates &gt; Hydrolase</td>\n",
       "      <td>Cu&gt;J</td>\n",
       "      <td>223</td>\n",
       "      <td>68</td>\n",
       "      <td>102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Compound &gt; downregulates &gt; Hydrolase</td>\n",
       "      <td>Cd&gt;J</td>\n",
       "      <td>215</td>\n",
       "      <td>85</td>\n",
       "      <td>114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>134</th>\n",
       "      <td>Growth factor &gt; interacts &gt; Developmental protein</td>\n",
       "      <td>Fi&gt;E</td>\n",
       "      <td>201</td>\n",
       "      <td>28</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>Ion channel &gt; interacts &gt; Hydrolase</td>\n",
       "      <td>Ki&gt;J</td>\n",
       "      <td>172</td>\n",
       "      <td>47</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>Cytokine &gt; interacts &gt; Hydrolase</td>\n",
       "      <td>Di&gt;J</td>\n",
       "      <td>168</td>\n",
       "      <td>57</td>\n",
       "      <td>109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Cytokine &gt; participates &gt; Biological Process</td>\n",
       "      <td>Dp&gt;BP</td>\n",
       "      <td>166</td>\n",
       "      <td>104</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>Hormone &gt; interacts &gt; Hydrolase</td>\n",
       "      <td>Ii&gt;J</td>\n",
       "      <td>159</td>\n",
       "      <td>16</td>\n",
       "      <td>151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87</th>\n",
       "      <td>Developmental protein &gt; interacts &gt; Ion channel</td>\n",
       "      <td>Ei&gt;K</td>\n",
       "      <td>151</td>\n",
       "      <td>24</td>\n",
       "      <td>96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>71</th>\n",
       "      <td>Compound &gt; downregulates &gt; Cytokine</td>\n",
       "      <td>Cd&gt;D</td>\n",
       "      <td>142</td>\n",
       "      <td>61</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>144</th>\n",
       "      <td>Heparin-binding &gt; interacts &gt; Developmental pr...</td>\n",
       "      <td>Hi&gt;E</td>\n",
       "      <td>132</td>\n",
       "      <td>25</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>112</th>\n",
       "      <td>Ion channel &gt; participates &gt; Biological Process</td>\n",
       "      <td>Kp&gt;BP</td>\n",
       "      <td>129</td>\n",
       "      <td>79</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Chaperone &gt; participates &gt; Biological Process</td>\n",
       "      <td>Bp&gt;BP</td>\n",
       "      <td>112</td>\n",
       "      <td>65</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62</th>\n",
       "      <td>Developmental protein &gt; interacts &gt; Heparin-bi...</td>\n",
       "      <td>Ei&gt;H</td>\n",
       "      <td>111</td>\n",
       "      <td>62</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>63</th>\n",
       "      <td>Heparin-binding &gt; interacts &gt; Hydrolase</td>\n",
       "      <td>Hi&gt;J</td>\n",
       "      <td>109</td>\n",
       "      <td>20</td>\n",
       "      <td>83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>Developmental protein &gt; interacts &gt; Cytokine</td>\n",
       "      <td>Ei&gt;D</td>\n",
       "      <td>106</td>\n",
       "      <td>46</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91</th>\n",
       "      <td>Ion channel &gt; interacts &gt; Ion channel</td>\n",
       "      <td>Ki&gt;K</td>\n",
       "      <td>97</td>\n",
       "      <td>33</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>108</th>\n",
       "      <td>Compound &gt; upregulates &gt; Cytokine</td>\n",
       "      <td>Cu&gt;D</td>\n",
       "      <td>90</td>\n",
       "      <td>42</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119</th>\n",
       "      <td>Hormone &gt; interacts &gt; Developmental protein</td>\n",
       "      <td>Ii&gt;E</td>\n",
       "      <td>88</td>\n",
       "      <td>26</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>Compound &gt; downregulates &gt; Developmental protein</td>\n",
       "      <td>Cd&gt;E</td>\n",
       "      <td>88</td>\n",
       "      <td>45</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>Compound &gt; upregulates &gt; Ion channel</td>\n",
       "      <td>Cu&gt;K</td>\n",
       "      <td>80</td>\n",
       "      <td>34</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>Compound &gt; upregulates &gt; Developmental protein</td>\n",
       "      <td>Cu&gt;E</td>\n",
       "      <td>79</td>\n",
       "      <td>34</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>Heparin-binding &gt; participates &gt; Biological Pr...</td>\n",
       "      <td>Hp&gt;BP</td>\n",
       "      <td>74</td>\n",
       "      <td>45</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>Growth factor &gt; interacts &gt; Hydrolase</td>\n",
       "      <td>Fi&gt;J</td>\n",
       "      <td>70</td>\n",
       "      <td>21</td>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>Growth factor &gt; interacts &gt; Heparin-binding</td>\n",
       "      <td>Fi&gt;H</td>\n",
       "      <td>69</td>\n",
       "      <td>18</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>Hydrolase &gt; interacts &gt; Cytokine</td>\n",
       "      <td>Ji&gt;D</td>\n",
       "      <td>65</td>\n",
       "      <td>39</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64</th>\n",
       "      <td>Protease inhibitor &gt; interacts &gt; Hydrolase</td>\n",
       "      <td>Mi&gt;J</td>\n",
       "      <td>59</td>\n",
       "      <td>13</td>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>68</th>\n",
       "      <td>Hormone &gt; participates &gt; Biological Process</td>\n",
       "      <td>Ip&gt;BP</td>\n",
       "      <td>58</td>\n",
       "      <td>44</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>Growth factor &gt; interacts &gt; Growth factor</td>\n",
       "      <td>Fi&gt;F</td>\n",
       "      <td>56</td>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70</th>\n",
       "      <td>Hydrolase &gt; interacts &gt; Heparin-binding</td>\n",
       "      <td>Ji&gt;H</td>\n",
       "      <td>55</td>\n",
       "      <td>40</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>Heparin-binding &gt; interacts &gt; Heparin-binding</td>\n",
       "      <td>Hi&gt;H</td>\n",
       "      <td>54</td>\n",
       "      <td>21</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>Compound &gt; downregulates &gt; Hormone</td>\n",
       "      <td>Cd&gt;I</td>\n",
       "      <td>54</td>\n",
       "      <td>35</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>115</th>\n",
       "      <td>Developmental protein &gt; interacts &gt; Hormone</td>\n",
       "      <td>Ei&gt;I</td>\n",
       "      <td>54</td>\n",
       "      <td>30</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>Translocase &gt; participates &gt; Biological Process</td>\n",
       "      <td>Np&gt;BP</td>\n",
       "      <td>52</td>\n",
       "      <td>39</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>Compound &gt; upregulates &gt; Hormone</td>\n",
       "      <td>Cu&gt;I</td>\n",
       "      <td>52</td>\n",
       "      <td>38</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Hydrolase &gt; interacts &gt; Protease inhibitor</td>\n",
       "      <td>Ji&gt;M</td>\n",
       "      <td>51</td>\n",
       "      <td>37</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>Hydrolase &gt; interacts &gt; Ion channel</td>\n",
       "      <td>Ji&gt;K</td>\n",
       "      <td>50</td>\n",
       "      <td>26</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>Growth factor &gt; participates &gt; Biological Process</td>\n",
       "      <td>Fp&gt;BP</td>\n",
       "      <td>50</td>\n",
       "      <td>29</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>125</th>\n",
       "      <td>Chaperone &gt; interacts &gt; Chaperone</td>\n",
       "      <td>Bi&gt;B</td>\n",
       "      <td>46</td>\n",
       "      <td>21</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>Developmental protein &gt; interacts &gt; Growth factor</td>\n",
       "      <td>Ei&gt;F</td>\n",
       "      <td>44</td>\n",
       "      <td>23</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>Heparin-binding &gt; interacts &gt; Growth factor</td>\n",
       "      <td>Hi&gt;F</td>\n",
       "      <td>42</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>Compound &gt; upregulates &gt; Growth factor</td>\n",
       "      <td>Cu&gt;F</td>\n",
       "      <td>40</td>\n",
       "      <td>27</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              metaedge abbreviation  edges  \\\n",
       "9    Developmental protein > interacts > Developmen...         Ei>E   1227   \n",
       "74       Developmental protein > interacts > Hydrolase         Ei>J    654   \n",
       "26       Hydrolase > participates > Biological Process        Jp>BP    583   \n",
       "2                    Hydrolase > interacts > Hydrolase         Ji>J    514   \n",
       "56   Developmental protein > participates > Biologi...        Ep>BP    312   \n",
       "15                     Cytokine > interacts > Cytokine         Di>D    264   \n",
       "150       Cytokine > interacts > Developmental protein         Di>E    263   \n",
       "76       Hydrolase > interacts > Developmental protein         Ji>E    252   \n",
       "82              Compound > downregulates > Ion channel         Cd>K    246   \n",
       "5                   Compound > upregulates > Hydrolase         Cu>J    223   \n",
       "16                Compound > downregulates > Hydrolase         Cd>J    215   \n",
       "134  Growth factor > interacts > Developmental protein         Fi>E    201   \n",
       "123                Ion channel > interacts > Hydrolase         Ki>J    172   \n",
       "57                    Cytokine > interacts > Hydrolase         Di>J    168   \n",
       "8         Cytokine > participates > Biological Process        Dp>BP    166   \n",
       "146                    Hormone > interacts > Hydrolase         Ii>J    159   \n",
       "87     Developmental protein > interacts > Ion channel         Ei>K    151   \n",
       "71                 Compound > downregulates > Cytokine         Cd>D    142   \n",
       "144  Heparin-binding > interacts > Developmental pr...         Hi>E    132   \n",
       "112    Ion channel > participates > Biological Process        Kp>BP    129   \n",
       "21       Chaperone > participates > Biological Process        Bp>BP    112   \n",
       "62   Developmental protein > interacts > Heparin-bi...         Ei>H    111   \n",
       "63             Heparin-binding > interacts > Hydrolase         Hi>J    109   \n",
       "128       Developmental protein > interacts > Cytokine         Ei>D    106   \n",
       "91               Ion channel > interacts > Ion channel         Ki>K     97   \n",
       "108                  Compound > upregulates > Cytokine         Cu>D     90   \n",
       "119        Hormone > interacts > Developmental protein         Ii>E     88   \n",
       "107   Compound > downregulates > Developmental protein         Cd>E     88   \n",
       "79                Compound > upregulates > Ion channel         Cu>K     80   \n",
       "98      Compound > upregulates > Developmental protein         Cu>E     79   \n",
       "93   Heparin-binding > participates > Biological Pr...        Hp>BP     74   \n",
       "47               Growth factor > interacts > Hydrolase         Fi>J     70   \n",
       "38         Growth factor > interacts > Heparin-binding         Fi>H     69   \n",
       "41                    Hydrolase > interacts > Cytokine         Ji>D     65   \n",
       "64          Protease inhibitor > interacts > Hydrolase         Mi>J     59   \n",
       "68         Hormone > participates > Biological Process        Ip>BP     58   \n",
       "80           Growth factor > interacts > Growth factor         Fi>F     56   \n",
       "70             Hydrolase > interacts > Heparin-binding         Ji>H     55   \n",
       "88       Heparin-binding > interacts > Heparin-binding         Hi>H     54   \n",
       "52                  Compound > downregulates > Hormone         Cd>I     54   \n",
       "115        Developmental protein > interacts > Hormone         Ei>I     54   \n",
       "136    Translocase > participates > Biological Process        Np>BP     52   \n",
       "46                    Compound > upregulates > Hormone         Cu>I     52   \n",
       "13          Hydrolase > interacts > Protease inhibitor         Ji>M     51   \n",
       "49                 Hydrolase > interacts > Ion channel         Ji>K     50   \n",
       "37   Growth factor > participates > Biological Process        Fp>BP     50   \n",
       "125                  Chaperone > interacts > Chaperone         Bi>B     46   \n",
       "103  Developmental protein > interacts > Growth factor         Ei>F     44   \n",
       "118        Heparin-binding > interacts > Growth factor         Hi>F     42   \n",
       "122             Compound > upregulates > Growth factor         Cu>F     40   \n",
       "\n",
       "     source_nodes  target_nodes  \n",
       "9             202           271  \n",
       "74            141           330  \n",
       "26            396            45  \n",
       "2             190           224  \n",
       "56            221            35  \n",
       "15             63            74  \n",
       "150            96            54  \n",
       "76             91           116  \n",
       "82             74            89  \n",
       "5              68           102  \n",
       "16             85           114  \n",
       "134            28            31  \n",
       "123            47            33  \n",
       "57             57           109  \n",
       "8             104            15  \n",
       "146            16           151  \n",
       "87             24            96  \n",
       "71             61            39  \n",
       "144            25            50  \n",
       "112            79            21  \n",
       "21             65            20  \n",
       "62             62            38  \n",
       "63             20            83  \n",
       "128            46            60  \n",
       "91             33            32  \n",
       "108            42            33  \n",
       "119            26            54  \n",
       "107            45            56  \n",
       "79             34            39  \n",
       "98             34            60  \n",
       "93             45            16  \n",
       "47             21            29  \n",
       "38             18            13  \n",
       "41             39            43  \n",
       "64             13            42  \n",
       "68             44            10  \n",
       "80             12            15  \n",
       "70             40            21  \n",
       "88             21            19  \n",
       "52             35            20  \n",
       "115            30            28  \n",
       "136            39            10  \n",
       "46             38            21  \n",
       "13             37            23  \n",
       "49             26            46  \n",
       "37             29            12  \n",
       "125            21            17  \n",
       "103            23            24  \n",
       "118            10            11  \n",
       "122            27            13  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metaedge_df.sort_values('edges', ascending=False).head(50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.barplot(x='abbreviation', y='edges', data=metaedge_df.sort_values('edges'))\n",
    "for tick in ax.get_xticklabels():\n",
    "    tick.set_rotation(90)\n",
    "ax.set_xlabel(''); ax.set_ylabel('edges');\n",
    "\n",
    "# change dimensions of plot to be wider\n",
    "#sns.set(rc={'figure.figsize':(100, 10)})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>metaedge</th>\n",
       "      <th>unicode_metaedge</th>\n",
       "      <th>standard_metaedge</th>\n",
       "      <th>abbreviation</th>\n",
       "      <th>standard_abbreviation</th>\n",
       "      <th>source</th>\n",
       "      <th>target</th>\n",
       "      <th>inverted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>261</th>\n",
       "      <td>Antimicrobial &lt; interacts &lt; Antimicrobial</td>\n",
       "      <td>Antimicrobial←interacts←Antimicrobial</td>\n",
       "      <td>Antimicrobial &gt; interacts &gt; Antimicrobial</td>\n",
       "      <td>A&lt;iA</td>\n",
       "      <td>AiA</td>\n",
       "      <td>Antimicrobial</td>\n",
       "      <td>Antimicrobial</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>Antimicrobial &lt; interacts &lt; Cytokine</td>\n",
       "      <td>Antimicrobial←interacts←Cytokine</td>\n",
       "      <td>Cytokine &gt; interacts &gt; Antimicrobial</td>\n",
       "      <td>A&lt;iD</td>\n",
       "      <td>DiA</td>\n",
       "      <td>Antimicrobial</td>\n",
       "      <td>Cytokine</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>241</th>\n",
       "      <td>Antimicrobial &lt; interacts &lt; Developmental protein</td>\n",
       "      <td>Antimicrobial←interacts←Developmental protein</td>\n",
       "      <td>Developmental protein &gt; interacts &gt; Antimicrobial</td>\n",
       "      <td>A&lt;iE</td>\n",
       "      <td>EiA</td>\n",
       "      <td>Antimicrobial</td>\n",
       "      <td>Developmental protein</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>229</th>\n",
       "      <td>Antimicrobial &lt; interacts &lt; Heparin-binding</td>\n",
       "      <td>Antimicrobial←interacts←Heparin-binding</td>\n",
       "      <td>Heparin-binding &gt; interacts &gt; Antimicrobial</td>\n",
       "      <td>A&lt;iH</td>\n",
       "      <td>HiA</td>\n",
       "      <td>Antimicrobial</td>\n",
       "      <td>Heparin-binding</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85</th>\n",
       "      <td>Antimicrobial &lt; interacts &lt; Hormone</td>\n",
       "      <td>Antimicrobial←interacts←Hormone</td>\n",
       "      <td>Hormone &gt; interacts &gt; Antimicrobial</td>\n",
       "      <td>A&lt;iI</td>\n",
       "      <td>IiA</td>\n",
       "      <td>Antimicrobial</td>\n",
       "      <td>Hormone</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              metaedge  \\\n",
       "261          Antimicrobial < interacts < Antimicrobial   \n",
       "39                Antimicrobial < interacts < Cytokine   \n",
       "241  Antimicrobial < interacts < Developmental protein   \n",
       "229        Antimicrobial < interacts < Heparin-binding   \n",
       "85                 Antimicrobial < interacts < Hormone   \n",
       "\n",
       "                                  unicode_metaedge  \\\n",
       "261          Antimicrobial←interacts←Antimicrobial   \n",
       "39                Antimicrobial←interacts←Cytokine   \n",
       "241  Antimicrobial←interacts←Developmental protein   \n",
       "229        Antimicrobial←interacts←Heparin-binding   \n",
       "85                 Antimicrobial←interacts←Hormone   \n",
       "\n",
       "                                     standard_metaedge abbreviation  \\\n",
       "261          Antimicrobial > interacts > Antimicrobial         A<iA   \n",
       "39                Cytokine > interacts > Antimicrobial         A<iD   \n",
       "241  Developmental protein > interacts > Antimicrobial         A<iE   \n",
       "229        Heparin-binding > interacts > Antimicrobial         A<iH   \n",
       "85                 Hormone > interacts > Antimicrobial         A<iI   \n",
       "\n",
       "    standard_abbreviation         source                 target  inverted  \n",
       "261                   AiA  Antimicrobial          Antimicrobial         1  \n",
       "39                    DiA  Antimicrobial               Cytokine         1  \n",
       "241                   EiA  Antimicrobial  Developmental protein         1  \n",
       "229                   HiA  Antimicrobial        Heparin-binding         1  \n",
       "85                    IiA  Antimicrobial                Hormone         1  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Summary of different styles for representing each metaedge\n",
    "metaedge_style_df = hetnetpy.stats.get_metaedge_style_df(metagraph)\n",
    "metaedge_style_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9316"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of edges in the network\n",
    "metaedge_df.edges.sum()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Metapath Extraction\n",
    "\n",
    "Here, we have functions which can extract metapaths:"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Extract metapaths from a given source node type (and, optionally, to a target node type):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "tgt_metapaths = metagraph.extract_metapaths(source='Compound', target='Biological Process', max_length=4)\n",
    "# print(tgt_metapaths)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We do not want any with a compound in the middle:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "CtoBP_metapaths = []\n",
    "\n",
    "for i in tgt_metapaths:\n",
    "    if '<' in repr(i):\n",
    "        continue\n",
    "    if (repr(i) != 'Ct>BP') & (Counter(re.split('>|<|[^A-Z]', repr(i)))['C'] == 1):\n",
    "        CtoBP_metapaths.append(i)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How many did it prune?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2590\n",
      "10932\n"
     ]
    }
   ],
   "source": [
    "print(len(CtoBP_metapaths))\n",
    "print(len(tgt_metapaths))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Metapaths Between Training, Validation, Testing\n",
    "\n",
    "Here, we'll look at the metapaths that occur between each of the sets of drug-BP pairs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_paths(\n",
    "    graph,\n",
    "    source,\n",
    "    target,\n",
    "    rule,\n",
    "):\n",
    "    try:\n",
    "        paths = paths_between(\n",
    "            graph,\n",
    "            source=('Compound', source),\n",
    "            target=('Biological Process', target),\n",
    "            metapath=tuple(rule),\n",
    "            duplicates=False,\n",
    "        )\n",
    "    except Exception as e:\n",
    "        print(e)\n",
    "        paths = []        \n",
    "    \n",
    "    return paths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read in train.txt, test.txt, and dev.txt from MARS_DIR\n",
    "train = pd.read_csv(osp.join(MOA_NET_PROTCLASSED, 'train.tsv'), sep='\\t')\n",
    "test = pd.read_csv(osp.join(MOA_NET_PROTCLASSED, 'test.tsv'), sep='\\t')\n",
    "dev = pd.read_csv(osp.join(MOA_NET_PROTCLASSED, 'dev.tsv'), sep='\\t')\n",
    "\n",
    "# for each, get a set of the source and target columns\n",
    "train = {(row['source'], row['target']) for i, row in train.iterrows()}\n",
    "test = {(row['source'], row['target']) for i, row in test.iterrows()}\n",
    "dev = {(row['source'], row['target']) for i, row in dev.iterrows()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_metapath_counts(drug_BP_pairs):\n",
    "\n",
    "    metapath_dict = {str(mpath): 0 for mpath in CtoBP_metapaths}\n",
    "\n",
    "    for source, bp in tqdm.tqdm(drug_BP_pairs):\n",
    "        # All metapaths going from compound to biological process\n",
    "        for mpath in CtoBP_metapaths:\n",
    "                    \n",
    "            paths = get_paths(\n",
    "                graph=graph,\n",
    "                source=source,\n",
    "                target=bp,\n",
    "                rule=mpath,\n",
    "            )\n",
    "            \n",
    "            if not paths:\n",
    "                continue\n",
    "                \n",
    "            metapath_dict[str(mpath)] += len(paths)\n",
    "\n",
    "    return metapath_dict"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This next step is really long, which is why I've made it possible to skip if the files already exist:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# check if a file exists\n",
    "if not osp.exists(osp.join(METAPATHS_DIR, 'train_metapaths.json')):\n",
    "    train_metapaths = get_metapath_counts(train)\n",
    "    with open(osp.join(METAPATHS_DIR, 'train_metapaths.json'), 'w') as f:\n",
    "        json.dump(train_metapaths, f)\n",
    "\n",
    "else:\n",
    "    with open(osp.join(METAPATHS_DIR, 'train_metapaths.json'), 'r') as f:\n",
    "        train_metapaths = json.load(f)\n",
    "\n",
    "if not osp.exists(osp.join(METAPATHS_DIR, 'test_metapaths.json')):       \n",
    "    test_metapaths = get_metapath_counts(test)\n",
    "    with open(osp.join(METAPATHS_DIR, 'test_metapaths.json'), 'w') as f:\n",
    "        json.dump(test_metapaths, f)\n",
    "\n",
    "else:\n",
    "    with open(osp.join(METAPATHS_DIR, 'test_metapaths.json'), 'r') as f:\n",
    "        test_metapaths = json.load(f)\n",
    "\n",
    "if not osp.exists(osp.join(METAPATHS_DIR, 'dev_metapaths.json')): \n",
    "    dev_metapaths = get_metapath_counts(dev)\n",
    "    with open(osp.join(METAPATHS_DIR, 'dev_metapaths.json'), 'w') as f:\n",
    "        json.dump(dev_metapaths, f)\n",
    "\n",
    "else:\n",
    "    with open(osp.join(METAPATHS_DIR, 'dev_metapaths.json'), 'r') as f:\n",
    "        dev_metapaths = json.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for each value, turn it into a ratio of value/ sum of all values\n",
    "test_ratios = {k: v/sum(test_metapaths.values()) for k, v in test_metapaths.items()}\n",
    "train_ratios = {k: v/sum(train_metapaths.values()) for k, v in train_metapaths.items()}\n",
    "dev_ratios = {k: v/sum(dev_metapaths.values()) for k, v in dev_metapaths.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get any keys where the value is zero\n",
    "zero_train = {k: v for k, v in train_ratios.items() if v == 0}\n",
    "zero_test = {k: v for k, v in test_ratios.items() if v == 0}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2332"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(zero_test.keys()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1349"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(zero_train.keys()))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Are there metapaths with zero instances in the training set but instances in the test set?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# nonzero keys in test\n",
    "nonzero_test = {k: v for k, v in test_ratios.items() if v != 0}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1214"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(zero_train.keys()) - set(nonzero_test.keys()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "135"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(zero_train.keys()) - set(zero_test.keys()))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Are the most common keys mostly the same?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "top_test = {tup[0] for tup in Counter(test_ratios).most_common(100)}\n",
    "top_train = {tup[0] for tup in Counter(train_ratios).most_common(100)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(top_test & top_train) / len(top_test)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It seems they are mostly the same, but not exactly."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# MARS Input Preparation\n",
    "\n",
    "What if we want to get these in a format similar to the MARS output?\n",
    "\n",
    "First let's give them all a uniform probability:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "#prob = 1 / (2 * len(CtoBP_metapaths))\n",
    "prob = 0.5"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can split the metapaths into a series of individual relations:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_forward_rule_body(mpath, conf, metagraph):\n",
    "    \"\"\"Gets the rule body for a given metapath in the format that MARS requires\n",
    "        :param mpath: the metapath instance, written in hetnetpy format\n",
    "        :param conf: the confidence value to use for the rule\n",
    "        :param metagraph: the metagraph instance, as a hetnetpy object\n",
    "    \"\"\"\n",
    "    st_edges = metaedges_from_metapath(mpath, standardize_by=metagraph)\n",
    "    directional_edges = metaedges_from_metapath(mpath)\n",
    "    rule_head = re.split('>|<|[^A-Z]', mpath)\n",
    "    rule_head = rule_head[0] + 't' + rule_head[-1]\n",
    "    rule_body = [conf, rule_head]\n",
    "    for st, dic in zip(st_edges, directional_edges):\n",
    "        if '>' in dic:\n",
    "            rule_body.append(st)\n",
    "        else:\n",
    "            rule_body.append('_' + st)\n",
    "    return rule_body"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'edge abbreviation not found: u>'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[38], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[39m# just testing the function\u001b[39;00m\n\u001b[0;32m----> 2\u001b[0m get_forward_rule_body(\u001b[39m'\u001b[39;49m\u001b[39mCu>G<uCd>Gp>BP\u001b[39;49m\u001b[39m'\u001b[39;49m, prob, metagraph)\n",
      "Cell \u001b[0;32mIn[37], line 7\u001b[0m, in \u001b[0;36mget_forward_rule_body\u001b[0;34m(mpath, conf, metagraph)\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mget_forward_rule_body\u001b[39m(mpath, conf, metagraph):\n\u001b[1;32m      2\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"Gets the rule body for a given metapath in the format that MARS requires\u001b[39;00m\n\u001b[1;32m      3\u001b[0m \u001b[39m        :param mpath: the metapath instance, written in hetnetpy format\u001b[39;00m\n\u001b[1;32m      4\u001b[0m \u001b[39m        :param conf: the confidence value to use for the rule\u001b[39;00m\n\u001b[1;32m      5\u001b[0m \u001b[39m        :param metagraph: the metagraph instance, as a hetnetpy object\u001b[39;00m\n\u001b[1;32m      6\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m----> 7\u001b[0m     st_edges \u001b[39m=\u001b[39m metaedges_from_metapath(mpath, standardize_by\u001b[39m=\u001b[39;49mmetagraph)\n\u001b[1;32m      8\u001b[0m     directional_edges \u001b[39m=\u001b[39m metaedges_from_metapath(mpath)\n\u001b[1;32m      9\u001b[0m     rule_head \u001b[39m=\u001b[39m re\u001b[39m.\u001b[39msplit(\u001b[39m'\u001b[39m\u001b[39m>|<|[^A-Z]\u001b[39m\u001b[39m'\u001b[39m, mpath)\n",
      "File \u001b[0;32m~/.local/lib/python3.10/site-packages/hetnetpy/abbreviation.py:121\u001b[0m, in \u001b[0;36mmetaedges_from_metapath\u001b[0;34m(abbreviation, standardize_by)\u001b[0m\n\u001b[1;32m    113\u001b[0m \u001b[39m\u001b[39m\u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m    114\u001b[0m \u001b[39mGet the abbreviated metaedges for an abbreviated metapath.\u001b[39;00m\n\u001b[1;32m    115\u001b[0m \u001b[39mPass a hetnetpy.MetaGraph object to `standardize_by` to standardize metaedge\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    118\u001b[0m \u001b[39mabbreviation. Default (`None`) does not standardize.\u001b[39;00m\n\u001b[1;32m    119\u001b[0m \u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m    120\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(standardize_by, hetnetpy\u001b[39m.\u001b[39mhetnet\u001b[39m.\u001b[39mMetaGraph):\n\u001b[0;32m--> 121\u001b[0m     metapath \u001b[39m=\u001b[39m standardize_by\u001b[39m.\u001b[39;49mmetapath_from_abbrev(abbreviation)\n\u001b[1;32m    122\u001b[0m     \u001b[39mreturn\u001b[39;00m [metaedge\u001b[39m.\u001b[39mget_standard_abbrev() \u001b[39mfor\u001b[39;00m metaedge \u001b[39min\u001b[39;00m metapath]\n\u001b[1;32m    123\u001b[0m \u001b[39m# Note that this is a valid regex module pattern but will not work in the\u001b[39;00m\n\u001b[1;32m    124\u001b[0m \u001b[39m# re module due to \"look-behind requires fixed-width pattern\".\u001b[39;00m\n",
      "File \u001b[0;32m~/.local/lib/python3.10/site-packages/hetnetpy/hetnet.py:437\u001b[0m, in \u001b[0;36mMetaGraph.metapath_from_abbrev\u001b[0;34m(self, abbrev)\u001b[0m\n\u001b[1;32m    435\u001b[0m metaedge_abbrevs \u001b[39m=\u001b[39m hetnetpy\u001b[39m.\u001b[39mabbreviation\u001b[39m.\u001b[39mmetaedges_from_metapath(abbrev)\n\u001b[1;32m    436\u001b[0m \u001b[39mfor\u001b[39;00m metaedge_abbrev \u001b[39min\u001b[39;00m metaedge_abbrevs:\n\u001b[0;32m--> 437\u001b[0m     metaedge_id \u001b[39m=\u001b[39m hetnetpy\u001b[39m.\u001b[39;49mabbreviation\u001b[39m.\u001b[39;49mmetaedge_id_from_abbreviation(\n\u001b[1;32m    438\u001b[0m         \u001b[39mself\u001b[39;49m, metaedge_abbrev\n\u001b[1;32m    439\u001b[0m     )\n\u001b[1;32m    440\u001b[0m     metaedges\u001b[39m.\u001b[39mappend(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mget_edge(metaedge_id))\n\u001b[1;32m    441\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mget_metapath_from_edges(\u001b[39mtuple\u001b[39m(metaedges))\n",
      "File \u001b[0;32m~/.local/lib/python3.10/site-packages/hetnetpy/abbreviation.py:154\u001b[0m, in \u001b[0;36mmetaedge_id_from_abbreviation\u001b[0;34m(metagraph, abbreviation)\u001b[0m\n\u001b[1;32m    152\u001b[0m         \u001b[39mbreak\u001b[39;00m\n\u001b[1;32m    153\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m--> 154\u001b[0m     \u001b[39mraise\u001b[39;00m \u001b[39mKeyError\u001b[39;00m(\u001b[39m\"\u001b[39m\u001b[39medge abbreviation not found: \u001b[39m\u001b[39m{}\u001b[39;00m\u001b[39m\"\u001b[39m\u001b[39m.\u001b[39mformat(edge_abbrev))\n\u001b[1;32m    155\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39m\"\u001b[39m\u001b[39m>\u001b[39m\u001b[39m\"\u001b[39m \u001b[39min\u001b[39;00m abbreviation:\n\u001b[1;32m    156\u001b[0m     direction \u001b[39m=\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mforward\u001b[39m\u001b[39m\"\u001b[39m\n",
      "\u001b[0;31mKeyError\u001b[0m: 'edge abbreviation not found: u>'"
     ]
    }
   ],
   "source": [
    "# just testing the function\n",
    "get_forward_rule_body('Cu>G<uCd>Gp>BP', prob, metagraph)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_reverse_rule_body(forward_body):\n",
    "    \"\"\"Gets the REVERSE rule body for a given metapath in the format that MARS requires\n",
    "        :param forward_body: the forward rule body as a list\n",
    "    \"\"\"\n",
    "    rev_body = [forward_body[0], '_' + forward_body[1]]\n",
    "    for i in reversed(forward_body[2:]):\n",
    "        if '_' in i:\n",
    "            rev_body.append(re.sub('_', '', i))\n",
    "        else:\n",
    "            rev_body.append('_' + i)\n",
    "    return rev_body"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.5, '_CtBP', '_GpBP', '_CdG', 'CuG', '_CuG']"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fw = get_forward_rule_body('Cu>G<uCd>Gp>BP', prob, metagraph)\n",
    "get_reverse_rule_body(fw)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we'll do it for all the compound to BP metapaths!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'CtBP': [[0.5, 'CtBP', 'CdA', 'ApBP'], [0.5, 'CtBP', 'CuA', 'ApBP'], [0.5, 'CtBP', 'CdG', 'GpBP'], [0.5, 'CtBP', 'CuG', 'GpBP'], [0.5, 'CtBP', 'CdB', 'BpBP'], [0.5, 'CtBP', 'CuB', 'BpBP'], [0.5, 'CtBP', 'CdD', 'DpBP'], [0.5, 'CtBP', 'CuD', 'DpBP'], [0.5, 'CtBP', 'CdA', 'AiA', 'ApBP'], [0.5, 'CtBP', 'CdA', 'AiG', 'GpBP'], [0.5, 'CtBP', 'CdA', 'AiB', 'BpBP'], [0.5, 'CtBP', 'CdA', 'AiD', 'DpBP'], [0.5, 'CtBP', 'CuA', 'AiA', 'ApBP'], [0.5, 'CtBP', 'CuA', 'AiG', 'GpBP'], [0.5, 'CtBP', 'CuA', 'AiB', 'BpBP'], [0.5, 'CtBP', 'CuA', 'AiD', 'DpBP'], [0.5, 'CtBP', 'CdG', 'GiA', 'ApBP'], [0.5, 'CtBP', 'CdG', 'GiG', 'GpBP'], [0.5, 'CtBP', 'CdG', 'GiB', 'BpBP'], [0.5, 'CtBP', 'CdG', 'GiD', 'DpBP'], [0.5, 'CtBP', 'CuG', 'GiA', 'ApBP'], [0.5, 'CtBP', 'CuG', 'GiG', 'GpBP'], [0.5, 'CtBP', 'CuG', 'GiB', 'BpBP'], [0.5, 'CtBP', 'CuG', 'GiD', 'DpBP'], [0.5, 'CtBP', 'CdB', 'BiA', 'ApBP'], [0.5, 'CtBP', 'CdB', 'BiG', 'GpBP'], [0.5, 'CtBP', 'CdB', 'BiB', 'BpBP'], [0.5, 'CtBP', 'CdB', 'BiD', 'DpBP'], [0.5, 'CtBP', 'CuB', 'BiA', 'ApBP'], [0.5, 'CtBP', 'CuB', 'BiG', 'GpBP'], [0.5, 'CtBP', 'CuB', 'BiB', 'BpBP'], [0.5, 'CtBP', 'CuB', 'BiD', 'DpBP'], [0.5, 'CtBP', 'CdD', 'DiA', 'ApBP'], [0.5, 'CtBP', 'CdD', 'DiG', 'GpBP'], [0.5, 'CtBP', 'CdD', 'DiB', 'BpBP'], [0.5, 'CtBP', 'CdD', 'DiD', 'DpBP'], [0.5, 'CtBP', 'CuD', 'DiA', 'ApBP'], [0.5, 'CtBP', 'CuD', 'DiG', 'GpBP'], [0.5, 'CtBP', 'CuD', 'DiB', 'BpBP'], [0.5, 'CtBP', 'CuD', 'DiD', 'DpBP'], [0.5, 'CtBP', 'CdA', 'AiA', 'AiA', 'ApBP'], [0.5, 'CtBP', 'CdA', 'AiA', 'AiG', 'GpBP'], [0.5, 'CtBP', 'CdA', 'AiA', 'AiB', 'BpBP'], [0.5, 'CtBP', 'CdA', 'AiA', 'AiD', 'DpBP'], [0.5, 'CtBP', 'CdA', 'AiG', 'GiA', 'ApBP'], [0.5, 'CtBP', 'CdA', 'AiG', 'GiG', 'GpBP'], [0.5, 'CtBP', 'CdA', 'AiG', 'GiB', 'BpBP'], [0.5, 'CtBP', 'CdA', 'AiG', 'GiD', 'DpBP'], [0.5, 'CtBP', 'CdA', 'AiB', 'BiA', 'ApBP'], [0.5, 'CtBP', 'CdA', 'AiB', 'BiG', 'GpBP'], [0.5, 'CtBP', 'CdA', 'AiB', 'BiB', 'BpBP'], [0.5, 'CtBP', 'CdA', 'AiB', 'BiD', 'DpBP'], [0.5, 'CtBP', 'CdA', 'AiD', 'DiA', 'ApBP'], [0.5, 'CtBP', 'CdA', 'AiD', 'DiG', 'GpBP'], [0.5, 'CtBP', 'CdA', 'AiD', 'DiB', 'BpBP'], [0.5, 'CtBP', 'CdA', 'AiD', 'DiD', 'DpBP'], [0.5, 'CtBP', 'CuA', 'AiA', 'AiA', 'ApBP'], [0.5, 'CtBP', 'CuA', 'AiA', 'AiG', 'GpBP'], [0.5, 'CtBP', 'CuA', 'AiA', 'AiB', 'BpBP'], [0.5, 'CtBP', 'CuA', 'AiA', 'AiD', 'DpBP'], [0.5, 'CtBP', 'CuA', 'AiG', 'GiA', 'ApBP'], [0.5, 'CtBP', 'CuA', 'AiG', 'GiG', 'GpBP'], [0.5, 'CtBP', 'CuA', 'AiG', 'GiB', 'BpBP'], [0.5, 'CtBP', 'CuA', 'AiG', 'GiD', 'DpBP'], [0.5, 'CtBP', 'CuA', 'AiB', 'BiA', 'ApBP'], [0.5, 'CtBP', 'CuA', 'AiB', 'BiG', 'GpBP'], [0.5, 'CtBP', 'CuA', 'AiB', 'BiB', 'BpBP'], [0.5, 'CtBP', 'CuA', 'AiB', 'BiD', 'DpBP'], [0.5, 'CtBP', 'CuA', 'AiD', 'DiA', 'ApBP'], [0.5, 'CtBP', 'CuA', 'AiD', 'DiG', 'GpBP'], [0.5, 'CtBP', 'CuA', 'AiD', 'DiB', 'BpBP'], [0.5, 'CtBP', 'CuA', 'AiD', 'DiD', 'DpBP'], [0.5, 'CtBP', 'CdG', 'GiA', 'AiA', 'ApBP'], [0.5, 'CtBP', 'CdG', 'GiA', 'AiG', 'GpBP'], [0.5, 'CtBP', 'CdG', 'GiA', 'AiB', 'BpBP'], [0.5, 'CtBP', 'CdG', 'GiA', 'AiD', 'DpBP'], [0.5, 'CtBP', 'CdG', 'GiG', 'GiA', 'ApBP'], [0.5, 'CtBP', 'CdG', 'GiG', 'GiG', 'GpBP'], [0.5, 'CtBP', 'CdG', 'GiG', 'GiB', 'BpBP'], [0.5, 'CtBP', 'CdG', 'GiG', 'GiD', 'DpBP'], [0.5, 'CtBP', 'CdG', 'GiB', 'BiA', 'ApBP'], [0.5, 'CtBP', 'CdG', 'GiB', 'BiG', 'GpBP'], [0.5, 'CtBP', 'CdG', 'GiB', 'BiB', 'BpBP'], [0.5, 'CtBP', 'CdG', 'GiB', 'BiD', 'DpBP'], [0.5, 'CtBP', 'CdG', 'GiD', 'DiA', 'ApBP'], [0.5, 'CtBP', 'CdG', 'GiD', 'DiG', 'GpBP'], [0.5, 'CtBP', 'CdG', 'GiD', 'DiB', 'BpBP'], [0.5, 'CtBP', 'CdG', 'GiD', 'DiD', 'DpBP'], [0.5, 'CtBP', 'CuG', 'GiA', 'AiA', 'ApBP'], [0.5, 'CtBP', 'CuG', 'GiA', 'AiG', 'GpBP'], [0.5, 'CtBP', 'CuG', 'GiA', 'AiB', 'BpBP'], [0.5, 'CtBP', 'CuG', 'GiA', 'AiD', 'DpBP'], [0.5, 'CtBP', 'CuG', 'GiG', 'GiA', 'ApBP'], [0.5, 'CtBP', 'CuG', 'GiG', 'GiG', 'GpBP'], [0.5, 'CtBP', 'CuG', 'GiG', 'GiB', 'BpBP'], [0.5, 'CtBP', 'CuG', 'GiG', 'GiD', 'DpBP'], [0.5, 'CtBP', 'CuG', 'GiB', 'BiA', 'ApBP'], [0.5, 'CtBP', 'CuG', 'GiB', 'BiG', 'GpBP'], [0.5, 'CtBP', 'CuG', 'GiB', 'BiB', 'BpBP'], [0.5, 'CtBP', 'CuG', 'GiB', 'BiD', 'DpBP'], [0.5, 'CtBP', 'CuG', 'GiD', 'DiA', 'ApBP'], [0.5, 'CtBP', 'CuG', 'GiD', 'DiG', 'GpBP'], [0.5, 'CtBP', 'CuG', 'GiD', 'DiB', 'BpBP'], [0.5, 'CtBP', 'CuG', 'GiD', 'DiD', 'DpBP'], [0.5, 'CtBP', 'CdB', 'BiA', 'AiA', 'ApBP'], [0.5, 'CtBP', 'CdB', 'BiA', 'AiG', 'GpBP'], [0.5, 'CtBP', 'CdB', 'BiA', 'AiB', 'BpBP'], [0.5, 'CtBP', 'CdB', 'BiA', 'AiD', 'DpBP'], [0.5, 'CtBP', 'CdB', 'BiG', 'GiA', 'ApBP'], [0.5, 'CtBP', 'CdB', 'BiG', 'GiG', 'GpBP'], [0.5, 'CtBP', 'CdB', 'BiG', 'GiB', 'BpBP'], [0.5, 'CtBP', 'CdB', 'BiG', 'GiD', 'DpBP'], [0.5, 'CtBP', 'CdB', 'BiB', 'BiA', 'ApBP'], [0.5, 'CtBP', 'CdB', 'BiB', 'BiG', 'GpBP'], [0.5, 'CtBP', 'CdB', 'BiB', 'BiB', 'BpBP'], [0.5, 'CtBP', 'CdB', 'BiB', 'BiD', 'DpBP'], [0.5, 'CtBP', 'CdB', 'BiD', 'DiA', 'ApBP'], [0.5, 'CtBP', 'CdB', 'BiD', 'DiG', 'GpBP'], [0.5, 'CtBP', 'CdB', 'BiD', 'DiB', 'BpBP'], [0.5, 'CtBP', 'CdB', 'BiD', 'DiD', 'DpBP'], [0.5, 'CtBP', 'CuB', 'BiA', 'AiA', 'ApBP'], [0.5, 'CtBP', 'CuB', 'BiA', 'AiG', 'GpBP'], [0.5, 'CtBP', 'CuB', 'BiA', 'AiB', 'BpBP'], [0.5, 'CtBP', 'CuB', 'BiA', 'AiD', 'DpBP'], [0.5, 'CtBP', 'CuB', 'BiG', 'GiA', 'ApBP'], [0.5, 'CtBP', 'CuB', 'BiG', 'GiG', 'GpBP'], [0.5, 'CtBP', 'CuB', 'BiG', 'GiB', 'BpBP'], [0.5, 'CtBP', 'CuB', 'BiG', 'GiD', 'DpBP'], [0.5, 'CtBP', 'CuB', 'BiB', 'BiA', 'ApBP'], [0.5, 'CtBP', 'CuB', 'BiB', 'BiG', 'GpBP'], [0.5, 'CtBP', 'CuB', 'BiB', 'BiB', 'BpBP'], [0.5, 'CtBP', 'CuB', 'BiB', 'BiD', 'DpBP'], [0.5, 'CtBP', 'CuB', 'BiD', 'DiA', 'ApBP'], [0.5, 'CtBP', 'CuB', 'BiD', 'DiG', 'GpBP'], [0.5, 'CtBP', 'CuB', 'BiD', 'DiB', 'BpBP'], [0.5, 'CtBP', 'CuB', 'BiD', 'DiD', 'DpBP'], [0.5, 'CtBP', 'CdD', 'DiA', 'AiA', 'ApBP'], [0.5, 'CtBP', 'CdD', 'DiA', 'AiG', 'GpBP'], [0.5, 'CtBP', 'CdD', 'DiA', 'AiB', 'BpBP'], [0.5, 'CtBP', 'CdD', 'DiA', 'AiD', 'DpBP'], [0.5, 'CtBP', 'CdD', 'DiG', 'GiA', 'ApBP'], [0.5, 'CtBP', 'CdD', 'DiG', 'GiG', 'GpBP'], [0.5, 'CtBP', 'CdD', 'DiG', 'GiB', 'BpBP'], [0.5, 'CtBP', 'CdD', 'DiG', 'GiD', 'DpBP'], [0.5, 'CtBP', 'CdD', 'DiB', 'BiA', 'ApBP'], [0.5, 'CtBP', 'CdD', 'DiB', 'BiG', 'GpBP'], [0.5, 'CtBP', 'CdD', 'DiB', 'BiB', 'BpBP'], [0.5, 'CtBP', 'CdD', 'DiB', 'BiD', 'DpBP'], [0.5, 'CtBP', 'CdD', 'DiD', 'DiA', 'ApBP'], [0.5, 'CtBP', 'CdD', 'DiD', 'DiG', 'GpBP'], [0.5, 'CtBP', 'CdD', 'DiD', 'DiB', 'BpBP'], [0.5, 'CtBP', 'CdD', 'DiD', 'DiD', 'DpBP'], [0.5, 'CtBP', 'CuD', 'DiA', 'AiA', 'ApBP'], [0.5, 'CtBP', 'CuD', 'DiA', 'AiG', 'GpBP'], [0.5, 'CtBP', 'CuD', 'DiA', 'AiB', 'BpBP'], [0.5, 'CtBP', 'CuD', 'DiA', 'AiD', 'DpBP'], [0.5, 'CtBP', 'CuD', 'DiG', 'GiA', 'ApBP'], [0.5, 'CtBP', 'CuD', 'DiG', 'GiG', 'GpBP'], [0.5, 'CtBP', 'CuD', 'DiG', 'GiB', 'BpBP'], [0.5, 'CtBP', 'CuD', 'DiG', 'GiD', 'DpBP'], [0.5, 'CtBP', 'CuD', 'DiB', 'BiA', 'ApBP'], [0.5, 'CtBP', 'CuD', 'DiB', 'BiG', 'GpBP'], [0.5, 'CtBP', 'CuD', 'DiB', 'BiB', 'BpBP'], [0.5, 'CtBP', 'CuD', 'DiB', 'BiD', 'DpBP'], [0.5, 'CtBP', 'CuD', 'DiD', 'DiA', 'ApBP'], [0.5, 'CtBP', 'CuD', 'DiD', 'DiG', 'GpBP'], [0.5, 'CtBP', 'CuD', 'DiD', 'DiB', 'BpBP'], [0.5, 'CtBP', 'CuD', 'DiD', 'DiD', 'DpBP']]}\n"
     ]
    }
   ],
   "source": [
    "rules = {}\n",
    "rules['CtBP'] = [get_forward_rule_body(str(mpath), prob, metagraph) for mpath in CtoBP_metapaths]\n",
    "print(rules)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['CtBP'])"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rules.keys()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's write it to a file we can use for MARS:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(osp.join(MARS_DIR, 'rules.txt'), 'w') as f:\n",
    "    json.dump(rules, f)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's get the input triples in a format consistent with MARS's input files:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def convert_to_mars(df):\n",
    "    \"\"\"Converts the KG into a format that MARS can use\n",
    "    \"\"\"\n",
    "    df['metaedge'] = [kind_to_abbev[row['source_node_type']] + kind_to_abbev[row['edge_type']] + \\\n",
    "                       kind_to_abbev[row['target_node_type']] for i, row in df.iterrows()]\n",
    "    return df[['source', 'metaedge', 'target']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = convert_to_mars(pd.read_csv(osp.join(MOA_NET_PROTCLASSED, 'train.tsv'), sep='\\t'))\n",
    "test = convert_to_mars(pd.read_csv(osp.join(MOA_NET_PROTCLASSED, 'test.tsv'), sep='\\t'))\n",
    "val = convert_to_mars(pd.read_csv(osp.join(MOA_NET_PROTCLASSED, 'dev.tsv'), sep='\\t'))\n",
    "\n",
    "kg_triples = convert_to_mars(pd.read_csv(osp.join(MOA_NET_PROTCLASSED, 'kg_with_train_smpls.tsv'), sep='\\t'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>source</th>\n",
       "      <th>metaedge</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ncbigene:6300</td>\n",
       "      <td>DiA</td>\n",
       "      <td>ncbigene:2354</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ncbigene:55920</td>\n",
       "      <td>GiG</td>\n",
       "      <td>ncbigene:79003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ncbigene:5925</td>\n",
       "      <td>AiA</td>\n",
       "      <td>ncbigene:3024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ncbigene:23396</td>\n",
       "      <td>DiG</td>\n",
       "      <td>ncbigene:6383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ncbigene:440275</td>\n",
       "      <td>DiA</td>\n",
       "      <td>ncbigene:4780</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            source metaedge          target\n",
       "0    ncbigene:6300      DiA   ncbigene:2354\n",
       "1   ncbigene:55920      GiG  ncbigene:79003\n",
       "2    ncbigene:5925      AiA   ncbigene:3024\n",
       "3   ncbigene:23396      DiG   ncbigene:6383\n",
       "4  ncbigene:440275      DiA   ncbigene:4780"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "kg_triples.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get the inverse triples of the KG:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inverse_triples = kg_triples[['source', 'target', 'metaedge']].copy()\n",
    "inverse_triples['inv_metaedge'] = ['_' + i for i in inverse_triples['metaedge']]\n",
    "inverse_triples = inverse_triples[['target', 'inv_metaedge', 'source']]\n",
    "\n",
    "# get rid of those with _CtBP\n",
    "#inverse_triples = inverse_triples.loc[inverse_triples['inv_metaedge'] != '_CtBP']"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Great, write to the MARS folder:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "kg_triples.to_csv(osp.join(MARS_DIR, 'graph_triples.txt'), sep='\\t', index=False, header=False)\n",
    "inverse_triples.to_csv(osp.join(MARS_DIR, 'graph_inverses.txt'), sep='\\t', index=False, header=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train.to_csv(osp.join(MARS_DIR, 'train.txt'), sep='\\t', index=False, header=False)\n",
    "test.to_csv(osp.join(MARS_DIR, 'test.txt'), sep='\\t', index=False, header=False)\n",
    "val.to_csv(osp.join(MARS_DIR, 'dev.txt'), sep='\\t', index=False, header=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we need dictionaries to map the entities and relations to unique IDs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "entities = {\"PAD\": 0, \"UNK\": 1}\n",
    "\n",
    "next_id = 2\n",
    "for df in kg_triples, train, test, val:\n",
    "    for i in df['source'].unique():\n",
    "        if i not in entities:\n",
    "            entities[i] = next_id\n",
    "            next_id += 1\n",
    "    for i in df['target'].unique():\n",
    "        if i not in entities:\n",
    "            entities[i] = next_id\n",
    "            next_id += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "relations = {\"PAD\": 0, \"DUMMY_START_RELATION\": 1, \"NO_OP\": 2, \"UNK\": 3}\n",
    "\n",
    "# combine the metaedges and inverse metaedges\n",
    "metaedges = set(kg_triples['metaedge']) | (set(inverse_triples['inv_metaedge']))\n",
    "\n",
    "next_id = 4\n",
    "for rel in metaedges:\n",
    "    relations[rel] = next_id\n",
    "    next_id += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# write dictionary to json file\n",
    "VOCAB_DIR = osp.join(MARS_DIR, 'vocab')\n",
    "os.makedirs(VOCAB_DIR, exist_ok=True)\n",
    "\n",
    "with open(osp.join(VOCAB_DIR, 'entity_vocab.json'), 'w') as f:\n",
    "    json.dump(entities, f)\n",
    "\n",
    "with open(osp.join(VOCAB_DIR, 'relation_vocab.json'), 'w') as f:\n",
    "    json.dump(relations, f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "meta_mapping = {v: k for k, v in kind_to_abbev.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(osp.join(VOCAB_DIR, 'meta_mapping.json'), 'w') as f:\n",
    "    json.dump(meta_mapping, f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 'Compound',\n",
       " 'G': 'Gene',\n",
       " 'BP': 'Biological Process',\n",
       " 'u': 'upregulates',\n",
       " 'd': 'downregulates',\n",
       " 'i': 'interacts',\n",
       " 'p': 'participates',\n",
       " 't': 'induces',\n",
       " 'A': 'DNA-binding',\n",
       " 'B': 'Receptor',\n",
       " 'D': 'Transferase'}"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "meta_mapping"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "moanet_env",
   "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": "3.10.13"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}