NetworkAnalysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "28930fd4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import networkx as nx\n",
    "from pyvis.network import Network\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "63f432ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_gene_drug = pd.read_csv(\"Data/Munged_Data/gene_drug.csv\")\n",
    "df_disease_drug = pd.read_csv(\"Data/Munged_Data/disease_drugbank_assocations.tsv\", sep='\\t')\n",
    "df_drug_disease = pd.read_csv(\"Data/Munged_Data/drug_disease_accs.tsv\", sep='\\t')\n",
    "df_disease_go_bp_ida_assocations = pd.read_csv(\"Data/Munged_Data/disease_go_bp_ida_assocations.tsv\", sep='\\t')\n",
    "\n",
    "drug_list = list(set(df_disease_drug[\"DrugBankID\"]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4902e83c",
   "metadata": {},
   "outputs": [],
   "source": [
    "G = nx.from_pandas_edgelist(df_disease_drug, 'diseaseId', 'DrugBankID')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2348c84c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "nx.draw(G)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "39b1eecb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Graph with 7288 nodes and 630953 edges\n",
      "Network density: 0.023761282177469856\n"
     ]
    }
   ],
   "source": [
    "print(nx.info(G))\n",
    "print(\"Network density:\", nx.density(G))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "728d5059",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"500px\"\n",
       "            height=\"500px\"\n",
       "            src=\"disease_drug.html\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x7ff4d58dcbe0>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#pyvis visualization for a random subset\n",
    "G_sample = nx.from_pandas_edgelist(df_disease_drug.sample(n=1000), 'diseaseId', 'DrugBankID')\n",
    "net = Network(notebook=True, directed=False)\n",
    "net.from_nx(G_sample)\n",
    "net.show_buttons(filter_=True)\n",
    "net.show( \"disease_drug_sample.html\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "458479fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "F = nx.from_pandas_edgelist(df_disease_go_bp_ida_assocations, 'diseaseId', 'GO', \"Edge_Strenght\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ca587f0f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Graph with 12988 nodes and 322097 edges\n",
      "Network density: 0.003819139790805944\n"
     ]
    }
   ],
   "source": [
    "print(nx.info(F))\n",
    "print(\"Network density:\", nx.density(F))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "76c8cc0c",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"500px\"\n",
       "            height=\"500px\"\n",
       "            src=\"disease_disease_sample.html\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x7ff4e3698f40>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#pyvis visualization for a random subset\n",
    "F_sample = nx.from_pandas_edgelist(df_disease_go_bp_ida_assocations.sample(n=1000), 'diseaseId', 'GO', \"Edge_Strenght\")\n",
    "net = Network(notebook=True, directed=False)\n",
    "net.from_nx(F_sample)\n",
    "net.show_buttons(filter_=True)\n",
    "net.show( \"disease_disease_sample.html\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0eccb48e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Graph with 15100 nodes and 953050 edges\n",
      "Network density: 0.00836027472544342\n"
     ]
    }
   ],
   "source": [
    "H = nx.compose(G,F)\n",
    "print(nx.info(H))\n",
    "print(\"Network density:\", nx.density(H))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "cf7bb69d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"500px\"\n",
       "            height=\"500px\"\n",
       "            src=\"Network_sample.html\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x7f8ee6aa50d0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#pyvis visualization for a random subset\n",
    "H_sample = nx.compose(G_sample, F_sample)\n",
    "net = Network(notebook=True, directed=False)\n",
    "net.from_nx(H_sample)\n",
    "net.show_buttons(filter_=True)\n",
    "net.show( \"network_sample.html\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "34f71301",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['C0494261', 'GO:0006974', 'C0007102', 'DB00907']\n"
     ]
    }
   ],
   "source": [
    "#example - finding a path between disease and drug\n",
    "print(nx.single_source_shortest_path(H, \"C0494261\")[\"DB00907\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "72bffd0d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def shortest_path_all_targets(disease_name, graph, drug_list):\n",
    "    paths = []\n",
    "    path_lengths = []\n",
    "    for drug in drug_list:\n",
    "        try:\n",
    "            temp_path = nx.single_source_shortest_path(graph, disease_name)[drug]\n",
    "            if len(temp_path)>1:\n",
    "                paths.append(temp_path)\n",
    "                path_lengths.append(len(temp_path))\n",
    "        except:\n",
    "            pass\n",
    "    return paths, path_lengths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "927fbc97",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['C1956147', 'DB00641']\n"
     ]
    }
   ],
   "source": [
    "#finding a shortest target drug\n",
    "paths, path_lengths = shortest_path_all_targets(\"C1956147\", H, drug_list)\n",
    "shortest_target = paths[np.argmin(path_lengths)]\n",
    "print(shortest_target)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b151300",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "365311cc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}