{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sqlalchemy import create_engine\n",
    "from sklearn import preprocessing\n",
    "import mysql.connector\n",
    "from pandas import DataFrame\n",
    "from sklearn.metrics import jaccard_score\n",
    "from numpy import logical_and as l_and, logical_not as l_not"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "disnet_db_ares = mysql.connector.connect(\n",
    "  host=\"138.4.130.153\",\n",
    "  port = \"30602\",\n",
    "  user=\"disnet_user\",\n",
    "  password=\"tYkX4JxV8p79\",\n",
    "  database=\"disnet_drugslayer\"\n",
    ")\n",
    "\n",
    " \n",
    "\n",
    "disnet_mysql_cursor = disnet_db_ares.cursor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_dummies(data,col_index,col_col):\n",
    "    \n",
    "    df_final= pd.get_dummies(data.set_index(col_index),col_col).reset_index()\n",
    "    df_final=df_final.drop_duplicates()\n",
    "    df_final=df_final.groupby(col_index).sum().reset_index()\n",
    "    \n",
    "    return df_final\n",
    "\n",
    "def convert(lista):\n",
    "    return tuple(i for i in lista)\n",
    "\n",
    "\n",
    "def get_normal_diseases(rare_disease):\n",
    "    '''\n",
    "    Esta funcion devuelve un dataframe con la enfermedad, los medicamentos y los targets que son utiles\n",
    "    para la enfermedad rara que se le pase como parametro'''\n",
    "    \n",
    "    \n",
    "    q= f'''SELECT DISTINCT\n",
    "        ds.disease_id, drug_name,dg.gene_id\n",
    "        FROM\n",
    "        disnet_biolayer.disease_gene dg \n",
    "        JOIN disnet_biolayer.disease ds ON ds.disease_id = dg.disease_id\n",
    "        JOIN disnet_biolayer.encodes e_uno ON dg.gene_id = e_uno.gene_id\n",
    "        JOIN disnet_drugslayer.has_code hc ON hc.code = e_uno.protein_id\n",
    "        JOIN disnet_biolayer.ppi ppi on ppi.protein1_id = e_uno.protein_id\n",
    "        JOIN disnet_biolayer.encodes e_dos on e_dos.protein_id = ppi.protein2_id\n",
    "        JOIN disnet_drugslayer.drug_target dt ON hc.id = dt.target_id\n",
    "        JOIN disnet_drugslayer.drug d ON dt.drug_id = d.drug_id\n",
    "        WHERE\n",
    "             hc.entity_id = 3\n",
    "             \n",
    "        and ds.ddf_type = \"disease\"\n",
    "        and ppi.quality = \"GOLD\"\n",
    "        AND dg.gene_id in (SELECT distinct(gene_id) FROM disnet_biolayer.disease_gene\n",
    "        where disease_id = \"{rare_disease}\")\n",
    "        ;'''\n",
    "    \n",
    "    df = pd.read_sql(q, con=disnet_db_ares)\n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "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>rare_dis_id</th>\n",
       "      <th>C0000731</th>\n",
       "      <th>C0000737</th>\n",
       "      <th>C0000744</th>\n",
       "      <th>C0000768</th>\n",
       "      <th>C0000771</th>\n",
       "      <th>C0000772</th>\n",
       "      <th>C0000786</th>\n",
       "      <th>C0000822</th>\n",
       "      <th>C0000832</th>\n",
       "      <th>...</th>\n",
       "      <th>C4540327</th>\n",
       "      <th>C4540342</th>\n",
       "      <th>C4540395</th>\n",
       "      <th>C4540493</th>\n",
       "      <th>C4540496</th>\n",
       "      <th>C4540534</th>\n",
       "      <th>C4540535</th>\n",
       "      <th>C4540536</th>\n",
       "      <th>C4540602</th>\n",
       "      <th>C4545381</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C0011195</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.023256</td>\n",
       "      <td>0.025641</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C0023944</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C0024901</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>C0027877</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C0036391</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.016949</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>C0265202</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.017241</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>C0751337</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>C0869083</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>9 rows × 12311 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  rare_dis_id  C0000731  C0000737  C0000744  C0000768  C0000771  C0000772  \\\n",
       "0    C0011195         0         0         0  0.000000         0         0   \n",
       "1    C0023944         0         0         0  0.000000         0         0   \n",
       "2    C0024901         0         0         0  0.000000         0         0   \n",
       "3    C0027877         0         0         0  0.000000         0         0   \n",
       "4    C0036391         0         0         0  0.016949         0         0   \n",
       "5    C0265202         0         0         0  0.017241         0         0   \n",
       "6    C0549463         0         0         0  0.000000         0         0   \n",
       "7    C0751337         0         0         0  0.000000         0         0   \n",
       "8    C0869083         0         0         0  0.000000         0         0   \n",
       "\n",
       "   C0000786  C0000822  C0000832  ...  C4540327  C4540342  C4540395  C4540493  \\\n",
       "0  0.023256  0.025641         0  ...         0         0         0         0   \n",
       "1  0.000000  0.000000         0  ...         0         0         0         0   \n",
       "2  0.000000  0.000000         0  ...         0         0         0         0   \n",
       "3  0.000000  0.000000         0  ...         0         0         0         0   \n",
       "4  0.000000  0.000000         0  ...         0         0         0         0   \n",
       "5  0.000000  0.000000         0  ...         0         0         0         0   \n",
       "6  0.000000  0.000000         0  ...         0         0         0         0   \n",
       "7  0.000000  0.000000         0  ...         0         0         0         0   \n",
       "8  0.000000  0.000000         0  ...         0         0         0         0   \n",
       "\n",
       "   C4540496  C4540534  C4540535  C4540536  C4540602  C4545381  \n",
       "0         0         0         0         0         0         0  \n",
       "1         0         0         0         0         0         0  \n",
       "2         0         0         0         0         0         0  \n",
       "3         0         0         0         0         0         0  \n",
       "4         0         0         0         0         0         0  \n",
       "5         0         0         0         0         0         0  \n",
       "6         0         0         0         0         0         0  \n",
       "7         0         0         0         0         0         0  \n",
       "8         0         0         0         0         0         0  \n",
       "\n",
       "[9 rows x 12311 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rare_diseases= ['C0011195', 'C0023944', 'C0024054', 'C0024901', 'C0027877', 'C0036391', 'C0265202', 'C0268059', 'C0549463', 'C0751337', 'C0869083', 'C1852146', 'C0796280']\n",
    "matrix_jac_ppi = pd.read_excel((\"mat_jacc_pip.xlsx\"),engine='openpyxl')\n",
    "matrix_jac_ppi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "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>disease_id</th>\n",
       "      <th>drug_name</th>\n",
       "      <th>gene_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C0002448</td>\n",
       "      <td>SORAFENIB TOSYLATE</td>\n",
       "      <td>673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C0002448</td>\n",
       "      <td>VEMURAFENIB</td>\n",
       "      <td>673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C0002448</td>\n",
       "      <td>SORAFENIB</td>\n",
       "      <td>673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>C0002448</td>\n",
       "      <td>REGORAFENIB</td>\n",
       "      <td>673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C0002448</td>\n",
       "      <td>DABRAFENIB</td>\n",
       "      <td>673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2710</th>\n",
       "      <td>C0699790</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>22858</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2711</th>\n",
       "      <td>C1306460</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>22858</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2712</th>\n",
       "      <td>C1691215</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>22858</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2713</th>\n",
       "      <td>C2675227</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>22858</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2714</th>\n",
       "      <td>C2981150</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>22858</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2715 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     disease_id           drug_name  gene_id\n",
       "0      C0002448  SORAFENIB TOSYLATE      673\n",
       "1      C0002448         VEMURAFENIB      673\n",
       "2      C0002448           SORAFENIB      673\n",
       "3      C0002448         REGORAFENIB      673\n",
       "4      C0002448          DABRAFENIB      673\n",
       "...         ...                 ...      ...\n",
       "2710   C0699790        FOSTAMATINIB    22858\n",
       "2711   C1306460        FOSTAMATINIB    22858\n",
       "2712   C1691215        FOSTAMATINIB    22858\n",
       "2713   C2675227        FOSTAMATINIB    22858\n",
       "2714   C2981150        FOSTAMATINIB    22858\n",
       "\n",
       "[2715 rows x 3 columns]"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_useful_diseases = get_normal_diseases(\"C0869083\")\n",
    "df_useful_diseases"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "352"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "useful_diseases = get_normal_diseases(\"C0869083\")['disease_id'].drop_duplicates().tolist()\n",
    "len(useful_diseases)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "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>rare_dis_id</th>\n",
       "      <th>C0000731</th>\n",
       "      <th>C0000737</th>\n",
       "      <th>C0000744</th>\n",
       "      <th>C0000768</th>\n",
       "      <th>C0000771</th>\n",
       "      <th>C0000772</th>\n",
       "      <th>C0000786</th>\n",
       "      <th>C0000822</th>\n",
       "      <th>C0000832</th>\n",
       "      <th>...</th>\n",
       "      <th>C4540327</th>\n",
       "      <th>C4540342</th>\n",
       "      <th>C4540395</th>\n",
       "      <th>C4540493</th>\n",
       "      <th>C4540496</th>\n",
       "      <th>C4540534</th>\n",
       "      <th>C4540535</th>\n",
       "      <th>C4540536</th>\n",
       "      <th>C4540602</th>\n",
       "      <th>C4545381</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>C0869083</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 12311 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  rare_dis_id  C0000731  C0000737  C0000744  C0000768  C0000771  C0000772  \\\n",
       "8    C0869083         0         0         0       0.0         0         0   \n",
       "\n",
       "   C0000786  C0000822  C0000832  ...  C4540327  C4540342  C4540395  C4540493  \\\n",
       "8       0.0       0.0         0  ...         0         0         0         0   \n",
       "\n",
       "   C4540496  C4540534  C4540535  C4540536  C4540602  C4545381  \n",
       "8         0         0         0         0         0         0  \n",
       "\n",
       "[1 rows x 12311 columns]"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = matrix_jac_ppi[matrix_jac_ppi.rare_dis_id == 'C0869083']\n",
    "df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "scrolled": true
   },
   "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>rare_dis_id</th>\n",
       "      <th>C0869083</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>C0346360</th>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C3263719</th>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0019624</th>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C1275336</th>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C1332969</th>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0600139</th>\n",
       "      <td>0.001592</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0376358</th>\n",
       "      <td>0.001548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C2239176</th>\n",
       "      <td>0.001522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0678222</th>\n",
       "      <td>0.001031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0006142</th>\n",
       "      <td>0.001025</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>352 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "rare_dis_id  C0869083\n",
       "C0346360     0.500000\n",
       "C3263719     0.500000\n",
       "C0019624     0.500000\n",
       "C1275336     0.500000\n",
       "C1332969     0.500000\n",
       "...               ...\n",
       "C0600139     0.001592\n",
       "C0376358     0.001548\n",
       "C2239176     0.001522\n",
       "C0678222     0.001031\n",
       "C0006142     0.001025\n",
       "\n",
       "[352 rows x 1 columns]"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_useful = df[['rare_dis_id'] + useful_diseases].set_index('rare_dis_id').transpose()\n",
    "df_useful.sort_values(by='C0869083', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_normal_dis_jacc=pd.DataFrame(df_useful['C0869083'].nlargest(5)).reset_index().rename(columns={'index':'disease_id', 'C0869083':'jacc_idx'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_normal_dis_jacc.to_excel(\"dis_jacc_C0869083_ppi.xlsx\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_normal_rare_drug=pd.merge(df_useful_diseases,df_normal_dis_jacc,on=\"disease_id\",how=\"inner\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "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>disease_id</th>\n",
       "      <th>drug_name</th>\n",
       "      <th>gene_id</th>\n",
       "      <th>jacc_idx</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C0019624</td>\n",
       "      <td>SORAFENIB TOSYLATE</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C0019624</td>\n",
       "      <td>VEMURAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C0019624</td>\n",
       "      <td>SORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>C0019624</td>\n",
       "      <td>REGORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C0019624</td>\n",
       "      <td>DABRAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>C0019624</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>C0019624</td>\n",
       "      <td>DABRAFENIB MESYLATE</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>C0019624</td>\n",
       "      <td>ENCORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>C0278600</td>\n",
       "      <td>SORAFENIB TOSYLATE</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>C0278600</td>\n",
       "      <td>VEMURAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>C0278600</td>\n",
       "      <td>SORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>C0278600</td>\n",
       "      <td>REGORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>C0278600</td>\n",
       "      <td>DABRAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>C0278600</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>C0278600</td>\n",
       "      <td>DABRAFENIB MESYLATE</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>C0278600</td>\n",
       "      <td>ENCORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>C0278851</td>\n",
       "      <td>SORAFENIB TOSYLATE</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>C0278851</td>\n",
       "      <td>VEMURAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>C0278851</td>\n",
       "      <td>SORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>C0278851</td>\n",
       "      <td>REGORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>C0278851</td>\n",
       "      <td>DABRAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>C0278851</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>C0278851</td>\n",
       "      <td>DABRAFENIB MESYLATE</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>C0278851</td>\n",
       "      <td>ENCORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>C0278882</td>\n",
       "      <td>SORAFENIB TOSYLATE</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>C0278882</td>\n",
       "      <td>VEMURAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>C0278882</td>\n",
       "      <td>SORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>C0278882</td>\n",
       "      <td>REGORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>C0278882</td>\n",
       "      <td>DABRAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>C0278882</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>C0278882</td>\n",
       "      <td>DABRAFENIB MESYLATE</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>C0278882</td>\n",
       "      <td>ENCORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>C0334438</td>\n",
       "      <td>SORAFENIB TOSYLATE</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>C0334438</td>\n",
       "      <td>VEMURAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>C0334438</td>\n",
       "      <td>SORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>C0334438</td>\n",
       "      <td>REGORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>C0334438</td>\n",
       "      <td>DABRAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>C0334438</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>C0334438</td>\n",
       "      <td>DABRAFENIB MESYLATE</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>C0334438</td>\n",
       "      <td>ENCORAFENIB</td>\n",
       "      <td>673</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   disease_id            drug_name  gene_id  jacc_idx\n",
       "0    C0019624   SORAFENIB TOSYLATE      673       0.5\n",
       "1    C0019624          VEMURAFENIB      673       0.5\n",
       "2    C0019624            SORAFENIB      673       0.5\n",
       "3    C0019624          REGORAFENIB      673       0.5\n",
       "4    C0019624           DABRAFENIB      673       0.5\n",
       "5    C0019624         FOSTAMATINIB      673       0.5\n",
       "6    C0019624  DABRAFENIB MESYLATE      673       0.5\n",
       "7    C0019624          ENCORAFENIB      673       0.5\n",
       "8    C0278600   SORAFENIB TOSYLATE      673       0.5\n",
       "9    C0278600          VEMURAFENIB      673       0.5\n",
       "10   C0278600            SORAFENIB      673       0.5\n",
       "11   C0278600          REGORAFENIB      673       0.5\n",
       "12   C0278600           DABRAFENIB      673       0.5\n",
       "13   C0278600         FOSTAMATINIB      673       0.5\n",
       "14   C0278600  DABRAFENIB MESYLATE      673       0.5\n",
       "15   C0278600          ENCORAFENIB      673       0.5\n",
       "16   C0278851   SORAFENIB TOSYLATE      673       0.5\n",
       "17   C0278851          VEMURAFENIB      673       0.5\n",
       "18   C0278851            SORAFENIB      673       0.5\n",
       "19   C0278851          REGORAFENIB      673       0.5\n",
       "20   C0278851           DABRAFENIB      673       0.5\n",
       "21   C0278851         FOSTAMATINIB      673       0.5\n",
       "22   C0278851  DABRAFENIB MESYLATE      673       0.5\n",
       "23   C0278851          ENCORAFENIB      673       0.5\n",
       "24   C0278882   SORAFENIB TOSYLATE      673       0.5\n",
       "25   C0278882          VEMURAFENIB      673       0.5\n",
       "26   C0278882            SORAFENIB      673       0.5\n",
       "27   C0278882          REGORAFENIB      673       0.5\n",
       "28   C0278882           DABRAFENIB      673       0.5\n",
       "29   C0278882         FOSTAMATINIB      673       0.5\n",
       "30   C0278882  DABRAFENIB MESYLATE      673       0.5\n",
       "31   C0278882          ENCORAFENIB      673       0.5\n",
       "32   C0334438   SORAFENIB TOSYLATE      673       0.5\n",
       "33   C0334438          VEMURAFENIB      673       0.5\n",
       "34   C0334438            SORAFENIB      673       0.5\n",
       "35   C0334438          REGORAFENIB      673       0.5\n",
       "36   C0334438           DABRAFENIB      673       0.5\n",
       "37   C0334438         FOSTAMATINIB      673       0.5\n",
       "38   C0334438  DABRAFENIB MESYLATE      673       0.5\n",
       "39   C0334438          ENCORAFENIB      673       0.5"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_normal_rare_drug"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_normal_rare_drug=df_normal_rare_drug.drop_duplicates()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "prueba=df_normal_rare_drug.groupby('drug_name').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "scrolled": true
   },
   "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>disease_id</th>\n",
       "      <th>gene_id</th>\n",
       "      <th>jacc_idx</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>drug_name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DABRAFENIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DABRAFENIB MESYLATE</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENCORAFENIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FOSTAMATINIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>REGORAFENIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SORAFENIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SORAFENIB TOSYLATE</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VEMURAFENIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     disease_id  gene_id  jacc_idx\n",
       "drug_name                                         \n",
       "DABRAFENIB                    5        5         5\n",
       "DABRAFENIB MESYLATE           5        5         5\n",
       "ENCORAFENIB                   5        5         5\n",
       "FOSTAMATINIB                  5        5         5\n",
       "REGORAFENIB                   5        5         5\n",
       "SORAFENIB                     5        5         5\n",
       "SORAFENIB TOSYLATE            5        5         5\n",
       "VEMURAFENIB                   5        5         5"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prueba"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "drug=prueba[prueba[\"disease_id\"]>4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "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>disease_id</th>\n",
       "      <th>gene_id</th>\n",
       "      <th>jacc_idx</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>drug_name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DABRAFENIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DABRAFENIB MESYLATE</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENCORAFENIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FOSTAMATINIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>REGORAFENIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SORAFENIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SORAFENIB TOSYLATE</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VEMURAFENIB</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     disease_id  gene_id  jacc_idx\n",
       "drug_name                                         \n",
       "DABRAFENIB                    5        5         5\n",
       "DABRAFENIB MESYLATE           5        5         5\n",
       "ENCORAFENIB                   5        5         5\n",
       "FOSTAMATINIB                  5        5         5\n",
       "REGORAFENIB                   5        5         5\n",
       "SORAFENIB                     5        5         5\n",
       "SORAFENIB TOSYLATE            5        5         5\n",
       "VEMURAFENIB                   5        5         5"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drug"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "drug.to_excel(\"drugs_C0869083_ppi.xlsx\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "drug_all=df_normal_rare_drug[df_normal_rare_drug[\"disease_id\"]==\"C0869083\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "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>disease_id</th>\n",
       "      <th>drug_name</th>\n",
       "      <th>gene_id</th>\n",
       "      <th>jacc_idx</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [disease_id, drug_name, gene_id, jacc_idx]\n",
       "Index: []"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drug_all"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "drug_all.to_excel(\"drugs_all_C0549463_ppi.xlsx\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}