{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "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": 6,
   "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": 14,
   "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",
    "        d.disease_id, drug_name,dg.gene_id\n",
    "        FROM\n",
    "        disnet_biolayer.disease_variant dv\n",
    "            JOIN disnet_biolayer.disease d on dv.disease_id = d.disease_id\n",
    "            JOIN disnet_biolayer.variant v on v.variant_id = dv.variant_id\n",
    "            JOIN disnet_biolayer.disease_gene dg ON d.disease_id = dg.disease_id\n",
    "            JOIN disnet_biolayer.encodes e ON dg.gene_id = e.gene_id\n",
    "            JOIN disnet_drugslayer.has_code hc ON hc.code = e.protein_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",
    "             and v.chromosome is not null\n",
    "        and d.ddf_type = \"disease\"\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": 4,
   "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>C0000744</th>\n",
       "      <th>C0000846</th>\n",
       "      <th>C0000889</th>\n",
       "      <th>C0001080</th>\n",
       "      <th>C0001127</th>\n",
       "      <th>C0001193</th>\n",
       "      <th>C0001206</th>\n",
       "      <th>C0001311</th>\n",
       "      <th>C0001339</th>\n",
       "      <th>...</th>\n",
       "      <th>C4540135</th>\n",
       "      <th>C4540251</th>\n",
       "      <th>C4540293</th>\n",
       "      <th>C4540389</th>\n",
       "      <th>C4540439</th>\n",
       "      <th>C4540496</th>\n",
       "      <th>C4540534</th>\n",
       "      <th>C4540536</th>\n",
       "      <th>C4540602</th>\n",
       "      <th>C4543822</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</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>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",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C0027877</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>...</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>C0036391</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>...</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>C0549463</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>...</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>C0751337</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>...</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>C1852146</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>...</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>6 rows × 6953 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  rare_dis_id  C0000744  C0000846  C0000889  C0001080  C0001127  C0001193  \\\n",
       "0    C0011195         0         0         0         0         0         0   \n",
       "1    C0027877         0         0         0         0         0         0   \n",
       "2    C0036391         0         0         0         0         0         0   \n",
       "3    C0549463         0         0         0         0         0         0   \n",
       "4    C0751337         0         0         0         0         0         0   \n",
       "5    C1852146         0         0         0         0         0         0   \n",
       "\n",
       "   C0001206  C0001311  C0001339  ...  C4540135  C4540251  C4540293  C4540389  \\\n",
       "0         0         0         0  ...         0         0         0         0   \n",
       "1         0         0         0  ...         0         0         0         0   \n",
       "2         0         0         0  ...         0         0         0         0   \n",
       "3         0         0         0  ...         0         0         0         0   \n",
       "4         0         0         0  ...         0         0         0         0   \n",
       "5         0         0         0  ...         0         0         0         0   \n",
       "\n",
       "   C4540439  C4540496  C4540534  C4540536  C4540602  C4543822  \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",
       "\n",
       "[6 rows x 6953 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rare_diseases= ['C0011195', 'C0023944', 'C0024054', 'C0024901', 'C0027877', 'C0036391', 'C0265202', 'C0268059', 'C0549463', 'C0751337', 'C0869083', 'C1852146', 'C0796280']\n",
    "\n",
    "matrix_jac_gen = pd.read_excel((\"mat_jacc_variant.xlsx\"),engine='openpyxl')\n",
    "matrix_jac_gen"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "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",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [disease_id, drug_name, gene_id]\n",
       "Index: []"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_useful_diseases = get_normal_diseases(\"C1852146\")\n",
    "df_useful_diseases"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "358"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "useful_diseases = get_normal_diseases(\"C1852146\")['disease_id'].drop_duplicates().tolist()\n",
    "len(useful_diseases)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "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>C0000744</th>\n",
       "      <th>C0000846</th>\n",
       "      <th>C0000889</th>\n",
       "      <th>C0001080</th>\n",
       "      <th>C0001127</th>\n",
       "      <th>C0001193</th>\n",
       "      <th>C0001206</th>\n",
       "      <th>C0001311</th>\n",
       "      <th>C0001339</th>\n",
       "      <th>...</th>\n",
       "      <th>C4540135</th>\n",
       "      <th>C4540251</th>\n",
       "      <th>C4540293</th>\n",
       "      <th>C4540389</th>\n",
       "      <th>C4540439</th>\n",
       "      <th>C4540496</th>\n",
       "      <th>C4540534</th>\n",
       "      <th>C4540536</th>\n",
       "      <th>C4540602</th>\n",
       "      <th>C4543822</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>C0549463</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>...</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 × 6953 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  rare_dis_id  C0000744  C0000846  C0000889  C0001080  C0001127  C0001193  \\\n",
       "3    C0549463         0         0         0         0         0         0   \n",
       "\n",
       "   C0001206  C0001311  C0001339  ...  C4540135  C4540251  C4540293  C4540389  \\\n",
       "3         0         0         0  ...         0         0         0         0   \n",
       "\n",
       "   C4540439  C4540496  C4540534  C4540536  C4540602  C4543822  \n",
       "3         0         0         0         0         0         0  \n",
       "\n",
       "[1 rows x 6953 columns]"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = matrix_jac_gen[matrix_jac_gen.rare_dis_id == 'C1852146']\n",
    "df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "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>C0549463</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>C0549463</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0002871</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0149925</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0085786</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0085435</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0812413</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0751781</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0740457</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C0699885</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C4082937</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>358 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "rare_dis_id  C0549463\n",
       "C0549463          1.0\n",
       "C0002871          0.0\n",
       "C0149925          0.0\n",
       "C0085786          0.0\n",
       "C0085435          0.0\n",
       "...               ...\n",
       "C0812413          0.0\n",
       "C0751781          0.0\n",
       "C0740457          0.0\n",
       "C0699885          0.0\n",
       "C4082937          0.0\n",
       "\n",
       "[358 rows x 1 columns]"
      ]
     },
     "execution_count": 68,
     "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='C1852146', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_normal_dis_jacc=pd.DataFrame(df_useful['C1852146'].nlargest(5)).reset_index().rename(columns={'index':'disease_id', 'C1852146':'jacc_idx'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_normal_dis_jacc.to_excel(\"dis_jacc_C1852146_v.xlsx\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "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": 73,
   "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>C0002871</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C0002871</td>\n",
       "      <td>PAZOPANIB HYDROCHLORIDE</td>\n",
       "      <td>3702</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C0002871</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>3702</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>C0002871</td>\n",
       "      <td>PAZOPANIB</td>\n",
       "      <td>3702</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C0002871</td>\n",
       "      <td>DEQUALINIUM</td>\n",
       "      <td>331</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>C0002871</td>\n",
       "      <td>EMAPALUMAB</td>\n",
       "      <td>3458</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>C0002871</td>\n",
       "      <td>OLSALAZINE</td>\n",
       "      <td>3458</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>C0002871</td>\n",
       "      <td>GLUCOSAMINE</td>\n",
       "      <td>3458</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>C0003868</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>C0004135</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>PAZOPANIB HYDROCHLORIDE</td>\n",
       "      <td>3702</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>3702</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>PAZOPANIB</td>\n",
       "      <td>3702</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>ZINC ACETATE</td>\n",
       "      <td>325</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>DEQUALINIUM</td>\n",
       "      <td>331</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>EMAPALUMAB</td>\n",
       "      <td>3458</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>OLSALAZINE</td>\n",
       "      <td>3458</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>GLUCOSAMINE</td>\n",
       "      <td>3458</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>IXEKIZUMAB</td>\n",
       "      <td>3605</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>SECUKINUMAB</td>\n",
       "      <td>3605</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>PAZOPANIB HYDROCHLORIDE</td>\n",
       "      <td>3702</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>3702</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>PAZOPANIB</td>\n",
       "      <td>3702</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>ZINC ACETATE</td>\n",
       "      <td>325</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>DEQUALINIUM</td>\n",
       "      <td>331</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>EMAPALUMAB</td>\n",
       "      <td>3458</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>OLSALAZINE</td>\n",
       "      <td>3458</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>GLUCOSAMINE</td>\n",
       "      <td>3458</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>IXEKIZUMAB</td>\n",
       "      <td>3605</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>SECUKINUMAB</td>\n",
       "      <td>3605</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   disease_id                drug_name  gene_id  jacc_idx\n",
       "0    C0002871           MERCAPTOPURINE     3251       0.0\n",
       "1    C0002871  PAZOPANIB HYDROCHLORIDE     3702       0.0\n",
       "2    C0002871             FOSTAMATINIB     3702       0.0\n",
       "3    C0002871                PAZOPANIB     3702       0.0\n",
       "4    C0002871              DEQUALINIUM      331       0.0\n",
       "5    C0002871               EMAPALUMAB     3458       0.0\n",
       "6    C0002871               OLSALAZINE     3458       0.0\n",
       "7    C0002871              GLUCOSAMINE     3458       0.0\n",
       "8    C0003868           MERCAPTOPURINE     3251       0.0\n",
       "9    C0004135           MERCAPTOPURINE     3251       0.0\n",
       "10   C0006142           MERCAPTOPURINE     3251       0.0\n",
       "11   C0006142  PAZOPANIB HYDROCHLORIDE     3702       0.0\n",
       "12   C0006142             FOSTAMATINIB     3702       0.0\n",
       "13   C0006142                PAZOPANIB     3702       0.0\n",
       "14   C0006142             ZINC ACETATE      325       0.0\n",
       "15   C0006142              DEQUALINIUM      331       0.0\n",
       "16   C0006142               EMAPALUMAB     3458       0.0\n",
       "17   C0006142               OLSALAZINE     3458       0.0\n",
       "18   C0006142              GLUCOSAMINE     3458       0.0\n",
       "19   C0006142               IXEKIZUMAB     3605       0.0\n",
       "20   C0006142              SECUKINUMAB     3605       0.0\n",
       "21   C0549463           MERCAPTOPURINE     3251       1.0\n",
       "22   C0549463  PAZOPANIB HYDROCHLORIDE     3702       1.0\n",
       "23   C0549463             FOSTAMATINIB     3702       1.0\n",
       "24   C0549463                PAZOPANIB     3702       1.0\n",
       "25   C0549463             ZINC ACETATE      325       1.0\n",
       "26   C0549463              DEQUALINIUM      331       1.0\n",
       "27   C0549463               EMAPALUMAB     3458       1.0\n",
       "28   C0549463               OLSALAZINE     3458       1.0\n",
       "29   C0549463              GLUCOSAMINE     3458       1.0\n",
       "30   C0549463               IXEKIZUMAB     3605       1.0\n",
       "31   C0549463              SECUKINUMAB     3605       1.0"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_normal_rare_drug"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_normal_rare_drug=df_normal_rare_drug.drop_duplicates()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "prueba=df_normal_rare_drug.groupby('drug_name').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "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>DEQUALINIUM</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EMAPALUMAB</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FOSTAMATINIB</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GLUCOSAMINE</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IXEKIZUMAB</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MERCAPTOPURINE</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OLSALAZINE</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAZOPANIB</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAZOPANIB HYDROCHLORIDE</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECUKINUMAB</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZINC ACETATE</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         disease_id  gene_id  jacc_idx\n",
       "drug_name                                             \n",
       "DEQUALINIUM                       3        3         3\n",
       "EMAPALUMAB                        3        3         3\n",
       "FOSTAMATINIB                      3        3         3\n",
       "GLUCOSAMINE                       3        3         3\n",
       "IXEKIZUMAB                        2        2         2\n",
       "MERCAPTOPURINE                    5        5         5\n",
       "OLSALAZINE                        3        3         3\n",
       "PAZOPANIB                         3        3         3\n",
       "PAZOPANIB HYDROCHLORIDE           3        3         3\n",
       "SECUKINUMAB                       2        2         2\n",
       "ZINC ACETATE                      2        2         2"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prueba"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "drug=prueba[prueba[\"disease_id\"]>4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "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>MERCAPTOPURINE</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",
       "MERCAPTOPURINE           5        5         5"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drug"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "drug.to_excel(\"drugs_C0549463_V.xlsx\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "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>C0002871</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>C0003868</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>C0004135</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>C0006142</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   disease_id       drug_name  gene_id  jacc_idx\n",
       "0    C0002871  MERCAPTOPURINE     3251       0.0\n",
       "8    C0003868  MERCAPTOPURINE     3251       0.0\n",
       "9    C0004135  MERCAPTOPURINE     3251       0.0\n",
       "10   C0006142  MERCAPTOPURINE     3251       0.0\n",
       "21   C0549463  MERCAPTOPURINE     3251       1.0"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_normal_rare_drug[df_normal_rare_drug[\"drug_name\"]==\"MERCAPTOPURINE\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "drug_all=df_normal_rare_drug[df_normal_rare_drug[\"disease_id\"]==\"C0549463\"]"
   ]
  },
  {
   "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",
       "      <th>jacc_idx</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>MERCAPTOPURINE</td>\n",
       "      <td>3251</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>PAZOPANIB HYDROCHLORIDE</td>\n",
       "      <td>3702</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>FOSTAMATINIB</td>\n",
       "      <td>3702</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>PAZOPANIB</td>\n",
       "      <td>3702</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>ZINC ACETATE</td>\n",
       "      <td>325</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>DEQUALINIUM</td>\n",
       "      <td>331</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>EMAPALUMAB</td>\n",
       "      <td>3458</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>OLSALAZINE</td>\n",
       "      <td>3458</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>GLUCOSAMINE</td>\n",
       "      <td>3458</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>IXEKIZUMAB</td>\n",
       "      <td>3605</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>C0549463</td>\n",
       "      <td>SECUKINUMAB</td>\n",
       "      <td>3605</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   disease_id                drug_name  gene_id  jacc_idx\n",
       "21   C0549463           MERCAPTOPURINE     3251       1.0\n",
       "22   C0549463  PAZOPANIB HYDROCHLORIDE     3702       1.0\n",
       "23   C0549463             FOSTAMATINIB     3702       1.0\n",
       "24   C0549463                PAZOPANIB     3702       1.0\n",
       "25   C0549463             ZINC ACETATE      325       1.0\n",
       "26   C0549463              DEQUALINIUM      331       1.0\n",
       "27   C0549463               EMAPALUMAB     3458       1.0\n",
       "28   C0549463               OLSALAZINE     3458       1.0\n",
       "29   C0549463              GLUCOSAMINE     3458       1.0\n",
       "30   C0549463               IXEKIZUMAB     3605       1.0\n",
       "31   C0549463              SECUKINUMAB     3605       1.0"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drug_all"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "drug_all.to_excel(\"drugs_all_C0549463_v.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
}