survival_analysis_jupyter.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Author**: Vinay Barnabas\n",
    "\n",
    "*Tech Stack*: VSCode + Jupyter Notebooks <br>\n",
    "*Presentation*: Jupyter Notebooks <br>\n",
    "*Version Control*: Git <br>\n",
    "*Github repository*: https://github.com/VinylBr/InterviewTechTestSol <br>\n",
    "*App link*: https://vinylbr-interviewtechtestsol-app-bc-ds9ega.streamlit.app/\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Table of Contents**:\n",
    "\n",
    "1. Data Exploration and Sanity Checks\n",
    "2. Approach\n",
    "3. Data Maturity (Age at diagnosis and Follow-up time)\n",
    "4. Survival Analysis (Stratify by age at diagnosis)\n",
    "5. HER2 Status\n",
    "6. App for Survival Analysis\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {},
   "outputs": [],
   "source": [
    "# importing necessary libraries\n",
    "import sqlite3\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns #plotting\n",
    "import matplotlib.pyplot as plt #plotting\n",
    "from lifelines import KaplanMeierFitter #survival analysis\n",
    "from lifelines.statistics import logrank_test #statistical test in survival analysis\n",
    "from lifelines.utils import median_survival_times #summary statistic for median survival times\n",
    "from datetime import datetime\n",
    "\n",
    "# muting FutureWarnings. These are warnings about features that will be removed in a future version\n",
    "import warnings\n",
    "warnings.simplefilter(action = \"ignore\", category = FutureWarning)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Plotting conventions for the notebook*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define plot style\n",
    "def set_plot_style():\n",
    "    '''\n",
    "    function to set figure parameters\n",
    "    Inputs: None\n",
    "    Return: None\n",
    "    '''\n",
    "\n",
    "    # define style: plots will not have gridlines. X- and y- ticks would be present\n",
    "    sns.set_style('ticks')\n",
    "\n",
    "    # parameter to control scaling of plot elements.\n",
    "    sns.set_context('notebook')\n",
    "\n",
    "    # fix figure parameters\n",
    "    plt.rcParams['figure.figsize'] = (10,6)\n",
    "    plt.rcParams['axes.labelsize'] = 12\n",
    "    plt.rcParams['axes.titlesize'] = 12\n",
    "    plt.rcParams['xtick.labelsize'] = 10\n",
    "    plt.rcParams['ytick.labelsize'] = 10\n",
    "    plt.rcParams['legend.fontsize'] = 10\n",
    "    plt.rcParams['lines.linewidth'] = 2\n",
    "    \n",
    "    # fix color palette\n",
    "    sns.set_palette('muted')\n",
    "\n",
    "def frame_off(ax):\n",
    "    '''\n",
    "    function to remove outer frame in plots\n",
    "    Inputs: ax = plot axes\n",
    "    Return: Nothing\n",
    "    '''\n",
    "    for _, spines in ax.spines.items():\n",
    "        spines.set_visible(False)\n",
    "\n",
    "\n",
    "# fix figure paraeters by calling the function\n",
    "set_plot_style()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Exploration and Sanity Checks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Establish connection to sqlite3 database and define a cursor\n",
    "con = sqlite3.connect(\"techTest.db\")\n",
    "cur = con.cursor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Table patient with 1000 rows\n",
      "             name  data_type  notnull default_value  primary_key\n",
      "index                                                           \n",
      "0      patient_id       TEXT        0          None            0\n",
      "1             dob  TIMESTAMP        0          None            0\n",
      "2             dod  TIMESTAMP        0          None            0\n",
      "3          gender       TEXT        0          None            0\n",
      "--------------------------------------------------\n",
      "\n",
      "\n",
      "Table diagnosis with 3026 rows\n",
      "                      name  data_type  notnull default_value  primary_key\n",
      "index                                                                    \n",
      "0               patient_id       TEXT        0          None            0\n",
      "1           diagnosis_date  TIMESTAMP        0          None            0\n",
      "2           diagnosis_code       TEXT        0          None            0\n",
      "3      diagnosis_code_type       TEXT        0          None            0\n",
      "--------------------------------------------------\n",
      "\n",
      "\n",
      "Table biomarker with 3049 rows\n",
      "                      name  data_type  notnull default_value  primary_key\n",
      "index                                                                    \n",
      "0               patient_id       TEXT        0          None            0\n",
      "1                test_date  TIMESTAMP        0          None            0\n",
      "2           biomarker_name       TEXT        0          None            0\n",
      "3      biomarker_test_type       TEXT        0          None            0\n",
      "4              test_result       TEXT        0          None            0\n",
      "--------------------------------------------------\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# query to extract table info from sqlite_master\n",
    "sql_master_query = '''--sql\n",
    "                SELECT name\n",
    "                FROM sqlite_master\n",
    "                WHERE type = \"table\";\n",
    "            '''\n",
    "# Execute the query and fetchall the results\n",
    "cur.execute(sql_master_query)\n",
    "list_table_info = cur.fetchall()\n",
    "\n",
    "cols_pragma = [\"index\", \"name\", \"data_type\", \"notnull\", \"default_value\", \"primary_key\"] #from PRAGMA info\n",
    "\n",
    "# for each table display PRAGMA table information\n",
    "for item in list_table_info:\n",
    "    # get number of rows for each table\n",
    "    sql_row_query = f\"\"\"--sql\n",
    "                      SELECT COUNT(*) \n",
    "                      FROM {item[0]};\n",
    "                      \"\"\"\n",
    "    cur.execute(sql_row_query)\n",
    "    num_rows = cur.fetchone()[0] # [0] is to obtain the first element of the output tuple\n",
    "    print(f\"Table {item[0]} with {num_rows} rows\")\n",
    "    \n",
    "    # get number of columns, their names, type, whether NULL is allowed, whether primary key\n",
    "    sql_column_query = f\"PRAGMA table_info({item[0]})\"\n",
    "    cur.execute(sql_column_query)\n",
    "    column_info = cur.fetchall()\n",
    "\n",
    "    # create a dataframe for each column info\n",
    "    df_temp = pd.DataFrame(column_info, columns = cols_pragma).set_index(\"index\")\n",
    "    print(df_temp)\n",
    "\n",
    "    print(50*\"-\")\n",
    "    print(\"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**OBSERVATION**: <br>\n",
    "- Tables diagnosis and biomarkers have more than 1000 records since each patient_id has multiple diagnosis dates and biomarker test dates\n",
    "- patient_id can be used as a key to join the three tables"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Sanity Checks\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Is date of birth later than date of death?*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 219,
   "metadata": {},
   "outputs": [],
   "source": [
    "# query to count number of patients where date of birth is later than date of death\n",
    "sql_query = '''--sql\n",
    "            \n",
    "            SELECT COUNT(*)\n",
    "            FROM Patient\n",
    "            WHERE dob > dod\n",
    "            \n",
    "            ;'''\n",
    "\n",
    "cur.execute(sql_query)\n",
    "\n",
    "# assert statement returns an error if False\n",
    "assert cur.fetchall()[0][0] == 0, \"Date of Birth is later than Date of death\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Is date of death later than 2024?*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of records in Patient table where date of death is later than current date: 174\n"
     ]
    }
   ],
   "source": [
    "#query to identify patients whose year of death is later than 2024\n",
    "sql_death_query = '''--sql\n",
    "            \n",
    "                    SELECT COUNT(*)\n",
    "                    FROM patient\n",
    "                    ---converting dod to year format and comparing to 2024--- \n",
    "                    WHERE STRFTIME(\"%Y\",dod) > STRFTIME(\"%Y\",\"now\");\n",
    "                \n",
    "                '''\n",
    "cur.execute(sql_death_query)\n",
    "\n",
    "num_dodbiggernow = cur.fetchone()[0]\n",
    "\n",
    "print(f\"Number of records in Patient table where date of death is later than current date: {num_dodbiggernow}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Is date of death earlier than earliest date of diagnosis?*\n",
    "\n",
    "I plot distributions of \n",
    "1. DOD - earliest date of diagnosis\n",
    "2. DOD - latest date of diagnosis\n",
    "\n",
    "Here, <br>\n",
    "earliest date of diagnosis = MIN(diagnosis_date) for a patient <br>\n",
    "latest date of diagnosis = MAX(diagnosis_date) for a patient <br>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 221,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7kAAAIpCAYAAABntv83AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAACdZ0lEQVR4nOzdeZzO5f7H8fcwgxlLlHAqJ2KGQpnMNIRUVCh0SjoVqdNpG1k6IZ0T2tCi5Uec0I5TSRGZopBkG4REk7GMpRHZzZh9vr8/rmbuGWOZ5Z77upfX8/GYx1z3fX/nvt8zlms+97UFOY7jCAAAAAAAP1DBdgAAAAAAANyFIhcAAAAA4DcocgEAAAAAfoMiFwAAAADgNyhyAQAAAAB+gyIXAAAAAOA3KHIBAAAAAH6DIhcAAAAA4DcocgEAAAAAfsN6kXvw4EHFxsYqKipKMTExGjVqlLKzs0957ZIlS9StWze1bNlSXbp00eLFi/Mfy8jI0KhRo3TNNdeoVatWuuOOO7Ry5cr8x0+cOKGnnnpKMTExatWqlYYOHarU1NT8x3fs2KG+ffsqMjJS7dq101tvvVV+3zQAAF6G/hgA4C+sF7mDBg1SWFiYli5dqpkzZ2rFihV6//33i1yXlJSk/v37a+DAgVqzZo369++vQYMGad++fZKksWPH6scff9Qnn3yi+Ph43XHHHXrkkUeUnJwsSXr++ee1d+9ezZ8/XwsWLNDevXs1duxYSVJWVpYeeeQRtWjRQqtWrdLkyZM1ffp0ffXVVx77OQAAYBP9MQDAX1gtcnfu3Kn4+HgNGTJEoaGhql+/vmJjYzV9+vQi186aNUtRUVHq1KmTgoOD1bVrV0VHR+uTTz6RZN45HjBggP7yl7+oYsWK6tWrlypVqqRNmzYpLS1Nc+fO1YABA1SzZk2dd955Gjx4sD7//HOlpaVp9erV2r9/vwYMGKBKlSrpsssuU58+fU6ZAwAAf0N/DADwJ8E2XzwxMVE1a9ZU3bp18+9r1KiRkpOTdezYMdWoUSP//q1btyoiIqLQ1zdu3FgJCQmSpOeee67QYytWrNDx48fVtGlT7dy5U1lZWYW+vlGjRkpPT1dSUpISExPVsGFDVapUqdBzT548+bTZO3bseMbvbeHChWd8HAAAb0F/DADwJ1ZHclNTUxUaGlrovrzbJ06cOOu1VapUKXKdJK1fv16DBg3SY489pvr16yslJUWSFBYWVuR1UlNTT5vjVM8NAIC/oT8GAPgTqyO5YWFhSktLK3Rf3u2qVasWuj80NFTp6emF7ktPTy9y3aeffqrRo0drwIABuv/++/NfJ++5867Pe51q1aqdNsfJz10Q7wwDAPwF/TEAwJ9YLXLDw8N15MgRHThwQLVr15Ykbdu2TfXq1VP16tULXRsREaFNmzYVum/r1q1q3ry5JCknJ0fPPvusFixYoAkTJujqq6/Ov65hw4YKCQnR1q1bdcUVV+S/TkhIiBo0aKCDBw8qKSlJ2dnZCg4Ozn/u8PDwcvveAQDwFvTHAAB/YnW6coMGDdSqVSuNHj1aKSkp2r17tyZOnKiePXsWubZ79+6Kj49XXFycsrOzFRcXp/j4ePXo0UOSNGbMGH3//ff67LPPCnWoknnXuUuXLho7dqwOHTqkQ4cOaezYsbrllltUpUoVxcTEqFatWnr11VeVkZGhhIQETZ069ZQ5AADwN/THAAB/EuQ4jmMzwIEDB/Tcc89p1apVqlChgm699VYNHjxYFStWVGRkpJ599ll1795dkrR06VKNHTtWu3bt0oUXXqghQ4aoQ4cOOnTokNq2bauKFSsqJCSk0PPnfX1KSopeeuklLVq0SFlZWerYsaOGDx+eP3Vq586deu6557RhwwaFhYWpd+/eeuihhzz+8wAAwAb6YwCAv7Be5AIAAAAA4C5WpysDAAAAAOBOFLkAAAAAAL9BkQsAAAAA8BsUuQAAAAAAv0GRCwAAAADwGxS5AAAAAAC/QZELAAAAAPAbFLkAAAAAAL9BkQsAAAAA8BsUuQAAAAAAv0GRCwAAAADwGxS5AAAAAAC/QZELAAAAAPAbFLkAAAAAAL9BkQsAAAAA8BsUuQAAAAAAv0GRCwAAAADwGxS5AAAAAAC/QZELAAAAAPAbFLkAAAAAAL9BkQsAAAAA8BsUuQAAAAAAv0GRCwAAAADwGxS5AAAAAAC/QZELAAAAAPAbFLkAAAAAAL9BkQsAAAAA8BsUuQAAAAAAv0GRCwAAAADwGxS5AAAAAAC/QZELAAAAAPAbFLkAAAAAAL9BkQsAAAAA8BsUuQAAAAAAv0GRCwAAAADwGxS5AAAAAAC/QZELAAAAAPAbFLkAAAAAAL9BkQsAAAAA8BsUuQAAAAAAv0GRCwAAAADwGxS5AAAAAAC/QZELAAAAAPAbFLkAAAAAAL9BkQsAAAAA8BsUuQAAAAAAv0GRCwAAAADwGxS5AAAAAAC/QZELAAAAAPAbFLkAAAAAAL9hvcg9ePCgYmNjFRUVpZiYGI0aNUrZ2dmnvHbJkiXq1q2bWrZsqS5dumjx4sWnvO6FF17QsGHD8m+vWbNGkZGRhT6aN2+uJk2aaN++fZKkyZMnq1mzZoWuef31193/DQMA4IXojwEA/iLYdoBBgwapbt26Wrp0qQ4cOKBHH31U77//vv75z38Wui4pKUn9+/fXa6+9pmuvvVYLFizQoEGDtGDBAtWtW1eSdPjwYY0aNUpz587V3/72t/yvjYqK0rp16/Jvp6Sk6M4779TNN9+c/7U///yzHn30UT322GMe+K4BAPAu9McAAH9hdSR3586dio+P15AhQxQaGqr69esrNjZW06dPL3LtrFmzFBUVpU6dOik4OFhdu3ZVdHS0PvnkE0lSamqqOnfurBo1auimm2464+u+8MILqlu3rmJjY/Pv27hxo5o3b+7ebxAAAB9AfwwA8CdWi9zExETVrFkz/91bSWrUqJGSk5N17NixQtdu3bpVERERhe5r3LixEhISJEmVK1fWvHnzNGLECIWFhZ32NdesWaO4uDg9//zz+fcdPHhQycnJmjFjhtq1a6frr79eL7/8sjIyMtzxbQIA4NXojwEA/sTqdOXU1FSFhoYWui/v9okTJ1SjRo0zXlulShWdOHFCkhQcHKzatWuf9TXHjx+vu+66SxdeeGH+fX/88YeioqJ022236Y033tDu3bs1aNAgpaWlaeTIkad8no4dO57xdRYuXHjWLAAAeAP6YwCAP7E6khsWFqa0tLRC9+Xdrlq1aqH7Q0NDlZ6eXui+9PT0Itedya5duxQfH68+ffoUur9p06aaPn26OnXqpEqVKqlRo0aKjY1VXFxcSb4dAAB8Ev0xAMCfWB3JDQ8P15EjR3TgwIH8d323bdumevXqqXr16oWujYiI0KZNmwrdt3Xr1hKt25k/f76uvPJKXXTRRYXuj4+P17p16/Twww/n35eZmakqVaqc9rl4ZxgA4C/ojwEA/sTqSG6DBg3UqlUrjR49WikpKdq9e7cmTpyonj17Frm2e/fuio+PV1xcnLKzsxUXF6f4+Hj16NGj2K+3du1aRUVFFbk/NDRU48eP19y5c5Wbm6vExERNnDhRd955Z5m+PwAAfAH9MQDAn1g/J3fcuHHKzs5Wx44d1atXL7Vv3z5/l8XIyEjNmTNHktkAY8KECZo0aZKio6M1ceJEjR8/Xg0bNiz2a+3Zs6fQphp5WrRooddee01vv/22WrVqpQceeEDdunXTI4884p5vEgAAL0d/DADwF0GO4zi2QwAAAAAA4A5W1+QCQFls3SolJkpZWdIVV0gXX2w7EQAAAGyzPl0ZAErqhx+kjh2l8HCpa1epRw+pQQPpqqukr76ynQ4AAAA2MV0ZgE955x3p4YelnJzTX3P33dJbb0knbQoLAACAAECRC8BnjB8vDRjgut24sdSrl2l/+aX000+ux668Upo3T6pXz7MZAQAAYBdFLgCfsHKl1L69lJ1tbg8aJI0dK1WsaG47jvThh9LAgdLRo+a+xo2lpUspdAEAAAIJRS4Ar3fkiNSypbRzp7k9ZIj08sunvnbTJqlLF2n3bnO7RQtpyRKpVi1PJAUAAIBtbDwFwOuNHu0qcNu2NbdPp1kzadky6a9/Nbc3bpRuv93swAwAgDdwHGndOmnaNGnNGtcsJQDuwUguAK+2c6fUpImUkSFVriwlJJidlM8mMVFq107av9/cHjBA+r//K9eoAACc1aefmhlJeW/eSma20dix0j/+YS8X4E8YyQXg1YYPNwWuZNbhFqfAlczxQrNmSSEh5va4cdJ775VHQgAAzi43Vxo50myYWLDAlaTDh6UHHpAeeoiZR4A7MJILwGtt3242j3Ic6dxzpW3bpJo1S/YcU6aYXxokqVIlsz63dWu3RwUA4IyGD5deeOHs1z32mDlNAEDpMZILwGu9+aYpcCXpiSdKXuBK0oMPSv36mXZmpnTbbdLevW6LCADAWX39dfEKXMn0fZ9+Wr55AH/HSC4Ar5SSIl14oXTsmFSlitktuXbt0j1XVpZ0ww1mFFeSrr5aWrzYjOwCAFCe9u41O/0fPFj8r6lSRfr1V9cmigBKhpFcAF7pww9NgStJ99xT+gJXMutyZ8yQ6tc3t5cvlx5/vOwZAQA4m2HDSlbgSlJ6Ov0UUBaM5ALwSldeaY5XkKT166Urrij7c65ZY3ZcztvI6p132MkSAFB+1qyRoqNL//WrV0tRUe7LAwQKRnIBeJ1ffnEVuFFR7ilw857rrbdctx99VIqPd89zAwBQkONI//pX2Z7j4Ydde1MAKD6KXABeZ/p0V/uee9z73PfdV3Qjqn373PsaAAAsWCAtXVq25/jxR2nZMvfkAQIJ05UBeBXHkRo1knbskCpUkH77TapXz72vkZkpdewo/fCDud2+vfTNN1Llyu59HQBA4Lr+erPJYVl16CB9913ZnwcIJIzkAvAqK1eaAlcyhai7C1zJ7Kr86afSBReY20uXmhHe3Fz3vxYAIPCsWuWeAlcyJwMkJrrnuYBAQZELwKvMmuVq3313+b1OvXrS7NlSaKi5/fHHZgdMAADK6qWX3Pt8zz3n3ucD/B3TlQF4laZNzdmAFSpI+/dL551Xvq83d650662uUdxx46T+/cv3NQEA/mvnTqlhQ/duGBUSYo4hql7dfc8J+DNGcgF4jS1bTIErSW3bln+BK0ndukkTJ7puDxwoffZZ+b8uAMA/TZni/h2Rs7IKb8oI4MwocgF4jblzXe3u3T33ug8/LD31lGk7jpkmzSYfAICSysyU3n67fJ779dfL53kBf0SRC8BrzJnjanuyyJWkUaOkvn1NOzNT6tFD2rDBsxkAAL7tiy/K71i6LVvMOfIAzo4iF4BXOHLEdaRPRIT58KSgIDPFrGtXc/vYMalzZ9dOzwAAnM3kyeX7/IzmAsVDkQvAK3z3nWvzpy5d7GQICZFmzJBatza3f/9duvFGswEWAABnsnu3tHBh+b7Gxx9LOTnl+xqAP6DIBeAVvv3W1e7UyV6OqlWlL780uzxL0tatZnQ3JcVeJgCA95s2zf0bTp3s+HHp++/L9zUAf0CRC8ArfPON+RwcLHXoYDfLeedJ8+dLF15obq9dKz3wQPn/8gIA8E2OI334oWdea8IEz7wO4MsocgFYt2uX2VBDMlOFveEcwL/+1RS6eVlmzJBeftluJgCAd1qzRkpI8MxrzZtnNkgEcHoUuQCsK7iGyeZU5ZM1a2amn+X5z3+klSvt5QEAeCdPjeJKUnq6eRMWwOlR5AKwzlvW455K9+7S8OGmnZMj9e7N+lwAgEt2tpnt40mTJnn29QBfE+Q4rDIDYI/jmKnBe/ZIYWHmKKGQENupCsvOltq3d43iPvqoNHGi3UwAAO+wcKHn36CtXl06dMjsYwGgKEZyAVi1c6cpcCWzHtfbClzJ/BIxdarZeVmS3npLWr3abiYAgHf4+GPPvya7LANnRpELwKqlS13ta66xl+NsGjeWnnnGtB1Hio3lrEIACHSZmdJnn9l57c8/t/O6gC+gyAVgVcEit317ezmKY+BAsxmVZHbSnDrVbh4AgF0LFkiHD9t57dmzpdxcO68NeDuKXABW5RW5wcFmurI3CwmR3nzTdfvZZznGAQAC2cyZ9l77t99YOgOcDkUuAGv273edK9iqldl4yttde610002mnZQkvf22zTQAAFuysqQ5c+xmmDXL7usD3ooiF4A1y5a52t4+VbmgF15wtUeNkjIy7GUBANjx3Xf2pipX+PM3+M8/N/tEACiMIheANXlH8khS27b2cpRUVJR0662mnZwsffKJ1TgAAAtsbTglSZdeaj4nJkqbN9vLAXgrilwA1hQscmNi7OUojSFDXO3XXuOddAAIJDk5ZuMnWwruYcGUZaAoilwAVmRnmx2KJenii6W//MVunpJq08ZVmG/YIC1ebDcPAMBzli+X9u2z9/oFi1yOEgKKosgFYMXPP0snTpi2t++qfCpBQdK//uW6PW6cvSwAAM+yXVief77ZsFGS1q0zGyECcKHIBWBFwanKvljkStJtt0kXXGDaX34p/f673TwAgPLnOPaLXMm1N4Rkf5dnwNtQ5AKwYtUqV9tXi9zgYKlvX9POyZGmTrWbBwBQ/taskXbtsp2icJH7xRfWYgBeiSIXgBV5I7khIVLLllajlMk//uFqv/MOG1ABgL/zhlFcSWrWTLrkEtNessTecUaAN6LIBeBxR49KCQmm3bKlVKWK1Thl0rix1KGDaf/6q7Rihd08AIDy4zh2jw4qKChI6tHDtHNypHnz7OYBvAlFLgCPW7/e1Y6OthbDbQqO5n78sb0cAIDy9csv5mxab5FX5EpMWQYKosgF4HE//uhqX3mlvRzucuutUuXKpv3pp+YddQCA//G2DZ7atpXOPde0v/5aysiwmwfwFhS5ADxu7VpX2x+K3Bo1pM6dTfv336UffrCbBwBQPrxttDQ4WLrlFtNOSZEWLbKbB/AWFLkAPC5vJDckxGyc4Q/uvNPV/uQTezkAAOXj998LnwzgLZiyDBRFkQvAo1JTXZtOtWghVapkN4+7dOvm2kBr5kymLAOAv5k3zzt30L/xRteSmTlzpNxcu3kAb0CRC8CjNmxw/ZLQqpXdLO5UrZrUpYtp//GHd77bDwAoPW8dJa1WTerUybT37jXn+AKBznqRe/DgQcXGxioqKkoxMTEaNWqUsrOzT3ntkiVL1K1bN7Vs2VJdunTR4sWLT3ndCy+8oGHDhhW6b8OGDWratKkiIyPzP+655578x3fs2KG+ffsqMjJS7dq101tvveW+bxJAPn/bdKqg7t1d7blz7eUASoP+GDi9Eyekb76xneL0br3V1fbWYhzwJOtF7qBBgxQWFqalS5dq5syZWrFihd5///0i1yUlJal///4aOHCg1qxZo/79+2vQoEHat29f/jWHDx/W4MGDNXXq1CJfv3HjRkVHR2vdunX5H9OnT5ckZWVl6ZFHHlGLFi20atUqTZ48WdOnT9dXX31Vbt83EKj8ucjt2tWcWyhR5ML30B8Dp/ftt1J6uu0Up9etm6v/mT3bahTAK1gtcnfu3Kn4+HgNGTJEoaGhql+/vmJjY/M7u4JmzZqlqKgoderUScHBweratauio6P1yZ87vKSmpqpz586qUaOGbrrppiJfv3HjRjVv3vyUOVavXq39+/drwIABqlSpki677DL16dPnlDkAlE3ezsoVK5o1uf6kTh2pdWvT3rRJ2r7dbh6guOiPgTPztqODTla3rqv/2bxZ2rrVbh7AtmCbL56YmKiaNWuqbt26+fc1atRIycnJOnbsmGrUqJF//9atWxUREVHo6xs3bqyEP3ewqVy5subNm6fatWsXmRolmU61du3auvHGG5WSkqKrrrpKw4YNU7169ZSYmKiGDRuqUoEdcBo3bqzJkyefNnvHjh3P+L0tXLjwzN88EIDS003xJ0mXXSaFhtrNUx66d5dWrDDtuXOlgQPt5gGKg/4YOL2cHN+YndOjh6v/+eIL6Ykn7OYBbLI6kpuamqrQk37Lzbt94sSJs15bpUqV/OuCg4NVu3btU75OTk6O6tSpo3bt2umzzz7Tl19+qaCgID300EPKyck5bY6TMwAom40bXbsO+9tU5TzdurnacXH2cgAlQX8MnF58vLR/v+0UZ8dRQoCL1ZHcsLAwpaWlFbov73bVqlUL3R8aGqr0kxZDpKenF7nuVCpWrFhkXdHw4cPVpk0bbdu27bQ5zvTcvDMMlFzB9bj+tLNyQZddJl10kbRnj/T992b0Ou9oIcBb0R8Dp+ftU5XzNG0qRURIW7ZIy5ZJBw5Ip3m/CfB7Vkdyw8PDdeTIER04cCD/vm3btqlevXqqXr16oWsjIiKUmJhY6L6tW7cqPDz8rK+zd+9ejRkzRqmpqfn3ZWZmSjLvPoeHhyspKanQLpLFfW4AxefPm07lCQoyZxZKpsD94Qe7eYDioD8GTu/LL20nKL680dzcXN/KDbib1SK3QYMGatWqlUaPHq2UlBTt3r1bEydOVM+ePYtc2717d8XHxysuLk7Z2dmKi4tTfHy8ehScm3EatWrV0rx58/T6668rIyNDhw4d0rPPPqs2bdror3/9q2JiYlSrVi29+uqrysjIUEJCgqZOnXrKHABKL6/IDQqSrrjCbpbylFfkStKCBfZyAMVFfwyc2q5d0s8/205RfAWPEvrsM2sxAOusHyE0btw4ZWdnq2PHjurVq5fat2+v2NhYSVJkZKTm/DlHpFGjRpowYYImTZqk6OhoTZw4UePHj1fDhg3P+hpVqlTR22+/rW3btqldu3a66aabVK1aNb3xxhuSzPqhd999V1u2bFHbtm310EMPqU+fPrrtttvK7fsGAk1mpvTTT6bdpIk5vN5fdezoOsqBIhe+gv4YKMrX9lZo3Vq64ALTXrBAOnrUbh7AliDHcRzbIQD4v/XrpchI0777bsnfTwSJjpbWrDHtvXulevXs5gEAlFz37t65s/Ls2YU3mipowABp/HjTnjpV6t3bY7EAr2F9JBdAYFi3ztXOK3b9WcEpy+yLAwC+Jz3dN///Lji7f+ZMezkAmyhyAXhEwTVN/rweN0/Bozu/+85aDABAKS1ZIvni6VVt27pmD339tXTsmN08gA0UuQA8YuNGV7tFC3s5PKV1a6lSJdOmyAUA3zNvnu0EpVOxopS3jD0jw3e/D6AsKHIBeERekVu7tlS3rt0snhAWJsXEmPbWrebcXACAb3Ac3y4O77jD1f70U3s5AFsocgGUuwMHpN9/N+3mzV07D/u7a691tZcssRYDAFBCW7ZI27fbTlF67dtLdeqY9ldfSSkpdvMAnkaRC6DcBdpU5TwFi1ymLAOA7/C1o4NOVnDKcnq6b49KA6VBkQug3AVqkdumjWtd7uLFdrMAAIrPH4pCpiwjkFHkAih3BXdWDqQiNzTUtS532zbXlG0AgPc6flz6/nvbKcrummvMPhiSGZlOTbWbB/AkilwA5a7gSG6zZvZy2NC2rau9fLm9HACA4lm0SMrKsp2i7IKDXVOW09Kk2bOtxgE8iiIXQLnKzXWN5DZoIFWvbjWOx119tau9bJm9HACA4vnmG9sJ3Kd3b1f7ww/t5QA8jSIXQLnaudO1q2MgTVXO06aNq81ILgB4P38qctu2NW8wS9K330rJyVbjAB5DkQugXAXqplN5ateWmjQx7bVrzZQxAIB32rXLHB/kLypUkPr0Me3cXOl//7ObB/AUilwA5SpQN50qKG9dblaWKXQBAN7Jn0Zx8+QVuZI0daq9HIAnUeQCKFeBPpIrsS4XAHyFPxa54eFS69am/dNP0oYNdvMAnkCRC6Bc5RW5ISFSRITdLLYULHJZlwsA3ik3V1q40HaK8sFoLgINRS6AcpOZKf36q2k3bWoK3UDUpIl07rmmvXy55Dh28wAAilq/XjpwwHaK8nHnna4+ePp0KTvbbh6gvFHkAig3CQmujjRQpypLZuOPvF2WDxyQEhPt5gEAFOWPU5XznHeedPPNpv377/47Yg3kocgFUG5Yj+uSt/mUxJRlAPBG/lzkStK997raH3xgLwfgCRS5AMoNOyu7sPkUAHivtDTphx9spyhfXbu6ls589pn/Ts0GJIpcAOWo4Ehu8+b2cniD6GgpONi0GckFAO+ydKmUkWE7RfmqXFm6/37TzsyU3n3Xbh6gPFHkAig3eUVujRrSX/9qN4ttYWFSZKRpb94sHTpkNw8AwMXfpyrneeQRV/utt8yO0oA/osgFUC6OHpV27TLt5s2loCC7ebxBwSnLK1faywEAKCxQitzGjaUbbzTtHTuk+fPt5gHKC0UugHLBetyiWrd2tePj7eUAALgcOiT99JPtFJ4TG+tqT5xoLwdQnihyAZQLityirrrK1V692l4OAIDLsmWBdX75zTdL9eub9rx5UlKS1ThAuaDIBVAu2HSqqIYNzVmFkhnJDaRfqgDAW33/ve0EnhUcLD38sGk7jjR5st08QHmgyAVQLjgjt6igILPLsmSObti5024eAEDgFbmS9MADrh3/337b/3eWRuChyAXgdo7jKnIvuMB1Lh9cRa7EulwAsO34cWntWtspPK9ePen22037jz+kGTPs5gHcjSIXgNslJ0uHD5s2o7iFsS4XALzHihVSTo7tFHb06+dqv/QSxwnBv1DkAnA7piqfHiO5AOA9AnGqcp527aQ2bUx70yazCRXgLyhyAbhdwZ2V2XSqsLp1pb/+1bTXrg3cEQQA8AaBXOQGBUlPPeW6PWYMGyLCf1DkAnA7RnLPLG80NzVV+uUXu1kAIFClp0urVtlOYdfNN7vejF6xIrCLfvgXilwAbpdX5FaoIF16qd0s3oh1uQBgX3y8lJlpO4VdFSpIw4a5bo8ZYy8L4E4UuQDcKjtb2rzZtMPDpdBQu3m8EetyAcA+Ri2NO++UGjQw7fnzpR9/tBoHcAuKXAButXWr67w9piqfWqtWZi2UxEguANhCkWsEB0tDhrhuv/iivSyAu1DkAnCrgutx2XTq1GrUkJo2Ne2ffjLrwgAAnpOVJS1fbjuF97j/fqlOHdOeOdP0TYAvo8gF4FYFd1ZmJPf08tblZmVJGzbYzQIAgWbdOrP5H4zQUGnoUNN2nMLrdAFfRJELwK3YWbl4Cq7LZcoyAHgWU5WL6tfPdcTdV19JixfbzQOUBUUuALfKK3JDQ6VLLrGbxZsV3GGZzacAwLNWrLCdwPtUqSI9/7zr9tChnJsL30WRC8BtUlOlbdtMu1kzqWJFu3m82eWXSyEhps1ILgB4VqCfj3s699xj+idJWrNG+vRTu3mA0qLIBeA2mze73vVl06kzq1xZatnStBMSpKNHrcYBgIDx22/mA0VVrFh4d+V//5uzhOGbKHIBuA2bTpVMwXW5a9faywEAgYRR3DPr3Fm67jrT3rZNeustu3mA0qDIBeA2bDpVMqzLBQDPo8g9s6Ag6eWXXbeHD5f27rWXBygNilwAbkORWzLssAwAnkeRe3ZRUebsXEk6dkx6/HG7eYCSosgF4DZ5RW7t2lLdunaz+IImTaTq1U2bkVwAKH/Z2WZDJZzdyy9L551n2p98Is2fbzcPUBIUuQDc4o8/pH37TLt5czPdCWdWsaJ5t1yS9uxhOhgAlLdNm8xJADi72rWlsWNdt2NjpbQ0e3mAkqDIBeAWTFUuHaYsA4DnMFW5ZPr2lTp0MO3t26UXXrCbByguilwAbsHOyqVDkQsAnkORWzJBQdJ//+s61/2VV6SffrKbCSgOilwAbsFIbulQ5AKA51Dkltyll0pDh5p2VpbUu7eUkWE3E3A2FLkA3KJgkdusmb0cvuavf5XOP9+0V6+WHMduHgDwV8eOSZs3207hm4YPd72BvXGjNGKE3TzA2VDkAiiz3FxXkXvJJa4dg3F2QUGu0dxDh6QdO+zmAQB/tWYNbySWVuXK0rRpUqVK5vYrr0jff283E3Am1ovcgwcPKjY2VlFRUYqJidGoUaOUnZ19ymuXLFmibt26qWXLlurSpYsWL158yuteeOEFDRs2rNB9e/bs0WOPPabWrVsrJiZGsbGx2r17d/7jkydPVrNmzRQZGZn/8frrr7vvGwX82Pbt0okTpn355Xaz+CKmLMMb0B/D3zFVuWwuv1x6/nnTdhyzKdWxY3YzAadjvcgdNGiQwsLCtHTpUs2cOVMrVqzQ+++/X+S6pKQk9e/fXwMHDtSaNWvUv39/DRo0SPvyziyRdPjwYQ0ePFhTp04t8vX9+vXTOeeco0WLFmnRokWqWbOmYmNj8x//+eef9eijj2rdunX5H49z8jVQLAU3oWA9bslR5MIb0B/D31Hklt0TT0jt25t2UpI0cKDVOMBpWS1yd+7cqfj4eA0ZMkShoaGqX7++YmNjNX369CLXzpo1S1FRUerUqZOCg4PVtWtXRUdH65NPPpEkpaamqnPnzqpRo4ZuuummQl979OhR1a5dWwMHDlRYWJiqVq2qe++9V1u2bNHRo0clSRs3blTz5s3L/5sG/FDB9biM5JYcRS5soz+Gv3Mcilx3qFhR+vBD17Kk99+XPv3UaiTglIJtvnhiYqJq1qypunXr5t/XqFEjJScn69ixY6pRo0b+/Vu3blVEREShr2/cuLESEhIkSZUrV9a8efNUu3btIlOjzjnnHL3zzjuF7ps/f74uvPBCnXPOOTp48KCSk5M1Y8YMPf3006pUqZI6d+6sgQMHqnLlyqfM3rFjxzN+bwsXLjz7DwDwEwVHcilyS65OHbMB1a5d0tq1Uk6O+UUC8BT6Y/i7PXuk33+3ncI/NGggjR8v3Xefuf3QQ1JMjOnHAG9hdSQ3NTVVoaGhhe7Lu30ib4HfGa6tUqVK/nXBwcGqXbt2sV73o48+0rvvvqsX/jzR+o8//lBUVJRuu+02LVq0SFOmTNHSpUv14osvlur7AgJNXpEbGio1amQ3i6/KG81NTZX+rBUAj6E/hr9bu9Z2Av9y773SnXea9pEj5lihnByrkYBCrI7khoWFKS0trdB9eberVq1a6P7Q0FClp6cXui89Pb3IdWeSmZmpMWPGKC4uTpMmTVLr1q0lSU2bNi00JatRo0aKjY3VM888o5EjR57yuXhnGDBSU6Vt20y7WTNGIEsrOlr67DPTXr2aY5jgWfTH8Hfr1tlO4F+CgqS33pJWrDCzkJYulV58UfrPf2wnAwyrI7nh4eE6cuSIDhw4kH/ftm3bVK9ePVU/6QySiIgIJSYmFrpv69atCg8PL9ZrHTp0SH369NH69es1c+bM/A5VkuLj4zVp0qRC12dmZqpKlSol/ZaAgLNpk+tIBqYql17Bdbnx8fZyIDDRH8PfUeS6X82a0vTpUoU/q4mRI6WVK61GAvJZLXIbNGigVq1aafTo0UpJSdHu3bs1ceJE9ezZs8i13bt3V3x8vOLi4pSdna24uDjFx8erR48eZ32drKws/fOf/1S1atX00UcfqX79+oUeDw0N1fjx4zV37lzl5uYqMTFREydO1J158zAAnBbrcd2jVStXm82n4Gn0x/B3FLnlo1071+htTo50zz0cKwTvYP0IoXHjxik7O1sdO3ZUr1691L59+/yjBCIjIzVnzhxJZsrShAkTNGnSJEVHR2vixIkaP368GjZseNbXWLx4sTZt2qTVq1erTZs2hc7eS05OVosWLfTaa6/p7bffVqtWrfTAAw+oW7dueuSRR8r1ewf8QcGdlTk+qPTOOUdq0sS0N2yQMjLs5kHgoT+Gv/rjD7PxFMrHiBFSmzamvX279NhjdvMAkhTkOHkTDQGg5K67TvruO9Pev186/3yrcXxanz7StGmmHR9feAozAKB0FiyQTjrNyufNni0VY/KEx+zYIV1xhXT8uLn96afSKSaCAB5jfSQXgO9yHNd05b/8hQK3rDgvFwDcj6nK5a9hQ2niRNftfv2kAkv8AY+jyAVQanv3SocOmTbrccuOIhcA3I8i1zPuucc1urx/vzRwoN08CGwUuQBKreCmU6zHLbuWLaXgPw92o8gFAPf48UfbCQJDUJD03/9KtWqZ2//7n5kqDthAkQug1NhZ2b1CQ11vFvzyi5SSYjcPAPi648elk068Qjn6y1+k115z3e7fn40UYQdFLoBSK7izMkWue+RNWc7NZfQBAMpqwwbbCQLPvfdKbdua9pYt0quv2s2DwESRC6DU8kZyK1aUmja1m8VfsC4XANyH9bieV6GC2YSqYkVze/Ro6fff7WZC4KHIBVAqmZlmSq1kznetXNluHn9BkQsA7kORa8fll0uPPmraqanS88/bzYPAQ5ELoFQ2bZKyskw7MtJuFn/SrJlZmytR5AJAWbHsw57hw6Vq1Ux78mTWRsOzKHIBlErBd8evvNJeDn8THOx602D7dungQbt5AMBXZWSYN2RhR5060pAhpp2dbYpewFMocgGUSsF3xxnJda+CU5bXrLGXAwB82aZNpriCPf/6lyl2JWnGDLMRFeAJFLkASqXgSG7LltZi+CXW5QJA2bEe175q1UyhK0mOI730kt08CBwUuQBKLCfHdSxDw4aug9/hHhS5AFB2rMf1Do8+Kp1zjml/+KG0a5fdPAgMFLkASiwx0eyWKDFVuTw0buz6hYAiFwBKZ/162wkgSTVqSP37m3Z2tvTGG1bjIEBQ5AIoMTadKl8VKkhRUaa9d6/022928wCAr3Ec6eefbadAngEDXEcNvveedOKE3TzwfxS5AEqMTafKH1OWAaD09uyRjh2znQJ5zj9fuusu0z5yRPrf/6zGQQCgyAVQYozklj+KXAAoPY4O8j79+rnab75pRtuB8kKRC6BEHMdV5NarZz7gfgWL3Ph4ezkAwBcxVdn7REVJV11l2hs2SMuX280D/0aRC6BEdu2SDh0ybaYql5+LLpLq1jXtNWuk3Fy7eQDAl1DkeqeCo7nvv28tBgIARS6AEmGqsmcEBUkxMaZ95IjZ0RoAUDxMV/ZOt99uzs6VpBkzpLQ0u3ngvyhyAZQIm055TuvWrvbKlfZyAIAvyc2lyPVWVatKPXua9rFj0hdf2M0D/0WRC6BEGMn1HIpcACi5HTsYIfRmffu62h9+aC8H/BtFLoASyStya9aUGjSwmcT/RUWZM3MlilwAKC5Gcb3bNddIf/2rac+fL/3+u9088E8UuQCKbf9+6bffTDsy0qwbRfmpXl1q3ty0f/pJSkmxmwcAfAGbTnm3ChWk3r1NOzdXmjXLbh74J4pcAMXGelzPy5uynJtrdlkGAJwZRa73y1uXK0mffWYvB/wXRS6AYit4XmurVvZyBBLW5QJAyTBd2fu1bCldcolpf/eddOCAzTTwRxS5AIpt1SpXu2DxhfJDkQsAxZeVJSUk2E6BswkKMscJSVJODrssw/0ocgEUi+O4itzataWGDe3mCRRNmphNviRT5DqO1TgA4NW2bpUyM22nQHEwZRnliSIXQLFs2yYdPGjaMTFsOuUpFSqYn7ck7dsn7dxpNw8AeDOmKvuO6Gipfn3T/vZbc24u4C4UuQCKpeBUWaYqexZTlgGgeNh0yncEBUk9eph2Vpa0cKHdPPAvFLkAiqXgety8kUV4BkUuABQPRa5v6drV1Z43z14O+J9SFbnJycnuzgHAyxUscqOj7eUIRFdd5WpT5KIg+mOgMKYr+5Zrr5VCQ007Lo59J+A+pSpyO3bsqPvvv19z585VRkaGuzMB8DLp6dL69aYdHu7aCAmece65ZgMqSVq3TuK/XeShPwZc0tOlxETbKVASoaHS9deb9t69rt81gLIqVZE7duxYBQcHa9iwYWrbtq1GjBih9fytBPzWunVmvYwkVaxoN0ugypuynJkp/fij3SzwHvTHgMuvv5rjaOBbCk5ZjouzlwP+pVRF7s0336wpU6bou+++08MPP6wff/xRf//739W5c2dNnjxZ+/btc3dOABYVnKp83nn2cgSyNm1c7eXL7eWAd6E/BlyYquybKHJRHsq08dT555+vBx98UF9++aVmzZqlOnXq6PXXX9f111+vRx99VGvXrnVXTgAWFVwHWqeOvRyBrG1bV/uHH+zlgHeiPwbYdMpXNWggNW1q2qtWcZQQ3KPMuyuvWbNGw4cP13333ac1a9aobdu2+ve//63s7Gz17t1b7733njtyArCo4EjuuefayxHILrvMtRZ62TI250BR9McIdBS5vqtjR/M5J0f6/nu7WeAfghyn5L8q7dy5U1988YXmzJmj3377TRdeeKH+9re/6fbbb1e9evXyrxs8eLC+//57xcfHuzU0AM/Zt08q8M9aDz4oTZ5sL08g69pV+uor096yxWwChsBGfwy4NGokbd9uO4VnzJ7tOmPWH8yaJd12m2kPGiS9/rrVOPADwaX5optuukmVK1dWp06d9Pzzz6tNwcViBVxyySVKSkoqSz4AlhUcxYVdbdu6itxlyyhyQX8M5ElNDZwC1x9de61UoYKUmystXGg7DfxBqUZyp0+fru7du6t69erlkQmAF/nPf6TRo123Gcm1Z8kS84uAJP3zn9KUKVbjwAvQHwPG6tWFzxT3d/42kiuZP7/Vq0173z72AEHZlGpN7vz587V///5TPpaQkKBu3bqVKRQA77FsmavN8UF2RUdLwX/Ov2HzKUj0x0Aedlb2fXnrciVp0SJ7OeAfij1dec2aNcob9I2Pj9fq1at16NChItctXrxYu3fvdl9CANZkZBSerhwUZC8LpLAw6corpfh4KSFBOniQI50CEf0xUBSbTvm+jh2lF1807UWLpL//3W4e+LZiF7kzZ87U7NmzFRQUpKCgID377LNFrsnrdG+55Rb3JQRgzdq1Unq67RQoqG1bU+RK5rxcBuoCD/0xUBRFru9r21YKCZGysthhGWVX7CL3P//5j2677TY5jqO+fftqxIgRaty4caFrKlSooBo1aiic3VAAv7B0qe0EOFnbtq5dJ5cto8gNRPTHQFFMV/Z9oaFSVJS0YoX066/S/v2sy0XpFbvIrV69uq76c0X/hx9+qGbNmqlq1arlFgyAfRS53qdtW1e74HppBA76Y6CwI0ekPXtsp4A7XHONKXIl8zvI7bfbzQPfVewid/bs2erQoYNq1aql5ORkJScnn/H6W2+9tazZAFiUm0sR5Y3q1TNnQW7bZnahzMiQKle2nQqeRH8MFMYorv9o31566SXTpshFWRS7yB02bJhmzJihWrVqadiwYWe8NigoiE4V8HE//2zeHYf3advWFLkZGWbd9NVX204ET6I/BgqjyPUfbduaTS4dh9lkKJtiF7kLFy7U+eefn98G4N+++852ApxO27bShx+a9rJlFLmBhv4YKIxNp/xHzZrS5ZdLGzZI69dLx45JNWrYTgVfVOwi98ILLzxlO092drZSUlJUs2ZNtwQDYNfixbYT4HTatXO1ly6VhgyxlwWeR38MFEaR61/atzdFbt6yqS5dbCeCL6pQmi/Kzs7Wm2++qTlz5kiSVqxYoauvvlpt2rRR3759dfToUbeGBOBZOTmM5HqzSy+Vatc27aVLzS8CCEz0xwDTlf1N+/au9vLl9nLAt5WqyB0/frz++9//6vjx45Kk0aNHq1atWnrqqae0a9cuvfrqq24NCcCzNmxgPa43Cwpy/RJw5AijGIGM/hiBbv9+8wH/UXAJTt5Oy0BJlarI/fLLL/Wvf/1L99xzj7Zv367ExEQ9+uijuvfee/X4449r0aJF7s4JwIP4J+z9OnRwtZcssZcDdtEfI9Axiut/LrrIfEjSqlVmdhlQUqUqcvfv368rrrhCkvT999+rQoUKuuaaayRJ9erVy39HuTgOHjyo2NhYRUVFKSYmRqNGjVJ2dvYpr12yZIm6deumli1bqkuXLlp8mkWDL7zwQpEdJ0+cOKGnnnpKMTExatWqlYYOHarU1NT8x3fs2KG+ffsqMjJS7dq101tvvVXs7wHwN6zH9X5//pcrSfr+e3s5YBf9MQIdRa5/at3afE5J4c8YpVOqIrdOnTra8+ep2998840uvfRSnXvuuZKkdevWqV69esV+rkGDBiksLExLly7VzJkztWLFCr3//vtFrktKSlL//v01cOBArVmzRv3799egQYO0b9++/GsOHz6swYMHa+rUqUW+/vnnn9fevXs1f/58LViwQHv37tXYsWMlSVlZWXrkkUfUokULrVq1SpMnT9b06dP11VdfleTHAviFzEyKJl9w+eXSOeeY9vffm+MWEHjojxHoWK7hn9q0cbWZsozSKFWR2717d40ZM0YPPPCA1q5dq9v/PKl51KhRGj9+vLp161as59m5c6fi4+M1ZMgQhYaGqn79+oqNjdX06dOLXDtr1ixFRUWpU6dOCg4OVteuXRUdHa1PPvlEkpSamqrOnTurRo0auummmwp9bVpamubOnasBAwaoZs2aOu+88zR48GB9/vnnSktL0+rVq7V//34NGDBAlSpV0mWXXaY+ffqcMgfg71auNO+cwrtVrOjaZXn/funXX+3mgR30xwh0FLn+iSIXZVXsI4QKGjBggKpUqaLVq1friSee0N133y1J2rhxo/7xj3/o0UcfLdbzJCYmqmbNmqpbt27+fY0aNVJycrKOHTumGgUOxtq6dasiIiIKfX3jxo2VkJAgSapcubLmzZun2rVrF5katXPnTmVlZRX6+kaNGik9PV1JSUlKTExUw4YNValSpULPPXny5NNm79ix4xm/N84uhK9asMB2AhRXhw7SvHmmvWSJ1LSp3TzwPPpj+uNA5jhMZfVXV14pVapkZpetXGk7DXxRqYrcoKAgPfzww3r44YcL3f/xxx+X6HlSU1MVGhpa6L682ydOnCjUqZ7q2ipVqujEiROSpODgYNXOO1PjJCl/DkuFhYUVeZ3U1NTT5sh7biCQUOT6jpPX5Z70XzICAP0xAllyMicB+KvKlU2hu3Klmal06JD050oMoFhKVeRK0vHjx7Vy5UqdOHFCzikWg916661nfY6wsDClpaUVui/vdtWqVQvdHxoaqvT09EL3paenF7nudK+T99x51+e9TrVq1U6b40zPzTvD8EcHDkhr1thOgeK68kqpalUpNdWM5DqOOV4IgYX+mP44UDFV2b+1bu0axV21SurSxW4e+JZSFblLlizRoEGDinREeYKCgorVqYaHh+vIkSM6cOBA/ru+27ZtU7169VS9evVC10ZERGjTSXNStm7dqubNm5/1dRo2bKiQkBBt3bo1fxfKbdu2KSQkRA0aNNDBgweVlJSk7OxsBQcH5z93eHj4WZ8b8CcLF7KBkS8JCTHnCX7zjfTbb9KOHdIll9hOBU+iP0YgY6qyf7vqKld77VqKXJRMqTaeeu2113TJJZdo2rRp+uabb7Rw4cJCH99++22xnqdBgwZq1aqVRo8erZSUFO3evVsTJ05Uz549i1zbvXt3xcfHKy4uTtnZ2YqLi1N8fLx69Ohx1tcJDQ1Vly5dNHbsWB06dEiHDh3S2LFjdcstt6hKlSqKiYlRrVq19OqrryojI0MJCQmaOnXqKXMA/mz+fNsJUFKclxvY6I8RyBjJ9W9RUa42s8xQUqUqcrdv365BgwYpKipK9evX14UXXljko7jGjRun7OxsdezYUb169VL79u0VGxsrSYqMjNScOXMkmY0pJkyYoEmTJik6OloTJ07U+PHj1bBhw2K9zsiRI9WgQQN169ZNnTt31kUXXaQRI0ZIMuuH3n33XW3ZskVt27bVQw89pD59+ui2224r4U8G8F05OdKXX9pOgZLivNzARn+MQEaR698aNXIdlUeRi5IKck61gOcsbrrpJg0aNEhdmDcA+I0VK8zU1zMJDpbuv186w0an8LCMDPNLQEaG1LChtH277UTwJPpjBKrcXKlGDbMnQSCaPVsqxuQJn9exo7RokWknJ0t/+YvdPPAdpRrJffjhhzVhwoT8A+gB+L65c20nQGlUruw6T3DHDmnnTrt54Fn0xwhUO3cGboEbSApOWV671l4O+J5SbTw1d+5c7du3TzfccIPOPfdcValSpdDjQUFBxV4HBMA7/DkTET7o+uul774z7cWLpfvus5kGnkR/jEDFVOXAULDIXb1auuUWe1ngW0pV5NarV0/16tVzdxYAlmzfzi6Vvuz666U/lzRq0SKK3EBCf4xARZ8VGNh8CqVVqiJ3zJgx7s4BwCKmKvu26GgpLEw6ccIUuZyXGzjojxGoGMkNDA0aSOeeKx06ZIpc+jcUV6nW5ObZtm2bPvzwQ40dO1b79u3TmjVrlJKS4q5sADyEIte3VaoktW9v2r/9JiUm2s0Dz6M/RqChyA0MQUGu0dz9+yW2H0BxlWokNycnRyNHjtRnn30mx3EUFBSkLl26aMKECdq9e7emTZvG9CnARxw9yvmq/uD6613nHC9aJEVE2M0Dz6A/RiDKzpYSEmyngKdERUkLFpj2mjVS/fp288A3lGok97///a/mzp2rF154QcuWLVPeKURPPvmkcnNz9frrr7s1JIDy8/XX5hcG+Lbrr3e1845bgP+jP0Yg2rbNHJuGwMC6XJRGqYrczz77TAMGDNDtt9+umjVr5t/ftGlTDRgwQMuWLXNXPgDljF2V/UNkpDkvVzI7LOfm2s0Dz6A/RiBiqnJgochFaZSqyD1w4IAuvfTSUz5Wt25dHTt2rEyhAHhGZqYUF2c7BdyhYkXp2mtN+8ABdh4NFPTHCET8/xZYLrpIqlPHtPM2nwLOplRF7sUXX6wlp1nEFx8fr4svvrhMoQB4xrffSkeO2E4Bd2HKcuChP0YgYiQ3sAQFmVMEJLPLclKS1TjwEaXaeKpv374aMWKEsrKydN111ykoKEg7d+7UqlWr9O6772rYsGHuzgmgHMyYYTsB3Om661ztRYukgQPtZYFn0B8jEFHkBp6oKGnePNNes0Zq2NBuHni/IMcp3aD/pEmT9NZbbyktLS3/vpCQEP3zn//UQH6zArxeRoZUt67ZXbm4goOl+++XJk8uv1wovdxcqV496Y8/pBo1pIMHzZ8Z/Bv9MQJJRoZUrRobJs6eLfXoYTuF53z5pdStm2kPHSq99JLdPPB+pf7158EHH1S3bt0UHx+v4OBgVa9eXVdccUWhjS8AeK8FC0pW4ML7VahgRnNnzJCOHZPWrXNN8YL/oj9GINmyhQI3ELVq5WqvXm0vB3xHiYvcL7/8Uh9//LE2bNig7D//l6lSpYquvPJK3XXXXerUqZPbQwJwP6Yq+6frr3f92S5aRJHrz+iPEYiYqhyY/vIX6YILpORk8wau45i1usDpFLvIzc3N1eDBgxUXF6c6deqoa9euql27tiRp3759io+PV//+/dWjRw+9+OKL5RYYQNmlp0tffGE7BcrDyZtPPfmkvSwoH/THCGTsrBy4IiNNkXvkiLRzp9Sgge1E8GbFLnL/97//6euvv9awYcN07733qkKFwhsz5+bm6qOPPtLo0aPVvn173XzzzW4PC8A9vv5aOn7cdgqUh8aNzXELe/ZIP/xgjomqVMl2KrgT/TECGSO5gatlS9fmU+vXU+TizIp9hNDnn3+uO++8U/fdd1+RDlWSKlSooHvuuUe9evXSDOZBAl6Nf6L+KyjItcvyiRNSfLzdPHA/+mMEMorcwNWypau9fr2tFPAVxS5yk5KS1KFDh7Ne1759e23fvr1MoQCUn7Q0ac4c2ylQnjgv17/RHyNQnTgh8Vc6cEVGutrr1tnLAd9Q7CI3LS1N55xzzlmvq1Wrlg4dOlSmUADKz5dfSqmptlOgPJ18Xi78C/0xAtUvv5gNhxCYGjaUqlc3bUZycTbFLnIdx1HFihXP/oQVKig3N7dMoQCUnw8/tJ0A5e3ii6VGjUx7xQoz+gH/QX+MQMVU5cBWoYJ0xRWmvWuXxHt4OJNiF7kAfN/+/dJXX9lOAU/Im7KcmSktW2Y3CwC4Azsro+C63A0brMWADyjRObnPPPOMqlWrdsZrUlJSyhQIQPn5+GMpJ8d2CnhCx47SlCmmvWiRdMMNdvPAveiPEYgYyUXBInfdusLLc4CCil3kRkdHSzLTpM6katWqioqKKlsqAOWCqcqBo2DHv3ChvRxwP/pjBCqKXBTcfIp1uTiTYhe5U6dOLc8cAMrZpk3S2rW2U8BT6tSRWrSQNm40f+5Hjkg1a9pOBXegP0YgOnZM2r3bdgrYdtllUnCwlJ1NkYszY00uECD4vTjw5K3Lzc2VliyxmwUAyoL1uJCkKlWkSy817V9+kdLT7eaB96LIBQJATo40bZrtFPC0jh1dbaYsA/BlTFVGnrx1udnZ0ubNVqPAi1HkAgHgu++k336znQKeds015sgFifNyAfi2jRttJ4C3KLgud906ezng3ShygQDAhlOB6ZxzpD/3KNKmTdLvv9vNAwCl9dNPthPAWxTcYZl1uTgdilzAz6WkSJ99ZjsFbMlblytJixfbywEApeU4jOTC5YorXG2KXJwORS7g52bNklJTbaeALazLBeDrkpOlQ4dsp4C3OPdc6a9/Ne0NG8zmisDJKHIBP8dU5cB29dVS5cqmzbpcAL6IUVycLG9d7vHj0vbtdrPAO1HkAn5szx5G7wJdaKgpdCVpxw7zAQC+hPW4OBnrcnE2FLmAH/vf/8xaJgS2gutyGc0F4GsYycXJKHJxNhS5gJ9yHOmDD2yngDdgXS4AX8ZILk5GkYuzocgF/NS6dRySDiMqSqpWzbQXLWJ0H4DvyMqSfvnFdgp4m4svlmrWNG3OysWpUOQCfooNp5AnJETq0MG09+3jzQ8AvuPXX02hCxQUFOQazU1OlvbvtxoHXogiF/BDWVlmPS6Qp+C6XKYsA/AVrMfF6RScsrxhg7UY8FIUuYAfWrBA+uMP2yngTQquy2XzKQC+gvW4OB3W5eJMKHIBPzR1qu0E8DYtWki1a5v2d99J2dlW4wBAsTCSi9MpWOSyLhcno8gF/MzRo9IXX9hOAW9ToYJ03XWmffQovxAA8A2M5OJ0Lr1UqlTJtBnJxckocgE/M3OmlJ5uOwW8EetyAfiSI0ek3bttp4C3qlRJatbMtH/9VTpxwm4eeBeKXMDPMFUZp8O6XAC+hKnKOJu8Kcu5ufx9QWEUuYAf2blTWrLEdgp4q8aNpYsuMu0ffpAyMuzmAYAzoWjB2bD5FE6HIhfwI9On204AbxYU5BrNTUuTVq60mwcAzoT1uDibyEhXm70mUBBFLuAnHIepyjg71uUC8BWM5OJsrrjC1WYkFwVR5AJ+Yu1aKSHBdgp4O9blAvAFjkORi7OrUUNq1Mi0f/pJysmxmwfegyIX8BOM4qI4LrxQatLEtFetklJS7OYBgFPZsUM6ftx2CviCvCnLaWlml2VAosgF/EJWlvTRR7ZTwFfkTVnOzpaWLrWbBQBOhfWVKC42n8KpUOQCfmD+fOmPP2yngK8oOGWZdbkAvNGPP9pOAF/B5lM4FYpcwA8wVRklce21ZqdliSIXgHeiWEFxFSxyGclFHutF7sGDBxUbG6uoqCjFxMRo1KhRys7OPuW1S5YsUbdu3dSyZUt16dJFixcvLvT4lClTdM0116hly5bq06ePtm/fLklas2aNIiMjC300b95cTZo00b59+yRJkydPVrNmzQpd8/rrr5fvNw+4wfHj0pw5tlPAl5x3nmt61/r10sGDNtPAW9Afw5swkoviqldPqlPHtNetM5uWAXIs6927t/PEE084J06ccHbt2uXcfPPNzpQpU4pct2PHDqdFixbON99842RlZTnz5s1zLr/8cuf33393HMdxPv/8c6d9+/bOli1bnPT0dGfMmDHOzTff7OTm5hZ5ruPHjztdu3Z1JkyYkH9f//79nfHjx5ffNwqUk6lTHcf8l17+H8HBjvPgg7a/Y7jDE0+4/lw/+8x2GngD+mN4i+Rkz/Vrvvoxe7btPyXvctNNrp/Nrl2208AbWB3J3blzp+Lj4zVkyBCFhoaqfv36io2N1fTp04tcO2vWLEVFRalTp04KDg5W165dFR0drU8++USSNGPGDN19990KDw9X5cqV9cQTTyg5OVmrVq0q8lwvvPCC6tatq9jY2Pz7Nm7cqObNm5ffNwuUk48/tp0AvqhDB1ebzadAfwxvwlRllBRTlnGyYJsvnpiYqJo1a6pu3br59zVq1EjJyck6duyYatSokX//1q1bFRERUejrGzdurIQ/DwbdunWrHnzwwfzHQkJC1KBBAyUkJKh169b5969Zs0ZxcXH66quv8u87ePCgkpOTNWPGDD399NOqVKmSOnfurIEDB6py5cqnzN6x4M4tp7CQhW7wgEOHzKZTQEm1a2fW5TqO9P33ttPANvpjeBOmKqOkCu6wvG6d1K2btSjwElZHclNTUxUaGlrovrzbJ06cOOu1VapUyb/ubI/nGT9+vO666y5deOGF+ff98ccfioqK0m233aZFixZpypQpWrp0qV588cWyfYNAOfv8c3MMDFBStWpJLVqY9vr10rFjVuPAMvpjeBNGclFS7LCMk1kdyQ0LC1NaWlqh+/JuV61atdD9oaGhSk9PL3Rfenp6/nVne1ySdu3apfj4eI0aNarQdU2bNi00JatRo0aKjY3VM888o5EjR54yO+8MwxswVRll0b699NNPUm6utHy51Lmz7USwhf4Y3oQiBSXVuLFUtaqUmsp0ZRhWR3LDw8N15MgRHThwIP++bdu2qV69eqpevXqhayMiIpSYmFjovq1btyo8PDz/uQo+npWVpaSkpEJTqubPn68rr7xSF110UaHniY+P16RJkwrdl5mZqSpVqpTtGwTK0e+/SydtaAqUyDXXuNpMWQ5s9MfwFocPSzt22E4BX1OhgnTFFaadlGT+HiGwWS1yGzRooFatWmn06NFKSUnR7t27NXHiRPXs2bPItd27d1d8fLzi4uKUnZ2tuLg4xcfHq0ePHpKk22+/XdOmTVNCQoIyMjL06quvqnbt2oqKisp/jrVr1xa6nSc0NFTjx4/X3LlzlZubq8TERE2cOFF33nln+X3zQBl9+qkZgQNKq317V5siN7DRH8NbMAqH0io4ZXnDBns54B2sn5M7btw4ZWdnq2PHjurVq5fat2+fv8tiZGSk5vx5AGijRo00YcIETZo0SdHR0Zo4caLGjx+vhg0bSpJ69uyp++67T/369VPr1q21efNmTZo0SSEhIfmvtWfPnkKbauRp0aKFXnvtNb399ttq1aqVHnjgAXXr1k2PPPKIB34CQOkwVRll9Ze/mClekrR6tXTSbFUEGPpjeAM2nUJpnbz5FAJbkONwZDLga3bulBo08PzrBgdL998vTZ7s+ddG+XjgAendd037u+8KHy0EAJ7Wu7d0ipOrcJLZs6U/J0/gT2vXSnkTRPr0kT780G4e2GV9JBdAyX32me0E8BesywXgTRiBQ2k1ayZVrGjaTHsHRS7gg2bOtJ0A/qLgutylS+3lAIATJ6Q/j1sGSqxKFemyy0x782bppE3eEWAocgEfs2ePtGKF7RTwFw0bSnnHlC5fLmVl2c0DIHDlHWkGlFbe5lM5OdLPP9vNArsocgEf8/nnthPAnwQFuaYsp6YyVRCAPfz/g7IquPkUU5YDG0Uu4GNYjwt34yghAN6AnZVRVgWPEeJNk8BGkQv4kN9/Z90k3K9gkbtsmb0cAALb6tW2E8DXMZKLPBS5gA+ZNUvi0C+422WXSeecY9orVvB3DIDnpaZKGzfaTgFfV7Om64jFDRvM2lwEJopcwIewqzLKQ4UKUuvWpr1vn5SUZDUOgAC0Zg2bTsE98qYsp6ZKW7fazQJ7KHIBH/HHH9J339lOAX/Vpo2rvXy5vRwAAtOqVbYTwF8wZRkSRS7gM2bP5l1ulJ+CRS5HVAHwNIpcuAubT0GiyAV8xqxZthPAn8XEmOOEJIpcAJ5HkQt3KVjksmN34KLIBXzAsWPSwoW2U8CfnXOO1KyZaW/YYNYyAYAn7Nkj/fab7RTwFxdeKNWpY9pr17KZYqCiyAV8wNdfS5mZtlPA3+VNWc7J4SgPAJ7DKC7cKShIiooy7UOHpB077OaBHRS5gA9gqjI84eqrXW2mLAPwFIpcuFt0tKvNm7aBiSIX8HIZGdK8ebZTIBCw+RQAGyhy4W55I7mSOZ4KgYciF/By330nHT9uOwUCQUSEdO65pr1iBeuYAJS/7GyKELhfwSKXkdzARJELeLnZs20nQKAICnKN5h44IG3dajcPAP/388/SiRO2U8Df1KsnXXSRaa9dyxGMgYgiF/BiubnSF1/YToFAwpRlAJ7EVGWUl7x1uSkp0q+/2s0Cz6PIBbzY6tXS3r22UyCQFCxyly+3lwNAYKDIRXkpuPkUU+IDD0UuvJrjSJMnS5dfLlWrJl1yiTRokDk39mRZWVJMjPTMM8V77oQEqXt3qUYN6bzzpL/9Tdq+vfA1w4ebs9Yuvlh6//2i2aKipP/9rxTfWDExVRmedtVVUoU/ewZGcgH3O3FCqljRLA8o+FGlyqmv/7//M48nJZ3+OZOSij5fwY/773dda7tfOxlFLsoL63IDW7DtAMCZvPKK9O9/S0OGSB07mjWCw4ebNTzffGM6b0lKS5N695bi46UuXc7+vLt3S23bSk2amM48LU16+mnpxhuljRul0FCzo/Err0jvvGPOWXvwQfOuYLNm5jk+/thsmHHXXeX3/XN0EDytWjWpRQtpwwbz7yw1Vapa1XYqwH/89JNZivLRR1KDBq77K5xi2CExUXrqqbM/51/+cuo3pSZMkD75RHrgAXPbG/q1go4elX75xTOvhcDDDsuBjSIXXis3VxozRnr4YfNZkjp1MqOuvXqZjQSioqSlS6V+/aTffiv+c48cKVWvLn37rRQWZu5r2NCM7K5ZI7Vvbx674QbpnnvM42+/bXY6btZMysyU/vMfaeJEV6HtbgkJrCGBHdHRpsjNzZXWrZPatbOdCPAf69dLlSpJt98uhYSc/rqcHKlvX9Pn7dlz5uesXFlq3brwfWvWmAJ39GjXv2Hb/drJVq9mF3eUn1q1pEaNpG3bTF+WlXXmf3PwL0xXhtc6dsyMzt59d+H7IyLM523bzOfu3c20qx9/LN7zOo70+efmne28AlcyBXNysilwJdPJh4a6Hq9UyfzSIZlfAi6+WOrcueTfV3ExVRm2FFzHxBQvwL3Wr5cuu+zsv2yPHSvt2ycNG1by13AcKTZWuvRS6fHHXffb7tdO9sMPnnstBKa8/iw9Xdq82W4WeBZFLrxWzZrS+PFmWnFBn39uPjdvbj5//700d67pnIsjKclMkWrQwIwAn3eeWQvVrZu0a5frujZtzDvcW7aYNUMbN5osx45Jo0ZJL71Utu/vbChyYQtFLlB+1q83U5NvuMEsBTj3XDNjqeB56Js2mf0l3n23dMsFPvrI/Nv9v/8z63/z2O7XTrZ0qWdfD4GHdbmBiyIXPmX5ctMJ33qraw1RixYle44//jCfn3zSTHH+6CMzZWv9eum668waREnq2VO67TbzOtddJz3/vNSqlZk63aGDaT/xhNS0qfT3v5tzRd0lOZnNOGBP8+auTXDi4+1mAfxJbq4pLBMTTf/y1VdmivBHH0ldu5rHs7PNNOV//tP0NaUxdqwpXq+9tvD9Nvu1k2VlSStXlt/zAxI7LAcy1uTCZyxdakZbGzUym2aUVmam+Vy3rhkVztvso3Fj8y73tGnmXfWgIOmtt8w74cHB5t3w334zG3msXm0+L1ggffaZWfP06KPSp5+W/fuUpDlz3PM8QGmEhEiRkWYjm23bzAY1555rOxXg+xzHbP5Ur54pJCXpmmvM7d69pfnzzRtLhw9LL75YutdYtsysPzzVbCCb/drJ1q0zO00D5Sky0vy9dxxGcgMNI7nwCR9/bKZ2XXyxtHBh2X7hrl7dfO7SpfBulq1bmynS69cXvr5yZdd0rxEjzK6TTZpIM2dKffqYd8QHDjQ7IeetbSorpirDNt79BtyvYkUzuppX4Oa5+Wbzef16U1xOnmz6nuxsM7ormf6lOH3MzJlmw52uXU9/jY1+7WRMVYYnVK9u1qZLZhZFerrdPPAcilx4vVdeMZtPtW5t1t/Wq1e252vUyBS3GRlFH8vKKrwpR0GbNkkzZpidmSVp/35XsV2rlvlFwB1Tu44elRYtKvvzAGXBulzA/X77TZoypehuyWlp5vOECWa2UadOZkZFSIjr+J/Gjc1Remfz5ZdmSU9xdpH1VL92KhS58JS8/iwryxzhhcBAkQuvNmmSNHSodMcdZgrVOeeU/TmrVTM7KH/+eeFCd+FCsx43b3flkz35pNS/v3TBBeZ2nTrS77+b9t695l3x884re76vvjL/EQM2UeQC7peRIT30kBmpLeiTT8ybr9Onm39vBT/yCtA5c0yfeCaHDpnz5E/esPF0PNWvncxx2FkZnlOwP2O/k8DBmlx4rd9/N0cfXHyx6YRPPiKoUSPp/POL91wrV5prGzUyt8eMMVPGunaVBg82xzQ8+aQUE2OOJDrZkiVmfeK0aa77br7ZHLkQGSmNG2eeK9gN/6K++KLszwGUVXi4VKOG2XWVIhdwj0suMdOBX3rJdbbtDz+YKcqxsafeaOrnn83nFi3MqQB5Tu7XJDMdUzJHFJ2NJ/u1k23eLB086P7nBU6lTRtXe/ly8zsl/B8jufBacXFmCtfOnWZ0tU2bwh/z5hX/udq0MbtIFry9eLFZ63T77abQ7dZN+vrrwsct5Bk6VHrqKbNmN8/AgeYXkrvvNiOv//1vqb/VfBkZJfu+gPJSoYLr6IXkZPMBoOwmT5aeflr64ANTVH74ofTss9Ibb5TseU7u1yTzhq1kphqfjaf6tVNZvLh8nhc4lcsvl8LCTHv5crtZ4DlBjuM4tkMAMObPlzp3tp3i9IKDpfvvLzrVDv5p2DDXuZmzZ0s9eliNA8BP3H6768x7lA7/J5fM9de73lzZvVu66CK7eVD+GMkFvAhTleFNWJcLwN1yc6XvvrOdAoHm6qtd7RUr7OWA51DkAl4iN5ciF96FIheAu23caDbIAjypYJHLlOXAQJELeIk1a1j3CO9Sv77ZbVUyfz9Z3AKgrFiPCxtat3a1KXIDA0Uu4CUYxYW3CQpyjeYeOiRt3243DwDfR5ELG849V7r0UtP+8UfX2dTwXxS5gJeYPdt2AqCoglOW1661lwOA78vONkcXATbkTVnOzjazk+DfKHIBL7Blizk3EPA2V17pap98VjUAlMSaNdLRo7ZTIFAVXJe7bJm9HPAMilzACzBVGd6KIheAuyxYYDsBAhmbTwUWilzAC1DkwltdcIFr86l169h8CkDpffON7QQIZBERZm2uZIpc+jP/RpELWLZvH+8ownsFBblGcw8ckPbssZsHgG86dozzSWFXhQpSmzamffCglJhoNw/KF0UuYNncubybCO8WGelqM2UZQGksXizl5NhOgUDHlOXAQZELWMZUZXg71uUCKCumKsMbUOQGDopcwKKUFDp+eD+KXABl4ThSXJztFIA5Fq9iRdNmh2X/RpELWDR/vpSRYTsFcGYNG0rnnGPa69bZzQLA9yQkSDt22E4BSFWrupbgbN5s9pqAf6LIBSxiqjJ8QcHNp377zWyWBgDFNW+e7QSAS4cOrvaSJfZyoHxR5AKWZGVJX35pOwVQPAU3n2I0F0BJUOTCm1x3nau9eLG9HChfFLmAJd9/Lx0+bDsFUDysywVQGkePSj/8YDsF4NK+vWtdLkWu/6LIBSyZOdN2AqD4Cha5jOQCKK4FC6TsbNspAJcaNaRWrUx782aW4PgrilzAgpwcadYs2ymA4ouIkMLCTJuRXADFNXeu7QRAUQWnLH/3nbUYKEfWi9yDBw8qNjZWUVFRiomJ0ahRo5R9mrf8lixZom7duqlly5bq0qWLFp80x2DKlCm65ppr1LJlS/Xp00fbt2/Pf2zDhg1q2rSpIiMj8z/uueee/Md37Nihvn37KjIyUu3atdNbb71VPt8wILNtPe8cwpdUrCi1bGna27cz1d4f0R/D3dh7At6Kdbn+z3qRO2jQIIWFhWnp0qWaOXOmVqxYoffff7/IdUlJSerfv78GDhyoNWvWqH///ho0aJD2/VkpzJo1S1OnTtU777yjVatWqVmzZhowYIAcx5Ekbdy4UdHR0Vq3bl3+x/Tp0yVJWVlZeuSRR9SiRQutWrVKkydP1vTp0/XVV1957OeAwMJUZfiigptPrV9vLQbKCf0x3G3pUt4Qg3dq21YKDjZtilz/ZLXI3blzp+Lj4zVkyBCFhoaqfv36io2Nze/sCpo1a5aioqLUqVMnBQcHq2vXroqOjtYnn3wiSZoxY4buvvtuhYeHq3LlynriiSeUnJysVatWSTKdavPmzU+ZY/Xq1dq/f78GDBigSpUq6bLLLlOfPn1OmQMoq9xc6fPPbacASo7Np/wX/THKA8ty4K2qVZOuusq0t2yRkpPt5oH7Bdt88cTERNWsWVN169bNv69Ro0ZKTk7WsWPHVKNGjfz7t27dqoiIiEJf37hxYyUkJOQ//uCDD+Y/FhISogYNGighIUGtW7fWxo0bVbt2bd14441KSUnRVVddpWHDhqlevXpKTExUw4YNValSpULPPXny5NNm79ix4xm/t4ULFxbvh4CAs2qVOWsU8DVsPuW/6I/hbo4jzZ5tOwVwetddJy1fbtqLF0sFVk3AD1gdyU1NTVVoaGih+/Junzhx4qzXVqlSJf+6Mz2ek5OjOnXqqF27dvrss8/05ZdfKigoSA899JBycnJOm+PkDIA7MFUZvuqyy6SQENOmyPUv9Mdwt7VrpT17bKcATo91uf7N6khuWFiY0tLSCt2Xd7tq1aqF7g8NDVV6enqh+9LT0/OvO9PjFStWLLKuaPjw4WrTpo22bdt22hwnZyiId4ZRGo4jffaZ7RRA6VSqJDVrZtbjJiRIaWnSSfUIfBT9MdyNN3Th7a6+2vRrmZkUuf7I6khueHi4jhw5ogMHDuTft23bNtWrV0/Vq1cvdG1ERIQSExML3bd161aFh4fnP1fBx7OyspSUlKSIiAjt3btXY8aMUWpqav7jmZmZksy7y+Hh4UpKSiq0i2TB5wbcZe1aaedO2ymA0svbfCo3V9q40W4WuA/9MdzJcaQZM2ynAM4sNFRq3dq0t2+Xdu2ymwfuZbXIbdCggVq1aqXRo0crJSVFu3fv1sSJE9WzZ88i13bv3l3x8fGKi4tTdna24uLiFB8frx49ekiSbr/9dk2bNk0JCQnKyMjQq6++qtq1aysqKkq1atXSvHnz9PrrrysjI0OHDh3Ss88+qzZt2uivf/2rYmJiVKtWLb366qvKyMhQQkKCpk6desocQFl8+qntBEDZ5B0jJDFl2Z/QH8Od1q6VduywnQI4O6Ys+y/rRwiNGzdO2dnZ6tixo3r16qX27dsrNjZWkhQZGak5c+ZIMhtgTJgwQZMmTVJ0dLQmTpyo8ePHq2HDhpKknj176r777lO/fv3UunVrbd68WZMmTVJISIiqVKmit99+W9u2bVO7du100003qVq1anrjjTckScHBwXr33Xe1ZcsWtW3bVg899JD69Omj2267zcrPBP7JcaQ/Nx8FfBbHCPkv+mO4C6O48BUFi9xvv7WXA+4X5OQdXAegXK1YYdZ/+LLgYOn++6UzbHQKP3fsmHTOOaYdEyOtXGk3DwDv4jhSw4YszSlvs2dLf06eQBlkZkrnnSelpEjnny/9/rtUwfoQINyBP0bAQz76yHYCoOxq1JAaNTLtn36ScnLs5gHgXVaupMCF76hUSco7heyPPzgD3p9Q5AIekJPD9C34j7wpy2lp0pYtdrMA8C7TptlOAJRMly6u9tdf28sB96LIBTzgu++kfftspwDco+C6XDafApAnM5O9J+B7Ond2tb/6yl4OuBdFLuABH39sOwHgPuywDOBU5s+XDh60nQIomYsvli691LRXrpQOHbKbB+5BkQuUs8xM6bPPbKcA3IcdlgGcClOV4avypizn5krffGM3C9yDIhcoZwsWSIcP204BuM9f/iLVrWva69aZ3VQBBLZDh6QvvrCdAigd1uX6H4pcoJwxVRn+KG/K8sGD0p49VqMA8ALTp0sZGbZTAKXTvr0UFmbaX39tRnTh2yhygXKUkmLOsgP8DVOWAeRxHGnKFNspgNKrXFm6/nrT/v13acMGu3lQdhS5QDmaNUtKTbWdAnA/dlgGkCc+Xtq40XYKoGyYsuxfKHKBcvTBB7YTAOWj4A7LjOQCge3tt20nAMqOo4T8C0UuUE5275YWLbKdAigfjRtLVauaNiO5QOA6flz66CPbKYCyu+QSKSLCtJcvl44csRoHZUSRC5ST6dPZdRb+q0IF6YorTDspiR3EgUD1yScsy4H/6NrVfM7JYcqyr6PIBcqB40gffmg7BVC+Cq7LZZMOIDCx4RT8SffurjYbh/o2ilygHKxdK/3yi+0UQPli8ykgsP30k9l0CvAX7dtLtWqZdlwcx2L5MopcoBwwiotAUHDzKYpcIPBMmmQ7AeBewcFSt26mffy4tHix3TwoPYpcwM0yM9mEA4GheXPzC4HEDstAoDl8WHr/fdspAPe79VZXmynLvosiF3CzefOkAwdspwDKX+XK0mWXmfbmzVJ6ut08ADznnXekEydspwDc78YbpSpVTPuLL6TcXLt5UDoUuYCbcV4gAknelOWcHOnnn61GAeAh2dnSm2/aTgGUj6pVTaErSb//zrpzX0WRC7jR7t1sOY/AUnDzKaYsA4Hhiy+knTttpwDKD1OWfR9FLuBG77/PtBYEFnZYBgLPG2/YTgCUr1tuMefBSxS5vooiF3CT3FyzRgkIJFdc4WpT5AL+b+1a6YcfbKcAytf550tt25r2r79KCQl286DkKHIBN1m4kOlbCDw1a0oNG5r2Tz+ZtbkA/Nf//Z/tBIBnMGXZt1HkAm7ChlMIVHlTllNTpa1b7WYBUH6Sk6WPP7adAvCMHj1c7Vmz7OVA6VDkAm5w4AD/ASJw5e2wLDFlGfBnr78uZWXZTgF4RqNG0uWXm3Z8vJSUZDUOSogiF3CDDz+k40fgYodlwP8dOiT997+2UwCe1auXqz1zpr0cKDmKXKCMHEeaNMl2CsAedlgG/N/48WZJAhBI7rjD1f70U3s5UHIUuUAZLVokbdliOwVgzwUXSLVrm/a6deaNHwD+4/hxNpxCYIqIcJ0iwJRl30KRC5TRxIm2EwB2BQW5RnP/+EPau9duHgDuNXmydPiw7RSAHQWnLDOa6zsocoEy+O036YsvbKcA7GPKMuCf0tOlV1+1nQKwp+CU5Rkz7OVAyVDkAmXw9tucCwpIhXdYZvMpwH988AGzMxDYwsNdb+SuWSNt3243D4qHIhcopawsM4ULACO5gD/KzJRefNF2CsC+gqO57LLsGyhygVKaO1dKTradAvAO4eFSWJhpU+QC/uH999loB5CYsuyLKHKBUmLDKcClYkXp8stNe/t26ehRu3kAlE1GhvTCC7ZTAN6hcWPpyitNe+1aads2u3lwdhS5QCls2iQtXGg7BeBdCk5Z3rDBXg4AZff229Lu3bZTAN6j4C7Ln3xiLweKhyIXKAXOCwSKKrj5FFOWAd+VliaNGmU7BeBdCha5//sfZ8J7O4pcoIQOHpSmTrWdAvA+BUdy2WEZ8F2TJrGjMnCyhg2lq6827U2bpI0b7ebBmVHkAiU0ebI5NxBAYc2bm7W5EiO5gK9KTZXGjLGdAvBOd9/tak+fbi8Hzo4iFyiBrCxpwgTbKQDvFBoqNW1q2ps2mY1rAPiW8eOl/fttpwC8U69erjdzP/pIys21mwenR5ELlMBnn0m//WY7BeC98qYsZ2dLmzfbzQKgZA4e5Fxc4EzOP1+66SbT3r1b+uEHu3lwehS5QAmw4RRwZgXX5TJlGfAto0dz/BdwNgWnLP/vf/Zy4MwocoFiWrVKWrnSdgrAu7HDMuCbdu6U3nzTdgrA+/XoIYWFmfann0qZmXbz4NQocoFiYhQXOLuCRS47LAO+Y/hwflkHiqNaNVPoStKhQ9L8+Xbz4NQocoFi+O03824dgDM791zp4otNe/16NuUAfMGGDdK0abZTAL7jnntcbXZZ9k4UuUAxTJxoNtIBcHZ5o7kpKdK2bVajACiGYcMkx7GdAvAdN94onXeeac+ZIx07ZjcPiqLIBc4iJUV66y3bKQDfUXDzKaYsA95t/nzp669tpwB8S0iIdOedpp2WJs2caTcPiqLIBc7inXfMmgsAxVOwyP3xR3s5AJxZdrb0r3/ZTgH4pr59Xe3337cWA6dBkQucQVaW9NprtlMAvqVVK1d79Wp7OQCc2eTJnGcNlFZ0tHTppaa9dCnLc7wNRS5wBh9/LO3aZTsF4FsuvFC64ALTXr2azacAb3T4sDRihO0UgO8KCpLuu891+8MPrUXBKVDkAqfhONLLL9tOAfimmBjz+dgx6ddf7WYBUNTzz0sHD9pOAfi23r2lCn9WUx98wJu63oQiFziNuDjp559tpwB801VXudqrVtnLAaCoLVuk8eNtpwB83wUXmJ2WJWnnTun77+3mgQtFLnAajOICpZc3kitJ8fH2cgAoavBgjsUD3IUNqLyT9SL34MGDio2NVVRUlGJiYjRq1Chln+Z/3iVLlqhbt25q2bKlunTposWLFxd6fMqUKbrmmmvUsmVL9enTR9u3b89/bM+ePXrsscfUunVrxcTEKDY2Vrt3785/fPLkyWrWrJkiIyPzP15//fXy+abh9Vau5N04oCxatTLrlSRGcn0F/XFg+OYbae5c2ykA/9Gjh3TOOaY9c6Y5ehL2WS9yBw0apLCwMC1dulQzZ87UihUr9P4p3gZJSkpS//79NXDgQK1Zs0b9+/fXoEGDtG/fPknSrFmzNHXqVL3zzjtatWqVmjVrpgEDBsj583Tzfv366ZxzztGiRYu0aNEi1axZU7GxsfnP//PPP+vRRx/VunXr8j8ef/xxj/wM4H1eesl2AsC31aghXXaZaf/0kzlHEN6N/tj/cWQQ4H6hodLf/27aqanSZ5/ZzQPDapG7c+dOxcfHa8iQIQoNDVX9+vUVGxur6dOnF7l21qxZioqKUqdOnRQcHKyuXbsqOjpan3zyiSRpxowZuvvuuxUeHq7KlSvriSeeUHJyslatWqWjR4+qdu3aGjhwoMLCwlS1alXde++92rJli44ePSpJ2rhxo5o3b+7R7x/eKSFB+uIL2ykA35e3Ljc7W1q3zm4WnBn9cWB4+232mgDKQ8Epy2+/bS8HXIJtvnhiYqJq1qypunXr5t/XqFEjJScn69ixY6pRo0b+/Vu3blVEREShr2/cuLESEhLyH3/wwQfzHwsJCVGDBg2UkJCg1q1b65133in0tfPnz9eFF16oc845RwcPHlRycrJmzJihp59+WpUqVVLnzp01cOBAVa5c+ZTZO3bseMbvbeHChcX7IcDrvPKK2VkZQNnExEjvvWfa8fHS1VfbzYPToz/2f0eOSMOH204B+KfWrc2Zub/8Iv3wg/mcd4Yu7LA6kpuamqrQ0NBC9+XdPnHixFmvrVKlSv51Z3u8oI8++kjvvvuuXnjhBUnSH3/8oaioKN12221atGiRpkyZoqVLl+rFF18s2zcIn7NzpzR1qu0UgH9gh2XfQX/s/154QTpwwHYKwD8FBUkPPeS6PXmyvSwwrI7khoWFKe2khVp5t6tWrVro/tDQUKWnpxe6Lz09Pf+6sz0uSZmZmRozZozi4uI0adIktW7dWpLUtGnTQlOyGjVqpNjYWD3zzDMaOXLkKbPzzrB/euklKSvLdgrAPzRvbtYqpaWxw7K3oz/2b4mJ0rhxtlMA/u3ee6Vhw6SMDHNm7pgxUpUqtlMFLqsjueHh4Tpy5IgOFHhrcdu2bapXr56qV69e6NqIiAglJiYWum/r1q0KDw/Pf66Cj2dlZSkpKSl/StWhQ4fUp08frV+/XjNnzszvUCUpPj5ekyZNKvTcmZmZqsLfzICyZ4900iw6AGUQEiJdeaVpb98u/fGH3Tw4Pfpj/zZkCG/gAuXt3HOlO+4w7cOHzU7LsMdqkdugQQO1atVKo0ePVkpKinbv3q2JEyeqZ8+eRa7t3r274uPjFRcXp+zsbMXFxSk+Pl49evSQJN1+++2aNm2aEhISlJGRoVdffVW1a9dWVFSUsrKy9M9//lPVqlXTRx99pPr16xd67tDQUI0fP15z585Vbm6uEhMTNXHiRN15550e+TnAO7z8spSZaTsF4F8Knpe7erW9HDgz+mP/tXAhmykCnvLww642U5btCnIcu1vsHDhwQM8995xWrVqlChUq6NZbb9XgwYNVsWJFRUZG6tlnn1X37t0lSUuXLtXYsWO1a9cuXXjhhRoyZIg6dOggSXIcR++9956mT5+uQ4cOqUWLFnr22WfVsGFDLViwQP3791flypVVsWLFQq8/b948XXDBBVqwYIEmTJigXbt2qXr16urVq5diY2NVoYL1U5bgAXv3Sg0bmikmOL3gYOn++/mPG8X3ySeuoxVGjJCefdZuHpwe/bH/yckxsyl++sl2Erjb7NnmfFZ4F8eRmjUzG09J0qZNruP04FnWi1zAG/zrX9Lrr9tO4f0oclFSSUnmDSRJ6txZ+uorq3GAgPL221KBja7hRyhyvdf//Z80aJBpDxrE75e28LYoAt7vv0tvvWU7BeCfLr5YqlPHtFeulHJz7eYBAsXx49LTT9tOAQSePn2kvBPPPvjAbL4Iz6PIRcAbPZr/gIDyEhTkOh/3yBFp82arcYCA8fLL0r59tlMAgefcc6VevUz78GHp44/t5glUFLkIaDt3MooLlLd27VztZcvs5QACxe7d0tixtlMAgSs21tV+4w2zVheeRZGLgPbssxyrAJS3gkXuDz/YywEEin//WzrpqGIAHtS6tet0gZ9+kr77zmqcgESRi4CVkGDWSgAoX5GRUt4xpxS5QPlas0aaNs12CgB5m09JbD5lA0UuAtbw4WyCA3hCpUqud7STkqTffrMaB/BbjmNOCwBg3+23SxddZNpffiklJtrNE2gochGQ1q6VZs60nQIIHKzLBcrfrFnS0qW2UwCQpJAQqX9/03Ycadw4u3kCDUUuAo7jSIMH204BBJaCRe7339vLAfirzExp6FDbKQAU9OCDUliYab/3njllAJ5BkYuA88UXbAAAeNrVV0sVK5r24sV2swD+aMIEads22ykAFFSrlnTffaadmiq9/bbVOAGFIhcBJSODUVzAhho1pFatTHvzZmn/frt5AH9y8KD03HO2UwA4lQEDXO1x48ysC5Q/ilwElDff5J1uwJbrrnO1mU0BuM/zzzMNEvBWTZpI3bqZ9u7d0vvvW40TMChyETD++MP8IgDAjoJFLlOWAffYssVMVQbgvYYPd7VHj2Y01xMochEwRoyQjh61nQIIXG3bSsHBpk2RC7jHsGFSdrbtFADOJDpa6tLFtHfulD780G6eQECRi4CwcqU0aZLtFEBgq1bNdPSS9OuvUnKy3TyAr/vhB3NsEADvN3Kkqz1qlJSVZS9LIKDIhd/LyjJbuDuO7SQArr/e1f72W3s5AF/nONITT9hOAaC4YmKkm24y7aQkaepUq3H8HkUu/N6rr0o//2w7BQBJuvFGV3v+fHs5AF83Y4YUH287BYCSYDTXcyhy4de2bZOefdZ2CgB52rSRqlc37QULpNxcu3kAX5SRIT31lO0UAEqqTRvphhtMe/t2RnPLE0Uu/JbjSI88IqWn204CIE9IiGvK8oED0rp1dvMAvmjCBGnHDtspAJRGwdHcESOkEyfsZfFnFLnwWxMmsOYP8EZ5a5IkpiwDJXXokPTCC7ZTACittm2lW24x7d9+k954w2ocv0WRC7/000/S4MG2UwA4lYJF7tdf28sB+KJRo6TDh22nAFAWL70kVfizCnvxRWn/frt5/BFFLvxOWpp0111mzRIA73PJJVJ4uGkvX25GpgCc3fbt0ptv2k4BoKwuu0z65z9N+/hx6d//tpvHH1Hkwu888YS0ebPtFADOpHt38zknR5o3z24WwFf8+99SZqbtFADc4bnnpBo1TPudd6RVq+zm8TcUufArn34q/fe/tlMAOJu8IleS5syxlwPwFatWSZ98YjsFAHepW9cUunn69TNv/MI9KHLhN9askfr2tZ0CQHFcfbV03nmm/fXXLC8AzsRx2GcC8Ef9+knNm5v22rVm01S4B0Uu/MKePWZkKC3NdhIAxREc7NpdMiVFWrzYbh7Am82eLf3wg+0UANwtOFiaONF1+9//lnbutJfHn1DkwuelppoCd+9e20kAlETBKcszZ9rLAXiztDSz1wQA/9S+vfTII6admio99JCZvYGyociFT8vIkO68U1q3znYSACXVubNUtappf/45G+oApzJ2rLRjh+0UAMrTSy9JF11k2gsWFB7dRelQ5MJnZWZKvXqxMyvgq8LCpB49TPvwYdOxA3DZuVMaM8Z2CgDlrUYN6e23XbcHD5Z++cVeHn9AkQuflJVlRnDZlRXwbXfd5Wp/9JG9HIA3GjyYvSaAQHHTTWYjKklKTzf9I//+S48iFz4nPd0UuLNn204CoKxuvFGqVcu0v/jCrEcCIC1cyFp1INC8/LLUtKlpb9ggDRhgN48vo8iFTzlwQOrYUZo1y3YSAO5QqZJ0++2mnZoqffaZ3TyAN8jK4pdbIBCFhUmffiqFhprbb78tvfee3Uy+iiIXPiMxUWrTRlq+3HYSAO50//2u9pQp9nIA3uLll6XNm22nAGBD8+bSW2+5bj/yiLRsmb08vooiFz7h22+l1q2lrVttJwHgbm3aSJddZto//MBmGwhsmzZJzz1nOwUAm+6917U+NzNT+tvfpO3b7WbyNRS58Gq5udLzz5t1e4cO2U4DoDwEBUkPPui6XXCHSSCQ5ORI//gHx2kBkN54Q+rUybT/+MNsTLV/v9VIPoUiF17r4EHpllukESM4FBvwd717m/W5kll/lJJiNw9gw+uvS/HxtlMA8AbBwdKMGdKll5rbW7dKXbuaI/dwdhS58ErffCNdfrn01Ve2kwDwhNq1XccJHT7MRhsIPFu2SMOH204BwJvUqiXNny9ddJG5vXat2YD1wAG7uXwBRS68Snq69PjjZnpycrLtNAA86YknXO3XX5eys+1lATwpO9tswJaebjsJAG9Tv74pdOvUMbfXrZOuu07at89uLm9HkQuv8dNPUlSUWYMAIPC0aGHWHEnSjh0cJ4TAMXIkJwcAOL3LLpOWLJEuuMDc/vln6ZprpD177ObyZhS5sC4nR3rlFSk62uwqCSBwDRniao8cyWgu/N+330pjxthOAcDbNW0qff+99Ne/mttbtkgxMdLKlXZzeSuKXFiVkCC1aycNHcpukgCk66+X2rc37V9/ld5/32ocoFzt2CH9/e9srgigeBo1kpYuNZ8ls7SvQwfp3Xft5vJGFLmwIidHevVVqWVL3oEC4BIUVHhU65lnpNRUa3GAcpOaKt16qzlJAACK669/NcsbOnQwtzMzpQceMOfqZmTYzeZNKHLhcVu2mHUEgwfzjxFAUW3bSt26mfZvv0nPPms3D+BuWVlSr15mLwoAKKk6dcxJJP37u+6bOFFq1crswAyKXHhQTo7ZMfWKK9hgA8CZjR3rOjf3tdek9eutxgHcxnGkhx+W4uJsJwHgy0JCpHHjzFTlvP5y0yazTnfECJYBUuTCI9auNf/o/vUvjkgAcHYREdLTT5t2To50331SWprVSECZ5eZKsbGcAw3Afe6/X1q9WoqMNLdzcqTnnzdLAr/6ymo0qyhyUa6OHpUGDJCuuorpEwBK5sknzbEJkrRhgzRokNU4QJlkZ0sPPSS99ZbtJAD8zeWXS6tWmeU9wcHmvl9+kbp2NUfzbdxoN58NFLkoF44jzZghXXqpNH68efcaAEqiUiXp44+l0FBze/Jk8wH4muPHpe7dpXfesZ0EgL8KCTHTlOPjpdatXfcvWGBGde++O7CKXYpcuN2yZeZYoDvvlPbutZ0GgC9r0cJsppHn0UelmTPt5QFK6uefzXKdQJ42CMBzIiPN3jcffyxdfLG5LzdX+ugjM+LbrZv0ww/+f3QZRS7cZv16qUcPU+CysRQAd7nvPumJJ0w7N1e66y7Oz4X3y8mR3nhDio420wYBwFOCgsxgU0KC9NJLUu3arse+/NKcR3/FFdKECWZpoT+iyEWZOI707bfSjTead47mzLGdCIA/euUVs7mGZNY23n+/WbMb6LtHwjt9/72ZLvj442y2CMCeKlWkoUOlpCTp//5Puugi12MbN0qPPSZdcIHUu7f5Hd6fjvakyEWp/Pab9PLLUvPm0g03mLO6AKC8BAVJU6aYDjnPyy+bTe1++MFeLiCP45i+8KabpA4dpDVrbCcCAKNqVbMR7LZt0gcfSG3auB47cUKaPt3MxqxbV7r3XjPV+eBBe3ndwXqRe/DgQcXGxioqKkoxMTEaNWqUsrOzT3ntkiVL1K1bN7Vs2VJdunTR4sWLCz0+ZcoUXXPNNWrZsqX69Omj7du35z924sQJPfXUU4qJiVGrVq00dOhQpaam5j++Y8cO9e3bV5GRkWrXrp3eYvvDQhzHHFr/yivS9ddL9eubUZTNm20nAxAoKlY0ZwK+8YbZYEMyuy63by/dfLPZXINN7kqP/rjkHMesuX3uOalJEzOracEC26kA4NQqVTJF7PLlpv/s10+qWdP1+NGj0tSpZlnQ+eeb/QSeekr64gtp3z5rsUvHsax3797OE0884Zw4ccLZtWuXc/PNNztTpkwpct2OHTucFi1aON98842TlZXlzJs3z7n88sud33//3XEcx/n888+d9u3bO1u2bHHS09OdMWPGODfffLOTm5vrOI7jDBs2zOnbt69z+PBh58CBA07v3r2dZ555xnEcx8nMzHRuvPFG55VXXnEyMjKcTZs2Oe3atXPi4uI894PwIunpjvPrr47zxReOM3Kk43Tr5jh16zqO6c75COSP4GDHefBB239DAcf58UfHadGi6N/Rv/zFcR56yHHee89xNm1ynOxs20l9B/3xmeXmOs6+fY7z7beO8+KLjtOzp+NceKH9/5f5CMyP2bNt/4uAv8jIcJwvv3ScPn0cp3r1M/+9a9DAcbp3d5xhwxznww8dZ+VKx0lOdpycHNvfRVFBjuM4tgrsnTt36sYbb9T333+vunXrSpLi4uL0yiuvFHlX+PXXX9fGjRv17rvv5t/3z3/+U5dffrkGDBigu+66Sx06dNAjjzwiScrKylJMTIwmTpyoK664QtHR0frwww915ZVXSpI2bNige++9VytXrtS6devUr18/rVq1SpUqVZIkTZ48Wd9//72mTZvmiR+F2/38s5lSnJl5+o/jx81UhEOHzMfBg9KePVJysvmrDJwsONisheQYF3iD7GyzAdVzz0m7d5/6mqpVpQYNzOyT+vWl886TqlWTqlc3n2vUMEsuzjnHk8m9T6D0xwkJ5u9KdraUlVX0c2qqGcko+HHwoLRrl/lISytzBMAtZs8200sBd0pPN0uA5s+Xvv7a1BPFERIiXXihGf0991ypVi3zce65pr+tXLnwR6VKrs8VK5qPChXMUUcFR5bLItg9T1M6iYmJqlmzZn6HKkmNGjVScnKyjh07pho1auTfv3XrVkVERBT6+saNGyshISH/8QcffDD/sZCQEDVo0EAJCQmqWbOmsrKyCn19o0aNlJ6erqSkJCUmJqphw4b5HWrec08+w2/yHTt2POP3tnDhwrN89+XnjTfMZhcA4M+Cg6V//tPsvvzFF9KHH5qOueDGGamp0qZN5uN0LrpI2rHDPF+gCoT+eNIk6c+6GwBwClWqSJ06mY9XXjFHga5cKa1YYT7/+KPpV0+WlWU2t0pKKtvrV60qbdliNsMqK6trclNTUxUaGlrovrzbJ06cOOu1VapUyb/uTI+npKRIksLCwoq8Tmpq6mlznJyhuPbu3XvWTrc8DRokefESJviw7GwpLu4L2zFKrGPHjlb/TZYWuYsnOFi6/XZT6Kanl3zi3+7d0k03+ebP2l0CoT9++GGz4QrgD0aMeMrn/s+iT/Msd+T+y1+kv/3NbPT4/fdSSkr5TsRPSZH69HHPz9vq+9ZhYWFKO2nuT97tqlWrFro/NDRU6Sftw5+enp5/3Zkez+tM09LS8q/Pe51q1aqdNsfJGQo600itN/xDePhh81ESebltjkKXBrk9x/V3mzlSgD8JlP743nvNh7fz5f7BlzJLvpw73nYEwKtZHckNDw/XkSNHdODAgfz7tm3bpnr16ql69eqFro2IiFBiYmKh+7Zu3arw8PD85yr4eFZWlpKSkhQREaGGDRsqJCREW7duLfQ6eVOowsPDlZSUVGgXyYLPDQCAP6M/BgD4E6tFboMGDdSqVSuNHj1aKSkp2r17tyZOnKiePXsWubZ79+6Kj49XXFycsrOzFRcXp/j4ePX4c9X97bffrmnTpikhIUEZGRl69dVXVbt2bUVFRSk0NFRdunTR2LFjdejQIR06dEhjx47VLbfcoipVqigmJka1atXSq6++qoyMDCUkJGjq1KmnzAEAgL+hPwYA+BPr5+SOGzdO2dnZ6tixo3r16qX27dsrNjZWkhQZGak5c+ZIMhtTTJgwQZMmTVJ0dLQmTpyo8ePHq2HDhpKknj176r777lO/fv3UunVrbd68WZMmTVLIn4cpjhw5Ug0aNFC3bt3UuXNnXXTRRRoxYoQkKTg4WO+++662bNmitm3b6qGHHlKfPn102223WfiJAADgefTHAAB/YX0vydq1a2vcuHGnfGzdunWFbrdv317t27c/5bVBQUH6xz/+oX/84x+nfLxatWp6/vnn9fzzz5/y8YsvvljvvPNOCZIDAOA/6I8BAP7C+kguAAAAAADuQpELAAAAAPAbQY7jOLZDAAAAAADgDozkAgAAAAD8BkUuAAAAAMBvUOQCAAAAAPwGRS4AAAAAwG9Q5AIAAAAA/AZFrhvt3r1bDz74oK666iq1adNGQ4cO1bFjx/If37Fjh/r27avIyEi1a9dOb731lsW0LhkZGXrhhRfUtm1btWrVSn379tW2bdvyH/fW3AUNGTJEffr0KXSft+bes2ePHnvsMbVu3VoxMTGKjY3V7t278x/31twHDx5UbGysoqKiFBMTo1GjRik7O9t2rCISEhJ0//3366qrrlLbtm01dOhQHTp0SJK0YcMG3XHHHYqMjNT111+vTz/91HLawnJyctSnTx8NGzYs/z5vznzkyBENHTpUMTExio6OVmxsrPbv3y/Ju3Nv2rRJ99xzj6KiotSuXTu98MILyszMlOTduVF8vtgf0xd7Fn1x+fLlvliiP/aEcu+LHbhNz549nRdffNHJzMx0Dh8+7Nxzzz3OU0895TiO42RmZjo33nij88orrzgZGRnOpk2bnHbt2jlxcXGWUzvOsGHDnL///e/Ovn37nIyMDOfZZ591br75ZsdxvDt3nk8//dRp2rSp07t37/z7vDl39+7dnX//+99Oamqqk5KS4jz11FPOLbfc4jiOd+fu3bu388QTTzgnTpxwdu3a5dx8883OlClTbMcqJC0tzWnbtq3zf//3f05GRoZz6NAh58EHH3Qefvhh58iRI85VV13lTJs2zcnKynKWL1/uREZGOhs2bLAdO98bb7zhNG3a1HnyyScdx3G8PnPv3r2dfv36OUePHnWOHz/uPPbYY85DDz3k1blzcnKctm3bOh988IGTk5Pj7N2717npppucN99806tzo2R8sT+mL/Ys+uLy4+t9sePQH5c3T/TFFLluFBkZ6YwZMyb/H3Tv3r2d5557znEcx1m2bJnTsmVLJyMjI//6SZMmOffcc4+tuI7jOM6BAwecSy+91NmxY0f+fampqc7PP//s5Obmem3uPImJic51113nDB8+vFDH6q25jxw54vzjH/9w9u3bl3/fL7/84kRERDhHjhzx2txJSUlORESE8/vvv+ffN2/ePOfaa6+1mKqobdu2OQ888ICTnZ2df9+3337rXHnllc6MGTOcG2+8sdD1I0aMcIYOHerpmKe0fPlyp2vXrs6AAQPyO1Vvzrxx40anRYsWzvHjx/PvO3z4sLNlyxavzn3o0CEnIiLCee+995zs7Gxn7969TpcuXZx33nnHq3OjZHytP6Yv9iz64vLly32x49Afe4In+mKmK5dAenq6du7cecqPEydOqH///po2bZpatmyp1q1bKzMzU4MHD5YkJSYmqmHDhqpUqVL+8zVu3FgJCQlWc2/cuFHVq1fX+vXrdfPNN+dP66pVq5aCgoK8NveJEyeUnp6uxx9/XCNHjtT5559f6Gu9NXdISIjeeecd1alTJ//6+fPn68ILL9Q555xjNfeZJCYmqmbNmqpbt27+fY0aNVJycnKhKYC2XXLJJXr77bdVsWLF/Pvmz5+vZs2aKTExUREREYWu94afrWSmn/3nP//Rq6++qtDQ0Pz7vTnzTz/9pMaNG2vGjBm64YYb1K5dO7300ks6//zzvTp3rVq1dN999+mll15SixYt1KFDBzVo0ED33XefV+dGYb7YH9MXe09u+uLy5at9sUR/7Cme6IuD3RnY323YsEH33nvvKR+bMGGCgoKC9Oijj+r+++/X4cOH9a9//UsjRozQK6+8otTU1EL/WCQpNDRUJ06csJr7lVde0fHjx7VgwQJNnTpVISEheu655/TII49o1qxZXpt7woQJWrRokdq2basOHTrop59+KvS4N+fu1KlT/u2PPvpI7777rv773/9Kspv7TE6XS5JOnDihGjVq2Ih1Ro7j6I033tDixYs1bdo0ffjhh0W+hypVqlj/2ebm5mrIkCG6//771bRp00KPnern7g2ZJeno0aP69ddf1bx5c82aNUvp6ekaOnSonnzySdWuXdtrc+fm5qpKlSoaPny4evbsqZ07d+qxxx7TuHHjvPrnjcJ8sT+mLza8JTd9sWf4Sl8s0R97kif6YorcEoiJidGvv/56ysd+/vlnDRkyRKtXr1ZwcLDCwsI0dOhQ3XPPPRo5cqTCwsKUlpZW6GvS0tJUtWpVq7m//vpr5eTk6Mknn9S5554rSXrqqafUpk0b7dixw2tzz5kzRwkJCfr4449P+bi35s6TmZmpMWPGKC4uTpMmTVLr1q0l2c19JqfLJcl6tlNJSUnRU089pU2bNmnatGlq0qSJQkNDdfz48ULXpaenW88/adIkVapUqchmLZK8NrOk/BGO//znP6pcubKqVaumQYMGqVevXrrtttuUnp5e6Hpvyf3NN99o/vz5+vrrryVJ4eHh6tevn0aNGqVu3bp57c8bhflif0xfbHhD7jz0xeXLl/piif7YkzzRF1PkusnevXuVk5Oj3Nzc/PtCQkIUFBSkihUrKjw8XElJScrOzlZwsPmxb926VeHh4bYiSzLD/5LydzOTzI5yknn3zVtzf/HFF9qxY4euvvpqSWZXypycHEVFRWnOnDlem1uSDh06pEcffVSZmZmaOXOm6tevn/+Yt+YODw/XkSNHdODAAdWuXVuStG3bNtWrV0/Vq1e3mu1ku3bt0oMPPqgLLrhAM2fOzP+FMSIiQsuWLSt0rTf8bL/44gvt379fUVFRkpTfGX377bcaOnSoV2aWzP8dubm5ysrKUuXKlSUp//+/Sy+9VP/73/8KXe8tuffu3Vvo/ztJCg4OVkhIiNf+HUHJ+GJ/TF/sefTF5cvX+mKJ/tiTPNIXu2sBcaA7ePCgc9VVVzkjR4500tPTnQMHDjj33nuv079/f8dxHCcrK8u5/vrrnRdffNFJT093fvnlF6ddu3bOZ599Zjm549xzzz3O3//+d+fgwYNOSkqK869//cv529/+5jiOd+cuaNy4cYU2u/DW3JmZmc7f/vY35x//+IeTlpZW5HFvze04jnPXXXc5jz/+uHP8+PH8HR3HjRtnO1YhR44cca699lpn2LBhTk5OTqHHDh065ERFRTnvvfeek5mZ6axYscKJjIx0VqxYYSntqT355JP5G114c+bMzEznhhtucPr37++kpKQ4Bw8edO69916nX79+Xp07MTHRad68ufPf//7Xyc7Odnbt2uXccsstzosvvujVuVF8vtof0xd7Dn1x+fKHvthx6I/Lkyf6YopcN9q4caPTt29fJzo62mnXrp0zYsSIQjudJSUlOf/4xz+cVq1aOe3bt3cmTZpkMa3LsWPHnOHDhzvt27d3IiMjnUceecTZu3dv/uPemrugkztWx/HO3PPnz3ciIiKcFi1aOC1btiz08dtvvzmO4525Hcdx/vjjD6d///7OVVdd5bRu3dp58cUXC+2c6A3effddJyIiwrniiiuK/Hwdx3F++ukn584773QiIyOdjh07esUvLCcr2Kk6jndn/v33351BgwY5bdu2daKiopyhQ4c6R48edRzHu3MvW7bMueOOO5xWrVo51157rfPaa6/l76LqzblRfL7YH9MXew59cfnyh77YceiPy1t598VBjuM4ZR1yBgAAAADAG3CEEAAAAADAb1DkAgAAAAD8BkUuAAAAAMBvUOQCAAAAAPwGRS4AAAAAwG9Q5AIAAAAA/AZFLlDOOKXL9/FnCAC+j//LfRt/figJitwAN3jwYDVp0kQ//PDDKR9funSpmjRpopdeesnDyYrvgw8+ULt27XT55Zdr4sSJp7ymSZMmhT4uu+wyxcTE6P7779eSJUvKLdvWrVt11113Fckyfvx4t7/WgQMHFBMToxtuuEHp6emnvObxxx9X8+bNtXnzZre/vjukpKTo0Ucf1RVXXKHo6GglJSUVuWbVqlVF/jwLftx3332SpM8//1xNmjTRnj17ypRp7dq1evjhh894Td5rFfxo0aKFrr/+eg0fPlz79u0rdP348ePVpEmTMuWyzR++B8Cb0B/TH3uT4vTH7upni4O++NT84XsoL8G2A8Cup59+WitXrtTIkSP15ZdfKjQ0NP+x1NRUjRw5UuHh4Xr88cctpjy9lJQUvfjii+rQoYMeeOABXXTRRae9tmfPnrrjjjskSVlZWfrjjz80c+ZMPfTQQxo+fLh69+7t9nxfffWV1q1b5/bnPZXatWvr6aef1uDBgzVu3DgNHTq00OOLFi1SXFycBg0apMsuu8wjmUpq9uzZWrRokUaMGKHw8PAz/nmOGDFCzZo1K3J/9erVJUnXXnutPvnkE9WpU6dMmT799FNt3bq1WNe++eabOv/88yVJaWlpSkxM1KRJk7Ro0SJ9/PHHql+/viTpjjvuUPv27cuUyzZ/+B4Ab0J/TH/sTUrSH3sCffGp+cP3UF4ocgNczZo19cwzz6hfv35644039NRTT+U/9uqrr2r//v369NNPValSJYspT+/o0aPKzc3VDTfcoOjo6DNeW69ePbVs2bLQfV26dFG/fv304osv6tprr7X+n3hZdevWTV9//bU++OAD3XLLLfmdZ0pKip555hm1bNlSDz30kOWUp3fkyBFJ0t13362goKAzXtu4ceMif54FnXvuuTr33HPdmO7sLr300kJ/h9q0aaPrrrtOt912m0aMGKH33ntPkvm7WK9ePY9mczd/+B4Ab0J/TH/sTUrSH3sb+mJITFeGpE6dOumWW27R1KlTtXHjRknSjz/+qP/973967LHHdOmll0qSkpOT9a9//UtXXXWVrrjiCvXt27fINJs9e/Zo6NChateunZo1a6Y2bdpo6NChOnz4cP41119/vUaPHq2+ffvqyiuv1IgRI06bbdmyZbr77rvVqlUrxcTE6IknntDevXslmWkp119/vSTp3//+d6mmawQFBemJJ55QVlaWZs6cmX9/RkaGXn75ZXXo0EHNmzdXt27dFBcXV+hr09PT9eqrr+rGG29U8+bNdeWVV+r+++/XL7/8IslMIXnzzTclFZ0SlZKSov/85z+66qqrFBkZqQEDBujgwYMlzn8qzzzzjKpVq6ann35aOTk5kqSXX35Zx44d00svvaSKFStKkr799lvddtttatGihdq2basXXnhBJ06cKPRc3377re6++25FRkaqefPm6ty5s6ZNm5b/eN7U4Y8//ljXXXedrr766tNOtcvIyNCECRPUuXNntWjRQjfeeKMmT56s3NxcSVKfPn3yf0ZNmzbVsGHDyvRzOHka1bBhw9S3b1+NHDlSUVFR+tvf/qbs7GwtX75cd955pyIjIxUdHa3Y2Fht3749/2tmzZql3377TU2aNNHnn39e4hz169dXr169tHz5cu3atUtS0elFOTk5mjx5sm655RZdfvnlatmypf7+979rxYoVhZ7ru+++023/3969B1VR/n8Afx9AQMHQJMHRFBQXAT0oCRJyEVAp00rCmgiV8EKKlwDTar4oqDhkogMCiigg4pXwgkSiJqB4SAfCtEkNBREnAbkoJgbC+fz+YM7qkdvB6Fcyn9cMM57d5+w+u3vc93n2PPusuzukUinc3NyQnp6OKVOmiPtNcTzy8vLg4+MDS0tL2NnZ4ZtvvkFTU5O4nM6OBQCUlZVh0aJFmDBhAiwtLfHRRx8pdSV8fhs6K88Y6xznMedxT8zja9euYcmSJbC1tYWFhQUcHBywfv16pa7cnMWcxd2OGCOimpoasrOzIw8PD2psbKQZM2bQhx9+SE1NTUREVF1dTQ4ODjR16lRKS0ujU6dOkZeXF40dO5Zu3LhBRET19fXk7OxM7u7udPLkScrLy6OoqCgyMzOj//3vf+K6nJ2dydzcnEJDQyk3N5fy8/PbrNPRo0dJEAT6/PPPKTs7m44cOULOzs7k4OBAVVVVVF1dTSdPniRBEGjLli1UWFjY7vYJgkCRkZHtzndyciIvLy8iIpLL5TRv3jwaN24cJSQk0NmzZykoKIgEQaAjR46I71m6dCnZ2tpSSkoKXbhwgQ4ePEh2dnbk5uZGcrmc7t69S19//TUJgkCFhYV09+5dsS6jRo2iL774gmQyGe3evZssLCxo6dKlKh0rVSj2XXJyMl26dIlMTU0pOTlZnJ+WlkaCIFBgYCDl5OTQvn37yNramubOnUtyuZyIiLKyskgQBFq/fj3JZDI6c+YM+fj4kCAIVFBQQEREP/30EwmCQDY2NvTDDz/Q0aNH6eHDh63qI5fLydvbm8aOHUtxcXGUm5tL4eHhSp+NoqIipf1VWlra5rYp1imTyejJkydKf4rPKxFRamoqCYJAZWVlRES0atUqMjc3p7lz55JMJqNTp07R7du3SSqVUkhICOXl5dGJEyfIzc2NXF1dqbm5mUpLS2nBggU0ceJEKiwspOrq6jbr9Py6nieTyUgQBDp8+DAREUVGRpIgCOL8sLAwkkqllJSURBcuXKBjx47R1KlTydramh49ekRERHl5eWRmZkaLFy+m7Oxs2r17N1lZWZGFhYX42VbsGzs7O4qKiiKZTEYbNmwgQRBo//79Kh+L5uZmevvtt2nOnDmUnZ1Nubm5tHDhQjI3N6dbt2612gZVyjPGVMN5zHn8suRxZ9lHRFRRUUFWVlbk4+NDWVlZdP78eQoNDSVBEGjbtm1ERJzFnMX/CG7kMpEioLy8vMjS0pJKSkrEeZs3b6YxY8bQnTt3xGkNDQ3k6uoqhsFvv/1GH3/8cauToa+vL02dOlV87ezsTJMmTaLm5uZ269Lc3EwTJ04kb29vpemlpaVkYWFBGzduJCKisrIyEgSBUlNTO9y2zkL1gw8+oLfeeouIiHJzc0kQBPr++++VyqxYsYImTpxIT548oYaGBvLx8WlVJj4+ngRBoIqKCiJqfQJV1GXWrFlK0wIDA8na2rrDbeiqzz77jKytrWn69Onk4+MjTpfL5eTo6Ejz5s1TKq84+WdlZRERUVxcHK1cuVKpTG1tLQmCQNu3byeipyfyzZs3d1iX7OxsEgSBjh07pjQ9OjqaBEGgoqIiImp7fz1Psc62/iZPniyWa6uRKwiC0ok+PT2dBEGg8vJycdovv/xCmzdvFr8crFq1ipydnTusU2fBWlRURIIgUGxsbJvbGRAQQAkJCUrvyczMJEEQ6OeffyYiIk9PT5oxY4b4pefZ+j8frFu2bFFalouLC/n6+hKRaseisrKyVZm6ujrasGEDXb9+vdU2qFKeMaY6zmPO45chj1Vp5J47d44++eSTVg3uZ/cFZ/FTnMXdh+/JZaIpU6Zg2rRpyMjIwOrVq2FkZCTOy8vLg5mZGQwMDMSuFmpqanB0dERaWhqAlnsg9u3bB7lcjrKyMty6dQtFRUUoLi5W6p4BACNGjICaWvu95UtKSnDv3j0EBAQoTR86dCjGjRuHCxcudNNWP6W45yQvLw8SiQROTk5K9XZxcUFaWhqKiopgZmaGXbt2AQAqKytRWlqK4uJiZGVlAWgZSKMjb7zxhtLr119/HXV1de2Wb25uVho6XyKRiN2c2hMSEoJ33nkH5eXl2Llzpzi9uLgY5eXl8PX1Vdo+a2tr6Orq4vz585g0aRLmz58PAKivr8ft27dRUlIidp97fvs665p28eJFqKurY9q0aUrT3333XURERODChQswMTHpcBltbd/zA09paWl1+B5tbW0MHTpUfG1paQktLS14eHhg2rRpcHJywvjx4yGVSrtUF1W1d19TeHg4AKCmpgalpaUoKSnBmTNnALTs68bGRhQWFsLPz09pGW5ubtDQaH0aHzdunNJrQ0NDseubKsfC09MTJiYmCAoKgkwmg6OjI+zt7ZXuEXyWvr5+l8ozxjrGecx5/DLlcUfs7e1hb2+PJ0+eoKSkBLdu3cL169dRU1ODfv36AeAsfhZncffhRi5T4uDggIyMDDg5OSlNv3//PkpLS9sczRZoGb2ud+/eSEhIQGxsLGpra6Gvrw8LCwv07t0bDx8+VCqvr6/fYT0UAx60VU5fX7/bh9yvqKjAyJEjxXUTEaysrNosW1lZCTMzM5w7dw4bNmxAcXExdHR0YGpqCh0dHQCdP8utT58+Sq/V1NQ6fI+3tzcuXrwovraxscGePXs6XMfAgQMxatQoAICBgYE4XbFvQ0JCEBIS0ub2AS0n+TVr1uD06dOQSCQYNmyY+GXg+boOGDCgw7o8ePAA/fv3bxUCitEPn/98qMLY2Bhjxozp0nsGDBigFExDhgxBcnIyduzYgUOHDiExMRGvvPIKPD09sXz58g6/+HWF4rEF7Q0OceXKFYSEhODKlSvQ1taGiYkJBg8eDKBlX9+/fx/Nzc2t9rOGhgb69+/fanna2tpKr5/9fKlyLCQSCeLj47Ft2zacOnUKR44cQa9evTB58mQEBweLX0wUulqeMdY5zmPO45cljzsil8uxefNm7N27F/X19Rg0aBCkUqnSRWnO4qc4i7sPN3KZSvr27QsbG5tWw+AraGpq4vjx4wgLC0NgYCA8PDzEkW2XL18uXnFUleI/YlVVVat59+7da/Nk8qJu3ryJyspKeHp6AmjZ1j59+iApKanN8sOGDcPt27fh5+cHV1dXxMbGir8O7t27F+fOneu2uimEhITg0aNH4mtFeL+IV155BQCwcuVK2NjYtJqvp6cHoOWZjTdv3kRCQgKsrKygqamJx48fIyUlpcvr1NPTQ21tLZqampRO6IoA787j2VVSqRRRUVFobGxEQUEBDh48iO3bt8PU1LTVFdYXJZPJIJFIMH78+Fbz/vzzT8yfPx+mpqZIT08Xf1XJyclBZmYmgJYvLb169Wo1GIpcLlcaREYVqh4LAwMDBAcHY82aNbh27RpOnDiBuLg46OnptfllrKvlGWMvhvP4Kc7j/34e79ixA4mJiQgODoabm5v4mD8PDw+lcpzFLTiLuw+PrsxUYmNjg5KSEvHXM8VfWloaUlJSoK6ujoKCAvTt2xcLFy4UA/XRo0coKChQGilOFcbGxnjttddw/PhxpellZWW4dOlSu1d1X0RkZCS0tbUxc+ZMAC3bWl9fDyJS2taioiJER0ejqakJv/76KxoaGuDr66vU/VURqIordd119XH48OFKdRk+fPjfWtaAAQNw584dpWUaGhoiPDxcvCpfUFAANzc32Nraio+sOHv2LAB0+Xja2Nigubm51YiYiq51z3cX+/+SmJgIFxcXNDY2QlNTE2+++SbWrVsHAOKooX/3GJaXlyMlJQWTJk3CoEGDWs0vLi7G/fv3MWfOHIwcOVJc37P7Wl1dHVZWVjh9+rTSe8+cOdOq62FnVDkWhYWFsLOzw+XLlyGRSGBmZgZ/f38IgoDy8vJWy+xqecbYi+M85jwGXp48LigogImJCTw8PMQGbkVFBX7//Xex7pzFT3EWdx/+JZepxNvbG8eOHYO3tzd8fHzQv39/ZGRk4NChQ2Jff6lUiv379yMsLAzOzs6orKzErl27UFVVJV6NVJWamhoCAgLw1Vdfwd/fH++//z5qa2sRFRUFPT09fPrpp13ehvLycly6dAkA0NTUhIqKChw5cgS5ublYu3at2H3FyclJHL5+8eLFGDFiBC5fvoytW7fC3t4er776KiwsLKChoYFvv/0WPj4+aGxsxOHDh5GdnQ0A4j0Xiqu06enpsLS0FB9A/m9SV1eHv78/Vq9eDXV1dTg7O6Ourg4xMTGoqKgQu8BJpVIcP34cFhYWMDQ0RGFhIWJjYyGRSPD48eMurdPR0RETJkzAmjVrUFlZCXNzc1y8eBFxcXGYOXNmt97/0xW2trbYtGkT/Pz84OXlBXV1dRw4cACamppwdnYG0HIMq6qqkJOTAzMzMwwcOLDd5V29elX8tePx48e4fv06EhMToaWl1e6jOYyNjaGrq4vt27dDQ0MDGhoayMzMFB+hodjXy5Ytw+zZs7Fs2TJ4eHjgjz/+QEREBID27y9qiyrHoqGhAdra2li5ciWWLl0KfX19yGQyXL16FXPmzGm1THNz8y6VZ4y9OM5jzuP/Wh6npqa2+bny9vaGVCpFTEwMduzYgbFjx6K0tBSxsbFobGwU685ZzFn8T+BGLlOJgYEBDhw4gPDwcAQHB6OhoQFGRkYIDQ0Vu5zMnDkTd+7cQWpqKvbt2wcDAwM4OTnB09MTQUFBuHHjRpdOnu7u7tDR0UFsbCz8/Pygq6sLBwcHBAQEiPcsdMV3330nnqx69eqFgQMHYvTo0UhOTlbquqKmpoYdO3YgIiICsbGxqK6uhoGBAby9veHn5wegpYtUeHg4oqKisGjRIujp6WHs2LHYs2cPZs+ejfz8fJiammLq1Kk4duwYvvzyS3h4eCA4OLjL9f4nzJo1Czo6Oti5cycOHjyIPn36wMrKCps2bRKDPywsDOvWrROvphoZGSEkJARpaWnIz8/v0vokEgliY2MRGRmJpKQk1NTUYMiQIfD393+hL0jdZdSoUdi+fTuio6MREBCA5uZmjB49GvHx8eLVeXd3d+Tk5MDPzw/Lli3DwoUL213ekiVLxH/r6upi0KBBeO+99zB79ux273vr27cvYmJisHHjRixfvhw6OjowMzNDcnIyFixYgPz8fLi4uGD8+PHYunUrIiIisHjxYgwePBhBQUHw9/fvUnc5VY6FlpYW4uPjER4ejtDQUNTV1cHIyAhr166Fu7t7q2V2tTxj7MVxHnMe/9fyOCYmps3p3t7e8PX1RW1tLZKSkhAdHS3moqIeDx484CzmLP5HSKizO/IZY4z963788UcYGhoqDTZTVFSE6dOnIyYmBq6urv9i7RhjjLGej7P45cG/5DLG2EsgNzcXGRkZWLFiBYyNjVFeXo5t27Zh+PDhsLe3/7erxxhjjPV4nMUvD/4llzHGXgJ//fUXIiIikJmZicrKSvTr1w8ODg4IDAzs9BEgjDHGGPv7OItfHtzIZYwxxhhjjDHWY/AjhBhjjDHGGGOM9RjcyGWMMcYYY4wx1mNwI5cxxhhjjDHGWI/BjVzGGGOMMcYYYz0GN3IZY4wxxhhjjPUY3MhljDHGGGOMMdZjcCOXMcYYY4wxxliPwY1cxhhjjDHGGGM9BjdyGWOMMcYYY4z1GP8Hld2zz19CC9oAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1000x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#query to calculate difference between date of death (dod) and the date of first diagnosis and dod and the date of last diagnosis\n",
    "query = '''--sql\n",
    "            SELECT \n",
    "                STRFTIME(\"%Y\", p.dod) - STRFTIME(\"%Y\",MIN(d.diagnosis_date)) as dod_firstdiagnosis, \n",
    "                STRFTIME(\"%Y\", p.dod) - STRFTIME(\"%Y\",MAX(d.diagnosis_date)) as dod_lastdiagnosis\n",
    "            \n",
    "            ---LEFT JOIN Patient and Diagnosis table on key patient_id---\n",
    "            FROM \n",
    "                Patient as p \n",
    "            LEFT JOIN \n",
    "                Diagnosis as d \n",
    "                    ON p.patient_id = d.patient_id \n",
    "                    \n",
    "            ---BC patients have diagnosis code starting with C50---\n",
    "            WHERE \n",
    "                d.diagnosis_code LIKE \"C50%\"\n",
    "            \n",
    "            GROUP BY \n",
    "                p.patient_id\n",
    "        \n",
    "            ;\n",
    "        '''\n",
    "\n",
    "# generating a dataframe using sqlquery\n",
    "df_dod_diagnosis = pd.read_sql_query(query, con)\n",
    "\n",
    "# plot distribution of (dod - date of firstdiagnosis) and (dod - date of lastdiagnosis)\n",
    "fig, ax = plt.subplots(1,2)\n",
    "kde_dod_firstdiagnosis = sns.kdeplot(df_dod_diagnosis, \n",
    "                                     x = \"dod_firstdiagnosis\",\n",
    "                                     color = \"b\", \n",
    "                                     ax = ax[0])\n",
    "\n",
    "kde_dod_lastdiagnosis = sns.kdeplot(df_dod_diagnosis, \n",
    "                                    x = \"dod_lastdiagnosis\",\n",
    "                                    color = \"b\", \n",
    "                                    ax = ax[1])\n",
    "\n",
    "\n",
    "# shade the area under kdeplot where difference (dod - date of first diagnosis) and (dod - date of last diagnosis) is negative\n",
    "l1 = kde_dod_firstdiagnosis.lines[0]\n",
    "l2 = kde_dod_lastdiagnosis.lines[0]\n",
    "# get x and y coordinates of kernel-density function\n",
    "x1, y1 = l1.get_xydata()[:,0], l1.get_xydata()[:,1]\n",
    "x2, y2 = l2.get_xydata()[:,0], l2.get_xydata()[:,1]\n",
    "\n",
    "# fill area using x and y coordinates\n",
    "kde_dod_firstdiagnosis.fill_between(x1, y1, \n",
    "                                    where = x1<=0, #where x-axis is negative\n",
    "                                    color = \"b\")\n",
    "kde_dod_lastdiagnosis.fill_between(x2, y2, \n",
    "                                   where = x2<=0, #where x-axis is negative\n",
    "                                   color = \"b\")\n",
    "\n",
    "# label the shaded area which is the fraction of cases where difference is negative\n",
    "x_pos_first, y_pos_first = -45, 0.006\n",
    "x_pos_last, y_pos_last = -65, 0.006\n",
    "ax[0].text(x_pos_first, y_pos_first, f\"{100*(df_dod_diagnosis['dod_firstdiagnosis'] < 0).mean():0.1f}%\", color = \"b\")\n",
    "ax[1].text(x_pos_last, y_pos_last, f\"{100*(df_dod_diagnosis['dod_lastdiagnosis'] < 0).mean():0.1f}%\", color = \"b\")\n",
    "\n",
    "# provide axes labels set their limits\n",
    "ax[0].set_xlabel(\"Year of Death - Year of First Diagnosis\")\n",
    "ax[1].set_xlabel(\"Year of Death - Year of Last Diagnosis\")\n",
    "ax[0].set_xlim([-80, 80])\n",
    "ax[0].set_ylim([0, 0.02])\n",
    "ax[1].set_xlim([-80, 80])\n",
    "ax[1].set_ylim([0, 0.02])\n",
    "# remove the frames for both axes\n",
    "frame_off(ax[0])\n",
    "frame_off(ax[1])\n",
    "\n",
    "# increase padding between the subplots\n",
    "fig.tight_layout(pad = 2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**OBSERVATIONS**: <br>\n",
    "1. More than 50% of patients have date of death sooner than date of last diagnosis\n",
    "2. More than 20% of patients have date of death sooner than date of first diagnosis "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "metadata": {},
   "outputs": [],
   "source": [
    "#c close the connection to the sqlite database\n",
    "con.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Approach: <br>\n",
    "----\n",
    "\n",
    "**Assumptions**: <br> <br>\n",
    "Assume that diagnosis_dates and biomarker_test_dates are more reliable than date of death column. \n",
    "\n",
    "Patients where date of death < (date of diagnosis, biomarker test dates); set the date of death to NA <br>\n",
    "<br>\n",
    "Patient where date of death > (date of diagnosis, biomarker test dates); accept date of death <br> <br>\n",
    "\n",
    "**Definitions**: <br> <br>\n",
    "\n",
    "_date_first_diagnosis_ = MIN(diagnosis_date (only BC)) <br>\n",
    "\n",
    "_Follow-up time_ =  date_first_diagnosis - Max(diagnosis_date (all), biomarker_test_date (all), date of death) <br>\n",
    "\n",
    "_Age at diagnosis_ = date_first_diagnosis - DOB\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data maturity \n",
    "1. Calculate the summary statistics and plot the distribution of follow up time for BC patients.\n",
    "2. Calculate the summary statistics and plot the distribution of age at initial diagnosis for BC patients"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Size of the table: 941 x 6\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>patient_id</th>\n",
       "      <th>dod</th>\n",
       "      <th>age_at_diagnosis</th>\n",
       "      <th>first_diagnosis_date</th>\n",
       "      <th>last_diagnosis_date</th>\n",
       "      <th>last_test_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0004477e-31f7-4704-a123-2b68dd64aaed</td>\n",
       "      <td>2008-03-20</td>\n",
       "      <td>59</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2011-12-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0022fa14-8260-4b11-88cd-3fa066ede851</td>\n",
       "      <td>2008-03-04</td>\n",
       "      <td>50</td>\n",
       "      <td>1993-04-07</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>2004-12-19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0183e236-7892-4aa9-8b7f-10f12676f1ec</td>\n",
       "      <td>2038-03-11</td>\n",
       "      <td>21</td>\n",
       "      <td>1986-06-02</td>\n",
       "      <td>2021-10-13</td>\n",
       "      <td>2023-12-29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>019d3689-e49a-4f23-8ca1-65379907230b</td>\n",
       "      <td>2042-06-15</td>\n",
       "      <td>30</td>\n",
       "      <td>1996-10-05</td>\n",
       "      <td>2020-06-13</td>\n",
       "      <td>2018-03-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>01d468fc-55b8-45e3-8eca-dcb7a6185f28</td>\n",
       "      <td>2007-07-30</td>\n",
       "      <td>26</td>\n",
       "      <td>1977-01-16</td>\n",
       "      <td>1981-10-28</td>\n",
       "      <td>1995-12-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>022ff2bf-19d7-4c4d-8410-a91773ad24f2</td>\n",
       "      <td>2004-02-09</td>\n",
       "      <td>26</td>\n",
       "      <td>1956-06-03</td>\n",
       "      <td>2005-01-07</td>\n",
       "      <td>2021-07-29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>023f0abd-4675-4b59-96e9-5b5ca052560a</td>\n",
       "      <td>2027-08-10</td>\n",
       "      <td>25</td>\n",
       "      <td>1991-07-11</td>\n",
       "      <td>2009-04-15</td>\n",
       "      <td>2014-06-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>02c9feee-9e65-499f-bb93-e8aa1aba0160</td>\n",
       "      <td>2013-10-16</td>\n",
       "      <td>27</td>\n",
       "      <td>1993-08-01</td>\n",
       "      <td>2020-12-05</td>\n",
       "      <td>2023-07-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>030f8523-82f4-44d0-bb76-e0cd71237b88</td>\n",
       "      <td>1967-09-18</td>\n",
       "      <td>45</td>\n",
       "      <td>1980-02-07</td>\n",
       "      <td>2003-09-09</td>\n",
       "      <td>1954-10-24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0337f270-8ffe-432f-99e4-64060cd2a360</td>\n",
       "      <td>2032-05-22</td>\n",
       "      <td>41</td>\n",
       "      <td>2007-09-02</td>\n",
       "      <td>2018-12-29</td>\n",
       "      <td>2017-06-11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             patient_id         dod  age_at_diagnosis  \\\n",
       "0  0004477e-31f7-4704-a123-2b68dd64aaed  2008-03-20                59   \n",
       "1  0022fa14-8260-4b11-88cd-3fa066ede851  2008-03-04                50   \n",
       "2  0183e236-7892-4aa9-8b7f-10f12676f1ec  2038-03-11                21   \n",
       "3  019d3689-e49a-4f23-8ca1-65379907230b  2042-06-15                30   \n",
       "4  01d468fc-55b8-45e3-8eca-dcb7a6185f28  2007-07-30                26   \n",
       "5  022ff2bf-19d7-4c4d-8410-a91773ad24f2  2004-02-09                26   \n",
       "6  023f0abd-4675-4b59-96e9-5b5ca052560a  2027-08-10                25   \n",
       "7  02c9feee-9e65-499f-bb93-e8aa1aba0160  2013-10-16                27   \n",
       "8  030f8523-82f4-44d0-bb76-e0cd71237b88  1967-09-18                45   \n",
       "9  0337f270-8ffe-432f-99e4-64060cd2a360  2032-05-22                41   \n",
       "\n",
       "  first_diagnosis_date last_diagnosis_date last_test_date  \n",
       "0           2019-11-04          2019-11-04     2011-12-03  \n",
       "1           1993-04-07          2019-05-20     2004-12-19  \n",
       "2           1986-06-02          2021-10-13     2023-12-29  \n",
       "3           1996-10-05          2020-06-13     2018-03-10  \n",
       "4           1977-01-16          1981-10-28     1995-12-08  \n",
       "5           1956-06-03          2005-01-07     2021-07-29  \n",
       "6           1991-07-11          2009-04-15     2014-06-08  \n",
       "7           1993-08-01          2020-12-05     2023-07-09  \n",
       "8           1980-02-07          2003-09-09     1954-10-24  \n",
       "9           2007-09-02          2018-12-29     2017-06-11  "
      ]
     },
     "execution_count": 223,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#connect to the sqlite database\n",
    "con = sqlite3.connect(\"techTest.db\")\n",
    "cur = con.cursor()\n",
    "\n",
    "query_patient = '''--sql\n",
    "                SELECT \n",
    "                    p.patient_id as patient_id, \n",
    "                    date(p.dod) as dod,  \n",
    "                    STRFTIME(\"%Y\", MIN(d.diagnosis_date)) - STRFTIME(\"%Y\",p.dob) as age_at_diagnosis, \n",
    "                    date(MIN(d.diagnosis_date)) as first_diagnosis_date, \n",
    "                    sub_dates.last_diagnosis_date as last_diagnosis_date, \n",
    "                    sub_dates.last_test_date as last_test_date\n",
    "\n",
    "                ---Patient joined to Diagnosis---\n",
    "                FROM \n",
    "                    patient as p \n",
    "                LEFT JOIN \n",
    "                    diagnosis as d\n",
    "                        ON p.patient_id = d.patient_id\n",
    "\n",
    "                ---Left JOIN to sub-query that calculates last_diagnosis_date and last_test_date---\n",
    "                ---last_diagnosis_date and last_test_date are calculated without considered BC status---\n",
    "                ---This is because last encounter in the database should be disease independent---                        \n",
    "                LEFT JOIN\n",
    "                    (SELECT \n",
    "                        sub_p.patient_id, \n",
    "                        date(MAX(sub_d.diagnosis_date)) as last_diagnosis_date,\n",
    "                        date(MAX(sub_b.test_date)) as last_test_date\n",
    "\n",
    "                    FROM\n",
    "                        patient as sub_p\n",
    "                    LEFT JOIN\n",
    "                        diagnosis as sub_d \n",
    "                            ON sub_p.patient_id = sub_d.patient_id\n",
    "                    LEFT JOIN\n",
    "                        biomarker as sub_b\n",
    "                            ON sub_p.patient_id = sub_b.patient_id\n",
    "                    \n",
    "                    GROUP BY\n",
    "                        sub_p.patient_id\n",
    "                    ) as sub_dates\n",
    "                        ON p.patient_id = sub_dates.patient_id \n",
    "\n",
    "                WHERE  \n",
    "                    d.diagnosis_code LIKE \"C50%\"\n",
    "                        \n",
    "                GROUP BY \n",
    "                    p.patient_id;\n",
    "\n",
    "                '''\n",
    "\n",
    "df = pd.read_sql_query(query_patient, con)\n",
    "\n",
    "print(f\"Size of the table: {df.shape[0]} x {df.shape[1]}\")\n",
    "df.head(n = 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*What about patients with a test date that is lower than the first diagnosis date?*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of patients with the last test date sooner than the first diagnosis date: 76\n"
     ]
    }
   ],
   "source": [
    "# records where last_test_date is sooner than first_diagnosis_date\n",
    "df_diagnostic_biomarker = df[df[\"last_test_date\"] < df[\"first_diagnosis_date\"]]\n",
    "\n",
    "# biomarker table\n",
    "df_biomarker = pd.read_sql(\"SELECT * FROM biomarker\", con)\n",
    "\n",
    "\n",
    "condition = df_biomarker[\"patient_id\"].isin(df_diagnostic_biomarker[\"patient_id\"])\n",
    "positive_biomarker = (df_biomarker.loc[condition, \"test_result\"] == \"Positive\").sum()\n",
    "print(f\"Number of patients with the last test date sooner than the first diagnosis date: {positive_biomarker}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Definition of first diagnosis is the first encounter in the diagnosis table. <br>\n",
    "Biomarkers may be used to accompany biopsy testing and are not usually used as a diagnostic tool [1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 941 entries, 0 to 940\n",
      "Data columns (total 6 columns):\n",
      " #   Column                Non-Null Count  Dtype \n",
      "---  ------                --------------  ----- \n",
      " 0   patient_id            941 non-null    object\n",
      " 1   dod                   941 non-null    object\n",
      " 2   age_at_diagnosis      941 non-null    int64 \n",
      " 3   first_diagnosis_date  941 non-null    object\n",
      " 4   last_diagnosis_date   941 non-null    object\n",
      " 5   last_test_date        941 non-null    object\n",
      "dtypes: int64(1), object(5)\n",
      "memory usage: 44.2+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*DOD that is later than today is set to today's date*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 226,
   "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>patient_id</th>\n",
       "      <th>dod</th>\n",
       "      <th>age_at_diagnosis</th>\n",
       "      <th>first_diagnosis_date</th>\n",
       "      <th>last_diagnosis_date</th>\n",
       "      <th>last_test_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0004477e-31f7-4704-a123-2b68dd64aaed</td>\n",
       "      <td>2008-03-20</td>\n",
       "      <td>59</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2011-12-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0022fa14-8260-4b11-88cd-3fa066ede851</td>\n",
       "      <td>2008-03-04</td>\n",
       "      <td>50</td>\n",
       "      <td>1993-04-07</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>2004-12-19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0183e236-7892-4aa9-8b7f-10f12676f1ec</td>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>21</td>\n",
       "      <td>1986-06-02</td>\n",
       "      <td>2021-10-13</td>\n",
       "      <td>2023-12-29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>019d3689-e49a-4f23-8ca1-65379907230b</td>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>30</td>\n",
       "      <td>1996-10-05</td>\n",
       "      <td>2020-06-13</td>\n",
       "      <td>2018-03-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>01d468fc-55b8-45e3-8eca-dcb7a6185f28</td>\n",
       "      <td>2007-07-30</td>\n",
       "      <td>26</td>\n",
       "      <td>1977-01-16</td>\n",
       "      <td>1981-10-28</td>\n",
       "      <td>1995-12-08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             patient_id        dod  age_at_diagnosis  \\\n",
       "0  0004477e-31f7-4704-a123-2b68dd64aaed 2008-03-20                59   \n",
       "1  0022fa14-8260-4b11-88cd-3fa066ede851 2008-03-04                50   \n",
       "2  0183e236-7892-4aa9-8b7f-10f12676f1ec 2024-07-01                21   \n",
       "3  019d3689-e49a-4f23-8ca1-65379907230b 2024-07-01                30   \n",
       "4  01d468fc-55b8-45e3-8eca-dcb7a6185f28 2007-07-30                26   \n",
       "\n",
       "  first_diagnosis_date last_diagnosis_date last_test_date  \n",
       "0           2019-11-04          2019-11-04     2011-12-03  \n",
       "1           1993-04-07          2019-05-20     2004-12-19  \n",
       "2           1986-06-02          2021-10-13     2023-12-29  \n",
       "3           1996-10-05          2020-06-13     2018-03-10  \n",
       "4           1977-01-16          1981-10-28     1995-12-08  "
      ]
     },
     "execution_count": 226,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dod_mod = df.copy()\n",
    "\n",
    "# Converting columns with date-time to pandas date-time\n",
    "time_format_cols = [\"first_diagnosis_date\", \"last_diagnosis_date\", \"last_test_date\", \"dod\"]\n",
    "for col in time_format_cols:\n",
    "    df_dod_mod[col] = pd.to_datetime(df_dod_mod[col])\n",
    "\n",
    "today = datetime(2024, 7, 1)\n",
    "df_dod_mod.loc[df_dod_mod[\"dod\"] > today, \"dod\"] = today\n",
    "df_dod_mod.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*DOD that is sooner than either diagnosis dates of test_dates is set to NA*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "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>patient_id</th>\n",
       "      <th>dod</th>\n",
       "      <th>age_at_diagnosis</th>\n",
       "      <th>first_diagnosis_date</th>\n",
       "      <th>last_diagnosis_date</th>\n",
       "      <th>last_test_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0004477e-31f7-4704-a123-2b68dd64aaed</td>\n",
       "      <td>NaT</td>\n",
       "      <td>59</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2011-12-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0022fa14-8260-4b11-88cd-3fa066ede851</td>\n",
       "      <td>NaT</td>\n",
       "      <td>50</td>\n",
       "      <td>1993-04-07</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>2004-12-19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0183e236-7892-4aa9-8b7f-10f12676f1ec</td>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>21</td>\n",
       "      <td>1986-06-02</td>\n",
       "      <td>2021-10-13</td>\n",
       "      <td>2023-12-29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>019d3689-e49a-4f23-8ca1-65379907230b</td>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>30</td>\n",
       "      <td>1996-10-05</td>\n",
       "      <td>2020-06-13</td>\n",
       "      <td>2018-03-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>01d468fc-55b8-45e3-8eca-dcb7a6185f28</td>\n",
       "      <td>2007-07-30</td>\n",
       "      <td>26</td>\n",
       "      <td>1977-01-16</td>\n",
       "      <td>1981-10-28</td>\n",
       "      <td>1995-12-08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             patient_id        dod  age_at_diagnosis  \\\n",
       "0  0004477e-31f7-4704-a123-2b68dd64aaed        NaT                59   \n",
       "1  0022fa14-8260-4b11-88cd-3fa066ede851        NaT                50   \n",
       "2  0183e236-7892-4aa9-8b7f-10f12676f1ec 2024-07-01                21   \n",
       "3  019d3689-e49a-4f23-8ca1-65379907230b 2024-07-01                30   \n",
       "4  01d468fc-55b8-45e3-8eca-dcb7a6185f28 2007-07-30                26   \n",
       "\n",
       "  first_diagnosis_date last_diagnosis_date last_test_date  \n",
       "0           2019-11-04          2019-11-04     2011-12-03  \n",
       "1           1993-04-07          2019-05-20     2004-12-19  \n",
       "2           1986-06-02          2021-10-13     2023-12-29  \n",
       "3           1996-10-05          2020-06-13     2018-03-10  \n",
       "4           1977-01-16          1981-10-28     1995-12-08  "
      ]
     },
     "execution_count": 227,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def replace_dod(row):\n",
    "    '''\n",
    "    function to replace dod with NA if dod < diagnosis dates or test dates\n",
    "    Input: row values\n",
    "    Return: modified row value\n",
    "    '''\n",
    "    if (row[\"dod\"] < row[\"first_diagnosis_date\"]) | (row[\"dod\"] < row[\"last_diagnosis_date\"]) | (row[\"dod\"] <  row[\"last_test_date\"]):\n",
    "        row[\"dod\"] = np.nan\n",
    "    \n",
    "    return row[\"dod\"]\n",
    "\n",
    "df_dod_mod[\"dod\"] = df_dod_mod.apply(replace_dod, axis = 1)\n",
    "df_dod_mod.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**OBSERVATION**: <br>\n",
    "NaT is \"Not a Time\". It is the equivalent of NA for date, time context in pandas\n",
    "\n",
    "**Next Step**: <br>\n",
    "Calculate the last_encounter_date"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*last_encounter_date = MAX(dad, last_diagnosis_date, last_test_date)*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "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>patient_id</th>\n",
       "      <th>dod</th>\n",
       "      <th>age_at_diagnosis</th>\n",
       "      <th>first_diagnosis_date</th>\n",
       "      <th>last_diagnosis_date</th>\n",
       "      <th>last_test_date</th>\n",
       "      <th>last_encounter_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0004477e-31f7-4704-a123-2b68dd64aaed</td>\n",
       "      <td>NaT</td>\n",
       "      <td>59</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2011-12-03</td>\n",
       "      <td>2019-11-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0022fa14-8260-4b11-88cd-3fa066ede851</td>\n",
       "      <td>NaT</td>\n",
       "      <td>50</td>\n",
       "      <td>1993-04-07</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>2004-12-19</td>\n",
       "      <td>2019-05-20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0183e236-7892-4aa9-8b7f-10f12676f1ec</td>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>21</td>\n",
       "      <td>1986-06-02</td>\n",
       "      <td>2021-10-13</td>\n",
       "      <td>2023-12-29</td>\n",
       "      <td>2024-07-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>019d3689-e49a-4f23-8ca1-65379907230b</td>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>30</td>\n",
       "      <td>1996-10-05</td>\n",
       "      <td>2020-06-13</td>\n",
       "      <td>2018-03-10</td>\n",
       "      <td>2024-07-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>01d468fc-55b8-45e3-8eca-dcb7a6185f28</td>\n",
       "      <td>2007-07-30</td>\n",
       "      <td>26</td>\n",
       "      <td>1977-01-16</td>\n",
       "      <td>1981-10-28</td>\n",
       "      <td>1995-12-08</td>\n",
       "      <td>2007-07-30</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             patient_id        dod  age_at_diagnosis  \\\n",
       "0  0004477e-31f7-4704-a123-2b68dd64aaed        NaT                59   \n",
       "1  0022fa14-8260-4b11-88cd-3fa066ede851        NaT                50   \n",
       "2  0183e236-7892-4aa9-8b7f-10f12676f1ec 2024-07-01                21   \n",
       "3  019d3689-e49a-4f23-8ca1-65379907230b 2024-07-01                30   \n",
       "4  01d468fc-55b8-45e3-8eca-dcb7a6185f28 2007-07-30                26   \n",
       "\n",
       "  first_diagnosis_date last_diagnosis_date last_test_date last_encounter_date  \n",
       "0           2019-11-04          2019-11-04     2011-12-03          2019-11-04  \n",
       "1           1993-04-07          2019-05-20     2004-12-19          2019-05-20  \n",
       "2           1986-06-02          2021-10-13     2023-12-29          2024-07-01  \n",
       "3           1996-10-05          2020-06-13     2018-03-10          2024-07-01  \n",
       "4           1977-01-16          1981-10-28     1995-12-08          2007-07-30  "
      ]
     },
     "execution_count": 228,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dod_mod[\"last_encounter_date\"] = np.max(df_dod_mod[[\"last_diagnosis_date\", \"last_test_date\", \"dod\"]], axis = 1)\n",
    "\n",
    "df_dod_mod.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*follow_up_time = (last_encounter_date - first-diagnosis date) in years*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "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>patient_id</th>\n",
       "      <th>dod</th>\n",
       "      <th>age_at_diagnosis</th>\n",
       "      <th>first_diagnosis_date</th>\n",
       "      <th>last_diagnosis_date</th>\n",
       "      <th>last_test_date</th>\n",
       "      <th>last_encounter_date</th>\n",
       "      <th>follow_up_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0004477e-31f7-4704-a123-2b68dd64aaed</td>\n",
       "      <td>NaT</td>\n",
       "      <td>59</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2011-12-03</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0022fa14-8260-4b11-88cd-3fa066ede851</td>\n",
       "      <td>NaT</td>\n",
       "      <td>50</td>\n",
       "      <td>1993-04-07</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>2004-12-19</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>26.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0183e236-7892-4aa9-8b7f-10f12676f1ec</td>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>21</td>\n",
       "      <td>1986-06-02</td>\n",
       "      <td>2021-10-13</td>\n",
       "      <td>2023-12-29</td>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>38.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>019d3689-e49a-4f23-8ca1-65379907230b</td>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>30</td>\n",
       "      <td>1996-10-05</td>\n",
       "      <td>2020-06-13</td>\n",
       "      <td>2018-03-10</td>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>27.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>01d468fc-55b8-45e3-8eca-dcb7a6185f28</td>\n",
       "      <td>2007-07-30</td>\n",
       "      <td>26</td>\n",
       "      <td>1977-01-16</td>\n",
       "      <td>1981-10-28</td>\n",
       "      <td>1995-12-08</td>\n",
       "      <td>2007-07-30</td>\n",
       "      <td>30.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             patient_id        dod  age_at_diagnosis  \\\n",
       "0  0004477e-31f7-4704-a123-2b68dd64aaed        NaT                59   \n",
       "1  0022fa14-8260-4b11-88cd-3fa066ede851        NaT                50   \n",
       "2  0183e236-7892-4aa9-8b7f-10f12676f1ec 2024-07-01                21   \n",
       "3  019d3689-e49a-4f23-8ca1-65379907230b 2024-07-01                30   \n",
       "4  01d468fc-55b8-45e3-8eca-dcb7a6185f28 2007-07-30                26   \n",
       "\n",
       "  first_diagnosis_date last_diagnosis_date last_test_date last_encounter_date  \\\n",
       "0           2019-11-04          2019-11-04     2011-12-03          2019-11-04   \n",
       "1           1993-04-07          2019-05-20     2004-12-19          2019-05-20   \n",
       "2           1986-06-02          2021-10-13     2023-12-29          2024-07-01   \n",
       "3           1996-10-05          2020-06-13     2018-03-10          2024-07-01   \n",
       "4           1977-01-16          1981-10-28     1995-12-08          2007-07-30   \n",
       "\n",
       "   follow_up_time  \n",
       "0             0.0  \n",
       "1            26.1  \n",
       "2            38.1  \n",
       "3            27.7  \n",
       "4            30.5  "
      ]
     },
     "execution_count": 229,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calculate follow-up_time in days, convert to years by dividing it by 365.25, and round the output to have 1 decimal place\n",
    "df_dod_mod[\"follow_up_time\"] = ((df_dod_mod[\"last_encounter_date\"] - df_dod_mod[\"first_diagnosis_date\"]).dt.days/365.25).round(1)\n",
    "\n",
    "df_dod_mod.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**OBSERVATION**: <br>\n",
    "Some patients have a follow-up time of 0 since they had only one encounter in the database <br>\n",
    "These are patients who have only 1 encounter in the diagnosis table and no later encounter in the biomarker table\n",
    "\n",
    "**NEXT STEPS**: <br>\n",
    "1. Sanity check for number of patients with DOD lower than first diagnosis <br>\n",
    "2. Sanity check for number of patients with DOD lower than last_diagnosis and test_dates <br>\n",
    "3. Calculate the proportion of patients where follow-up time is 0\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of patients with DOD < first diagnosis: 0\n",
      "Number of patients with DOD < last_diagnosis and test_dates: 0\n"
     ]
    }
   ],
   "source": [
    "print(f\"Number of patients with DOD < first diagnosis: {(df_dod_mod['dod'] < df_dod_mod['first_diagnosis_date']).sum()}\")\n",
    "\n",
    "print(f\"\"\"Number of patients with DOD < last_diagnosis and test_dates: {((df_dod_mod['dod'] < df_dod_mod['last_diagnosis_date'])\n",
    "                                                                       |(df_dod_mod['dod'] < df_dod_mod['last_test_date']))\n",
    "                                                                       .sum()}\"\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "patient_id                0\n",
       "dod                     656\n",
       "age_at_diagnosis          0\n",
       "first_diagnosis_date      0\n",
       "last_diagnosis_date       0\n",
       "last_test_date            0\n",
       "last_encounter_date       0\n",
       "follow_up_time            0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 231,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate number of empty row-values in each column\n",
    "df_dod_mod.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fraction of patients with 0 follow_up time:  5.53%\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>patient_id</th>\n",
       "      <th>dod</th>\n",
       "      <th>age_at_diagnosis</th>\n",
       "      <th>first_diagnosis_date</th>\n",
       "      <th>last_diagnosis_date</th>\n",
       "      <th>last_test_date</th>\n",
       "      <th>last_encounter_date</th>\n",
       "      <th>follow_up_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0004477e-31f7-4704-a123-2b68dd64aaed</td>\n",
       "      <td>NaT</td>\n",
       "      <td>59</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2011-12-03</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>0b6096c4-9ffd-4a69-8257-3f94cf82fa45</td>\n",
       "      <td>NaT</td>\n",
       "      <td>56</td>\n",
       "      <td>1987-05-20</td>\n",
       "      <td>1987-05-20</td>\n",
       "      <td>1974-11-22</td>\n",
       "      <td>1987-05-20</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>71</th>\n",
       "      <td>15591b85-3176-4982-8a31-68684b140b31</td>\n",
       "      <td>NaT</td>\n",
       "      <td>76</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>1966-01-15</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>188c9c32-5f88-4936-952f-358ad36d9cd6</td>\n",
       "      <td>NaT</td>\n",
       "      <td>70</td>\n",
       "      <td>2019-05-09</td>\n",
       "      <td>2019-05-09</td>\n",
       "      <td>2000-04-02</td>\n",
       "      <td>2019-05-09</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>114</th>\n",
       "      <td>2076ea86-235b-4639-9774-3c5c45d78942</td>\n",
       "      <td>NaT</td>\n",
       "      <td>61</td>\n",
       "      <td>1994-09-05</td>\n",
       "      <td>1994-09-05</td>\n",
       "      <td>1993-09-05</td>\n",
       "      <td>1994-09-05</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               patient_id dod  age_at_diagnosis  \\\n",
       "0    0004477e-31f7-4704-a123-2b68dd64aaed NaT                59   \n",
       "36   0b6096c4-9ffd-4a69-8257-3f94cf82fa45 NaT                56   \n",
       "71   15591b85-3176-4982-8a31-68684b140b31 NaT                76   \n",
       "81   188c9c32-5f88-4936-952f-358ad36d9cd6 NaT                70   \n",
       "114  2076ea86-235b-4639-9774-3c5c45d78942 NaT                61   \n",
       "\n",
       "    first_diagnosis_date last_diagnosis_date last_test_date  \\\n",
       "0             2019-11-04          2019-11-04     2011-12-03   \n",
       "36            1987-05-20          1987-05-20     1974-11-22   \n",
       "71            2010-04-16          2010-04-16     1966-01-15   \n",
       "81            2019-05-09          2019-05-09     2000-04-02   \n",
       "114           1994-09-05          1994-09-05     1993-09-05   \n",
       "\n",
       "    last_encounter_date  follow_up_time  \n",
       "0            2019-11-04             0.0  \n",
       "36           1987-05-20             0.0  \n",
       "71           2010-04-16             0.0  \n",
       "81           2019-05-09             0.0  \n",
       "114          1994-09-05             0.0  "
      ]
     },
     "execution_count": 232,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"fraction of patients with 0 follow_up time: {100*(df_dod_mod['follow_up_time']==0).mean(): 0.2f}%\")\n",
    "df_dod_mod[df_dod_mod[\"follow_up_time\"]==0].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In subsequent analysis, since these patients are not at risk any time, we ignore patients who have their follow-up time = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_final = df_dod_mod[df_dod_mod[\"follow_up_time\"]!=0].copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Summary statistics for follow-up time and age_at_diagnosis*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 234,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plot the distribution of follow_up_time and age_at_diagnosis\n",
    "fig, axes = plt.subplots(1,2, figsize = (10,5))\n",
    "dataframe = df_final\n",
    "\n",
    "# kernel-density estimation plot for follow_up_time\n",
    "x_follow_up_max = dataframe[\"follow_up_time\"].max()\n",
    "x_follow_up_min = dataframe[\"follow_up_time\"].min()\n",
    "p1 = sns.kdeplot(\n",
    "    dataframe,\n",
    "    x = \"follow_up_time\",\n",
    "    fill = True,\n",
    "    color = \"c\",\n",
    "    clip = (x_follow_up_min,x_follow_up_max), # to remove artifacts of kernel-density estimation\n",
    "    ax = axes[0]\n",
    "\n",
    ")\n",
    "\n",
    "# Kernel-density estimation plot for age_at_diagnosis\n",
    "x_age_diagnosis_max = dataframe[\"age_at_diagnosis\"].max()\n",
    "x_age_diagnosis_min = dataframe[\"age_at_diagnosis\"].min()\n",
    "p2 = sns.kdeplot(\n",
    "    dataframe,\n",
    "    x = \"age_at_diagnosis\",\n",
    "    fill = True,\n",
    "    color = \"c\",\n",
    "    clip = (x_age_diagnosis_min,x_age_diagnosis_max), # to remove artifacts of kernel-density estimations\n",
    "    ax = axes[1]\n",
    ")\n",
    "\n",
    "# define x-axis labels\n",
    "axes[0].set_xlabel(\"Follow-up time (years)\")\n",
    "axes[1].set_xlabel(\"Age at diagnosis (years)\")\n",
    "axes[0].set_xlim([0, 75])\n",
    "axes[1].set_xlim([10, 85])\n",
    "\n",
    "def summary_stats(x, ax):\n",
    "    '''\n",
    "        function to plot mean, median, and 2.5% and  97.5% quantile as vertical lines\n",
    "        Input: ax = axes\n",
    "               x = column name\n",
    "        \n",
    "        Return: nothing. plots vertical lines \n",
    "    '''\n",
    "    # calculate summary statistics: mean, median, and interquantile range\n",
    "    mu_x, med_x = dataframe[x].mean(), dataframe[x].median()\n",
    "    alpha_x, oneminusalpha_x = dataframe[x].quantile(q = [0.025, 0.975])\n",
    "    \n",
    "    # calculate y-axis text label position\n",
    "    y_limit_plot = ax.get_ylim()\n",
    "    y_pos = np.mean(y_limit_plot)\n",
    "\n",
    "    # plot mean and its text label\n",
    "    ax.axvline(mu_x, lw = 1, color = \"dimgray\")\n",
    "    ax.text(mu_x*1.05, y_pos, \n",
    "            f\"$\\mu$ = {np.round(mu_x,0):0.0f}\", \n",
    "            horizontalalignment = \"left\", \n",
    "            rotation = \"vertical\", \n",
    "            color = \"dimgray\")\n",
    "    \n",
    "    # plot median and its label\n",
    "    ax.axvline(med_x, lw = 1, color = \"m\")\n",
    "    ax.text(med_x*0.99, y_pos, \n",
    "            f\"$q_{{50}}$ = {np.round(med_x,0):0.0f}\", \n",
    "            horizontalalignment = \"right\", \n",
    "            rotation = \"vertical\", \n",
    "            color = \"m\")\n",
    "    \n",
    "    # plot 2.5% quantile and its text label\n",
    "    ax.axvline(alpha_x, lw = 1, ls = '--', color = \"m\")\n",
    "    ax.text(alpha_x*0.99, y_pos, \n",
    "            f\"$q_{{2.5}}$ = {np.round(alpha_x,0):0.0f}\", \n",
    "            horizontalalignment = \"right\", \n",
    "            rotation = \"vertical\", \n",
    "            color = \"m\")\n",
    "    \n",
    "    # plot 97.5% quantile and its text label\n",
    "    ax.axvline(oneminusalpha_x, lw = 1, ls = \"--\", color = \"m\")\n",
    "    ax.text(oneminusalpha_x*1.02, y_pos, \n",
    "            f\"$q_{{97.5}}$ = {np.round(oneminusalpha_x,0):0.0f}\", \n",
    "            horizontalalignment = \"left\", \n",
    "            rotation = \"vertical\", \n",
    "            color = \"m\")\n",
    "\n",
    "# summary vertical lines for mean, median, and 2.5% and 97.5% quantiles\n",
    "summary_stats(\"follow_up_time\", axes[0])\n",
    "summary_stats(\"age_at_diagnosis\", axes[1])\n",
    "\n",
    "# increase padding between figures\n",
    "fig.tight_layout(pad = 2)\n",
    "\n",
    "# Remove the outerframe of each plot\n",
    "frame_off(axes[0])\n",
    "frame_off(axes[1])\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**OBSERVATION**: <br> \n",
    "1. Follow-up time follows an almost symmetric distribution (mean = median). This is after removing patients whose follow-up time is 0. When patients with follow-up time of 0 were included there was a smaller peak at x = 0.\n",
    "2. Age at diagnosis has a right-skewed distribution. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting distribution of age_at_diagnosis and follow_up_time\n",
    "sns.displot(df_dod_mod, \n",
    "            x = \"age_at_diagnosis\", \n",
    "            y = \"follow_up_time\", \n",
    "            kind = \"kde\", \n",
    "            color = \"gray\")\n",
    "\n",
    "# set axis labels\n",
    "plt.xlabel(\"Age at diagnosis (years)\")\n",
    "plt.ylabel(\"Follow-up time (years)\")\n",
    "\n",
    "plt.box(False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**OBSERVATIONS**: <br>\n",
    "- As age at diagnosis increases, the follow-up time decreases "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {},
   "outputs": [],
   "source": [
    "# close the connection to the database\n",
    "con.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Survival Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Exploration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Proportion of patients with a censored status: 67.9 %\n"
     ]
    }
   ],
   "source": [
    "# For Survival Analysis\n",
    "# Event has occured if, for a patient, date of deat (dod) is known\n",
    "# a patient is censored if the last encounter in the database is either a diagnostic or biomarker test and the data of death is NA\n",
    "df_final[\"status\"] = df_final[\"dod\"].notna().map({\n",
    "                                                    True: \"event\",\n",
    "                                                    False: \"censor\"\n",
    "                                                })\n",
    "print(f\"Proportion of patients with a censored status: {100*df_final['status'].value_counts(normalize = True)['censor']:0.1f} %\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#age (in years) at diagnosis to stratify the patients\n",
    "age_to_stratify = 60\n",
    "mask = df_final[\"age_at_diagnosis\"] > age_to_stratify\n",
    "df_final[\"stratified_age\"] = mask\n",
    "df_final[\"stratified_age\"] = df_final[\"stratified_age\"].map({\n",
    "                                                                True: \"above\",\n",
    "                                                                False: \"below\"\n",
    "                                                            })\n",
    "\n",
    "custom_order = [\"below\", \"above\"]\n",
    "\n",
    "fig, ax = plt.subplots(1,2, figsize = (10,5))\n",
    "k = sns.kdeplot(df_final, \n",
    "            x = \"follow_up_time\", \n",
    "            hue = \"stratified_age\", \n",
    "            hue_order = custom_order,\n",
    "            fill = True, \n",
    "            common_norm = False, \n",
    "            clip = (0,80), # we clip the extremities to prevent artifacts by kernet density estimation\n",
    "            alpha = 0.8,\n",
    "            linewidth = 0,\n",
    "            legend = False,\n",
    "            ax = ax[1])\n",
    "\n",
    "\n",
    "\n",
    "c = sns.countplot(df_final,\n",
    "            x = \"status\",\n",
    "            hue = \"stratified_age\",\n",
    "            hue_order = custom_order,\n",
    "            alpha = 0.9,\n",
    "            ax = ax[0])\n",
    "\n",
    "# label count values on top of each bar\n",
    "for container in ax[0].containers:\n",
    "    ax[0].bar_label(container)\n",
    "\n",
    "# fix axes parameters\n",
    "ax[0].legend_.remove()\n",
    "ax[1].set_ylim([0, 0.06])\n",
    "ax[0].set_ylim([0, 700])\n",
    "ax[0].set_ylabel(\"Count\")\n",
    "\n",
    "# remove outer frame\n",
    "frame_off(ax[1])\n",
    "frame_off(ax[0])\n",
    "\n",
    "# display a common legend\n",
    "fig.legend([f\"below {age_to_stratify}\", f\"above {age_to_stratify}\"], \n",
    "           fontsize = 12, \n",
    "           frameon = False,\n",
    "           loc = \"upper center\",\n",
    "           ncols = len(ax))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**OBSERVATION**: <br>\n",
    "- Only 6% of the patients were over 60 at the time of diagnosis\n",
    "- Hence, number of events (death) for patients who were over 60 at time of diagnosis is also skewed: ~40 times lower than for patients above 60\n",
    "- Follow-up times for patients who were over 60 at the time of diagnosis is much lower than the other group"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Kaplan Meier Plots\n",
    "\n",
    "Kaplan Meier (KM) plots present $S(t) = Pr(T>t)$, where $Pr(T>t)$ is the probability of a patient surviving for time greater than t <br>\n",
    "Here, I plot KM plots using Kaplan-Meier fitter from the lifelines package using follow-up times and status of each patient. <br><br>\n",
    "Status = 1, implying death, if a patient's date of death is the last encounter in the database <br>\n",
    "Status = 0, implying censored, if a patient's date of death is lower than the last encounter\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_final[\"status\"] = df_final[\"status\"].map({\n",
    "                                                \"event\": 1,\n",
    "                                                \"censor\": 0\n",
    "                                            })\n",
    "\n",
    "fig, axes = plt.subplots(figsize = (9,5))\n",
    "\n",
    "# Instantiating the Kaplan-Meier class\n",
    "km_lower = KaplanMeierFitter()\n",
    "km_upper = KaplanMeierFitter()\n",
    "\n",
    "# fitting Kaplan-Meier for patients below the stratify_age\n",
    "km_lowerage = km_lower.fit(df_final.loc[~mask,\"follow_up_time\"], df_final.loc[~mask,\"status\"])\n",
    "km_lowerage.plot(label = f\"below {age_to_stratify}\")\n",
    "\n",
    "# fitting Kaplan-Meier for patient above the stratify_age\n",
    "km_upperage = km_upper.fit(df_final.loc[mask,\"follow_up_time\"], df_final.loc[mask,\"status\"])\n",
    "km_upperage.plot(label = f\"above {age_to_stratify}\", \n",
    "                 xlabel = \"Survival Time (Years)\", \n",
    "                 ylabel = \"Estimated Probability of Survival\")\n",
    "\n",
    "# fixing figure properties\n",
    "plt.legend(fontsize = 12, frameon = False)\n",
    "plt.box(False)\n",
    "plt.title(f\"Kaplan Meier curve for patients stratified by age at time of diagnosis\")\n",
    "plt.ylim([0, 1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here, I calculate the median survival times from the output of Kaplan-Meier fitter. <br>\n",
    "Confidence interval around the median survival times were calculated using the median_survival_times method from lifelines.utils package"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Median survival time for patients over 60 years of age at time of diagnosis is inf (inf, inf)\n",
      "Median survival time for patients below 60 years of age at time of diagnosis is 50.5 (45.9, 52.5)\n"
     ]
    }
   ],
   "source": [
    "# Calculate median survival times and their confidence intervals (using median survival times package from lifelines)\n",
    "median_upper = km_upperage.median_survival_time_\n",
    "median_ci_upper = median_survival_times(km_upperage.confidence_interval_).to_numpy()[0]\n",
    "median_lower = km_lowerage.median_survival_time_\n",
    "median_ci_lower = median_survival_times(km_lowerage.confidence_interval_).to_numpy()[0]\n",
    "\n",
    "print(f\"Median survival time for patients over {age_to_stratify} years of age at time of diagnosis is {median_upper} {median_ci_upper[0],median_ci_upper[1]}\")\n",
    "print(f\"Median survival time for patients below {age_to_stratify} years of age at time of diagnosis is {median_lower} {median_ci_lower[0],median_ci_lower[1]}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**OBSERVATIONs**: <br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Patients younger than 60 have an overall higher survival time\n",
    "- Patient over 60 have been followed up for very short durations. There are not have enough events hence their survival curves never go below 0.5. This leads to the value of inf for median survival times\n",
    "- The fewer samples for patients with age at diagnosis above 60 also leads to the wider confidence intervals\n",
    "- Since the data was randomly generated, it does not agree with age-dependent breast cancer biology. In real-scenarios, late-onset breast cancer grows more slowly and is less aggressive [2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Statistical Analysis: Logrank test\n",
    "\n",
    "General form of log-rank test statistic is as follows:\n",
    "$$W = \\frac{X - \\mu}{\\sqrt{Var(X)}} $$\n",
    "Here, $X = \\sum_{k=1}^K q_{1k}$, where $q_{1k}$ is the number of patients who died at time $d_k$ in group 1 and $\\mu = \\sum_{k=1}^K \\frac{r_{1k}}{r_k}q_k$. This gives the following test statistic for log-rank test\n",
    "\n",
    "$$W = \\frac{\\sum_{k=1}^K (q_{1k} - \\mu_k)}{\\sqrt{\\sum_{k=1}^K Var(q_{1k})}} $$\n",
    "\n",
    "When sample size is large, $W$ follows an approximate normal distribution or that $W^2$ follows an approximate chi-square distribution\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "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",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>t_0</th>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>null_distribution</th>\n",
       "      <td>chi squared</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>degrees_of_freedom</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>test_name</th>\n",
       "      <td>logrank_test</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div><table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>test_statistic</th>\n",
       "      <th>p</th>\n",
       "      <th>-log2(p)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>14.12</td>\n",
       "      <td>&lt;0.005</td>\n",
       "      <td>12.51</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/latex": [
       "\\begin{tabular}{lrrr}\n",
       " & test_statistic & p & -log2(p) \\\\\n",
       "0 & 14.12 & 0.00 & 12.51 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "<lifelines.StatisticalResult: logrank_test>\n",
       "               t_0 = -1\n",
       " null_distribution = chi squared\n",
       "degrees_of_freedom = 1\n",
       "         test_name = logrank_test\n",
       "\n",
       "---\n",
       " test_statistic      p  -log2(p)\n",
       "          14.12 <0.005     12.51"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# logrank test between survival times of patients above and below the stratifyng age\n",
    "stats_results = logrank_test(df_final.loc[mask,\"follow_up_time\"], df_final.loc[~mask,\"follow_up_time\"], \n",
    "                             df_final.loc[mask,\"status\"], df_final.loc[~mask,\"status\"])\n",
    "\n",
    "# Printing summary of results\n",
    "stats_results.print_summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**OBSERVATION**: <br>\n",
    "- _p_-value < $\\alpha$ (=0.05), hence there is enough evidence to reject the null hypothesis.\n",
    "- Difference in the survival curves of patients stratified by age at diagnosis = 60 is statistically significant"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## HER2 status\n",
    "1. Calculate the intent to test rate for HER2. \n",
    "2. Calculate the tested rate for HER2.\n",
    "3. Calculate the negativity rate for HER2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Connect to the sqlite database\n",
    "con = sqlite3.connect(\"techTest.db\")\n",
    "cur = con.cursor()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Intent to Test Rate\n",
    "\n",
    "This is the percentage of patients who have an encounter in the biomarker table for the relevant biomarker.\n",
    "This problem was split into three steps:\n",
    "1. Subquery1: calculate the number of patients with BC status and HER2\n",
    "2. Subquery2: calculate the number of patients with BC status \n",
    "3. Divide the output (*1.0) from two subqueries to get floating point result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Intent to test rate is 81.3%\n"
     ]
    }
   ],
   "source": [
    "sql_query = '''--sql\n",
    "            SELECT \n",
    "            ---sub-query1 to calculate number of distinct patients who have BC status and HER2 biomarker---\n",
    "                (\n",
    "                SELECT \n",
    "                    COUNT(DISTINCT p1.patient_id)\n",
    "\n",
    "                FROM \n",
    "                    patient  as p1\n",
    "                LEFT JOIN \n",
    "                    diagnosis  as d1 \n",
    "                        ON p1.patient_id = d1.patient_id\n",
    "                LEFT JOIN \n",
    "                    biomarker b1 \n",
    "                        ON p1.patient_id = b1.patient_id\n",
    "                \n",
    "                WHERE \n",
    "                    d1.diagnosis_code LIKE \"C50%\" ---BC status---\n",
    "                    AND \n",
    "                    b1.biomarker_name = \"HER2\" ---HER2 biomarker---\n",
    "                    ) * 1.0 / \n",
    "                \n",
    "            ---sub-query2 to calculate number of distinct patients who have BC status\n",
    "                (\n",
    "                SELECT \n",
    "                    COUNT(DISTINCT p2.patient_id)\n",
    "\n",
    "                FROM \n",
    "                    patient as p2\n",
    "                LEFT JOIN \n",
    "                    diagnosis as d2 \n",
    "                        ON p2.patient_id = d2.patient_id\n",
    "                \n",
    "                WHERE \n",
    "                    d2.diagnosis_code LIKE \"C50%\"\n",
    "                    ) AS fraction;\n",
    "        '''\n",
    "cur.execute(sql_query)\n",
    "intent_to_test_rate = cur.fetchone()[0]\n",
    "\n",
    "print(f\"Intent to test rate is {100*intent_to_test_rate :0.1f}%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tested Rate\n",
    "The percentage of patients who have an encounter in the biomarker table for the relevant biomarker with a valid test result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test rate is 55.3%\n"
     ]
    }
   ],
   "source": [
    "sql_query = '''--sql\n",
    "            SELECT \n",
    "            ---sub-query1 to calculate number of distinct patients who have BC status, HER2 biomarker, and either Positive or Negative test---\n",
    "                (\n",
    "                SELECT \n",
    "                    COUNT(DISTINCT p1.patient_id)\n",
    "                \n",
    "                FROM \n",
    "                    patient  as p1\n",
    "                LEFT JOIN \n",
    "                    diagnosis  as d1 \n",
    "                        ON p1.patient_id = d1.patient_id\n",
    "                LEFT JOIN \n",
    "                    biomarker b1 \n",
    "                        ON p1.patient_id = b1.patient_id\n",
    "\n",
    "                WHERE \n",
    "                    d1.diagnosis_code LIKE \"C50%\" ---BC status---\n",
    "                    AND \n",
    "                    b1.biomarker_name = \"HER2\" ---biomarker HER2---\n",
    "                    AND \n",
    "                    b1.test_result in (\"Positive\", \"Negative\") ---Valid test result---\n",
    "                    ) * 1.0 / \n",
    "                \n",
    "            ---sub-query2 to calculate number of distinct patients who have BC status\n",
    "                (\n",
    "                SELECT \n",
    "                    COUNT(DISTINCT p2.patient_id)\n",
    "\n",
    "                FROM \n",
    "                    patient as p2\n",
    "                LEFT JOIN \n",
    "                    diagnosis as d2 \n",
    "                        ON p2.patient_id = d2.patient_id\n",
    "                \n",
    "                WHERE \n",
    "                    d2.diagnosis_code LIKE \"C50%\" ---BC status---\n",
    "                    ) AS fraction;\n",
    "        '''\n",
    "cur.execute(sql_query)\n",
    "test_rate = cur.fetchone()[0]\n",
    "\n",
    "print(f\"Test rate is {100*test_rate :0.1f}%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Negativity Rate\n",
    "\n",
    "The percentage of patients who have an encounter in the biomarker table for the relevant biomarker with a negative result."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Negative rate is 34.9%\n"
     ]
    }
   ],
   "source": [
    "sql_query = '''--sql\n",
    "            SELECT \n",
    "            ---sub-query1 to calculate number of distinct patients who have BC status, HER2 biomarker, and Negative---\n",
    "                (\n",
    "                SELECT \n",
    "                    COUNT(DISTINCT p1.patient_id)\n",
    "\n",
    "                FROM \n",
    "                    patient  as p1\n",
    "                LEFT JOIN \n",
    "                    diagnosis  as d1 \n",
    "                        ON p1.patient_id = d1.patient_id\n",
    "                LEFT JOIN \n",
    "                    biomarker b1 \n",
    "                        ON p1.patient_id = b1.patient_id\n",
    "                \n",
    "                WHERE \n",
    "                    d1.diagnosis_code LIKE \"C50%\" ---BC status---\n",
    "                    AND \n",
    "                    b1.biomarker_name = 'HER2' ---HER2 biomarker---\n",
    "                    AND \n",
    "                    b1.test_result == \"Negative\" ---Negative test result---\n",
    "                    ) * 1.0 / \n",
    "                \n",
    "            ---sub-query2 to calculate number of distinct patients who have BC status\n",
    "                (\n",
    "                SELECT \n",
    "                    COUNT(DISTINCT p2.patient_id)\n",
    "\n",
    "                FROM \n",
    "                    patient as p2\n",
    "                LEFT JOIN \n",
    "                    diagnosis as d2 \n",
    "                        ON p2.patient_id = d2.patient_id\n",
    "\n",
    "                WHERE \n",
    "                    d2.diagnosis_code LIKE \"C50%\" ---BC status---\n",
    "                    ) AS fraction;\n",
    "        '''\n",
    "cur.execute(sql_query)\n",
    "negative_rate = cur.fetchone()[0]\n",
    "\n",
    "print(f\"Negative rate is {100*negative_rate :0.1f}%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 246,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Close the connection to the database\n",
    "con.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "\n",
    "**Assumption**:\n",
    "1. Date of death is less reliable that date of diagnosis and date of biomarker test\n",
    "2. Biomarker test date is not a reliable predictor of date of first diagnosis\n",
    "3. Right censor dataset\n",
    "\n",
    "**Result**:\n",
    "1. Distribution of follow-up time is nearly symmetric with mean = 32 years\n",
    "2. Distribution of age at diagnosis is a long right tail with median = 32 years\n",
    "3. Difference in the survival curves of patients stratified by the age of first diagnosis = 60 is statistically significant\n",
    "4. Intention to test rate = 81.3%\n",
    "5. Test rate = 55.3 %\n",
    "6. Negative rate = 34.9%"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## References\n",
    "\n",
    "[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5242068/ <br>\n",
    "[2] https://www.sciencedirect.com/science/article/pii/S1040842807001795?via%3Dihub#fig3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Miscellaneous"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "metadata": {},
   "outputs": [],
   "source": [
    "con = sqlite3.connect(\"techTest.db\")\n",
    "cur = con.cursor()\n",
    "sql_query ='''\n",
    "            SELECT *\n",
    "            FROM biomarker\n",
    "'''\n",
    "df_biomarker = pd.read_sql_query(sql_query, con)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3026, 3)"
      ]
     },
     "execution_count": 248,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "query_inner = '''--sql\n",
    "                SELECT \n",
    "                    p.patient_id, \n",
    "                    d.diagnosis_code,\n",
    "                    d.diagnosis_date\n",
    "                    ---b.test_date---\n",
    "\n",
    "                FROM \n",
    "                    patient as p\n",
    "                LEFT JOIN \n",
    "                    diagnosis as d\n",
    "                    ON p.patient_id = d.patient_id\n",
    "                    ;\n",
    "                ---LEFT JOIN---\n",
    "                 ---   biomarker as b---\n",
    "                 ---   ON p.patient_id = b.patient_id;---\n",
    "                '''\n",
    "df_inner = pd.read_sql_query(query_inner, con)\n",
    "df_inner.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 249,
   "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>patient_id</th>\n",
       "      <th>diagnosis_code</th>\n",
       "      <th>diagnosis_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>019d3689-e49a-4f23-8ca1-65379907230b</td>\n",
       "      <td>C50.4</td>\n",
       "      <td>1996-10-05 00:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>019d3689-e49a-4f23-8ca1-65379907230b</td>\n",
       "      <td>C50.4</td>\n",
       "      <td>2004-10-30 00:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>019d3689-e49a-4f23-8ca1-65379907230b</td>\n",
       "      <td>C50</td>\n",
       "      <td>2020-06-13 00:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>81de4975-519f-48c6-a5fa-a57d891495a6</td>\n",
       "      <td>C50</td>\n",
       "      <td>1986-04-01 00:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>dbed4b3e-33ba-4856-a78d-3db6d211def8</td>\n",
       "      <td>I10</td>\n",
       "      <td>1993-09-08 00:00:00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             patient_id diagnosis_code       diagnosis_date\n",
       "0  019d3689-e49a-4f23-8ca1-65379907230b          C50.4  1996-10-05 00:00:00\n",
       "1  019d3689-e49a-4f23-8ca1-65379907230b          C50.4  2004-10-30 00:00:00\n",
       "2  019d3689-e49a-4f23-8ca1-65379907230b            C50  2020-06-13 00:00:00\n",
       "3  81de4975-519f-48c6-a5fa-a57d891495a6            C50  1986-04-01 00:00:00\n",
       "4  dbed4b3e-33ba-4856-a78d-3db6d211def8            I10  1993-09-08 00:00:00"
      ]
     },
     "execution_count": 249,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_inner.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(941, 3)"
      ]
     },
     "execution_count": 250,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_filter = df_inner[df_inner[\"diagnosis_code\"].str.contains('C50.*')]\n",
    "df_filter2 = df_filter.drop_duplicates(\"patient_id\")\n",
    "df_filter2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "941x6\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>patient_id</th>\n",
       "      <th>dod</th>\n",
       "      <th>age_at_diagnosis</th>\n",
       "      <th>first_diagnosis_date</th>\n",
       "      <th>last_diagnosis_date</th>\n",
       "      <th>last_test_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0004477e-31f7-4704-a123-2b68dd64aaed</td>\n",
       "      <td>2008-03-20</td>\n",
       "      <td>59</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>2011-12-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0022fa14-8260-4b11-88cd-3fa066ede851</td>\n",
       "      <td>2008-03-04</td>\n",
       "      <td>50</td>\n",
       "      <td>1993-04-07</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>2004-12-19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0183e236-7892-4aa9-8b7f-10f12676f1ec</td>\n",
       "      <td>2038-03-11</td>\n",
       "      <td>21</td>\n",
       "      <td>1986-06-02</td>\n",
       "      <td>2021-10-13</td>\n",
       "      <td>2023-12-29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>019d3689-e49a-4f23-8ca1-65379907230b</td>\n",
       "      <td>2042-06-15</td>\n",
       "      <td>30</td>\n",
       "      <td>1996-10-05</td>\n",
       "      <td>2020-06-13</td>\n",
       "      <td>2018-03-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>01d468fc-55b8-45e3-8eca-dcb7a6185f28</td>\n",
       "      <td>2007-07-30</td>\n",
       "      <td>26</td>\n",
       "      <td>1977-01-16</td>\n",
       "      <td>1981-10-28</td>\n",
       "      <td>1995-12-08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             patient_id         dod  age_at_diagnosis  \\\n",
       "0  0004477e-31f7-4704-a123-2b68dd64aaed  2008-03-20                59   \n",
       "1  0022fa14-8260-4b11-88cd-3fa066ede851  2008-03-04                50   \n",
       "2  0183e236-7892-4aa9-8b7f-10f12676f1ec  2038-03-11                21   \n",
       "3  019d3689-e49a-4f23-8ca1-65379907230b  2042-06-15                30   \n",
       "4  01d468fc-55b8-45e3-8eca-dcb7a6185f28  2007-07-30                26   \n",
       "\n",
       "  first_diagnosis_date last_diagnosis_date last_test_date  \n",
       "0           2019-11-04          2019-11-04     2011-12-03  \n",
       "1           1993-04-07          2019-05-20     2004-12-19  \n",
       "2           1986-06-02          2021-10-13     2023-12-29  \n",
       "3           1996-10-05          2020-06-13     2018-03-10  \n",
       "4           1977-01-16          1981-10-28     1995-12-08  "
      ]
     },
     "execution_count": 251,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "query_trial= '''--sql\n",
    "                SELECT \n",
    "                    p.patient_id, \n",
    "                    date(p.dod) as dod,  \n",
    "                    STRFTIME(\"%Y\", MIN(d.diagnosis_date)) - STRFTIME(\"%Y\",p.dob) as age_at_diagnosis, \n",
    "                    date(MIN(d.diagnosis_date)) as first_diagnosis_date, \n",
    "                    sub_dates.last_diagnosis_date, \n",
    "                    sub_dates.last_test_date\n",
    "\n",
    "                ---Patient joined to Diagnosis joined to Biomarker on key patient_id---\n",
    "                FROM \n",
    "                    patient as p \n",
    "                LEFT JOIN \n",
    "                    diagnosis as d\n",
    "                        ON p.patient_id = d.patient_id\n",
    "                LEFT JOIN\n",
    "                    (SELECT \n",
    "                        sub_p.patient_id, \n",
    "                        date(MAX(sub_d.diagnosis_date)) as last_diagnosis_date,\n",
    "                        date(MAX(sub_b.test_date)) as last_test_date\n",
    "                    FROM\n",
    "                        patient as sub_p\n",
    "                    LEFT JOIN\n",
    "                        diagnosis as sub_d \n",
    "                            ON sub_p.patient_id = sub_d.patient_id\n",
    "                    LEFT JOIN\n",
    "                        biomarker as sub_b\n",
    "                            ON sub_p.patient_id = sub_b.patient_id\n",
    "                    GROUP BY\n",
    "                        sub_p.patient_id\n",
    "                    ) as sub_dates\n",
    "                    ON p.patient_id = sub_dates.patient_id \n",
    "\n",
    "                WHERE  \n",
    "                    d.diagnosis_code LIKE \"C50%\"\n",
    "                        \n",
    "                GROUP BY \n",
    "                    p.patient_id;\n",
    "\n",
    "                '''\n",
    "df_patient = pd.read_sql_query(query_trial, con)\n",
    "#df_patient[[\"dob\", \"dod\"]] = pd.to_datetime(df_patient[[\"dob\",\"dod\"]])\n",
    "print(f\"{df_patient.shape[0]}x{df_patient.shape[1]}\")\n",
    "df_patient.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 252,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9258x2\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "7352"
      ]
     },
     "execution_count": 252,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "query_test = '''--sql\n",
    "                    SELECT \n",
    "                        sub_p.patient_id, sub_d.diagnosis_code\n",
    "                        ---date(MAX(sub_d.diagnosis_date)) as last_dianosis_date,\n",
    "                        ---date(MAX(sub_b.test_date)) as last_test_date\n",
    "                    FROM\n",
    "                        patient as sub_p\n",
    "                    LEFT JOIN\n",
    "                        diagnosis as sub_d \n",
    "                            ON sub_p.patient_id = sub_d.patient_id\n",
    "                    LEFT JOIN\n",
    "                        biomarker as sub_b\n",
    "                            ON sub_p.patient_id = sub_b.patient_id\n",
    "                    ---GROUP BY\n",
    "                    ---    sub_p.patient_id\n",
    "\n",
    "'''\n",
    "\n",
    "df_inner = pd.read_sql_query(query_test, con)\n",
    "print(f\"{df_inner.shape[0]}x{df_inner.shape[1]}\")\n",
    "\n",
    "df_inner[\"diagnosis_code\"].str.match(\"C50.*\").sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "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>diagnosis_code</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>patient_id</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0004477e-31f7-4704-a123-2b68dd64aaed</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0022fa14-8260-4b11-88cd-3fa066ede851</th>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0183e236-7892-4aa9-8b7f-10f12676f1ec</th>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>019d3689-e49a-4f23-8ca1-65379907230b</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>01d468fc-55b8-45e3-8eca-dcb7a6185f28</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fe0c8392-f97d-4097-b0f2-a726a17cadd5</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fe3690ce-1b13-4bfa-b65e-71d385aeb8c1</th>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fe3e2d80-8c8c-40ee-8fa2-f4cb257c1c65</th>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ff717a0c-6c00-4585-9d45-8e22f9268b29</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>941 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      diagnosis_code\n",
       "patient_id                                          \n",
       "0004477e-31f7-4704-a123-2b68dd64aaed               3\n",
       "0022fa14-8260-4b11-88cd-3fa066ede851              12\n",
       "0183e236-7892-4aa9-8b7f-10f12676f1ec              16\n",
       "019d3689-e49a-4f23-8ca1-65379907230b               3\n",
       "01d468fc-55b8-45e3-8eca-dcb7a6185f28               8\n",
       "...                                              ...\n",
       "fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8               5\n",
       "fe0c8392-f97d-4097-b0f2-a726a17cadd5               3\n",
       "fe3690ce-1b13-4bfa-b65e-71d385aeb8c1              20\n",
       "fe3e2d80-8c8c-40ee-8fa2-f4cb257c1c65              10\n",
       "ff717a0c-6c00-4585-9d45-8e22f9268b29               5\n",
       "\n",
       "[941 rows x 1 columns]"
      ]
     },
     "execution_count": 253,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_inner[df_inner[\"diagnosis_code\"].str.match(\"C50.*\")].groupby([\"patient_id\"]).count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Do patients have multiple date of births* - No, there are 1000 patients"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {},
   "outputs": [],
   "source": [
    "#query to group data by patient_id and compare min and max of date of birth\n",
    "\n",
    "cur.execute('''--sql\n",
    "            \n",
    "                SELECT SUM(same_dob) as total_same_dob, COUNT(*) as total_rows\n",
    "\n",
    "                ---Sub-query to count patients whesre MIN(dob) is equal to MAX(dob)---\n",
    "                FROM (\n",
    "                    SELECT COUNT(*) as same_dob\n",
    "                    FROM Patient\n",
    "                    GROUP BY patient_id\n",
    "                    ---HAVING MIN(dob) = MAX(dob)---\n",
    "                    );\n",
    "                \n",
    "            ''')\n",
    "\n",
    "total_same_dob, total_rows = cur.fetchone()\n",
    "\n",
    "# assert statement to check whether all patients have identical dob\n",
    "assert total_same_dob == total_rows, \"A patient has multiple dobs\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "query_age = '''\n",
    "                SELECT STRFTIME(\"%Y\",dod) - STRFTIME(\"%Y\",dob) as age, gender\n",
    "                FROM patient\n",
    "                WHERE gender != \"NULL\"\n",
    "                '''\n",
    "df_age = pd.read_sql_query(query_age, con)\n",
    "df_age.boxplot(by = \"gender\")\n",
    "plt.title(\"Distribution of age stratified by age\")\n",
    "plt.ylabel(\"Years\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'gender'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "File \u001b[1;32mc:\\Users\\vinay\\anaconda3\\Lib\\site-packages\\pandas\\core\\indexes\\base.py:3791\u001b[0m, in \u001b[0;36mIndex.get_loc\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   3790\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m-> 3791\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_engine\u001b[38;5;241m.\u001b[39mget_loc(casted_key)\n\u001b[0;32m   3792\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n",
      "File \u001b[1;32mindex.pyx:152\u001b[0m, in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "File \u001b[1;32mindex.pyx:181\u001b[0m, in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "File \u001b[1;32mpandas\\_libs\\hashtable_class_helper.pxi:7080\u001b[0m, in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[1;34m()\u001b[0m\n",
      "File \u001b[1;32mpandas\\_libs\\hashtable_class_helper.pxi:7088\u001b[0m, in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;31mKeyError\u001b[0m: 'gender'",
      "\nThe above exception was the direct cause of the following exception:\n",
      "\u001b[1;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[256], line 1\u001b[0m\n\u001b[1;32m----> 1\u001b[0m df_final[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mgender\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mcount_values()\n",
      "File \u001b[1;32mc:\\Users\\vinay\\anaconda3\\Lib\\site-packages\\pandas\\core\\frame.py:3893\u001b[0m, in \u001b[0;36mDataFrame.__getitem__\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   3891\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcolumns\u001b[38;5;241m.\u001b[39mnlevels \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m1\u001b[39m:\n\u001b[0;32m   3892\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_getitem_multilevel(key)\n\u001b[1;32m-> 3893\u001b[0m indexer \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcolumns\u001b[38;5;241m.\u001b[39mget_loc(key)\n\u001b[0;32m   3894\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m is_integer(indexer):\n\u001b[0;32m   3895\u001b[0m     indexer \u001b[38;5;241m=\u001b[39m [indexer]\n",
      "File \u001b[1;32mc:\\Users\\vinay\\anaconda3\\Lib\\site-packages\\pandas\\core\\indexes\\base.py:3798\u001b[0m, in \u001b[0;36mIndex.get_loc\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   3793\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(casted_key, \u001b[38;5;28mslice\u001b[39m) \u001b[38;5;129;01mor\u001b[39;00m (\n\u001b[0;32m   3794\u001b[0m         \u001b[38;5;28misinstance\u001b[39m(casted_key, abc\u001b[38;5;241m.\u001b[39mIterable)\n\u001b[0;32m   3795\u001b[0m         \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28many\u001b[39m(\u001b[38;5;28misinstance\u001b[39m(x, \u001b[38;5;28mslice\u001b[39m) \u001b[38;5;28;01mfor\u001b[39;00m x \u001b[38;5;129;01min\u001b[39;00m casted_key)\n\u001b[0;32m   3796\u001b[0m     ):\n\u001b[0;32m   3797\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m InvalidIndexError(key)\n\u001b[1;32m-> 3798\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01merr\u001b[39;00m\n\u001b[0;32m   3799\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m:\n\u001b[0;32m   3800\u001b[0m     \u001b[38;5;66;03m# If we have a listlike key, _check_indexing_error will raise\u001b[39;00m\n\u001b[0;32m   3801\u001b[0m     \u001b[38;5;66;03m#  InvalidIndexError. Otherwise we fall through and re-raise\u001b[39;00m\n\u001b[0;32m   3802\u001b[0m     \u001b[38;5;66;03m#  the TypeError.\u001b[39;00m\n\u001b[0;32m   3803\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_indexing_error(key)\n",
      "\u001b[1;31mKeyError\u001b[0m: 'gender'"
     ]
    }
   ],
   "source": [
    "df_final[\"gender\"].count_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_removenull = df[df[\"gender\"]!=\"NULL\"]\n",
    "df_removenull[[\"follow_up_time\", \"age_at_diagnosis\", \"gender\"]].boxplot(by = \"gender\", xlabel = \"Gender\")\n",
    "fig.tight_layout(pad = 2)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Testing HER2 status with pandas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2873, 4)\n",
      "denominator: 941\n",
      "Intention to treat: 1426.00\n",
      "Tested_rate: 900.00\n",
      "Tested_rate: 411.00\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array(['Positive', 'Negative', 'Unknown', 'Equivocal', 'Inconclusive',\n",
       "       'NA'], dtype=object)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sql_query = '''--sql\n",
    "                SELECT p.patient_id, b.biomarker_name, b.test_result, b.biomarker_test_type\n",
    "                \n",
    "                ---JOIN patient and biomarker table---\n",
    "                FROM \n",
    "                    patient as p \n",
    "                LEFT JOIN \n",
    "                    biomarker as b \n",
    "                        on p.patient_id = b.patient_id\n",
    "   \n",
    "                ---sub-query to select patients with breast cancer with diagnosis_code starting with C50---\n",
    "                WHERE \n",
    "                    p.patient_id IN (\n",
    "                        SELECT \n",
    "                            patient_id\n",
    "                        FROM \n",
    "                            diagnosis\n",
    "                        WHERE \n",
    "                            diagnosis_code LIKE \"C50%\"\n",
    "                        );\n",
    "            '''\n",
    "\n",
    "df_her2_status = pd.read_sql_query(sql_query, con)\n",
    "print(df_her2_status.shape)\n",
    "print(f\"denominator: {len(df_her2_status['patient_id'].unique())}\")\n",
    "print(f\"Intention to treat: {(df_her2_status['biomarker_name']=='HER2').sum():0.2f}\")\n",
    "print(f\"Tested_rate: {((df_her2_status['biomarker_name']=='HER2') & (df_her2_status['biomarker_test_type'].isin(['PCR','IHC']))).sum():0.2f}\")\n",
    "print(f\"Tested_rate: {((df_her2_status['biomarker_name']=='HER2') & (df_her2_status['test_result'] == 'Negative')).sum():0.2f}\")\n",
    "df_her2_status[\"test_result\"].unique()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>biomarker_name</th>\n",
       "      <th>test_result</th>\n",
       "      <th>biomarker_test_type</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>patient_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0004477e-31f7-4704-a123-2b68dd64aaed</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0022fa14-8260-4b11-88cd-3fa066ede851</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0183e236-7892-4aa9-8b7f-10f12676f1ec</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>01d468fc-55b8-45e3-8eca-dcb7a6185f28</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>022ff2bf-19d7-4c4d-8410-a91773ad24f2</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fe0c8392-f97d-4097-b0f2-a726a17cadd5</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fe3690ce-1b13-4bfa-b65e-71d385aeb8c1</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fe3e2d80-8c8c-40ee-8fa2-f4cb257c1c65</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ff717a0c-6c00-4585-9d45-8e22f9268b29</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>765 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      biomarker_name  test_result  \\\n",
       "patient_id                                                          \n",
       "0004477e-31f7-4704-a123-2b68dd64aaed               1            1   \n",
       "0022fa14-8260-4b11-88cd-3fa066ede851               1            1   \n",
       "0183e236-7892-4aa9-8b7f-10f12676f1ec               1            1   \n",
       "01d468fc-55b8-45e3-8eca-dcb7a6185f28               3            3   \n",
       "022ff2bf-19d7-4c4d-8410-a91773ad24f2               4            4   \n",
       "...                                              ...          ...   \n",
       "fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8               3            3   \n",
       "fe0c8392-f97d-4097-b0f2-a726a17cadd5               1            1   \n",
       "fe3690ce-1b13-4bfa-b65e-71d385aeb8c1               2            2   \n",
       "fe3e2d80-8c8c-40ee-8fa2-f4cb257c1c65               2            2   \n",
       "ff717a0c-6c00-4585-9d45-8e22f9268b29               3            3   \n",
       "\n",
       "                                      biomarker_test_type  \n",
       "patient_id                                                 \n",
       "0004477e-31f7-4704-a123-2b68dd64aaed                    1  \n",
       "0022fa14-8260-4b11-88cd-3fa066ede851                    1  \n",
       "0183e236-7892-4aa9-8b7f-10f12676f1ec                    1  \n",
       "01d468fc-55b8-45e3-8eca-dcb7a6185f28                    3  \n",
       "022ff2bf-19d7-4c4d-8410-a91773ad24f2                    4  \n",
       "...                                                   ...  \n",
       "fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8                    3  \n",
       "fe0c8392-f97d-4097-b0f2-a726a17cadd5                    1  \n",
       "fe3690ce-1b13-4bfa-b65e-71d385aeb8c1                    2  \n",
       "fe3e2d80-8c8c-40ee-8fa2-f4cb257c1c65                    2  \n",
       "ff717a0c-6c00-4585-9d45-8e22f9268b29                    3  \n",
       "\n",
       "[765 rows x 3 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_her2_status[df_her2_status[\"biomarker_name\"]==\"HER2\"].groupby(\"patient_id\").count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8129649309245484"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "765/941"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>biomarker_name</th>\n",
       "      <th>test_result</th>\n",
       "      <th>biomarker_test_type</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>patient_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0004477e-31f7-4704-a123-2b68dd64aaed</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0022fa14-8260-4b11-88cd-3fa066ede851</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>01d468fc-55b8-45e3-8eca-dcb7a6185f28</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>022ff2bf-19d7-4c4d-8410-a91773ad24f2</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>023f0abd-4675-4b59-96e9-5b5ca052560a</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fd0c8dae-b4ef-4ed6-917c-4ac27aa67635</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fd6ec646-f3ae-4552-8b63-aa10255f5095</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fe3690ce-1b13-4bfa-b65e-71d385aeb8c1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ff717a0c-6c00-4585-9d45-8e22f9268b29</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>520 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      biomarker_name  test_result  \\\n",
       "patient_id                                                          \n",
       "0004477e-31f7-4704-a123-2b68dd64aaed               1            1   \n",
       "0022fa14-8260-4b11-88cd-3fa066ede851               1            1   \n",
       "01d468fc-55b8-45e3-8eca-dcb7a6185f28               3            3   \n",
       "022ff2bf-19d7-4c4d-8410-a91773ad24f2               2            2   \n",
       "023f0abd-4675-4b59-96e9-5b5ca052560a               1            1   \n",
       "...                                              ...          ...   \n",
       "fd0c8dae-b4ef-4ed6-917c-4ac27aa67635               3            3   \n",
       "fd6ec646-f3ae-4552-8b63-aa10255f5095               2            2   \n",
       "fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8               2            2   \n",
       "fe3690ce-1b13-4bfa-b65e-71d385aeb8c1               1            1   \n",
       "ff717a0c-6c00-4585-9d45-8e22f9268b29               2            2   \n",
       "\n",
       "                                      biomarker_test_type  \n",
       "patient_id                                                 \n",
       "0004477e-31f7-4704-a123-2b68dd64aaed                    1  \n",
       "0022fa14-8260-4b11-88cd-3fa066ede851                    1  \n",
       "01d468fc-55b8-45e3-8eca-dcb7a6185f28                    3  \n",
       "022ff2bf-19d7-4c4d-8410-a91773ad24f2                    2  \n",
       "023f0abd-4675-4b59-96e9-5b5ca052560a                    1  \n",
       "...                                                   ...  \n",
       "fd0c8dae-b4ef-4ed6-917c-4ac27aa67635                    3  \n",
       "fd6ec646-f3ae-4552-8b63-aa10255f5095                    2  \n",
       "fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8                    2  \n",
       "fe3690ce-1b13-4bfa-b65e-71d385aeb8c1                    1  \n",
       "ff717a0c-6c00-4585-9d45-8e22f9268b29                    2  \n",
       "\n",
       "[520 rows x 3 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_her2_status[(df_her2_status[\"biomarker_name\"]==\"HER2\") & (df_her2_status[\"test_result\"].isin([\"Positive\", \"Negative\"])) ].groupby(\"patient_id\").count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5526036131774708"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "520/941"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>biomarker_name</th>\n",
       "      <th>test_result</th>\n",
       "      <th>biomarker_test_type</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>patient_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0004477e-31f7-4704-a123-2b68dd64aaed</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>01d468fc-55b8-45e3-8eca-dcb7a6185f28</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>022ff2bf-19d7-4c4d-8410-a91773ad24f2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>023f0abd-4675-4b59-96e9-5b5ca052560a</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>05720cba-7eda-4f7e-b0e8-bcc1022a1702</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fccd5a49-ba76-4d1f-a344-7df398462f60</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fd0c8dae-b4ef-4ed6-917c-4ac27aa67635</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fd6ec646-f3ae-4552-8b63-aa10255f5095</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ff717a0c-6c00-4585-9d45-8e22f9268b29</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>328 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      biomarker_name  test_result  \\\n",
       "patient_id                                                          \n",
       "0004477e-31f7-4704-a123-2b68dd64aaed               1            1   \n",
       "01d468fc-55b8-45e3-8eca-dcb7a6185f28               2            2   \n",
       "022ff2bf-19d7-4c4d-8410-a91773ad24f2               1            1   \n",
       "023f0abd-4675-4b59-96e9-5b5ca052560a               1            1   \n",
       "05720cba-7eda-4f7e-b0e8-bcc1022a1702               1            1   \n",
       "...                                              ...          ...   \n",
       "fccd5a49-ba76-4d1f-a344-7df398462f60               1            1   \n",
       "fd0c8dae-b4ef-4ed6-917c-4ac27aa67635               2            2   \n",
       "fd6ec646-f3ae-4552-8b63-aa10255f5095               2            2   \n",
       "fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8               1            1   \n",
       "ff717a0c-6c00-4585-9d45-8e22f9268b29               2            2   \n",
       "\n",
       "                                      biomarker_test_type  \n",
       "patient_id                                                 \n",
       "0004477e-31f7-4704-a123-2b68dd64aaed                    1  \n",
       "01d468fc-55b8-45e3-8eca-dcb7a6185f28                    2  \n",
       "022ff2bf-19d7-4c4d-8410-a91773ad24f2                    1  \n",
       "023f0abd-4675-4b59-96e9-5b5ca052560a                    1  \n",
       "05720cba-7eda-4f7e-b0e8-bcc1022a1702                    1  \n",
       "...                                                   ...  \n",
       "fccd5a49-ba76-4d1f-a344-7df398462f60                    1  \n",
       "fd0c8dae-b4ef-4ed6-917c-4ac27aa67635                    2  \n",
       "fd6ec646-f3ae-4552-8b63-aa10255f5095                    2  \n",
       "fdd8e833-ca01-4ce5-8ca9-30ec3ba404d8                    1  \n",
       "ff717a0c-6c00-4585-9d45-8e22f9268b29                    2  \n",
       "\n",
       "[328 rows x 3 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_her2_status[(df_her2_status[\"biomarker_name\"]==\"HER2\") & (df_her2_status[\"test_result\"]== \"Negative\") ].groupby(\"patient_id\").count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3485653560042508"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "328/941"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Biomarker test for diagnostics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "76"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "condition = df_biomarker[\"patient_id\"].isin(df_diagnostic_biomarker[\"patient_id\"])\n",
    "(df_biomarker.loc[condition, \"test_result\"] == \"Positive\").sum()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}