Visualize_expression_hormone.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import scipy\n",
    "\n",
    "from matplotlib import gridspec\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from google.cloud import bigquery as bq\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x72 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x72 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.palplot(sns.color_palette(\"Spectral_r\", n_colors=20))\n",
    "sns.palplot(sns.color_palette([\"#b7098b\",\"#ff639c\", \"#5bc1b5\", \"#d6e79a\", \"#eac435\", \"#d0213e\", \"#4c2c72\", \"#345995\"]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(53970, 19)\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>case_barcode</th>\n",
       "      <th>sample_barcode</th>\n",
       "      <th>aliquot_barcode</th>\n",
       "      <th>gene_symbol</th>\n",
       "      <th>normalized_count</th>\n",
       "      <th>transformed_count</th>\n",
       "      <th>sample_type</th>\n",
       "      <th>sample_type_name</th>\n",
       "      <th>initial_response</th>\n",
       "      <th>recurrence_status</th>\n",
       "      <th>recurrence_log</th>\n",
       "      <th>subtype</th>\n",
       "      <th>ER</th>\n",
       "      <th>PR</th>\n",
       "      <th>Her2_IHC</th>\n",
       "      <th>Her2_ISH</th>\n",
       "      <th>Her2</th>\n",
       "      <th>triple_negative</th>\n",
       "      <th>subtype_sub</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>TCGA-5L-AAT0</td>\n",
       "      <td>TCGA-5L-AAT0-01A</td>\n",
       "      <td>TCGA-5L-AAT0-01A-12R-A41B-07</td>\n",
       "      <td>ADRA2C</td>\n",
       "      <td>39.8885</td>\n",
       "      <td>5.353623</td>\n",
       "      <td>1</td>\n",
       "      <td>Primary solid Tumor</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LumA</td>\n",
       "      <td>Positive</td>\n",
       "      <td>Positive</td>\n",
       "      <td>1+</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LumA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>TCGA-5L-AAT0</td>\n",
       "      <td>TCGA-5L-AAT0-01A</td>\n",
       "      <td>TCGA-5L-AAT0-01A-12R-A41B-07</td>\n",
       "      <td>ANP32E</td>\n",
       "      <td>766.0947</td>\n",
       "      <td>9.583261</td>\n",
       "      <td>1</td>\n",
       "      <td>Primary solid Tumor</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LumA</td>\n",
       "      <td>Positive</td>\n",
       "      <td>Positive</td>\n",
       "      <td>1+</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LumA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>TCGA-5L-AAT0</td>\n",
       "      <td>TCGA-5L-AAT0-01A</td>\n",
       "      <td>TCGA-5L-AAT0-01A-12R-A41B-07</td>\n",
       "      <td>ARL4C</td>\n",
       "      <td>993.6941</td>\n",
       "      <td>9.958109</td>\n",
       "      <td>1</td>\n",
       "      <td>Primary solid Tumor</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LumA</td>\n",
       "      <td>Positive</td>\n",
       "      <td>Positive</td>\n",
       "      <td>1+</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LumA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>TCGA-5L-AAT0</td>\n",
       "      <td>TCGA-5L-AAT0-01A</td>\n",
       "      <td>TCGA-5L-AAT0-01A-12R-A41B-07</td>\n",
       "      <td>ASCL1</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>Primary solid Tumor</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LumA</td>\n",
       "      <td>Positive</td>\n",
       "      <td>Positive</td>\n",
       "      <td>1+</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LumA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>TCGA-5L-AAT0</td>\n",
       "      <td>TCGA-5L-AAT0-01A</td>\n",
       "      <td>TCGA-5L-AAT0-01A-12R-A41B-07</td>\n",
       "      <td>ASS1</td>\n",
       "      <td>673.4125</td>\n",
       "      <td>9.397487</td>\n",
       "      <td>1</td>\n",
       "      <td>Primary solid Tumor</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LumA</td>\n",
       "      <td>Positive</td>\n",
       "      <td>Positive</td>\n",
       "      <td>1+</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LumA</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   case_barcode    sample_barcode               aliquot_barcode gene_symbol  \\\n",
       "0  TCGA-5L-AAT0  TCGA-5L-AAT0-01A  TCGA-5L-AAT0-01A-12R-A41B-07      ADRA2C   \n",
       "1  TCGA-5L-AAT0  TCGA-5L-AAT0-01A  TCGA-5L-AAT0-01A-12R-A41B-07      ANP32E   \n",
       "2  TCGA-5L-AAT0  TCGA-5L-AAT0-01A  TCGA-5L-AAT0-01A-12R-A41B-07       ARL4C   \n",
       "3  TCGA-5L-AAT0  TCGA-5L-AAT0-01A  TCGA-5L-AAT0-01A-12R-A41B-07       ASCL1   \n",
       "4  TCGA-5L-AAT0  TCGA-5L-AAT0-01A  TCGA-5L-AAT0-01A-12R-A41B-07        ASS1   \n",
       "\n",
       "   normalized_count  transformed_count  sample_type     sample_type_name  \\\n",
       "0           39.8885           5.353623            1  Primary solid Tumor   \n",
       "1          766.0947           9.583261            1  Primary solid Tumor   \n",
       "2          993.6941           9.958109            1  Primary solid Tumor   \n",
       "3            0.0000           0.000000            1  Primary solid Tumor   \n",
       "4          673.4125           9.397487            1  Primary solid Tumor   \n",
       "\n",
       "  initial_response recurrence_status recurrence_log subtype        ER  \\\n",
       "0          Unknown               NaN            NaN    LumA  Positive   \n",
       "1          Unknown               NaN            NaN    LumA  Positive   \n",
       "2          Unknown               NaN            NaN    LumA  Positive   \n",
       "3          Unknown               NaN            NaN    LumA  Positive   \n",
       "4          Unknown               NaN            NaN    LumA  Positive   \n",
       "\n",
       "         PR Her2_IHC Her2_ISH Her2  triple_negative subtype_sub  \n",
       "0  Positive       1+      NaN  NaN              NaN        LumA  \n",
       "1  Positive       1+      NaN  NaN              NaN        LumA  \n",
       "2  Positive       1+      NaN  NaN              NaN        LumA  \n",
       "3  Positive       1+      NaN  NaN              NaN        LumA  \n",
       "4  Positive       1+      NaN  NaN              NaN        LumA  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hormone_expression_bigtable = pd.read_table(\"./datasets/hormone_expression_bigtable.IlluminaHiSeq.tsv\", sep=\"\\t\")\n",
    "\n",
    "print(hormone_expression_bigtable.shape)\n",
    "hormone_expression_bigtable.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sample types\n",
      "Primary solid Tumor    480\n",
      "Solid Tissue Normal     33\n",
      "Metastatic               1\n",
      "Name: sample_type_name, dtype: int64\n",
      "\n",
      "\n",
      "Initial responses\n",
      "Unknown               431\n",
      "Complete response      40\n",
      "Progressive/Stable      9\n",
      "Name: initial_response, dtype: int64\n",
      "\n",
      "\n",
      "Recurrence statuses\n",
      "Unknown     423\n",
      "Occured      42\n",
      "Low risk      8\n",
      "Name: recurrence_status, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(\"Sample types\")\n",
    "print(hormone_expression_bigtable.drop_duplicates(subset=\"sample_barcode\").sample_type_name.value_counts())\n",
    "print(\"\\n\")\n",
    "\n",
    "print(\"Initial responses\")\n",
    "print(hormone_expression_bigtable.drop_duplicates(subset=\"case_barcode\").initial_response.value_counts())\n",
    "print(\"\\n\")\n",
    "\n",
    "print(\"Recurrence statuses\")\n",
    "print(hormone_expression_bigtable.drop_duplicates(subset=\"case_barcode\").recurrence_status.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(105, 514)\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>aliquot_barcode</th>\n",
       "      <th>TCGA-5L-AAT0-01A-12R-A41B-07</th>\n",
       "      <th>TCGA-5L-AAT1-01A-12R-A41B-07</th>\n",
       "      <th>TCGA-5T-A9QA-01A-11R-A41B-07</th>\n",
       "      <th>TCGA-A1-A0SF-01A-11R-A144-07</th>\n",
       "      <th>TCGA-A1-A0SQ-01A-21R-A144-07</th>\n",
       "      <th>TCGA-A2-A04N-01A-11R-A115-07</th>\n",
       "      <th>TCGA-A2-A04R-01A-41R-A109-07</th>\n",
       "      <th>TCGA-A2-A04V-01A-21R-A034-07</th>\n",
       "      <th>TCGA-A2-A04X-01A-21R-A034-07</th>\n",
       "      <th>TCGA-A2-A04Y-01A-21R-A034-07</th>\n",
       "      <th>...</th>\n",
       "      <th>TCGA-S3-A6ZH-01A-22R-A32P-07</th>\n",
       "      <th>TCGA-S3-AA0Z-01A-11R-A41B-07</th>\n",
       "      <th>TCGA-S3-AA11-01A-31R-A41B-07</th>\n",
       "      <th>TCGA-S3-AA12-01A-11R-A41B-07</th>\n",
       "      <th>TCGA-S3-AA17-01A-11R-A41B-07</th>\n",
       "      <th>TCGA-V7-A7HQ-01A-11R-A33J-07</th>\n",
       "      <th>TCGA-W8-A86G-01A-21R-A36F-07</th>\n",
       "      <th>TCGA-WT-AB41-01A-11R-A41B-07</th>\n",
       "      <th>TCGA-WT-AB44-01A-11R-A41B-07</th>\n",
       "      <th>TCGA-XX-A899-01A-11R-A36F-07</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_symbol</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ADRA2C</th>\n",
       "      <td>5.353623</td>\n",
       "      <td>5.034955</td>\n",
       "      <td>8.948954</td>\n",
       "      <td>7.585473</td>\n",
       "      <td>4.197512</td>\n",
       "      <td>5.412802</td>\n",
       "      <td>1.061223</td>\n",
       "      <td>4.444813</td>\n",
       "      <td>3.688751</td>\n",
       "      <td>7.507771</td>\n",
       "      <td>...</td>\n",
       "      <td>5.335104</td>\n",
       "      <td>2.696283</td>\n",
       "      <td>6.970269</td>\n",
       "      <td>1.565500</td>\n",
       "      <td>6.750738</td>\n",
       "      <td>7.721103</td>\n",
       "      <td>6.547644</td>\n",
       "      <td>9.522046</td>\n",
       "      <td>5.137868</td>\n",
       "      <td>6.085958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ANP32E</th>\n",
       "      <td>9.583261</td>\n",
       "      <td>9.108360</td>\n",
       "      <td>6.086308</td>\n",
       "      <td>10.008464</td>\n",
       "      <td>10.175954</td>\n",
       "      <td>9.372962</td>\n",
       "      <td>11.066019</td>\n",
       "      <td>10.669437</td>\n",
       "      <td>8.792080</td>\n",
       "      <td>10.828133</td>\n",
       "      <td>...</td>\n",
       "      <td>11.234595</td>\n",
       "      <td>12.717716</td>\n",
       "      <td>9.483399</td>\n",
       "      <td>9.273418</td>\n",
       "      <td>9.429852</td>\n",
       "      <td>8.734465</td>\n",
       "      <td>9.307549</td>\n",
       "      <td>10.252020</td>\n",
       "      <td>9.942660</td>\n",
       "      <td>10.109748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ARL4C</th>\n",
       "      <td>9.958109</td>\n",
       "      <td>11.198834</td>\n",
       "      <td>7.288494</td>\n",
       "      <td>9.284177</td>\n",
       "      <td>8.181195</td>\n",
       "      <td>9.943622</td>\n",
       "      <td>8.949372</td>\n",
       "      <td>9.345226</td>\n",
       "      <td>9.395779</td>\n",
       "      <td>10.203331</td>\n",
       "      <td>...</td>\n",
       "      <td>9.326900</td>\n",
       "      <td>10.493760</td>\n",
       "      <td>8.256225</td>\n",
       "      <td>8.177847</td>\n",
       "      <td>10.446850</td>\n",
       "      <td>9.452429</td>\n",
       "      <td>9.865936</td>\n",
       "      <td>10.910269</td>\n",
       "      <td>9.778429</td>\n",
       "      <td>10.338403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ASCL1</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.417846</td>\n",
       "      <td>4.762912</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.711308</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.859811</td>\n",
       "      <td>2.085459</td>\n",
       "      <td>8.160099</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.425030</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.435522</td>\n",
       "      <td>5.553729</td>\n",
       "      <td>2.567229</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.767370</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.478920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ASS1</th>\n",
       "      <td>9.397487</td>\n",
       "      <td>10.356120</td>\n",
       "      <td>8.859190</td>\n",
       "      <td>9.084624</td>\n",
       "      <td>9.431415</td>\n",
       "      <td>10.309732</td>\n",
       "      <td>7.438101</td>\n",
       "      <td>8.386607</td>\n",
       "      <td>11.492383</td>\n",
       "      <td>12.477313</td>\n",
       "      <td>...</td>\n",
       "      <td>7.322312</td>\n",
       "      <td>11.420115</td>\n",
       "      <td>11.488817</td>\n",
       "      <td>4.463897</td>\n",
       "      <td>9.491997</td>\n",
       "      <td>12.288174</td>\n",
       "      <td>10.589305</td>\n",
       "      <td>8.588975</td>\n",
       "      <td>10.102970</td>\n",
       "      <td>12.120552</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 514 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "aliquot_barcode  TCGA-5L-AAT0-01A-12R-A41B-07  TCGA-5L-AAT1-01A-12R-A41B-07  \\\n",
       "gene_symbol                                                                   \n",
       "ADRA2C                               5.353623                      5.034955   \n",
       "ANP32E                               9.583261                      9.108360   \n",
       "ARL4C                                9.958109                     11.198834   \n",
       "ASCL1                                0.000000                      4.417846   \n",
       "ASS1                                 9.397487                     10.356120   \n",
       "\n",
       "aliquot_barcode  TCGA-5T-A9QA-01A-11R-A41B-07  TCGA-A1-A0SF-01A-11R-A144-07  \\\n",
       "gene_symbol                                                                   \n",
       "ADRA2C                               8.948954                      7.585473   \n",
       "ANP32E                               6.086308                     10.008464   \n",
       "ARL4C                                7.288494                      9.284177   \n",
       "ASCL1                                4.762912                      0.000000   \n",
       "ASS1                                 8.859190                      9.084624   \n",
       "\n",
       "aliquot_barcode  TCGA-A1-A0SQ-01A-21R-A144-07  TCGA-A2-A04N-01A-11R-A115-07  \\\n",
       "gene_symbol                                                                   \n",
       "ADRA2C                               4.197512                      5.412802   \n",
       "ANP32E                              10.175954                      9.372962   \n",
       "ARL4C                                8.181195                      9.943622   \n",
       "ASCL1                                7.711308                      0.000000   \n",
       "ASS1                                 9.431415                     10.309732   \n",
       "\n",
       "aliquot_barcode  TCGA-A2-A04R-01A-41R-A109-07  TCGA-A2-A04V-01A-21R-A034-07  \\\n",
       "gene_symbol                                                                   \n",
       "ADRA2C                               1.061223                      4.444813   \n",
       "ANP32E                              11.066019                     10.669437   \n",
       "ARL4C                                8.949372                      9.345226   \n",
       "ASCL1                                0.000000                      0.859811   \n",
       "ASS1                                 7.438101                      8.386607   \n",
       "\n",
       "aliquot_barcode  TCGA-A2-A04X-01A-21R-A034-07  TCGA-A2-A04Y-01A-21R-A034-07  \\\n",
       "gene_symbol                                                                   \n",
       "ADRA2C                               3.688751                      7.507771   \n",
       "ANP32E                               8.792080                     10.828133   \n",
       "ARL4C                                9.395779                     10.203331   \n",
       "ASCL1                                2.085459                      8.160099   \n",
       "ASS1                                11.492383                     12.477313   \n",
       "\n",
       "aliquot_barcode              ...               TCGA-S3-A6ZH-01A-22R-A32P-07  \\\n",
       "gene_symbol                  ...                                              \n",
       "ADRA2C                       ...                                   5.335104   \n",
       "ANP32E                       ...                                  11.234595   \n",
       "ARL4C                        ...                                   9.326900   \n",
       "ASCL1                        ...                                   0.000000   \n",
       "ASS1                         ...                                   7.322312   \n",
       "\n",
       "aliquot_barcode  TCGA-S3-AA0Z-01A-11R-A41B-07  TCGA-S3-AA11-01A-31R-A41B-07  \\\n",
       "gene_symbol                                                                   \n",
       "ADRA2C                               2.696283                      6.970269   \n",
       "ANP32E                              12.717716                      9.483399   \n",
       "ARL4C                               10.493760                      8.256225   \n",
       "ASCL1                                0.425030                      0.000000   \n",
       "ASS1                                11.420115                     11.488817   \n",
       "\n",
       "aliquot_barcode  TCGA-S3-AA12-01A-11R-A41B-07  TCGA-S3-AA17-01A-11R-A41B-07  \\\n",
       "gene_symbol                                                                   \n",
       "ADRA2C                               1.565500                      6.750738   \n",
       "ANP32E                               9.273418                      9.429852   \n",
       "ARL4C                                8.177847                     10.446850   \n",
       "ASCL1                                2.435522                      5.553729   \n",
       "ASS1                                 4.463897                      9.491997   \n",
       "\n",
       "aliquot_barcode  TCGA-V7-A7HQ-01A-11R-A33J-07  TCGA-W8-A86G-01A-21R-A36F-07  \\\n",
       "gene_symbol                                                                   \n",
       "ADRA2C                               7.721103                      6.547644   \n",
       "ANP32E                               8.734465                      9.307549   \n",
       "ARL4C                                9.452429                      9.865936   \n",
       "ASCL1                                2.567229                      0.000000   \n",
       "ASS1                                12.288174                     10.589305   \n",
       "\n",
       "aliquot_barcode  TCGA-WT-AB41-01A-11R-A41B-07  TCGA-WT-AB44-01A-11R-A41B-07  \\\n",
       "gene_symbol                                                                   \n",
       "ADRA2C                               9.522046                      5.137868   \n",
       "ANP32E                              10.252020                      9.942660   \n",
       "ARL4C                               10.910269                      9.778429   \n",
       "ASCL1                                7.767370                      0.000000   \n",
       "ASS1                                 8.588975                     10.102970   \n",
       "\n",
       "aliquot_barcode  TCGA-XX-A899-01A-11R-A36F-07  \n",
       "gene_symbol                                    \n",
       "ADRA2C                               6.085958  \n",
       "ANP32E                              10.109748  \n",
       "ARL4C                               10.338403  \n",
       "ASCL1                                0.478920  \n",
       "ASS1                                12.120552  \n",
       "\n",
       "[5 rows x 514 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "expression_matrix = hormone_expression_bigtable.pivot(index='gene_symbol', columns='aliquot_barcode', \n",
    "                                                      values='transformed_count')\n",
    "\n",
    "print(expression_matrix.shape)\n",
    "expression_matrix.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Normal: 33\n",
      "Complete response: 40\n",
      "Progressive/Stable: 9\n",
      "Recurrence occured: 42\n",
      "Low-risk recurrence: 8\n"
     ]
    }
   ],
   "source": [
    "labels_aliquots = {}\n",
    "\n",
    "labels_aliquots[\"Normal\"] = list(hormone_expression_bigtable.loc[hormone_expression_bigtable['sample_type'] == 11, \n",
    "                                                           'aliquot_barcode'].unique())\n",
    "\n",
    "labels_aliquots[\"Complete response\"] = list(hormone_expression_bigtable.loc[\n",
    "                                            (hormone_expression_bigtable['sample_type'] == 1) &\n",
    "                                            (hormone_expression_bigtable['initial_response'] == 'Complete response'), \n",
    "                                            'aliquot_barcode'].unique())\n",
    "\n",
    "labels_aliquots[\"Progressive/Stable\"] = list(hormone_expression_bigtable.loc[\n",
    "                                            (hormone_expression_bigtable['sample_type'] == 1) & \n",
    "                                            (hormone_expression_bigtable['initial_response'] == 'Progressive/Stable'), \n",
    "                                            'aliquot_barcode'].unique())\n",
    "\n",
    "labels_aliquots[\"Recurrence occured\"] = list(hormone_expression_bigtable.loc[\n",
    "                                            (hormone_expression_bigtable['sample_type'] == 1) & \n",
    "                                            (hormone_expression_bigtable['recurrence_status'] == 'Occured'), \n",
    "                                            'aliquot_barcode'].unique())\n",
    "\n",
    "labels_aliquots[\"Low-risk recurrence\"] = list(hormone_expression_bigtable.loc[\n",
    "                                            (hormone_expression_bigtable['sample_type'] == 1) & \n",
    "                                            (hormone_expression_bigtable['recurrence_status'] == 'Low risk'), \n",
    "                                            'aliquot_barcode'].unique())\n",
    "\n",
    "for k, v in labels_aliquots.items():\n",
    "    print(\"{}: {}\".format(k, len(v)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Initial response"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(105, 82)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "select = [item for key in ['Normal', 'Complete response', 'Progressive/Stable'] for item in labels_aliquots[key]]\n",
    "expression_matrix = expression_matrix.loc[:,select]\n",
    "expression_matrix.sort_values(by=select, inplace=True)\n",
    "\n",
    "df = pd.concat([expression_matrix.loc[:, labels_aliquots['Normal']].sort_values(by=list(expression_matrix.index), axis=1), \n",
    "                expression_matrix.loc[:, labels_aliquots['Complete response']].sort_values(by=list(expression_matrix.index), axis=1), \n",
    "                expression_matrix.loc[:, labels_aliquots['Progressive/Stable']].sort_values(by=list(expression_matrix.index), axis=1)],\n",
    "               axis=1)\n",
    "expression_matrix = df\n",
    "\n",
    "expression_matrix.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0\n",
      "17.9246517870353\n"
     ]
    }
   ],
   "source": [
    "print(expression_matrix.min().min())\n",
    "print(expression_matrix.max().max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "grid_kws = {\"width_ratios\": (0.40, 0.49, 0.11), \"wspace\": 0.05}\n",
    "fig, axes = plt.subplots(1, 3, figsize=(20,5), gridspec_kw=grid_kws)\n",
    "\n",
    "for label, ax in zip(['Normal', 'Complete response', 'Progressive/Stable'], axes):\n",
    "    ax.text(0.5, 0.5, label, va=\"center\", ha=\"center\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "grid_kws = {\"width_ratios\": (0.40, 0.49, 0.11), \"wspace\": 0.05}\n",
    "fig, axes = plt.subplots(1, 3, figsize=(20,10), gridspec_kw=grid_kws)\n",
    "\n",
    "for label, ax in zip(['Normal', 'Complete response', 'Progressive/Stable'], axes):\n",
    "    sns.heatmap(data=expression_matrix.loc[:, labels_aliquots[label]], \n",
    "                cmap=\"Spectral_r\", vmin=0, vmax=18, \n",
    "                ax=ax, cbar=False, \n",
    "                xticklabels=False, yticklabels=False)\n",
    "    ax.set_ylabel('')    \n",
    "    ax.set_xlabel(label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig.savefig(\"./plots/expressiondata.hormone.initial_response.png\", bbox_inches = 'tight', transparent=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Recurrence status"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(105, 83)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "select = [item for key in ['Normal', 'Recurrence occured', 'Low-risk recurrence'] for item in labels_aliquots[key]]\n",
    "expression_matrix = expression_matrix.loc[:,select]\n",
    "expression_matrix.sort_values(by=select, inplace=True)\n",
    "\n",
    "df = pd.concat([expression_matrix.loc[:, labels_aliquots['Normal']].sort_values(by=list(expression_matrix.index), axis=1), \n",
    "                expression_matrix.loc[:, labels_aliquots['Recurrence occured']].sort_values(by=list(expression_matrix.index), axis=1), \n",
    "                expression_matrix.loc[:, labels_aliquots['Low-risk recurrence']].sort_values(by=list(expression_matrix.index), axis=1)],\n",
    "               axis=1)\n",
    "expression_matrix = df\n",
    "\n",
    "expression_matrix.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0\n",
      "17.9246517870353\n"
     ]
    }
   ],
   "source": [
    "print(expression_matrix.min().min())\n",
    "print(expression_matrix.max().max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "grid_kws = {\"width_ratios\": (0.4, 0.5, 0.1), \"wspace\": 0.05}\n",
    "fig, axes = plt.subplots(1, 3, figsize=(20,5), gridspec_kw=grid_kws)\n",
    "\n",
    "for label, ax in zip(['Normal', 'Recurrence occured', 'Low-risk recurrence'], axes):\n",
    "    ax.text(0.5, 0.5, label, va=\"center\", ha=\"center\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABHEAAAI/CAYAAADwakVnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzs3XeYXlW9//3v9N5bksmkQhJIQi9BVKSIiEgVVAQBERQVFVGsgO14LNiOCCqCIoII0lSKFJEmEClSQiA9mUwmmUmmZlqmPX/8Ls51PUefx7zXbxK5Oe/X3/c3+957r73W2mvufFbW+Ph4SJIkSZIk6bUt+9/9BSRJkiRJkvSvuYgjSZIkSZKUAVzEkSRJkiRJygAu4kiSJEmSJGUAF3EkSZIkSZIygIs4kiRJkiRJGcBFHEmSJEmSpAzgIo4kSZIkSVIGcBFHkiRJkiQpA+SSD59x/HXjO+qL6PVta2UBrintGtoB30TS/0bX3n561kT8O3Qc7C/Lx8co7t2GazbMqsQ1U1Z1oc931RbjY1Ru7sc1r2XNc6pxTdOyDvT5ndVmtHOkzH9yhsdwTfYYn6IXDIzgGmpnzf/aG8twTV1LL65payrHNfXNPbhmZ0gZN8o7BnDNSG4OrkkZO9bOq8E1AyV56PPznt6Ij5Hq3zVvyQRDRWgJ47/tjD7vVSljeUTEzde9e7vuu7/EkSRJkiRJygAu4kiSJEmSJGUAF3EkSZIkSZIygIs4kiRJkiRJGSAtFUg73M4IqdyZUkLqaNhYRERvdRGu2ZmBcynn9P5vDOOamz+BS2LZ3g24Zs6zm/iBEryeggT/b6S0n6I+3n5ey0HkKd/t3yklcHblwjp+oHr+N5nmXBbSmz+fZyPmz9g5gbsvPMOv2cJ92nFN7ipcEtNOZqGeL/8XP0bXe3gf2bKmFNfUN/Mg2NFc3jYnre3GNfQatL3C5ws7a8xLkRLY2X4Sn2v2dLE+uGwZD8JNGU9+9V3eZk65pBbXjMzgr06rEwJODzy2E31+6bW8PXfX8pqUYOM5H+T389J9eZB+73AzrvnaMywQe9NKPgf57TUVuGYiLX3jFFzT9OIWXNOT8A4WEVHZzkOs+47k7SMiYt4ufAx7/F7eT0RETNtra1Ld9vKXOJIkSZIkSRnARRxJkiRJkqQM4CKOJEmSJElSBnARR5IkSZIkKQMYbLwTrJ1Xg2tmHsGDlzbdxsK5IiLqWvhxmuewMMyIiKwxHog59/iEYNoreeBa63H8fCIixl8YwzUpAY/f+zEuiYpqHhL2x89uxDVn3MJDEWtu5uGDQ0Wvv66qp7oQ14xlJ4TXJjyvlZt3ThBlSmB1ce/OCVD+/9LeyPrZlLY7+wUeuJvilUMmo8833daKj7E0IaR5+iIeBjj3If7dtj2ES2JydOCaZxezAOFrfsYCTSMiPvNeHiBdGXyM/c5veFB6XRu/Zmd+ggftV97Izue2D9+Dj/GeZYfgmjd8il+z5y/h84vv3shrPnJ2H64pKBhEn//29/mmG2d+bxquufA9bbimtJ6PJ1X38vacYtUGFqB77U/4fP49X+DjU8pYv+znPIT2o9/k89iWY5pwzfdOZGPHd7t4SPXJH0zbaOO2a5LK/sFuj26YmH/oX0jd4GLaF1n4f0TEuv9I28znso8txjXrL1yYdKx3f3dGUl0ct30f85c4kiRJkiRJGcBFHEmSJEmSpAzgIo4kSZIkSVIGcBFHkiRJkiQpA7zm0kKXHzoJ1zRM5sFsERFtS4pwzS7P8eC0oj4eateTEJxV1MeDDVNCl/trWdhaRMTkqTyoct1aFgYZEbHb+fw4M7N40G5ERO+uvOaUWbwtdA7xwK+f/IAHxj6TEGabElJceCE/n12DH2fwu7gkSj7Nv1vfZaP8QBHRvYi377LFPOgvRX0zD+Fb8oZGXDP/ry24JiUMeSLRfrazngc6vvmb/O8rD3+Oh5rSMOCDvsG/16ReHgL69ibeT367sQLXXPXtTbhmfR8fmz9/OwsxL8rlmxSMnJOFay45mAdDPreFn//Uch6Af+i3eFt78LPsGTj+1rfiY3z3ii245nOf4G3zit/wPvjzT5bgmqkf4/3G6qvYOPlwK+8Dv3ou32jhVx/lbeYHl/H7+ZX3800JUvr0v6xhz/RTCXn4tYfy94aVM3no+MXH8/v5g2/z5+asQ3go/Kb+PFbwEf4OtEcNH9Mm0h5f5e3vjkdrcc0Zh23GNRERyxJew75yHX9niYjYkrcPrjnlSh6YHRHxyffxOQbhL3EkSZIkSZIygIs4kiRJkiRJGcBFHEmSJEmSpAzgIo4kSZIkSVIGyBofH9/uD59x/HXb/+FE7SfxALyIiGP35SFwj3+BB7ptreSBVjnD/DgpYcjtjTwMsa6lF9fsLAMlMGwsIk78SlrI7NJuHgo5s5Q/DmX8lOLKP1fx4/w1Iez7KP7lym/g7WfLZB7oO7KIf7eG23jA3dlX8mc1IuKSH9bhmukv80DFrloeEDmezdt2Sj9X08pDxYt7eahiqmtvP51fiH/i92t+jR78Yyfz53c0n4fHnngOHwPfceEgrnns8zA89vv8sv/mS3zPhZQxs3lONa4pXJgwb1jKz+ehc15Anz/zgsn4GClWLuR93T2X8PHooyfzEPdfXPQ8rjn4PhZu/OjuN+BjfOD6w3HNzlL2GR7mv6CKz30G4dSscxvvN9Z8jc//6j/Hz3/Ver4pyu8OXItr3vHNGbjmpPPZsza1hF+zG87n92bFnvW4JmUjmZRA9DPGVuOady2Zjz7/u32X42NERGTNvGhC5i2Hf/qP+KEtGBjBx5l3WtrmG7PLeZ+y+Itp8/Wd6aQfpN2+Y2ectl2Fr4tf4qQs4EivSlnAkV6VsoAjZSK6gKM0dAFHaegCjtLQBRyloQs4SkMXcKQd5XWxiCNJkiRJkvR65yKOJEmSJElSBnARR5IkSZIkKQO4iCNJkiRJkpQBdujuVD3VhfgLlXfw3TJSjpN6rJ1lwyy+S9eUVV24Zu28GlyTsqvOzpKyo1XKriY7U8oOKhWbeUJ8bsIuaim7C/WX8Z13dtYuRim7MpV2De2AbzJxOutLcE1VGw9ITNl1LGVHq1QTtTsV3eWhehO/lj3VfPeT2Sfza9nezsbOvvv5331Gc3lNSj+et42np6bsvtEzj+8QN3d+BzvGL3hTLU5o3isfKcc1Dc18I4mhIn4/U9rA0HQ2tsx+YBM+xpHf4e25LI/vvNI7zO/nHy7h323F3g24pqaF9TWHf4DvYnnT3ZNwTX0zP05pF38PWDO/FtdEPr+fTTPYs9a8jO9Qe8xRfAeo1R18R68XnuLXrA62s4iI8iP4ONDxEOtrphw1gI9x6PS094wz577v37Y7VXctn4fUN6btTlV3C39/nXExb4cREc//mI8twwV8h8mIiMZTeVuJiPjpm07937M7lSRJkiRJ0uudiziSJEmSJEkZwEUcSZIkSZKkDOAijiRJkiRJUgZIS+rZTt2LeNBl+V08ZCw1oHjFnvW4pmIzDylKCU6rSgjDXHtUHa4pWsJDWV/L4a87M6Q4JXgxRdMyFooZEbFxegWumbS2G9ekyD2Fdztbb+HZbmPZvGawhIcub5nE+7mItPDgsezX7rp70VYePp3Sl6QENU+kqSs60ecHz+Lft+8p3g43X8nbRl8lq6lr4cGhKfcrJXA45XlKCY0f7uL9St9VLFNy/Tz+vcae5N9r3gsbcU2K8tN4sO/Y5Qmh5yvYx1PGyet/wDcASHtueIB21QB/BmhIcUREcS+bzy25lF+zXYp44G5Kv5Fit8WtO+U48Sj7+G7B7+XKxbgkIhJC5Pfk96Y+IUQ9fsFLxo9hG7y0/YHPWx5sTnsHOvP2pLJ/0DuJhxTPf3TDxBx8O6RssjN0Wdo7S/kAfydP2bgjIqLlBn7dIyLiTdv3sdfuG4EkSZIkSZL+m4s4kiRJkiRJGcBFHEmSJEmSpAzgIo4kSZIkSVIGQAmjPdWF6B9vumsz+vzOtstzPDgtJYgzJWw3JTR3+j3tuGZnaW8swzV5Qzw8rXJzP65ZuZAHQkdEjObyNdA5z25KOha1s0KKm4+uxTVNV/N+gce8pkkJNk6VEsie0i/szLDvnSEl6Hsi0TEg+3f8+s/u5X15SjBg9hgLnG09jgfuTr7j33u//v/srLZEA5TrEsJmU4Kdd5axy3no+c7o61LGyQ2zKnHNsr0bcE1K2xwq2qF7lfw3OnalBIinhNqmzDNTNh4ZzeNzv65aHlRNQ/RTpLTn7loe1poyn99Zm6g0/nELrsk09St3zvOUEuKeWtdTnRYanPodU9S0JgT0A/4SR5IkSZIkKQO4iCNJkiRJkpQBXMSRJEmSJEnKAC7iSJIkSZIkZYAdmoJGg5AjIlp25wFo2TksnPFVI8N8DWv6SzyUdfCsElyzoImHrd37GA/P2zaUg2uaZvCArM1reEDWh47ZiGuWdfPzqRxNC7ccTWh2z9fV45r99uaha5OK+ZebWYpL4o8X8uchJQz5LfvxgL+/vlCOa3pX8TDNqjYeph2RFhK75248iHPpygpc07mMBwrmTeZtLiWMfckbGnHNRMo/ifUxVVU8aHFoG2+Hu+d34ZoTZoygz29NGDP/Nj+hTy7gbemiufx5/+4yPpbdezUfy849l92b3z7D5wwbH+PP+de/xMeWn7zAwyQ7fs3bQP1ZPBD8+T+zOVDKRgOfu5DPF5Z386n2rhXs2YyI+Mli3jbnNfDgzTXr2LP21nl83HrkKh4EXHLEGK45cDYP3L17Me9rKiv5ODByKHvW3rU7f294ZjMPec3L4TUbE6ZJR03l93N0PAvXfPPXrN/Y/SDeB3x4j7R54kQ55nx+/Dr+Ch9j42m/DfnxVbzv+vLH0zbz+Y/L+Nz7TaelhSFPK9mxv5XxlziSJEmSJEkZwEUcSZIkSZKkDOAijiRJkiRJUgZwEUeSJEmSJCkDZI2Pb3+A4AkfuBmlDfaX8XDMSWt5ANrO1N7Iw5fqWtICkXaGtfNqcM30l3kYovR6lfIMRfgc7WzX3n46Tzz8J844/rq0JP0dbKCEhyFvreTJhTtjPOus58G+VW19O+CbTIyUTR7KO1hIaX9ZPj5Gce82XPN6Q9vaa7md7Sw7oz2/3myZzHeOqGnlodPaObZWsvfb0i4ebB3x+p+3/N8YKkrbm6lggIfG72zbe9/9JY4kSfqXXst/kHg9+d/+wruzpCwWirM96/WELuBIO4qLOJIkSZIkSRnARRxJkiRJkqQM4CKOJEmSJElSBkCpQDSoOCVocfXutXHBOTzw82ff44HDx53HQ8Pu+zYPAtz7a3ytbFYZz6AqyRvFNZf8LAfXnHXFGK7Jzebnc+NKHtQYEbFHNT/Wogb+f7aXd/Fwv2u/WYRrLv/eZlzzgxd4iN6cCn7dHv8Cbwsfu4qHiq3s5m3hslsn45qIiE+f2AorOuPX3y3Gxzni27xfWNHLM+7aB3BJHDo5pf/hbeHqS/j/Lc89IS3MbqLQ+7a+n9+z8jx+/acU8+dq7VYa0pod+9UNo4pXuvg8oLqgH9ecMH0Krrl62UZcs7id38+zd+Hn8/ctrJ0/vpo/F+/ek89nOod4vzXIpyaxcYBf5zPn0Mym/tjQz8aWlLFoZUK/3cqbTJwxh4enNhSx5zkiYt1WfkP/1Myu21um8O81rZS355Y+fj+3DvNn4B3Za3HNhevrcU0J7AY6E7LNfzStDddsquH5U999ntcc0cjHweEx+nwOxTE1bM73qRf4u9ZEOvVHfE6xRw2fPN66mr97RERMKU6Y85SkZW3duprPOzsG0/Kl3zyZz4uJ19wvcXbWAo5en3bWAo5en/gCTiQt4EiZiC7gKA1dwFEauoCjNHQBR2noAo7S0AUcaUd5zS3iSJIkSZIk6R+5iCNJkiRJkpQBXMSRJEmSJEnKAC7iSJIkSZIkZYCs8fHtD4I96Cv3otTY4SKesnXsEXy3iOGEnQ8iIhpLeAjug8v57kJzpvDg3MdursA1xXvyCzFnFt+ha1tC2PaKFeW4pqKSx+YX/oLuthLR9R7+3SIitrTztlBYxJPzx5bxdpqyM9y0vXhbKK/ku2EsrOEN6LGV/FqnfLcZ5fxaP5Pw3SIiDtyVbz/y2I/4jmiVm/lxlh7Ad/YqruZtuyLhHh2b0E4jIs6e97607QX+h/2++QBqJLWT+Q4Pm1t5m5r/xAZcs7WS7dKQM8yf3Q2zKnFNWRdvF/XNPbhm9e61uKZ0N97OB19gfy/LG+Jj+VDCfGv6y3wjiS0n87nJ5jYeBHrZe9fjmq/9ZzX6/HAB3zGmrIPP50pP4ONKSzPf5eXKd/Nr9pm7+A5IIyOsPW9dwdtmXQvdaSyip5r3m73VfExtWtaBa0aO4jviDD/C2k3NO3nwfPZPeHteO68G1ww28HDr/ALeD1ZV8/OpvJGPHdTG6bzfjIj40w+PnZB5y+Gf/iPuhKau6JyIQ2+XJYv47pJVbQlb+EXEXqfwvmUgcZ1h00/SdiW7+bp3b9d995c4kiRJkiRJGcBFHEmSJEmSpAzgIo4kSZIkSVIGcBFHkiRJkiQpA6C0MRoCl5uXkICbYNkGHkwWEXHsG7pxzTV3TcI1n1zEQ7Pmn9uFayYV88DFL/++Add8/h1tuKZ6TjuuSVH0Jt7mppXyEM2IiKte5gGrb53KQ9fW9PBAvKmlPBR6dcJx7lrDQ7sOrOffrSkhhPyBDXyN+pDJ/Lv96XEe8hcRMW8/HsJ96H/yvuTxTTxQ8MN1PCBzVQ8P077paR7emTMhMX/pPnsyC99f2s2/cP3+fGz6fXMZrpl1KgsG/N4iHlJ82JdxSUx9Fw8s7PgF77+21fE2e+RePNT0oLexfuUT1zfhY5x42CZcs+RSXBIXHcbDLq9dyu/n2l5+b6jsOfzZ7B4pwTWHT+cB0nc8yOe1U0p4f9owmd+bV5awAOnz3s/nf1f8moeO/+j8Flwzp4IHzl70JJ9rt/+Uz2E6m9imG1cczOfmH2nnmxjEEl4yuZHPl7uf4n3Avov4vGrwC2weu6iev2d89Ubeb0ykqUcmhACv4CUrF9bxooj46ql8Y4Yv38TDkCMilq3h9+9nx6W9w77y7R07jvlLHEmSJEmSpAzgIo4kSZIkSVIGcBFHkiRJkiQpA7iII0mSJEmSlAFQsHFuAQvmyurh4UGP/4KHP1UfNYxrIiI+ehUPD8wt4ef0X4t54GR+wSiuaazkIX277sZDCr91Nw+uOupAHgZ579+qcE1JaVpbSNH2Cg/5vLuRB+I1TuOBcBvvZIF4ERGVh/Hv1trCn9ePPlWPawr6+H2t2IWHFN8w3otryl9ICIyLiB/k8kDkmbvwwFsaQhkR8WxCv9CxmQdx5t7DQ8XvyE4LCDxzblLZP7jxayyorrSLn+O9CeGAt17GgwFP/AYLBjznOzzcbzauiGidx8NGz/8a3wzgisd4COazF/M5wC4/RFOtaHqGB+HeVcT7k9t+y4/zzq/ztjn7Bd5uDriWbwLwlm+wsfKCc3k7e/fX+NzshmeKcc1vvsZDaj9+Cg/R3DidX4PcIjZXOGLqAD5Gzumbcc0PP5gSIsrH7z2+yv/+PekC3m46bmA1HUOsn4mI+NghvN888GQe0tzSz+dw372Bb8owp5zfmzuWsrnLWXP5+0zTMl4zkZY+z8eHYy7hYfk/XsjH1IiIYy7m3+/8D7BNJl516zd5uzptMd8EKCLicxfu2PvuL3EkSZIkSZIygIs4kiRJkiRJGcBFHEmSJEmSpAzgIo4kSZIkSVIGyBof3/6Aqjdf8SeeZgU1TefBoi3NpUnHKlrCAyezDuDrXlvW8MDP/Q/jQYBPLeaBsbm5PKRx+qweXLPqKR60O2Mf3hZqqvk9XbMmrf1U1fBjtaxLONYKfo8GSnjAX94QD97rL+cBYfOf4AGs2R/jx3npcR7om8Obaez6YFq42pI3NOKafQ/ix1q9ggdX1tTyUNFNrTy8M8VQS9rfHh77ytt48vs/8eMl16Nx8LcP1eJjpIyDa5/gfUvTASyEcJdJvM/r5PnisaWjANc0r+EbCBy6iIen9vH898iDTbZjkDfVtoTnr7aeB842r+XXeb/5PJB96Rp+PuU3sOdm5eE8rPLYN/O52V+e5wNLyiYNubl8il5cwo9TCzf4ePxpHlra38NDej94LB8ftwzxZ20woQ945Gk+H2mawdrzlnb+rtHTxfva8ko+DqS057VL+XPzxkN5IPjzS9gmJSnn8vY9+AYlEREf3v19EzJv+cLiG3DncP/DvH+srOFzx4i098rs7LQliZT3sIrKhIlMRGzbljZfvfmEE7frvvtLHEmSJEmSpAzgIo4kSZIkSVIGcBFHkiRJkiQpA7iII0mSJEmSlAF2aLDx5EYWmhgR0dpSgmvy83kga0REXQMP9evu4gGr83bhIZUlPNMtagp5yNPv/lKHa3bdrRPX9CRctxRDQzm4Zo9d0wLHnkkIXdt/dx7eVZbP7+vGfp6FtqmD36Pdp/Cwr1WdfO04JUwzJbT7fYv4/bn3M/w4EWlhzSnhrUmBcTm8zb30Zx7cWDaLBwRWJYSXR0T85p0nTUhAIB0HG5t4/5IyDqZ45xu24Jo7HmVBzWOj/LLn5vFnamSY9yvDXfy7HXsMD7a94wEeEDk2xr5bcULYZoqx1h2+v0VERMw9tAvXvPRIQpg9LNmWMMeobejHNVuX8klgf+XOCanNW8Lb2l6nsn7wgYcm4WOUlvP5yNAm3m9Mmc/fbZrX8Pnibo/yTSB6qlm48cAhRfgYJQlh2N0JAcr9fXyDjrqEZ61xKr+fL79UhT6fci4REU9ccMS/Zd4SETHYn/Aimmj+XnyTgdRruuZ5Po8uqktbZzj5TXyOFRHx0fnbF2jtL3EkSdK/RBdwlIYu4CgNXcBRGrqAozR0AUdp6AKOtKO4iCNJkiRJkpQBXMSRJEmSJEnKAC7iSJIkSZIkZQAXcSRJkiRJkjIAip4uLBpB/3jXn3mydfkBPCm/5uZuXBMRsTUhyX/rFH5OKx/ix6nczBPW792Xp/nnjvBdQHp+zUMXS7t2TrBdaULNk7ukhZRNXcF36Vp3Gz9O13v4rgaVN/IdibaV8fT1x4/iVzwltLPsIb57QGkX70vuvQWXxJbJKa0uon8VDyGsfYj3de2L+S5YXXvx3ZGGSvhOAcMtfKeXnoRdLibSzHvb0OdXH1mPj9Hfw8eZmXN527jz15Xo8yON/O8+uz64EdcsWTQF19Q3810g+xfyZ/DuW3kabuEo2zWnqI/v/lJ5GJufRUS0PFWMawoG+HG66/lxeq/GJbHL6ewZWPFkBT7G5Ga+a1ZvOT//mS/x3Vpeq1Y08zGy6WW+q8v3f8jnZO+6chdc05kwpkYl39Xt7Zex/vbuTw/iYwznJuwa9RAfB951Hr83p8zm16woh+88dMVL7DhP9fD7/+SpS3HN/3FEYt3/W/kL/J2yboCPQynz7oiILcv4nDPniLTfoYwX8bq6xzqSjnX7Mv7+FhHx0e9t3+f8JY4kSZIkSVIGcBFHkiRJkiQpA7iII0mSJEmSlAFcxJEkSZIkScoAPD0R6KnmoYEf2b8d16yfz8MxIyIObhjANVefx0ObLriah0MV5fDA4b6RZlxz7s+n45ovXsZDihtL+Pm09PE1xl3KeYDetjEeoBcRsbKHn1NJLq+ZXsZCMSMinngjD4SbW8mDUS8+jX+3S3/Fw/dyT+Dn87WnecDdefN5n3DelWnBxh8/lAXkRkQsfBe/dn3DvGZ0nAdjp/j+53jg59yP8GDVf6fPHcPv8xhv7vGrL/Bg6ao8di23ZPNQ8v4yHgKZX8T7yRTHH8z7/nfP5uPfpx6oRZ9vXsxDHr+8iAcvlr0pIZx/K7+fu1a04poUH7ukAX2+aTO/Zm/8PN8A4Lk2PgfsOJWPX7l5/LkpLuH96cY72dz+pPk8dPzuu3g7e9ePZ+Oa+U+04JoD/oPPTWeW8Xtz+UMseL7qHP5+snc9788+fhI/zitdfOy44N383e4jP0t4puGeMGedz/uzg66Yh2siIp64IKnsHxQczmu6nuDzs+Y5PPg/IuK6j6zBNT9awr9fRMTAAF/6KO7l7zoRER/+Jp9/E/4SR5IkSZIkKQO4iCNJkiRJkpQBXMSRJEmSJEnKAC7iSJIkSZIkZQCU7pMLQ1nnPLsJfT4iYpcP89C49f08AC0iYmoJDyraWlmBa3q28UCzadX8nFq38GtXWcsDyhqKeEhfblYRrplRxtcY+0d4OO+NK9PCsd67Cw/B/VtbAa4pz+fhk4c3zsI1g6M84C6CBxYOjfKAu7J8HkS2bx1/7p5q5/dn3tMbcU1ERNnp/DrsUl6Da9oGeLBu+yAPyS3P4/3Pxum8P909Ky2I/N/lkhum4Jq5C/kzP1TE21Pl5n70+dIuPl6kqF3C+5Wqtj5c8+zFuCSePJuFFEdE5F/NrvPsYJ+PiPjlMj5n2NjCA5RTQiFHlyQkdSeYupY/N9TTX+I1Wyt5f9pTy6/zeELweOmRfD43aS2bZ930GB+3Gjt4P1/fzPuNFItb+P1cfAXvOwtOZWNq7lV8TvrELBaeHBGx7j/4WL+zfPH3LNw8IqK/j93PkxI2pZj7UGK4+wQFG2/ZzN/Bpq/gGw2l+u1R/D2s9ztp7bComj8nqb7y4/qkusO/sX2f85c4kiRJkiRJGcBFHEmSJEmSpAzgIo4kSZIkSVIGcBFHkiRJkiQpA2SNj29/4NzR1/0BpdONPsXD7MrfMoJrshOXogYHchKOxc9p2lQeUvjs33kQ3Nz5HbhmMCGkcNsQv27dXTx0sXcVD4/b5WAebNyT8N0iIlpbSnFNxRreFgbm8rDd/Hwe+DV7bheuSWk/Kdr+wK9B326FuGbvvXiY4uN/5UF6EWnXu78vIVQ0IUi6pJT3w8UlPCBzbIx/t02taUHk95xxDD/YP/HmL9yNBoHBhJD6PffnwYkpXnimDn3+bW/m36uX7x8Qy1bxvrXrZX6dG/bkAYf5BTwwfWYDCzV9fjk//5RntrJ6ENcsfYHPTaoSjpNfwMeweVPZcVa18TYzpYY36A1bEkKnH+Tj14Gn9ODDWfnNAAAgAElEQVSaxc/zYNuSUtbXV97Iv9fYh/n5FxbxZ+C5v/Hg0aYX+Tyh+D18+FnxJAv+L5rMn5mFC3gY+PoNPCT3sN343PfaO/jcavIsHnDfOJXVtKzngfBnHpgWun3CzNMmZN5y1v2/xS+vJUW8PS1+PG0+fNzhfCOkB5/mfVdERNlfeRvpPKAs6Vh1DbzdR0TcfMKJ23Xf/SWOJEmSJElSBnARR5IkSZIkKQO4iCNJkiRJkpQBXMSRJEmSJEnKACglk4b6Vh3JA+Cm1vKanMTYp/YBXrjqDzzcaM58HobcePBmXFPOc4BjcJRf7+KELNvHW3jo6fRZPFTrjQ38Whfm8O8WEfH7cl63/1v59V7SuRXX7F3Dr0PbIH8eBgt4kODscv7d7jmMN7pptTx88IQZ/J6u38BDGyPSQhgn17KA1IiIhx+ZhGvee1Qrruka4u1nQy//O8L0WWnXe6LQoOIPvmMjPsYwz86Np1r5MzIyzK7/zgopTgm8PvZd/Hn/2yoe0Hncbjyk97fPsCDMugYeuDy1go+XL67mIeEL9uDXuXcrn5zQthkR8fJ6Foa7enlCMGbCBhIVFfzBGTskYTzmTSDm784D9rcOsnuz59f5vbzzKr6ZwSnn8evcMYOPJ+V78TlM+yYe1EyDisvK+flPK+Xn8vgy/h70cB4/TmkXn+/MmMZDa+m701A975+/dTufh0VEnHBBUtk/SNmEZOUy3j8Ww9DzV43y5pHUd0VEPN7Fw5dnN6Yd68y9eHsk/CWOJEmSJElSBnARR5IkSZIkKQO4iCNJkiRJkpQBXMSRJEmSJEnKACjpqAwGuW6+lwVARkR07sfDzCIitg3l4JqUYNHqg3lw2B2LeThUUcJ3632GB1eNzeLXrbSMX4OSUn4+KaFalfk8dDEi4tmXKnBNbi5PH11SwkO/nnuqHtfEfm24pK2Vh1w2r+IBd5UJ4byH78dDxZp5HnR89Y8J1zoiqqp52GlLMw98jSZ+UnMTgjgffIHf185lvO/O4flySf1PRES8Ka3sf2p6kfUxP4+0QEPqsyfyAOWLD1iJay74VDUrOBQfIupu4c977zw+lnV38TZ7+bW8n/zcB9txzbfvZH0Rv5MRnz464Xv9gfeRdc9245rWhXxM3hOOe9W1G2MG7OoObuBj+De/zs/loi/ya/aTC/mcu+kTPFF0/R3sGVgfERdd1Ilq3ngx39zjS5fCvikiNjfycbimlYeV5ozw1OmsEvZ39q2RE1ddwjYl+NCX+fhUvjsfg/ML+Pk3LeNzl/lV/LcJN19Vjms+fQFrzzf8F98sIiIiJijYeMtmHqz9g1NacM3PX+bHiYj48038fe/k09k9eFXjW/lc6aY/TU461s9v4u++ERFvu277Pve6+CVOygKO9KqUBRzpVSkLOFImwgs4SkIXcJSGLuAoDV3AURq6gKM0dAFH2lFeF4s4kiRJkiRJr3cu4kiSJEmSJGUAF3EkSZIkSZIygIs4kiRJkiRJGSBrfHz7k+lPv+dmFGO/9mGe+n71Z9fjmtZ+nsgfETG/KgvXfOo9vOayG3k6dVX+FFzTuW0Drjnj5lpc85t382T+4lyePD442otrinJ4yvzgaMI2RhFRlMuPNTDSg2tKc3mY6LaxAVwzOs533TjmczxU/Pf/mbC7WW4Vrrmrmfcl+9Xxa/CZ9+bhmoiI7/+W9yVlefx5zcniu9b1DPNda7KCn88HzuV/R3jPxfweRUScPPt9/Av+E20D16BxsL5oJj7GtlH+/G4d4Tt5FOeyUPctg3yMacydimvG8vmuUVmd/Lttq6jBNQXj/HnqHWf9/oY+vjPR1BI+TpQM80diUxbfAXJVD79mtYX8OZ9ZNg19/oqX+A5IH5/H++A/tW7CNW9rasI1a3qX45q8bL47VXn+jg84Hx3n8+b87KKE4/B2VjbEx637uvi8+aCGEvT5ghz2+YiIvCze167r4+3s5lX8u715Mp8rzq3gxxkYZf1zQ/50fIy/dzXjmoiIvWo+MCHzlr9suBY/6G+ZxMfupd2rcU1ExLzKWbgmazRtLtg3zp/FkuzEFPysxN/KZB++XffdX+JIkiRJkiRlABdxJEmSJEmSMoCLOJIkSZIkSRnARRxJkiRJkqQMgIKNz7r/tygYqX0TDxkrKBjFNW+Y049rIiL2r+PBadcu5SFgC+v4Oe1Vw7/brhWDuOb6FTx8urEYl8T86iFcMzzG87zGeEZflOfx+xMRccMKfiH2rxvDNbPK+bW7eTV/9g5MaKdDo/wevdzN144Pm8KvwZZBHqZZns+vwZf/WI9rIiKOXdSJa+ZX8X5hbHxCcvH+pdW9POS6fZB/t4VV/BmKiDh2xmkTciFOvu1W1Mu8dR4PKd4yxL9qThbv/G75PWu7+x/GA6//fg8PQT3tvTw895d/asA1FWv43GHR+3kQfg68nfMr+b0sTRjHnmznfeRdD/H+7uTDebBvyvyMBig/0sL7rJ5uvpHG7tP43GxmKW8DT21O6TdwSfTDQOy6In4uwwndfFtCgHbn73jNB7/A+43lPfxCtw2wmo3d/Fw+sx/fQOTSP/M+va+PbwBRVMT7gMqEd42TdmUBylNLeaDuVz6TFox72zUnT8i85eFWHmz8wIa0TTtSDPJbHfUJ/UpExC0P8jHsM8fyMSwi4kf38c0TIiL+fO7bDTaWJEmSJEl6vXARR5IkSZIkKQO4iCNJkiRJkpQBXMSRJEmSJEnKACgFKzubhQht28ZD435+Ag8PetfV03FNRMTDdTxwsm8rD3rKzunGNYc18pSncy7iwY6bZ/KwrUPeyO/R2DgPhF7axfO85lfxoKt1fWm5YSfM5GGF17xSiGvmbOU1W/hXiyOn8rCvd9/KA/4GBnj43sttvCYlWL0qIRQvNy8taPfx1fy+Do8l3NgEL23mfXdRAb8Oix/nfdbiqTxYNiLi2BlJZf/gt8eXo8+PB+9jb1/dhmt2reRtI+c4dpyUsNkjz96Ma95ZyoImIyJO/PA6XLO4jY/ne9fyPu8Lj1Whz188i5//TzfwZ7Ywh4+X1561BtfMGeDnM1A3Bddc9RwLHP3FkexZjoh4vqMF1yyomoRrfruSB99fui8fvwZGeEjr3zezTR0OqOdz7UljPDz3vq4+XHPEJ9bgmufKpuKanzxSh2vufCd7d+gu5uHmhTl83rf7dB6GvK6Tt7Nfzt+Ia7Imz8c14z2t6PPvf4KP6cdcxPvAiZSyacd9j/I++PZzm3FNRMTKHv6bksaStGv6qaoluKZ/0rSkY006Li0QeXv5SxxJkiRJkqQM4CKOJEmSJElSBnARR5IkSZIkKQO4iCNJkiRJkpQBssbHtz/Y7iOP3oBS8DZt5MGixaU8ODFVUzUP2mrtSQj8LOSBUkdN5SGhN7/Mw4Mryvg1SFHPm0LMKeehi10JOVc//z0PHYyIOPIwHlj1/PJSXFNUzJ+J3aclhJwm5Dvf9RAPxTvsYB7aOqeCt4W+hK7kqVYeoDw2lhaMffyuvLGWJoQoF+Xymqtf5A/stiHeNzav5QGBC/bYgmsiIq5446lpN+p/OOnMm1BjzD+JX5dp9TygdHMf/5tM6y0sXLunmodx5zfxZ/fYgzpwzSPf5SHFWyv4+eQdgEsiDz633V18LG+YzAOXP34ADyj92Qu8b/jwHvy7XbecX4On769Fny+azOdmWzfzdjZ3Hx5S/LMj8nHNyb/lXdyRe/I2cOdVrN/uncXbzNz5vA8Yu5yPqcsP5fO/whZ+nKZl/HxW787ac81GHvrfupCFrkek3ZuRYT4+zarn1/mhv9XgmvIKNt7uPYdf53X/wfuaiIhrbz99QuYtZ//5RjwQVxfzuWPbN9POc9vZLCw9ImLLZj5+R0Q0NvH7l1+Qdl5L/s6e4Vc9ev6R23Xf/SWOJEmSJElSBnARR5IkSZIkKQO4iCNJkiRJkpQBXMSRJEmSJEnKACjY+DNPsGDjth4eEjqtiqeRLtuQFm7U3cWD40pKeBBwZTUPqdy2jYdhpgSLpoQ1FSdcgxk8vzSWbd4516CiIiENOSKyE+LG8hOWTQv5KcVgQgZXTcJj9MDfeCjevN15yOOLz/OwutlzunBNSvspLEoLY+/v4wGZk2t5X9LezY/TuYWHiqYEcKf0jSkBiRERvzji3RMSEHjGvSzYeFIFvy4bOvk969/Kx9uREX4tyytZf1lWyseLZS9X4pqmGTygNeX8s7N5UHOKrg72DKY8Sy3reND+AXvwfjVlQ4iqEj6IdQ/w+zmljIV3ru9OGJAT7FbH+40XWvmcNqU99/fxvobKzeXfK2U+25PwHrDbDB7UvbKVT7A6YR8Qwd9R5kzn55KyacTQCB9+CxLaQHtC2G1uwuYPe05l/e0fH0sLuH34I2+bkHnL6ffcjC9m8xr+4vbWA3nodUREX8IeO6OJQ/ELy/h5dXWkrTPM3KU7qe76d7zLYGNJkjQx6AKO0tAFHKWhCzjSa1nKH5nF0QUcaUdxEUeSJEmSJCkDuIgjSZIkSZKUAVzEkSRJkiRJygAonWx2OUsRml0+HJ3beCZT81b2+ZqawagrwoeJkhkDuKac503Gg8v5l2t5qhjXzD+SB0q9cxoPgrv2mRJcs3xpwg2KiMZprDFk54zHhvvZsTZGYcw9vgfVREQ8fT8PKssZ4f8HP3+I36P+yTxTISXc7dITNuKaK/5ajmsGu3iY5JLFNXHGcZtg1Ug8s4kf6+gZ/B49sIFf7+du4deuaG9+nLJy/n/rZzbw/ydeWcCT6UoS2ulEar+DBWEuq+chvcMFCUGwbTygsryDjYH9ETGSy75bTxRE/hBLwtz7ozw89+H76nBNzcY+XDNYwicBeQl9+KS1PBCx6z2sf8h/lmcc/f1ZPjcZKuJBuMun88DZ2Q+w/n5dRAzg+zkea+azsb+8nl/njXfwPri9iQd25ubx/jSrh9dMmc+etW2jWTHwIH1/4P1mXQsPRH+2iY/DpV2DuGawic1lB6MgssfYmPrkM/mxtZqFttY182tW38zn2OsTrnPKcTbMYmP0fVGAx8666I6WPatRzURa+zAPsS/t5m122T0Ju6pExMbpFbims4m/i0ZEVGzgc6XG9rTA5m1PJZVFvGP7PrZDf4mzMxZwIiJpAUevfXQBJyLwAk5EJC3g6LWPL+DETlvAkTIRXcCJCLyAo52zgKOUBZzACzjiCzgRkbCAI7qAExF4AUf8jx8R8W9dwNHrl/+dSpIkSZIkKQO4iCNJkiRJkpQBXMSRJEmSJEnKAC7iSJIkSZIkZQC0ZUA2zBkryuEhWx/ZnadhbxtLC0DrHOIhjdf8nadhf+ogHlK44GhcEpsGeHjcZc/z8/nwATyZflE9LomtI/y6lR/Bd7R4ejPfOSIi4oKFa3FNXjZ/JoZG+VprVQGv6d7GA3q/9JcaXPOuvXk7fes7eTL8yh5+DeZU8GtwdzPfeSUi4rjpPPB1369vwTWFOXxXgmXd/NlrH+DX4a9tKdeOP0MT6c4vb0af/+RyvpvP0U28bfwyYWzq72O72F329jZ8jMuX8O91ymy+09nSlTxssvEtfJeKGr7xXxzeyHYa2reOj2P/9SLvV087mtes7uXfLWXe9IYZvA9asjeb02UP8rnjGZN5YP7Bk3h7/tEs3m/c/SZ+b25bz8eUv7WzsbWxhPfZZx3P28zv1/D2PK+K7wL2ciffdKOHHyYi2LznqCZ+/pc+wXcF2u1tCXOXWr6jWWken/f9cS3voEuG2DvNJxbwceNDN/57g+e/+jE+dl/3Mr+Wc2vSfhtyRBF/rzxhZtJDFat6+e6XX3uoKulYC6fzNQ3CX+JIkiRJkiRlABdxJEmSJEmSMoCLOJIkSZIkSRnARRxJkiRJkqQMkDU+vv2BY9ctux6lk/15HQ+t7OrgQUqDj6QFG9e+gwfavvJiNa6Zu4CHc42N8nPacyoPz7vp7km45pS3b8Q1y7v4euHS53lo7pEHs9DRiIjlbXm4JiKip4uHCLau4mF9hxzGA8kmFfMgwa4h3uYee5aHfb1jEX8e2gb4d9vQye9rcQkPSjtlVlq42gMt/Ptt2sqfo9X3lOGavU7goaJ5CX8SKEl49LYk5sT98A2npg0U/8MZx1+HHq7Zl/AA/ZQ+Kfsn/MI0z2HjWdMy/uwu27sB11Ru5iHF9c09uKa9kT8bYwfweU3uE6yPqGnlwamrj+S7B8y8l48tuR/nbXPNXTxcPcXUFZ07/BgDCZ1WUR8fI5a+cQquyc3j4bEpNUWvsLlmaTfvm8boTioRUd6xY0NEX9V8dC2uabqLz01p31kwwAOH8xI2tJiyigfDpvjRTfx98NRP8OdzqIj16X3TC/ExGp5Lu2a3XXPyhMxbjjnvNvxSkDIOrVxYh2siInIbE2pyed8VEdHbyttIeWda39I5ib/zRUQ8ddFh23Xf/SWOJEmSJElSBnARR5IkSZIkKQO4iCNJkiRJkpQBXMSRJEmSJEnKACjY+PrlLNi4Ip+HDh05lYfZruzZgGsiIkoSAt2e31KEaxqKedjYLuU8NPe5Lfw45fk81Kwsj9ekGBufkDyvf6l9kAdVRkRMKeZhhS928PazqIEHcD+7mZ/TEVMn45r2gTW4JsWybv48tA/wa3BoIw+EfmITr4mIqCnkz2uKaaW8/fytjYexNSQ8D1sSnr2hhND3iIgTZp42IR3K45t+iW74ooTE502FPKR+dS8P6xseY5dkRhn/XiPwGBERjSU81PWPa3lw5OwKfj57tvH5RsfMaejz+Tl8nChta8U1UcADhx8b4e15fhVvm5UvL8M1Y3u8EX0+Z5yfy/LeVbhm19IZuGb8xUdxTVTzjQbaankb2AQ3Glg4zsPdnxrj4+reNbzfeHozf54PrObH2TTcjGta+lj73KdmJj7GpsGVuIa8N76qsoBvolKYw9vmxv4VuKY4twJ9PqV/Lkzb/yKi8J0TMm9p7f8pvmlZCb/zWNGTtlHMwQkhxQOVfKOhiIiidv7MP5vHw6wjIvYeTwtfzqr7oMHGkiRJkiRJrxcu4kiSJEmSJGUAF3EkSZIkSZIygIs4kiRJkiRJGQClSpblsVykvGwefnXnug5c8+MHeMhYRERZOQ/8zL6L1yw4n4e65SREWZ08kydnvdTJA7pmlvMwyAc3FOCa/et48GtDEb8GC6vTAqvaBvixZiVcu6LcclxTkteHa0bGeNs+5ZcssDMi4or3rMc133+EBc9FRMxs6sc197bwB28sIbg1ImLbEH8m+vt4EHB+QsD8lBreFkbH+fk8sbgO1+y/fzuuiYg4gWc+/lNfuacWff7uYzbjY+Rl82u5aLAN12Q17Y0+P97Vgo/RUYxLoiB4kPlxM9h9iYgYHN2KayIhxzwnm4U99mzj86Cych5Mv7WQ/x1vv2x+bz50Eg+d/uVVTbgmp5ddt1XRg49RkMMbQMfIJlyTs9tuuGZDH2/PDdl8/rNHGduAZPylJ/Ax5iScf+c2Hlia8p4yvvoZXJPdxIN996lk7za9I3ysachqwDWRMOXpGR3ANe+7jb8D/O6k2bgmOtaxz2cN4kP0lvF3rYiIsqSqfzQ5O+E9eRufQ0+qreTHiYjx1qW4ZrQi7eq0V/I+75JP8dD4iIg//oCPl4S/xJEkSZIkScoALuJIkiRJkiRlABdxJEmSJEmSMoCLOJIkSZIkSRkga3x8+0O9/rj2OpQAtqiBB2q+1IlLIiJiWikPjO3axkNCc7N4CFp9EU8B2zLIr13K+fRs4+t4+9bxa7Bu6yiumV7Kz+e+9TxEOiLi0EYeurZpgN/Xld08sDTFgmp+Psu7edhXTsLzsE8dbwub+vnzsKaXB4q9cTJvcxERL3bw/iclvPyudfyc3j+Hh1yv6eXtdN1W/uwdN4MHFy7vTkiWjYiDGs5MS6D+H8556DdpXwA4ehp/RsbG+en9+qkSXPP2haw9/ep+Hl593lE8vPquZbzNXn80C2iNiPjxSzw8dVk3vzc9/ex5OqSJP0tHJYS/37SKJ1U/uZbfmw/tw/utz/6MhztP24uFAR8wk4+tZSzXOiIifvYHfi5z5/NA7HftzkNav/2relxz0ftZ8Pr3r+LP5gXnbME1f2nlc+DGEj4EXLgHv2ZvvYo/0wsXsJeoV5bzjTOO2qcb17T28z6wMGE69skF/N5873n+3Y6exjZ/uHV1WsDtdxadOiHzlhNuuh1fmN1m8PFhySo+p4iIOHtRL665Y1XafL2rg49HPzmGh/NHRHz8Pv58RUT85p0nbdd9f138EidlAUd6VcoCjvSqlAUcKRPRBRyloQs4SkMXcJSGLuAoDV3AURq6gCPtKK+LRRxJkiRJkqTXOxdxJEmSJEmSMoCLOJIkSZIkSRnARRxJkiRJkqQMsEN3p/rzBh7O94G5PGT2uS18V52IiPc18l0WbtnI07r3r+dhkMW5PD07N5unn6/p5dd7dnkZrinN5TsNZA3ytPIne/nuDAU5aZvN7Fk9A9es3boC18zI4bu7LB/ejGtmlc/ANXeu3YRr3jyF9wt5Wfx5eHQjD618Ww1v279vSwt7fWceb6svFPHk/7pCvsPF5OwpuOa2Vr5jxTOb+d8Rzpqbdr1nlZ8zIbs8vNhxNeow5lTwXWZ6E57fghzeNvpH2D1b3csvYV42719nlvHnfUUPD5scGOHtb14l3zlsPWyyw2P8Ou9ZU4trXurku4Bt6ONbLXUO8et8+jS+q8fSATYejSbs6PZiB59vHlDP540lefy73b++CNccM52Px6t72fnsVT0TH+PxtrW4Zo9q3mZe6uLj1sJq/qwtbuPjfV3hMPr8hn7+bB7UUIFrBkZ6cM3gKH/XKMjhz1rPNj4OzNzGnoGNxey+RET8uaUU10REvG/X903IvGV591V4IK4t5O/I2VlpofxZCb8p6RtJC/Lugff7/0ZDcdpuXZX57/3fszuVJEmSJEnS652LOJIkSZIkSRnARRxJkiRJkqQM4CKOJEmSJElSBsglH86G8UpzKnig4ZNtPMhq40Ba7tPjvTzQ7OGNPDx4aikPG1vRjW5NRETUwhC0iIhfL+FBru/ebRDXFOU245quIX4NpvHTicp8HvwaEXHFS624pr6Qh1zdM5QQipjLA8k6htbjmuU9/HkoyePBcxv6xnBNaR5fo364mwf29Y/wdhoRcc0gD/su4dmA8XwH7x8X1fO+8e8d/Hof1MBDYu9v4W07IuJcnnn5T92xloWHbhnswsd40yR+LW96EZfEyEgl+nx+Pn8ON7Xy+9WfEJ77nsPbcM31t9Xjmq+eyfv9P6xjfcQTj03Cx3j7oRtxzbPLeEDrjBk8MP6lJVW45qkFPLRy6SssRLyunneohUV8vtA3wp+Ba//M780Ri7bgmite4s/0nY9MRZ//4JEb8DEe28jDzasL+dz0gYem45oj38KftZ5h3qctg89nWRmfW902mc8vO7t4e96niYfJ3n8VDwOe/R6+8cGBdexdtWcTb5v3vpAabJxU9g9+tpS/eyxbw2u6u/i1iYj4xNv4Zg6PbEx44YuItS08AH5Ta1pA8ey5fP4XEXHtkdv3OX+JI0mSJEmSlAFcxJEkSZIkScoALuJIkiRJkiRlABdxJEmSJEmSMgBK2yvNYwFoh0weQp+PiNjQz8O/FlTzYLaIiLFxHvh59jwehFeex8M7ZyWEcPYN82Djs/fg4dPl+fx8qgt4GODwGA9pqyvi4WGrevg1iIjYp5Zf74Ic3lYXJNQMjfH12b0qGnHNhr52XFOXEAxZnhCoOruc9yVdQ7zNpbTtiIjNgzwQuaGIt7kF1byfS/lub57E+4WhUd5O51WmXe+JUpLL+os6lh0cERFTS/l93m86v8+FMGtyZQ8/xvVH8kDAsx/hAa33PsdDDvc9lPdfK3p4QOehk1mb6d6Dn38x7+6ioIA/s/U8EzIuPr0F17zYwU/o2OksHPNnT/E2c9buPDy3NSGo+9ADeRs4qJ73jdlZCXPAw9hzU13Aj3HcDH4uVz3OJ87vP5oHle9SnjIf4e9DP4SXrWU972uHE16fvncYD2td2smfgbbTeEjxbpW8rdUUsovQVMov2h+G+LgxkU6fw99db+f7lkQEv2cREVuH+dyiMi1DOXrreB8+Npa2gdI5e/DgcMJf4kiSJEmSJGUAF3EkSZIkSZIygIs4kiRJkiRJGcBFHEmSJEmSpAyA0iuLctMChIlppTxYtDjxew0nBBW1D/JwrvaE8M471/FEqQPqcUnMq+TXuz8hV/RvbTxwLTshR2rLEG8LRzTycMeIiPtbeKrWzDJ+rJTnLiUUcnopC4WMiDhuxiRc80rXGlzDIwEjHuV5hbGgmtfUF6UFnhXk8Gfv3vXFuCYljPbY6TxYd/cqHhY3NMq/28DIv/dvDwPwOx/WyAMFu4YSgqUn8/Z0+bOsX85PCMI94Tb+9L5tAT/Oz5/iofaTp/IQxn1q+XX+Sysbz394KL9mr3TxgbmDP7KxuoMHdOZm8fnMgfV8zvDRk1mI5Peu58HWP17Cw5Av3IPfm7ckzEs+fh8P9j1mHm8Ea3pZH3jIZH6MvoR+vrKaPze/e7gW13zgcD5XqqEp8hHxrQPZPC53ET5E3LqaBwH/dCmfh8wp58fpS3jX2CchtJYGj5fATX4iIr5/7EZcM5Fe6SzENfc8UodrjnoT71MjIkYT9pe5cCE/p4iIkiw+thzw7bRg6j9MTws2fuN2vlr5SxxJkiRJkqQM4CKOJEmSJElSBnARR5IkSZIkKQO4iCNJkiRJkpQBssbHtz9N6G/tv0DRQw1FPBwzPyHNNjsrLXCoMIcHIb7SzYMQ51dV4ZqxcR5ql1Kzbmsvrtm1gofZjo7zttA/0o1ryvJ4SF1BNg9pi4jYPLQO15Tn8aCwwdGtuCZFbjYPn9zYz0PM6ot4evDwGA+rGxlPCEnPqcA1L3bydhqRFuJekGrEPfUAACAASURBVMMD2bIiLXgZHyeL/01gaScPopxczK9bRMSMsnMn5EI833E1GgfL83i/XJbHA/vysnlNyng2s4zd5/YB3vcPjfG21DXE5wGVCUHNU0v4cfpHeDvPyWJJjykbNbzUWYRr9q3j59I+wL9bacJzk5fN0zF7trH7mbK5xdxK/gw83c43TnjDJB6IPjLG+9POIV4zBDf4mFVeg4/RO8wDh1M2gdiTf7Wk95R7m/lzs6CazZWaSvmcYu1WPm7sUTYF1/y+ZQuumVHO+6eUMXpS8WT0+f96sQcfIyLioj3fNyHzlpa+n+LOsSCHzyn+ujHtfXzfOt6minP5O3xERGkef4DvXsf7loiIvWrTgo2nlW7ffNVf4kiSpH+JLuAoDV3AURq6gKM0dAFHaegCjtLQBRxpR7FnlSRJkiRJygAu4kiSJEmSJGUAF3EkSZIkSZIyAAo2fmbzNeg/aucnhMwtyK/ENRERw8U8nCtviAdgPdPHg7b2qeD/f3J8y1pck1XegGtaow3XTBrggXsxlBDOO8CDrqIq7f+qZuXz9jPez9tCViEPzh1b/Cg/ztzZuCZFbwMPqyvr6uIHGh3BJVuq+D2t6U/LkHtunH+/vfLLcM14Qv8TAwmBekXluCSrqgnXPNrJg7EPrOfB6hERedlHTUhA4HjrFWxga9vEDzK5kdeM8LDR8cnzcE1W8/OsoDeh769MmAf0J4wXJfwZjGGeO5HVMBfXjG9ZzQpS+tWyhGDIXB6AnyShPUdCIHiMj7GPv7gUHyJr4XxcE/l8w4Ws3IS5WYLx7lZelANDl4sS+gB4LyMi6XmOTQnnP3Umr0mZN9PzGePXbPwV2DdFRNaig3FNUh/Q14FLRh94BtfkHP1mVpDSNiMia/JHJmbe0vYz/kK+aQMuyZq5ENdERIw38341StOCjZPGysnT0o6V0B4jIrKmf/p/T7BxygKO9KqUBRzpVSkLOFImwgs4SoIXcJQm8cVKEF3AUZqUBSlheAFH2kFeF4s4kiRJkiRJr3cu4kiSJEmSJGUAF3EkSZIkSZIygIs4kiRJkiRJGQCljeXC3aYGRvka0XgH35Upr3hPXPN/Cnno97Iunv6/d/vTuCYaeBL2eH8nrhkuTAg+37iO11RV85r8hJ0WujfikvHsxLXM0lp+rOZXcE32gQkhagmBjePDA7jmrN/n4ZrfHZCy48IwLimrn86PM7ASl1y5ireDiIif7s3bXVZpHa4Zb5jDj9OTsKPSIN8F67NX8t3kHrv03xuSefBNu6LP33VeFT5GZT7fgWvR90dxDXXggfx7Pf00fz6+ciLf/eXg39yHa9Z+/DhcM7eU3f+IiC0j7Hm6dA2/Zp9aWIRrfv4yD/X/xr68D3pgI9/lpGc0B9fcs5b1qU8/zttzZfUsXHPtwc24ZmUPn5sdVMX7mpws3p9ev4nNL94+jYf/f+95/r1euYLXjBzFr9nNJ/P5+YudfG46o4bt0LW8m8+TppbwHaD+c8ObcE3nY3xXuzWLeV/zo/NxSTS+9zD0+YZ8Prf8wUt8J+CIiE+mbbj7D5bk8v70mq187njeaNpuTDPn7o9r7lrHdzeNiPhDB9/JuWtF2s5/Nx+RsAMm4C9xJEmSJEmSMoCLOJIkSZIkSRnARRxJkiRJkqQM4CKOJEmSJElSBsgaH9/+sOLnO65GycaNxTz0dFUvD2dsKuGhaRERvTwDLHqHeThUdkJ28OoeHgJWkMOCpyMiyvN3fBhmRMSmft4WCnJ4OG/3Nr4uOas8oSFExNpefk6lefycxsZ5A3q5m9eUJOTF1iRkfe1ZwwOUX+4qxDX9I/waFOfyZyhVyvPaNsDbd30Rb3MjY/za1RXxfjjlGc+HAfuvWlB9dkJP/I/+suFa9AXW9PIHq7qAn+PUUh5QubSTPcA5CVcwpZ1vHeYHWtrFaxpL+HfrS5hulMOh4kP1Q/gY39vAO+MZpbgkVvbymqObBnFNa8KcYUYZu24p87n1W/ncrKGYzzHub+H385DJvA9IQeckeQl9dlFuylwJl8TaXn4/127lfc25u/EQ8e+/0I8+v18t75zohjURaWP9lGI+Dn7nOd4HLKrf8XPslOf5oVbeziIivrj3qRMyb3mghc1bIiI29PF7Npo4hS7kXXHS+1RExMAIn0c3FKetMzywgbfhiIiv7Lt9991f4kiSJEmSJGUAF3EkSZIkSZIygIs4kiRJkiRJGcBFHEmSJEmSpAyAUosKYADWU+3o4xER8bZyHmA6VFDODxQRk4KHwD3c34Nr3lTAc6n2rEgIbBrnNZuKeEBXQ3dCEHAuDzaM3IQgsMFuXjNSxmsiYlFtNS/qSXgogt/Xd1YlrM/m8WdvWRG/R3O28fOZl70R1wzO3BXXFHZswjWdFcW4JiKierwS14y3vMQPlFOES7IqG/hxsngQ56N9fbhmQV09rplI02CA8FsqE8anrJ3z95XHN21Fnz+v++/4GD+v2gvXfHAeb3/jwRMVf/lKG675SGkrrllRWYE+f00r71M+1cjH5cGKKlzTP8LH2NxsHupakMMTlBuK2HX+zUoeVvnWRj5vPLCMzxVWl7FnMyKirpCfz64VM3DN05ub0efnV/F2lpvF5xa5j92Pa3Y96CBcU5bF+/TxrXzud8481tYWt/EQ1bdV8vlBylwxCniK+rfm8Xnf2AvP4pqBAw9Gny9etQYfo25mwpxqAh22fimuGd3/UFyzoX8tromImFbQhGt6xhNS9iOib6QT10x6ZX3SsQ7ca1FS3fbylziSJEmSJEkZwEUcSZIkSZKkDOAijiRJkiRJUgZwEUeSJEmSJCkDZI2Pb38Y4PLuq1ByYEEODxoszEFZyxERMZ4Q/BoRkZWwhjU4yoPj8mAgdEREYQ4PARsZ54F7AyMJIcUJ8rN5uHN2Vg6uycvm4apZiSGi4wlB0qPjvP3kZ/PguW1jA7gmRVbw+9o7zEOuKwt4+PT6rf24JqUnaSzmYYIRaW2hOJcFd0ZEjI2P4prhMX6PcrL5dRgY4UHxqSYXf4g31n9iVQ8bB+uLeKhpa38HrplRxoMB6X3u3saDv8f48Bfl+fyabRvlfd7w2BCuyc3mgauj42ycrS2cjo+R8sxuHebtLGW+VZTDg2AHR3mwb/8Iu58DI3zsryrgNSl98OZBfpziXH5v+pOuARu78hLGhpRrljI+vtLNw/V3reDBvinzRTqPS3nOUubZKXPSoTF+nTf2d+Ga2kIeCk/fAcryavExtgyyMPBX1Rd9YELmLZsHr8UjcXleHT7O1hE+pkSkPe8p7TAiYnPCBjg1hWkb4ORkpb0flOadtF333V/iSJIkSZIkZQAXcSRJkiRJkjKAiziSJEmSJEkZwEUcSZIkSZKkDMBThIH71pfgmkGebRQLq3lAaETELat54NDZ83h44tUv8/Clp+/mwY4HvYMHSp27G7923/gbD09bu4rXNEzmwbQtzTwQ+tADt+CaiIi/X8nD7Tr34d+vfytvp7l5PERv+iweMnvxmzpxzce+PRXX1GzkIZc/voJfg4v/xu/P6XN4oHhERHUBT3z98Pd5nxr1fK3+vKPacc2kYn4dVvbw692bmMV+4R5pdf/Txz7Pwv42z+SBePst4oG7Lb/kwZHDBTTUsiEaTmDfbezylOeDh/S2N/Ix5i3n8mv2u/sm4Zq8XjrOjsWUVSzUc90cPmcYzeN98fSX+Xi5qYn3W+PZvG/Y5bk2XENtmFWJa/qm87lC7ZJeXFPcy/uNnqP5da67hQfONsP2WdjHO/q2Jt7XznqBzxN6q/mcvrSL92l95TxANxsmyQ8V8dfA8g4+HxvJ5QHKlZv5c/PyvvW4pryT3pvBKO9g74IbZvK+NiLi8UuTyv7BmRfw8XEsm4/deUNpYcODJfw9p3dW2rF2fZD3k1sm801zIiIaTubHioi4+rDt+5y/xJEkSf8SXcBRGrqAI72W0QUcpaELOEpDF3CkHcVFHEmSJEmSpAzgIo4kSZIkSVIGcBFHkiRJkiQpA+zQYOOIiEOnsHDau5uL46SZrGZZd34sqOahWbesjvjYfBZsODCaHVNLeMDRpfuy8LRj746491IWtnXpUxGf3pNdu55tEdNLG1FNRG98/83sfE5cFXH7mSyo70N/yolfHs2Czd7604h7zmX/X/Vbz0V8cS8eVnhyDMSX/pOFAV94c2l866RWVHP+tdPi0QtZO33LfxXFhcduQjW/e7Eovv4WFlo5NJodC0p40OcdX2XX7QPnZsc3fsCDlzcP8nXqqw5h5/Nw67rYrYqHI27qj6gr4v+H/fJPbkCf/9gNU+P2c1ib+1Nzbrx9Gvtuz26O2K+O9Y0re0bjlNnsHl398licM4+F4F31cmIa8j9x1w/YOR7ww4iHPsmCIC98IuI/9mfX/8xfRlx/JSqJUz4ZccP3t7+/vODh7PjFYSxs84zLN8S117F+/IzTB+PHNxejmlM+GXHnB59GNd+JeXH+fBbs+rv7Ih69gGUDHfrVnLj3S9vfh5/5gYibf8UCOg/8+kg8+kW2M8RB386OBz7FQsw/cG52XH81C5Q84ssRj36VPTcHf3lbPHAJC9b80EkR196w/ffzjFO3xrW3TkHHeOun++MmOO6/4yeFce9HWM2pH82Km69k1+yM04bil79m7fnEGyLuPI0FT595S1f86Kbt/27Hfyniz99kfcDR5w/HXZdXoJr9vjUWD17INvg4+8yIn1zDgsff9+mI237Axq3Tz4z4xTXsleuUT0bc9Z3tHweOuTDi1stZm3nHZ0bjvi+z94aTPp4Vv72G3ZuTzt0ad1zFspHOOKE/TrucjYNf+kXE3Rexue+7P18R9313+5+bk84aiBuvYUHBb/5axMMXs3nim782cb+z+OPlPMD86I/3xR8uZ33Dief0xB0/4ksLb/tcxB3fy0I1h10e8deP8rn3GQ9GXHYje+bPuoA/WxER5z00FFcewtrKeQ9t/1iRNT6+/Q/I8u6r0NOUw+5HREQUJywr5eekJVRvGeThVDWFaceiCnNSdm3ZjGuq8tkEJiKif7Qb12QHX2QryOE7WoyO85e13Kx8XBMR8VwHGygiIkbG+EOxd20NrvnDGr7gsVct3w1s10Lefrqz+K4bG/v5bgjDCdd6QfUsXNM+uAbXRESMjfOt+Fr7+aA+tYQ/e9lZvCYni3fe48EXsVL6koiI8vxTEkakf2L4bvSlB7L4Dg+Do7y952XznTxGxtl3q8zlO39EH981cWshb+clL/4d1wwu3A/X5GTxXTSGx9gfZEpyq/gxxnnodP8ID1BOGZcLE3aaGhzjz0DhENwFrJDv2NI5vBHXpDybKTXjwV9oCrL5/aRzzfJcPoeJLN4H9I7wOfDwGH9uinPZAkZExMgYHwdKh9k12FbAXzLz+/h8bKSU/9EzZZ59X0szrtm3jreb6nz2h+yUvjZVXvZREzNvGb0PT7ZGE9aQcgb5e0RExGA+P1jheNq7W/sI+0NoRERFXkPSsVLl5xy9Xffd/04lSZIkSZKUAVzEkSRJkiRJygAu4kiSJEmSJGUAF3EkSZIkSZIyAEqiLMhhuUj52TyPqWE120UlIiJr1p64JiKiOiEgc9UQD9ub2c6DHWOMB9QVbeOhvr278mC/ioSwsRTjq9fjmrxiHjydVVGGayIi9m1t40X5PBQzq41/v3clHCfW8OC9Zxr5M7R382pcU17OgzG7pk7FNeNLHsE1z1dOxzUREYeX8uDK+s2r+IG28ZDUlFDJGIOhohHRUc37n8pxtmvRRBt76C70+dzlvJ+omM7vWVYND5uMYtYGB268Dh+i8ENH4Jrip17ANfHGg3jNV36GS3IWsN25IiKyelnfOrZgGj5GvLgOl5Qf/1ZcM3zjrbwmj/cnKfHl40fsjT4/9sxSfIyKPP7NsnZNGCPgsxkRMd7KdhqLiIgcfj7jv2C7wI0eyjdAyKriY37JYELg7AjfYCDq2S5LEREFZfx8xrZ0os+PP8n7gLFDdsU12W0J7zRHHoNL3nzDbf8Pe3ceJWl2l3f+ubFHRu5LZWXtVd3VW7W61ZJoLQgJAQKErAGBZRkMNsaAsRl7OIYBe4zHB5vjY8YD48NhxmONGQYQw26BQYDQhkBIIHWr1fRaXb3U0rVlVu6ZsUfc+QM1BrV8yOd3ssuOnu/n7/j1jXzf+95731vRz7VrClN+uHN+413W50uB9XV+5Cm7RpL09V8dq/s8w/f/ol2TaoFrecctdo0kVT77uN/Woj8XS9LMuct2TQrsZ0hS4YvvCdXp+Nfs7b8f+68DAAAAAADgZmITBwAAAAAAYASwiQMAAAAAADAC2MQBAAAAAAAYASnnvYcVX9x5j5VsHAk2rhT9YNqtbtOukaSZqh+q+dy239aJCf9vqhb88M5+9oNpt7p+QNlE2Q/d3O55AW2SNFbyg+DGSn64Z9XPY5Uk9QIBh53Brl3T2PZr0uSiXZOLgdDloR8K2Fbbrmn1t+yazsB/VtsDL7xdkqarsWDsiEIg7rNW9J+jztDvc83+pl3T6vsB7sMcC5g7NfkdscLPt/OrVifJaxfsJvJRPwyv0PeD7fPWdevzkXFlULLOT5AUGydz9vtSYxAIqNxdtWtaU968VL963m5DGxt2ycqtfoByueCHXVYK/hpobDVwcMDAWwM1D/jh9+WCHzhcXvUPaUhzJ+2aiI3BDbtmuu9dg7zpH1jSWvDDkKvFhl0TGTdWO/7hHjMV/++p9M31SOBvUd9/b9CYv87uyZ+fyj1/fZmf/JRdo4UF6+Pp4J1+G5F7I0nFt+7PuiV/1F/cbvtj8GbNb0aSJncC/TD47p8W/PDljgLfT1J11z8MSZI0+e493Xd+iQMAAAAAADAC2MQBAAAAAAAYAWziAAAAAAAAjAA2cQAAAAAAAEaAlThYLXp7Phd3/BDO++a8gClJKiU/UDPqlbN+2NyNzkW7Zqu7bddMV/ywscMVP9hwK0e+m39fd/p+6HJh4Pe5atEPapak3rBl12T54WatST+A++FVP0jwzIwfph3RHvhBX7PVw3bNeNkPWOsGwpD/aNl/HiTpdQf8QORGIFT8WvNpuyYSXtoZ+H07Ei7fG3bsmn1V975z5/Bpu4nN9nm75uNX/fDYn/xRb85ofIU/5q2v+kGwE3630KXz/vz3zW/zx8m/cdofV374U964cuXfLdltfP+PTdk1//3/6gf7/tb/6F+zT6/4Y+R/uuB3graZg3rwuh+yXyv6z8DXHvf/vfRX/mTFrvmBe+fsmqmmHx57vuh9t5WSPza9OhDI/85f9APRu39ol+i7v8+fH3/joh88/n++wptTv+dxv5+5z4wkPfQH/hh48oy/tvq5d/hB1d+7e7td86WT3nX+kq4fVP6BS37AvyT9DX/58AX9nY96hxhI0tZmZX8a34N/9zZ/Xfcr12KHinzjrH84wdv//WyordvPxIKef/LL9vY5fokDAAAAAAAwAtjEAQAAAAAAGAFs4gAAAAAAAIwANnEAAAAAAABGQMp576E7y63/20roGSv5QXutwZZds9vr2TWSVCsmu+ZG29/3Wqj7wUblgh8G2er7oW7lQtmuKRX8sKth9tPTekM/dDDy3WqBAD1J6g78gMPdvv83zVT9MNtI/ykmvy80+35Y3yD37ZpIX3h2yw+SO9zo2jXDWG6ZCv7wo87AL1qs++NwP/vXIcn/bpF2ohZqfztwxV9su/fL1h0vJX9MenTdD3X/olk/pHZodsKVQOByZB0QCT9fqJ2waza7fthjZJzc7N6wPn946Ico7tT8tclHLvtz2Cvn/YDSsZIfBLvT84MuD415z0D5ylN2G1dm/ZDew/Vb7Zq8dsGuWRv357ztnj+3LtYXrc9H1iORcfMzq8/ZNYt1//2hN/SnkoNjflh5re890892rthtlArBBYzpWO24XbPS9/+etUBS8/EJrz/XCv47Q+SAG0mar/2tfVm3rLZ/xr7Rc0M/OPiDa7GDPu6Z89eClYI/FkvSTMcfJzdq/sEdklRO/twnSY3y1+/pvvNLHAAAAAAAgBHAJg4AAAAAAMAIYBMHAAAAAABgBLCJAwAAAAAAMAKsdJ+UvHyl3f669XlJOrDhhxvNS8oHb7PrUmAP67H1S3bNqUk/2HEsEM41aHjhXJK00jpv18wOGnZNXnnGrknHXm3XaHvZr9FQmjjglwXCN2e3/LBCtZ61S9Kp+/12tq7ZJdeyH6Z9+/Qtdo2afoByedr/bjNXIv1HemRqzq45PTlv19TW/CDWVPOf137ZH7Miga/XWufsmnpp0q7ZT+PPm+GZk/7Y8kV1/2/MVx6za1J1zPr8AUnrU17N+LXL1uclabh40K7p/vj/btcU/t677ZrGk4/YNa3TXqjnSnFHU2VvPh/v+wGlbznshzWOb/hBuAocAjD72YftmnS7P09oyutrS5/+jN1EvtsfGwfv/4hdU/jrb7NrZqr+WFNP3hig7Ieoa+C/C3xq2Q8RfedJ/xmY/sFftGsq73qlXXPjVXdbn5+sVDRfPmTV5G1/TaFNf624c8hfL8/v+GPawowfoJyXvXVIlqTs9ZuJ937Y+vyf+f6/Fav7PLOPPGnX9F/9ZrvmTUt+ILkkld//PrsmvfqeUFurc9N2zaVt/0AWSbo3+3OfpD/d2NiDl8UvcSIbOMCfiWzgAJ8T2cABRpG7gYMYdwMHQeYGDmLsDRyEuBs4CDI3cICXystiEwcAAAAAAODljk0cAAAAAACAEcAmDgAAAAAAwAhgEwcAAAAAAGAEeKdTmXs+Ky0//CkSMVsY+Cc5SVK+8KBd86ojR+ya+tk/sWtywd9fK+w07Zr6vX66d34scDrDrv/d8rn/aNeo5iejp6VYsHG+fsNvK5CKnlf9EzfyuV+yayLXoXzSfx6Gn/gtu0ZDfyyZfcVr7Jp8w7/WRw7FwgSrH/+oXRMZF3LJPwGqsLZp18g8gUeSdg76fe7gauA0DUk6Giv7fOm4d2pevvSQ3Ua+4Z/opEX/+mvFO23x400/2PgdVX9Mnkr+6Wi9ctGuiZyclbf9U+8W8qzXxoZ/Sl4OnOZT/90/sGvaf/2v2jXVBz5h16RX3WfX5Mcftz5feN2ddhuaGPdrrj9vl2xHTpoa+t8t767aNZryns/06B/ZTVw77a8tvnPNP4GncPJNdo1+9B/bJfmR37dr5i54/SZPbNttqOefNtc/6p2aJUmNLX+9fKnmnwh0bHfNrlHXfD8p1+wmyvcH5uf9NO+/e5QG/rq7eM1fb0pSnp/xi4qxk7Bmn/C/40zHn18lqf/oxVBd+du/bU+f45c4AAAAAAAAI4BNHAAAAAAAgBHAJg4AAAAAAMAIYBMHAAAAAABgBFjBxq5iynbN5pwfblTOftCgJI3NHbZrakU/OK53hx+wWhkmu6Zf9PfkWm0/dKl6z/12Ta1nl0hFv3vuqmXXpBTby6xP+H11MO+HmxVrj9g1OuG3k6b95+Fgxe+nhXu8kE9Jyn3/vp4NjAu3vuYtds1sa8eukSS9+Wvtknz1Mb+d2qRfs+oHvvZv8YNIj2Z/YNgZ84MYJWkiVPVi+fpZr6AZ+L6TgXt26Wm7JN3xWuvz/11ry25DORBYWPDH/me/8a12zW2Be5OO+uNkt1K2Pl9eD4wpl6/YJcUvusOvuXHVrhnc/2V+O2c/bdeku72DGvJyIITzVf7fokuftUtmun5o504gCLYRCILNTfMAgBk/qPzgMLBOCLw/5CcfsGsUOGCgd7c31kpSec17ptPEot3G8OO/a9eUZwMhvZWGXXK07Qfr5ucC62XzQJR08ha/jaXAYRH7KTDX5e0Vv51ux6+RlBr+oQmqBGokacwPpk5HYsHUpZnAWs7AL3EAAAAAAABGAJs4AAAAAAAAI4BNHAAAAAAAgBHAJg4AAAAAAMAIsNIDk7wQ01OTS9bnJakziIUUR+w26nbNePKDlJ7dPW/XfGrZb+feOT/89c7aAbvmsxt+2NVqu2jXHBrzQ0+f3/XDAO8/4NdI0vu3/aC2emvZrnn1gh/J+ull/7vdM/Dv64mWHyT4+ztNu6bn59vpjQf98efZrfN2za3P+uHgkpRf+Wa75je703bNiZofNFddWLBr5rt+346Eiq+0YsHGt/uX7gv6mw971+bvvtIfyzc6/nj5tObsmssPrluff+IZP9h/ccl/3q9f9eey3/imU3bNbsP7+yXpsXU/3Pn2gRdqOX3ojN3G9Rk/rPHijt/Pnlz358uD/et2zYe2/Pv5FRNesO+//E3/On/V66/ZNQs1P3D2ePKfgatN/342Sv48+UUHvHX6gyv+euQn3+f3s5991zG75vs/4bfzo2/0D235hh/wFzHf8V3eO8rvn92223jTidfZNW8p+SG95zb9sPq5qh/UrVMn7JIrTS94/tIlf07rDk/YNZL0naGqF/uxgR/G/N1TgVDeqcBBBpIeWbth1xyq+mOkJF2f88e8qOPHj4bq9rqa45c4AAAAAAAAI4BNHAAAAAAAgBHAJg4AAAAAAMAIYBMHAAAAAABgBKSc9x7QtdZ5r5XmVSn4wcGXdvygwaPjsSClSIhyueCHB1aK/nXY7PpBgLWiHzpZLfqBc7s9/x7daHftmojZmh/sV0qxYON+9v+mwdAPahtkPxDv3KbfT9948KBdUyr41+7j1y7YNXfN+KF4EZG+MDmohtq6mv2AzK2u37/Pbfrf73WLfp9rlPzk4ELy7+tu3x9/JGm2+s1eMv9/wadXfsqaB4+P+yGY5YJ/z2ZKfkj9B65ctT5/37w/fv2/T/t/y3RgSH7ghn97v+12PyT76LjfznTFC1IsZ/8578o/CKDS9Wv+03U/PPX0lH+dF+pe2Kgk7fS8EPerTb+jvS75836aPW7XaHfNLvl4yw+xv2fWD17/zA0vUPRLD/hri1z07/8fLfuHDPyHx/y10r96/Y5dsyj/Crl9OwAAIABJREFUsIDtktfXHl3z32leN/D/ljR92K4Z1P33k3/xGf+wjR+61R+fB5Pz1udbfT/cfrvnB/dK0tLY392XdUuz/2v2QiTyjhx5p5Skra5/r5/b9scISZqu+OuYuVrsNy/j5dlQ3Vjp6/Z03/klDgAAAAAAwAhgEwcAAAAAAGAEsIkDAAAAAAAwAtjEAQAAAAAAGAFWsPF65+esYKRhHthfaKLsBUxJsYBZKRbAFAkcni0t2jUq+oGfvaEfatcdNO2aRikQJN33gw1VDKRbBoKD1fJDyiSpPeYH4kWCc0tNP3gu0n9U8Gv6Zb+mtOtf72EgvDwSgBsZfyKBbJI0WfGDDuVn5MaeiUhN4Hld7/nhzuPlObtGksqFr96XgMDe8Hesu1Du+fNgs+gHzo7temGjkqRJP3B0s+fNgZHDAEI1TT+E8WrBD5tcSoFrVvTvTTF5oY3jRT9YvC//OW8N/PF7ohgId2z77eT6lF3j9uf+0F9vzid/XlHVDxyOhKdWC/46uB0Iw50seeN2J/trxkLyA8HbAz+oe6LpPzftcT/Yt9by+1rTPNwjciBKoeOPZ/2Kvz4YZP869wPvQaXAQQKV689Zn+8unrTbkKRq8e37sm7J+SP26jG1Nu12IgHWklQc+IdpRN/d8ri/huwM/fWFJNX8pdznCt9BsDEAANgf7gsvYtwNHMTQn28OdwMHMe4GDmLcDRzgpcImDgAAAAAAwAhgEwcAAAAAAGAEsIkDAAAAAAAwAgLpp3vXHfrhV1eaz4famqz4IXCRcLbOwA836pT8cLLq9oZds1X2r/daxw8Bm6ys+TVlP8S1JD/Pq5v9oGbVirq44/9Nc10/EC/LD+8aFvxg1N7Av3b9nl/z3A0/rO6OaT+wcKLnJ/qWCv53u9G+aNdI0uN+hrLunfPzGCKh4sVrZ+2a7sFb7BoFgigj96jZ98dGSZoK5KR/IZd2LlmfLyS/755o+Nd/s+5f/ykzPHZKdalqBhcGQrJ7gQTv1ZI/N/f6/pjX8jOXNZX8TJCz2xesz09W/FDbVt//d7xj48fsmpz865wCa7prrXN2jTsGRcbg1f6qXVPt+c9z4wMftGsKX/52u2ZYqVufb+emBsNowufeNQoTdk05cJhBnvTHp7L8ddzVwhWvoOs/0yttf/Fy57Qfhj059CfgTvbfT5oDP4x3oXzCKzh0Rh3zUJhzm/4hDpJ0dyAT/guJHPSxmwKHgwTePSQpZ//dKNdiv0N5dt3PNLp1MpjrVfRDsx0vi1/iRDZwgBdENnCAF0Q2cICR5G7gIMTdwEFMZBMZvpuxgYPYpix87gYO8FLhiQcAAAAAABgBbOIAAAAAAACMADZxAAAAAAAARgCbOAAAAAAAACPAOp2qkIrWf/z8tp/K3Bv6ydZHGv6pTJJ0pVm2a+6Y9o+meHbrql3zXODadQb+ntybD3mnDEjSSss/NeF6yz8549fO+9/tO+7wT4x6eDUW1jlZ8dPUL+749+juWT8U8MK237crgS3dNxz0T3i70vTDJMdL3ik6kjQbOEXmE9f8okONWGjjMPunVTy0ap5WIemZlp+qf/+u385kxb+vkWvQHvinHUn7dzrVD3zYu54//BY/OP3ZnWftmu/9Hf8EwKE5hL3ytH+i2sc+7R+v8X1vW7Frfvg/HrJrFpf8E61+8Mv9+/mH17xx8qF/5s//b/kRfwD/ifcdtGu++e3+KSuvmvef2cfX/cNTP/j4Aevz1x/21xg//Pf8Z2DgH2ak9533T+Y5eezNds07e8t2zT/7iHdC17vP+GvGf/3ri3bNMPD+MH/AXzMePuave5q7/ppsdcUb0yLf64sO+OvYH/qA/9w0xv2xdnzCX1vNNPyx8+tOeKeSjpf9dcv/9sCkXSNJP/vVobIX+emn/GdwteMfGrTRiYU+v/6AP0j+yVrsJKyJir/O/6UPxO7fP/mr/vu/JH3Dqb19jl/iAAAAAAAAjAA2cQAAAAAAAEYAmzgAAAAAAAAjgE0cAAAAAACAEWAlxxWTF8x1YmLT+nykjc9VBWqkO6b9UKTZ6mG75uHVy3bNkfGuXVMt+MFQk+Wjdk1ncN6uud7y7+vfPO2HtLX6/r5kObiVGblHkxW/r1aL/n19zYIf9n1+209+LRX8v+dioJ375v2wtMmyF3L5p+34oZVLY/6YIEmdgR/0d2zcv6+nJvwQvhttf2xc7/jPw1zND0gM5FXvq7GGF9L6K8/54YBfecQPNb37pP+MPPm8dzXnqv5YtHTY7+e/dckfV37vH/rPxv/wCf+7rbT8wN2tnvc8nX3zkt3GmbY/dr3lS/2a5ZY/NvwvZhCuJN1/pz//v/Uu7xl44pDfZx4LBC4fCAxaTz01Zde88yv9AyRW2v7fc2TeG+sv7fp95vSd63bNM2en7ZqlI/4Y8A/O+P3m3zzgX+dTp7zQ5Vsn/fG54X8tVav+muLtd/nX7GOX/C+33fX72mdueO2Mlfzv9R33xgJ/98ttU/49+zcf9A/F+JrXbNg1ktT2v54++bD/vEtSY9wPzH7Lm/xAf0m6Fjt3ac/4JQ4AAAAAAMAIYBMHAAAAAABgBLCJAwAAAAAAMALYxAEAAAAAABgBKee9B2Ht9H7VSs2qlybtL1Ts+uGY/XIgmUtSaTC0a1YGfrjRQvIDVjXwr4OKfmDsRtqya6YjQU1lP9kvd/3AuVSs2jUqxvpP5PuFdAOBaP1A/8n+86BDd/s1axf9msB325r0A2Wnhg27ZrMQ6weRttRrh9pyhZ69aT/guSO/n1YKfhiyJKX0ZX7i4ReQl9/jpUfuBoL+xmf9mpY/ltvjcuA5TJN+SG9ev2TXqBD4N6mBF1ItKTSXpQOnrc8PP/k7fhtH/eusqYN+O3U/pDgvn7Nr1PCfgVTxxtS8ddVuQ63A2Lh0m12Td1ftmsj8kMYX/HZK3loz76zYTaT5W+ya/Myn7BqN+wHSaf6k305grMlNM9z58rN2G5qb92uq43ZJmgrMA8884LezdKtdExnTXfnxT4fq0qt/eH/WLZf+rZ96HZlT67GwYW0v+zXmOPRnuoF1dLStYN9KB79rT/edX+IAAAAAAACMADZxAAAAAAAARgCbOAAAAAAAACOATRwAAAAAAIARYCW6Jnn5Su3+tvV5SerID3EtDwJhtpIKKto1taIf6NUr+u2Uq4FwqPaOXVIqBMKQ634Y6XjJD5ntlP0crrGSf93cfv2CtYLfv6tFP8x2OObfo87QD1+sFvzvVh7412AwOWHXFJL/DE2V/MDC/tDv26VBz66RYoHIlXos1NdVn/DDDrd7N+ya7sBPSS8XYkFxk8Fcus/3a7veWPaGg/6/lUTGid2xgV3zB1e9+eyOmY7dxuRwza55qOeHrb5u0Z//Hl/354u5sh9Qeqx3xfr8hw6+ym7jSxf8tdPPPe3PfV9++LpdU2r4Y/5ioD8X5I2pvWl/7Via9f+WYvLvTafhr3+6Q3+Qmwj05ws75nVOZbuNQ62n7ZrH6364/nTJ72fH+34g9kR5zq7ZrHnj7fqRI3YbK23/3pyo+vPAUsHvz4/M+WHIt9b8MW2s5K0p1jqX7TZWT52wayTpzlDVi7WW/PbrRf9wos2ePz9IUicQiLzhd0NJUqvmv8cv1GLr/HopcICJpL3G+vNLHAAAAAAAgBHAJg4AAAAAAMAIYBMHAAAAAABgBLCJAwAAAAAAMAKsYGNXseAHZlXlB+CmFNuLioQUD7IfAlfOfihrpJ1iIKypPPAD96rJD2oqNQOhy429Rjv9Z1l+eFpzsGnXSLGA1ayhXTMWCOidLPjBoBFb3RW7Zrzs39dCDoRP9/2Q4p7adk0khFaSqgo8RwO//6gcCAIOXLtICF4kSPq/trmaF4TZ6vv3bLzkB64+s+Xf55751R5Zrer4hDc3bXT8+e/0tP8cztUO2TUbnS27ZnHMDzjc6XkJjK9b7Ojnn/Hm89cN/LlvueWPq41AEOz1lr8WnDNDXSVpq+v2taQPXfbC4r/1Nj9cvjP011mRebI98NdZkcMgOoNlu+bYuDfWr7T915Odnv8uEBmfqkV/fErJD3j/8GX/fWiq4g3q01X/eb4RuDeTlXW75rE1fx6sBA4cOTjmHbAQOWjjI1diB/DcORMqe5HI2NAb+v38F54OrFElfcMp/5puBJePkWf+kdXY/TtQ958vSfqqo3v7HL/EAQAAfyl3Awcx7gYOYtwNHMS4GziIcTdwAIw2NnEAAAAAAABGAJs4AAAAAAAAI4BNHAAAAAAAgBHAJg4AAAAAAMAIsCLG3UTunP2QrUbRP4knp8DJNZJS9k9zKAwDbQVKit1AEFzVT4zvBk5NiJxCo1LFr2ltBGr8k6a6E7HU8Wkv0P5PRU4KKgeu3ZZ/ekQe+qeuNCv+RZgoz9s1Cnw3Ffznoe4fiKJW2U/wl6R6wQ8vzTfO+g0F+lyaPmzX9LM/Zk1WDtg1BfknC+ynzsAb0KerE3YbkRNjntzwn8Vvvc07+uLCzmW7jVOTJ+yatc7zds0/f8A/leT+wCF+Jyf8+e+5be8UrGLy1yYnGrfYNT9yv//veOvdK3ZNtbhr11xt+idanTTn8r92iz9mlQv+cTGREzCf3fYno9OT/txaWb9u19w3d9D6fHfoj00fu+LPq3fP+jVHx/1TIvvDwGlr5SW75ui496zNVv3g+chJU/ddvWjXPFbY43E7f07kBNjTU17flKRq0TsFbL3jj4HffZt/2tx+iryPP3DDH4O+/Q7/+kuxE5mv7PonnknSOxf9Z/7Hn/XflSXpq4/FTqfaK36JAwAAAAAAMALYxAEAAAAAABgBbOIAAAAAAACMADZxAAAAAAAARkDKRrhvZ/B+K22vUqj7X6gfSBYtBoJfg1rDHbumLj+cK++s2DWpMWfXdAp+2FU1+0FoannBjpKkoR90FQmzVcULNXtBfv5RuyYdusNvqOeH9eVeIHU50M7unJ8MOj7wn9fI87AzdXMCZbuDWOBZPZDVrBTYd+/4Y5YagRC+wDjcCQSruwH7LygXvjqWgP/58keteTCvXbCbSGV/7swr5/2aq14Aelrw+0U6drddk598wG/njtf47QSC8EPzklmTZvwQ0MjYkK/7Qen50F12TQrMlSoExroN8zCEWuBQgzF/Xoncm9VpfwyY/aOH7Jp0xn8+7b9nd+2lb0OSWoF1z5wfOBxaz1b9dWaaPW59Pq88Y7eRn3jSrkl3B/pMYH2QSoHnM7KeN98b8tZVu4n8xON2jSQV3vRj+7JuyX/8j+20/Nzy3wkKX/J2u0aS8uVH/KKVwLgiSfP+Oj8df3WoqfzEH4bq0p0/uKf7zi9xAAAAAAAARgCbOAAAAAAAACOATRwAAAAAAIARwCYOAAAAAADACHhJg42T/Dym5sAPGowEKEtSzn6ob6M8Y9c0+/7f1B36AW0T5Xm7ppj9fbwb3eftmlLBDzUb5oFdE7mnteK4XSNJZzfNEEVJE2X/b5qq+EGuyy07w0yLY35fWJDf51ol/x5t9fxg48WyHw7aSX5oaaRvS1Kr74cj7vT8ILfu0O8LU5VJuyaiVvRDQiPziiRVil+zLwGBN9o/bV3QrW7XbuPY+DG7phQYy9938br1+ZmqP369ueZf9oez/xx++LIfgjlRtkv0Tbf6BxWM97x784HVbbuNu2f9NcPDq34I6JFxvz+vtf0DB87M+vPEQsV7bvLVx+w2nh4PBA5X/b752VV/3L5rxg8iXRr6c/gfN72560jD7zOrHb/P1It+n3li3b+f75j023koEIj+zKbXbxbqfhtfuuTPNe3AAS/F5A+2v3nhhl1z75w/Dp4cen0g0GVUD6x1JKle+tp9Wbdc3HmPPaAElo5aaQcmVUkLNf+kjxONW0Jt/d41/x32xETg0CVJrX7stzJ3znw7wcYAAAAAAAAvF2ziAAAAAAAAjAA2cQAAAAAAAEYAmzgAAAAAAAAjwE8OM0QChzd7XtCiFA+mbQfCgze61+yayHWIhPpGaooFvwuMlabsmtDfI79msrJg12jgh+5J0tKYH7ZbLTbsmkj/WRzzQ3NDz1HHv3ZN7do1kesWCSnO8gMLix0/TFKSOoHrUC74AZmSH8jWG/o1M9VDdk1kXOgOmnaNJAXywb+gc5te2t/rDvjhe5F5sN33wybvmPYyE680/dDCf3/Nv/CNkl9zz6zflz56xZ//IgcVXO5648qblvx5rD3wr9l9836fubjjB7mXCn5CZuhgg43LXkHFD3aeqfrPQL3kB8X3h/4cXkh+HzifV+2aYfb6wHRgXTZW8tfnMxV/DvrsjSt2zfWaPz8eDay1SwVvbfHIqh+6ni89ZNfsHvDDsCPvDRFjkbdacxkbOUQmtnbbP7Wif2EiB6Q8uBI76ONdp/yxa73nv49LUiEQFX28HfvNy8r4S/tbGX6JAwAAAAAAMALYxAEAAAAAABgBbOIAAAAAAACMADZxAAAAAAAARkDKee/BRb3h71gpR5GQtYL8mqRASpFiwbmtvh82FwllLTX9wMFhww8Oi9yj/tAPsy31/JDZSOhglh/ElYxn4C/oBQJtsx+cm1eesWuaB4/ZNWMrfsBfOnTGrgkZ+P1ne7hh11QKfp8rFWJBboNhz65pD/xxYbI4Y9f0kj82lrM/lnTlX4PImCVJpcJXxSaKz9d8nzdgBMaJ4aQfBFrYWbdrlLx/x8nbfuBymj9p1+T1S3bNbzT9Z/cd03aJ1PTHlTTu3c986azdhip+4G5aus2uyZtmeLCkNHXYb6frB78PZ5aszxc2A/255B80oIL/76V2SLOkdOC0XROZW5/teGuFk1eX7TbS8XvsmtA1m/b7pvqxwzBcuRUY012B+++OZ5Kkqj8+D0p+GG9h/apdk8bMNVIpsOZr+e+OkqSJd+3LuiVvvNd+0dkKpERPbvprVElSN3BgRc0PjJekfO4Ju6Zw5jWhtiLvfJKkyXfv6b7zSxwAAAAAAIARwCYOAAAAAADACGATBwAAAAAAYASwiQMAAAAAADACrNQiN1AyRwJ9NvxQqiwpTQSCIDt+kNKw6v9NkVDf3N60awqB8DzV/WTHYiA4LA8DwcaR/mMGdUqf6z+BoLbcfOmDRCWFQtTGdgMhYYF2IkHftS2/b6vtt5MWvJBLSar2IiFkbXXK/n2tbPv9pxwIx8vdi3ZNacwfFyLjfWX+FrtGkbFE2rd/ssh1L0A+BUIw0/Wn7RpNLvo15n1O47PKD33Ea2PBv8dpbM6uecdwxa7RdiCcvtvxaya8e5OO3al85Smvjcj8H5ljIwcOrF3w2wnMR8Wu2W/qM3ZQc9685rUhKR1+hV8zc9SuCYWn1sbtklOVg17B8YPKO+bzGQmPjfTNq37IaZr1701ovRgIe3WDmnNn224jUpMC97Pw/CN2jSo1v6ZU9T7fbUplr51IULuk4LE9X0CgL01VAu9FCgYbDwNz0UX/0BdJSqfvtGvyjedibb3Eh7+8LH6JE9nAAV4QStoHPieygQOMInsDByH2Bg5CIidtwWdv4CAkdNIWfOYGDvBS4e0DAAAAAABgBLCJAwAAAAAAMALYxAEAAAAAABgBbOIAAAAAAACMAOt0Kldx7SaFxgVO/5CkHDgBpHLibr+di4/aNZr0T+fIG/71Htb9kwkKkUTwYfZrKoHuWfJrcnPDb0eKhZv1AqehLK/5NZWyXzPjnbojSa2BfxpELXC6mXb8ZP3xQCp8Xj1r1/TnDtg1klTZWvaLrt/wa2b9+5offNCuSfcErvfyObsmcsqCJKVjXxWqe9F/p2+eTtSY9RvpteySfO6zdk06eZf3+Ve80W5Dg8BpYvVJv2b7ul+yeMiumWgG1hvmKVDNU/7pGY2i/5xf7/gnbixWGnZNa8Y/PKB2NXAaiHmK4c7MjN3ExJQfHhsJ9s1T5glQktKmf2qQAjVp9rhXEDiZKG/664RknhwoSRrz+0DeCqxhbvjz/eDM663Pb/f89cHM+Em7Jq9f8mvOBdYUt32RXRM5pS9fM9d9bX9+3jh2zK6RJL93/hdETm5r+SfJtmbm7RpJGmtO2DU5srZS7HS5/ox/2q0klbYD72+StMcD0/glDgAAAAAAwAhgEwcAAAAAAGAEsIkDAAAAAAAwAtjEAQAAAAAAGAEpZyNwdvhhK512mPww20K/Z9eEAmalWMhsIKQxV/0gwN7Q/26Vvn+9e+WiXVNMfnhwoeMHgUXC8FTwv1tzGAgDlDTIgb4akAJ7reOB8M3dsT0maf051aLft/vZ/27F5Ac1l7uB+1MMhGkPvdDS/1wXCHxNfl/oV/znqJD8cSHST7P8a1fIya7508IvDxb+RZ9e+SlroD3h5/VpuuKH6D29+bxdM2tOnddb/iU82vDn56e3/DEicp3Pbvh/z62T/riyeNULN/zp/hG7jTOz/jXrD/2/f7Hu//3Nvj82zNf8seHSrjd+twLf65Vzdbsmsla4tOuvAQ+N+fNXozRt11xveSHijbI/f0fm/P7QfwZW234/m6/78+MwD+yaFXO8bQ38/jxZ9r/XwTH/QJSJsh94O8j+GulG+4JdUyt6f09vaB5uIOlTy4HDRiT9lePfsi/rlrXOe+0XxF96xm/6SCO2Hr5rxh/vrjRj1zQy77255N9zSdqcDixMJE1XvnFPX5Jf4gAAAAAAAIwANnEAAAAAAABGAJs4AAAAAAAAI4BNHAAAAAAAgBHwkgYba9cL85Ok3ZofGDYWCGaTpKH8QK/OYNeuGesFcqkCAb2qjtklnUHTbyb5QZWRkOuISHhcaWcj1FYzEAQcCYyttQKBlYHguUiQYOr4z0NIIHB4Wzt2zUQg67zfmPSLFAxuDIRCl5IfbBy6rxV//InMEaHwaUkae+e+BATm5fd4g1nb74fp6H12TV7zAx1T1QzdG5+124iEQEbG8WrLf3g7dX8uq1x/zq4ZLt1ufT7J76pp5Vm7pjV7wK5xQ0AlqT3wn4H61fN2jWredxvOH7ebKMifw/UnH7NLOne/xq6pFvyDBlLbP9ihU/HG4MrGit1GaszZNbm96bczfdhv5/IjfjtTfjvueLvbW7ebGLt63q7RWOCdq++vXToL/jWrXvPHZzXN9U7NDzdPR+61a/608C37sm6x39+l0EEavRwLAC5tLPtFXf/9VZLy4q12TeTdTZLyqr8uk6Q0960EGwMAAAAAALxcsIkDAAAAAAAwAtjEAQAAAAAAGAFs4gAAAAAAAIyAYELkHg37dkmx4AcNpn4sSKkZCD6NhDRWqgt2TSSMdJD9613tDe2aTtkPkKykQBDYoGfXFIr+ddOkH+4oSZvNp+yaifK8XbPj5yer17th11QK/j3KRT8rbTvw3arJD2ycHfg1nTG//2x3r9k1klQvmqGykuolP0Q5tfywx07Vvw7V9pZdMxj3AxLbfT+IU5L83vCFbUx5/6XpBT9Eb6Pn96lBw/83maG8e7bQ90Ntd4d+2Ga14Idk/9vn/HnpW077Ad7leX++KJp99tyWH7Z/atp/lq7u+GPxrWb/l6T6pv/39A7dZtc8tHrd+nx9/aLdxmTZD92eu+tuu6YeWANm+eu5zaIfDtrttqzP1yb8cWO7d8WuqdQC81bXD10eLvhr+nOb/vx4q7nWnqkestv4bMN/Nmer/rvGWMlfX/Y7gbXVjN/Ojbo3pt02tWS3sd5+xq6RpMX6W0J1n2+rv2rXtAZ+n22UZuwaSdK0P69G3iUkadi5HKqLmAz8XZK015GMX+IAAAAAAACMADZxAAAAAAAARgCbOAAAAAAAACOATRwAAAAAAIARkHI2gkkHH/RSTDt+cHDe8oLpJEnFYD7zhh9olk6+yq7JD33Mr+n5ob5pYdavOfUau2b46Q/ZNdrxw/NCAsF2KRC6F21LO14goCTl3cC1KxXtklT1E5SHr3qzXVN44KN2TR74YZLp0KJdo0D4oAJBp5KkdmB8vOqPWeoHrt1EIAZ43A+jTcfusWvydmCOkJTmvy2FCj+//Ss/Yc2DaSLQDwOHAqjhj/+72Q/2bRSn7BrXIPkBremTv2vXrL7qLrtmfs2/ZunAabsmX37Ea2Pxdr+NXT/sUtm/NyGlwFxe8/tm3vHG1NDzHLhmy9kf6xee9oOateiH4a5N+geQzF64ZH0+3Xq/3UbkXWCY/fmx0AqE65f9a5avnbVrBofvtD5f6viB8LkT+Pv7XbtkZ8o//GGs5Ae8F1uBtdiN56zPp2P32W38qbfsy7pFvd/2TyFpBgLpJ2LBxqXLT9o1kXlVktTz+7xqsffEvO6Ney9IC9++p/vOL3EAAMBf6mZs4MDfwEGMu4GDGHcDBzHuBg5i3A0c4KXCJg4AAAAAAMAIYBMHAAAAAABgBLCJAwAAAAAAMAKCicB7k9uB8KtAmJ0UDJubOerXBIKedOo2uyRFgi2LgWu3u2aXFM74YcihMMSbFaAohUJBQ98v+fumaeD3hdxa99sp+sHGxd1AEPBtr7BLUsUPzc3NwDWYWrJrJEmFwFA6vuDXNPzQPgX6jwLXO1X9EMJI+H10jtg35vXMPT/MPJX8Z1FdPwC9kcwA9P6O8vI5r2bDXwcUAqHxueCPrfO7gczI2qRdkq885hWkgnaXjlsl4yU/eDEFAh4jYchpzA+7zF0/QDo/85Bdo9l5r41gsLprIfA8a6zu1wTmh3rJewZat5zRWNfLVY3kQqX5k37NdXM8k5THA4eI1AOBr+aYVrx6Vu1Fb9zola2PS5JqTb/PRELHe0N/7iwGgpo19Nbyafa4H1i7479rSZKC5668iPk3StJGYDiZuhEIV5eUJv13+HzxT0Jtuc+VJKXFW2NtRfYMDC+LX+KENnCAF0Q2cIAXRDZwgBFkb+AgxN3AQZC5gYMYdwMHMe4GDmKiJw4B++1lsYkDAAAAAADwcscmDgAAAAAAwAhgEwcAAAAAAGAEsIkDAAAAAAAwAl7SRM7BieejAAAgAElEQVTBtB843Bn4JxJUiuYJG59TdE/mkFSo+ydTbFT9JPcnA4HWZ6Zqds1k34+mvzhcsWtaA3+/sFr009Qv7vinutxTj6XGf3q5Z9ccG+/aNc9sBU6NSn6Q4GTFf/Zef8AP0rvcfMquaXb9U4xOV/3A6t2qf1rNWClwYpSk3b5/etZTHX/IXu/5/ee+cX9sLCb/eZgo+8GiBcXG+/3y3pZ3qtjRoj/+H274p749eMU/UczVHvintxXrfk0hcGjUZxV4ds/77dw26Y+txeSthb5i4I93f7zs1zyy5o8nheSfQtcOnID0tccD8//UCevzz234Y8ltU/7fcnrKPzHn/Sv+aUZvOeSP9ds9/xpcueydOLdQ96/ZU23//eHW7U275iMbR+ya+8v+37Ox5feBJza9+fG2lj8G9AKHrE5V/HHjSPbvzXsf8d9p3nbUX1ctjplrl4Y/Bj58JXBqlqRv8Q83/oJ+9bL/UnnNPxxMvWHDL5J0amJg15Sq/rMrSesd/3309x70390k6WuOHQzVvXOPn+OXOAAAAAAAACOATRwAAAAAAIARwCYOAAAAAADACGATBwAAAAAAYASknI2Qvv4HvES/th+ypYYfRqpu06+JGvqBZsp+clhu+SFgadYPmQ1kQUrNQOpyIZChXQzUDAMpbZF7KkmBsMZQX9jxg6TTlB8mGrpHpYpf0/JDW1X2A+5C360YqNm54ddIUuWlD6KVpNzyg/5SY85vKHJfA0HxoX4qScW3Rka7F9v+ZWsejFx/lQL9fXvZLkmLt3sF/UC4X+Ae54IftpraXtiqJKnnh03mrh8A747H+eoTfhszR+2avLvqtzPpB85G5srQvOeugSL3v+EHDmvtol0SuZ8hu/7BDrntrU9TbcpuI7RW6vlJrKniB7HmTmCsmTvh16w/731+JhD0et0/aEKLftpuaHwOvNvlgX/AQip6B7xE5oDoe2o6/A/3Zd2SN97rJ/IH3qdSyQ9Xl2LPbmSMkCSlwO9XOoH9jGhbktLS39/TfeeXOAAAAAAAACOATRwAAAAAAIARwCYOAAAAAADACGATBwAAAAAAYAQEEyL3ZtDwAw2bfT8ktF4OhGNKKqVAiGkgSKmd/UCrUs0Pph0M/Xaq2e8Cw/qEXdMZ+EFg5YK/x5hK/t9TbAXCOiXtVv28sXopEIo47tekvD8Zrn+pQJhqd9wPOczyM9mqgZDiziDwDEUCuKVQsHEn8IxXpg7ZNb3sPxPDxrhdU0h+P02Khdnt12S3VfX6YmXMH8tbfT8kujThB9sPshfW1y/5/WJlyw92nq/7wcYLA38dsFHz+1LXy8CUJLU7l6zPHzx8h91GP/DMqn7QLtnq+oHDY1V/zC/U/ADlfvaem3bBD6ssd/y1zPTsYbum2fcDh3vDjl9T8sOdZQ71Y4HBNwVG7G7VrxkLhMg3K/51ngg8n2t1r2Z66IfDrk/6Y221e9WuKQUOmmjKf9aKyR+ga+ZasTweCOsPrlsCRxx8Qb0J/9Cg5iBwwE5grS5Jqeavh5t9//tJUqPsv8OWUuDdTbExWZL22sP4JQ4AAAAAAMAIYBMHAAAAAABgBLCJAwAAAAAAMALYxAEAAAAAABgBKWcjhKj3215i0aBvfh1JgfCrSNiwJCkF9rAiNb1AcFwkLHUYuA4363r3A6GLkWsQaae8X9FhexB5JgIBz+r4AbiqBwLCI89DN/DdIu1E+mmknZvZf9p+4K1qgfs6CDxHN6svRI29c3/Svjvv9+bBSP8IXMuhBnZNYWDWROaLyJgc6UsFf77oZT90sDwMdCMzODMSYF4uVO2aSGB8ceCPq/3A7YwElKaWGXQZeTYDgfmR/hzpA9WiHw4aWpOYc+uw6Ifn2mOTFBqfdnt+8HqjFxgDIuvzsWnr4zlwUEBk3ihGXrk6foh4p+rfz0Ly+5o9pgfGgMhcI0nlwlfvz7rFfX+XQu8RO/40JEkaL/nBy4pemcizeDP3GSSp8OV7+uv4JQ4AAAAAAMAIYBMHAAAAAABgBLCJAwAAAAAAMALYxAEAAAAAABgBXhKgGwZUCYSsRUJcmxt+jXTzAklvVkhx5O8ZBkLtAqFe3RQITwsEVaZSIAxx7Xm7RpLSzFG/KBC8ly89bNek+ZN2TSR4Lm9ctmvS7HG7JhSoGug/kaC0vHzOL1LwOkRCigOGgeeo0A0E95nBjZIUyGLdV3n1Oa8gEA6Yjt7r16xdsGtUm/JrzGcxb123m0gHb7dr1PJDv8uB0OXcNsNzJaVKw/p8VVJ+9nGvkVIgPPaO19o1ueQnV5Y6u34768/4NZG1lrlOTcVAgPSVp+yadOt9dk1vGAjqzoFJzwykL0jKl8z+vHSr93lJubNt15Rml+waBQJ38+ZVu6Y/5o0bylJp/ZpVUrh20WtDko6fsUtyyw+QrvYn/HbOPeLXTPvtaHLO+njZnAP+TGCK/oIic13geSrWZuwaScqXHvKLou/wW35f1FxgnJDCwcZpj92LX+IAAIC/XGQzFTZ7AwcxkX9ohM3ewEGIu4GDIHMDB3ipsIkDAAAAAAAwAtjEAQAAAAAAGAFs4gAAAAAAAIwANnEAAAAAAABGQMrZOOqj9eveuSBmgr0k5X7ghJNg+rOyfwJUqvtR4Xlnxa5R2z8pSHX/5Br31AxJyi3/dA4N/GT00AldkdPNAqdtSYr9TdVxv2Yz0H8q/gkakWuXJv3E9sgpBRr4p6il8QW7Jg/88SdVAycbSMrNwHVo+yfwhE7pukkBtilyOlLkRD1JmvrGwDEsL5Yv//hLfz5WJHA18IzYJzrVAuNXRGRO31z1a6YCAZW9tl9zM67bzppfUwmc7tEN/P3js35NpD+7Y0Pg5LjQ3xIZg28ETn48eMquyV3/5DB7fRqZTyI1W4ExoB44NSgyBjQC/cY9USiy/g2c6hdax0ZE+kDglFX7lKPIu0nkOktKJ79/f9YtV/8Pf90Suf7BvzM0568F1tCS0pk32DXRU2jDp1Mt/f093Xd+iQMAAAAAADAC2MQBAAAAAAAYAWziAAAAAAAAjAA2cQAAAAAAAEZAIG3N4IZFSVIg2DiVAiGukvKwZ9e0Kv6+Vy0SahcIXQ6F5wUCylIgqCn3WnZNKBAqEKat61f9Gkk6dNSvidyjNT9IOt32CrsmEnCY1y/ZNaGwNDfgT1JnzA+HrQa6TzQEOJXrflEkWH0z0L8jYd+7gWDV+Vv8mu1lv2Y/uYGr64Eg+KXA3BkJWncFQqXTgdN+O5HA2UA7+dJDfjvtwOELXa8mnzvvtxGQztzuF00e8Gsi65lIUPXQy+5MhwN9ZuW8XaOuv9bU0gm/ZiJwb248Y5ekKe9Ag3zprN/G7a+1a3Jzw29n9rjfzhOfsmsi40a69X6vYOeG3Ube9a9ZKDy4P/BrymW/JjLfVsx2dvz5Kb3mrXbNvoq8GwWkpTtDdfnRT/hFkXFVUn7wo35NJ7a+SlOxg0+0xyGWX+IAAAAAAACMADZxAAAAAAAARgCbOAAAAAAAACOATRwAAAAAAIAR8JIGGy+ndbvmwNicXZNbgSArSc+V/KCxascPf22W/b2yX1k/ZNd82wE/uOrA1SfsmvVD/nf7+Ut+QNm7Tvn354OrfojUzEzDrpGkT54v2jUnJrzgRUn6vz7+ZXbN6VX/2SuV/Gv3o1/iB7D+zFPJron4+tnLdk0/+9/t5x+Nhav90zv8fvfjz/iBgr/78BG75uf/mv/sTdRP2jWRMN7c2fbbkbRfve5frx20Pn9kbNFu47gbnizp1y76IfWbLW9u+qev8a/9zz/qj0VfvOg/U18a6Evv+Phtdk1j3L833/O6LevznznqP7PtQG7o4+f8OewVC35DgSWQ3rzkPWeS9IvPmGHxZ/2/5fnNE3bN6w/7feZD3+3fm9u/57pd88aD/hz+z3/OW6d/yetvtdt46lfads1bzniBy5L0a7/tr01nZl9t17zxDv/d4df/mddvvupd/vry8Nhhu+Zjl/xXxzsX/GfgDYF54BPy59vzq97q4Dvv9Pvmg2ev2DWS9HfuCJW9yPde8d/bFgJnb/zBh/z7LEmvvsMPRP7m22Nhzf/Th/x9ht3dQMi2pFfd6c39L/hXe/wcv8QBAAAAAAAYAWziAAAAAAAAjAA2cQAAAAAAAEYAmzgAAAAAAAAjIOVsBGE13+elZqXAHlGp4tf0/UBDSVLZD3RTc8OvifxNhUDmdM8P28oDP8A0NfxQqMh3U9G/brnlh2im+oxdIykWytr1w+26E/73Kya//5QiOefDQIhZILRVbT8cLA8CgcOFm7evnSqBQO3KmF0yKPp/U7EfuEetQIBbZGyMGnvnvmQb57Wf8dMjXX1/vOzN+WG45ecftz6f5gLh1cXAuBIYI4aNKbumsOPPF3lnxa5JU2bgauA5D2n7QelqzNol+eKDfjtj03aJO5fn5qrfxpQfBBtZl2jzml9Tn7RLIn9Pt+QNpZWuPxc3y/4wWyn4SaylFHnn8MfnbsH/e8rba15BYA5OM0ftGnX9QNnBhD9uFNb8wynssVay3+t64/5cUzLn2hekY/9of9Yty++xO2Bovo+sAyXlbT+UPXSvFXsPS4VYsHEexg4+SdPfvKf7zi9xAAAAAAAARgCbOAAAAAAAACOATRwAAAAAAIARwCYOAAAAAADACPASB80A4bx51fq8JKWFW+yaUAiwFArObTf8wMFaKxC83AsEDlbH7ZIUCMLLG4GwsfEFuyYSMhtqJxJWLcWu97jffyrLz/jtRJ6jtn8dIiGfkTDbVPYDC1uTft8eaw/smkjQp6RQWHMkLD4SUpx3/cDP0HN0s8LvJaV9yop1++LNClsvr1y0a7S2aX08T7XsJtK4H5w6yH6fLaaiXRPqS4FQT7efR9ZOuhiYJ+54jd9OQDr8CrsmX37ErzEPUEjTft9UNTCQPB0Idg7MX8MFP4i02PFDesuPPmB9Pk/6If5js37gbu4H5qBAH8jL5+yaSiRAuGYG6AYOSwjN9Ts37JJCJLh26Yxdk689Ydeo5QXdlrb99WhoTb6P0sSiX7RrBmtLyuuX/Hak0PtUvvxkrK2xwKEis8djba0H3pEkaY+5/vwSBwAAAAAAYASwiQMAAAAAADAC2MQBAAAAAAAYAWziAAAAAAAAjICUc97zh/Paz+z9w5LS1JL9hdzw5D9jBtpJ0qDhB8cVrvihWYOl2+2alfZ5u2ax7gdnpSuP+zWH/JDC9e4Vu6ZR8sM9mwMvqPMF02mPKVJ/zvrQD/1KgSDXsaIZbqfYdagX/eehsu2HtmrqkF2y2/fbGW8P7ZpmvWzXSFJ93Q/6S405u6ZV8ftPe+CHpM8M/ZA5DfwwWmX/Hqlc82skqfaOFCv8i260f9qaB2tF/1r2hv58dmnXrzlY9y9JMXkHCTwQyPX74oN+8GDkeb9W9INQZ6r++NUfduya58yx9cyMH2q71nneromETj/v5YZKku6eOWjXPLPlrzN+7YIXVPxtt/vB1gdqp+yaZiCkt5/99XOrv23X1Er+mDYY9qzPV4p+eGwx+fN3s++vldY7/nU+3Thm16z2/TDguaoXoPx7V/1A/Nce8NfLn7gePEDE9PpFf7283fPXb+51joybklQtvn1f1i1b3V+y1i2SNDn0n8E/3vKvpSTdMumvbd178ILI+8TVpv++J0mHG4HDdiSNlb5uT/f95fFLnMAGDvCCyAYO8ILIBg4witwNHMS4GziIcTdwEONu4CAm+lILD9cZ/614eWziAAAAAAAAvMyxiQMAAAAAADAC2MQBAAAAAAAYAWziAAAAAAAAjADvdKrV/8dLt46cPLLiny6gseBpJd1A2NrMol+z7ifMh75bIbAnt+gn5uvqeb+mP/BrCoFQ9m4gAX56wq+RlC8H7mupaJekatWuGZ71Txwo3HncrtHx03ZJfvxhv51A306n/NPatLrst3PEvwaSlJ/1T4bT0D5gQKpV/JqImt9P1fBPklDXPx1GktKxf7QvpzwM/+gHvFMaG/4JD3nVP8kjLcVOQbBE7nFkLqv51ywdudeuGX7s1+2a0FhU9Z7BvO6fmJPuCozFT5/322n4YcDDi9f8dhYCBw40zUMuZv3xp/eH5+yaypedsWs27rrDrpl+/Em7JvR8jpt9ILL+q/gnTeVLV29KO+m0f9pYfsQ/2bbwpq+0Pj/8+IfsNux7KSkV/XWsTt3q11w8b5fknt+f05T5DhB4ZvJz/kmAklT4up/cn3XLb32Xv3hs+6cqpvvutGuk2PVJi/Ohtm7a2CL5c9LnFP7Ke/5/dDoVAAAAAADAyxybOAAAAAAAACOATRwAAAAAAIARwCYOAAAAAADACChZnx6aobFlP3A4r67bNWlsya6RJAUCY+1rIMUChyNhSJF2Bl2/JhBSl1uxcCdbJIjr4IFYW82WX1PyHjlJUt1/jgqv9YMUQ6HLE36Yao5cg8B9DYk8d8VgcHDgeuf1LbsmVQLXO2LHDxxO836wet71Q1/30/DsZevzhdOB+SnS3wPPVb5+w/r88Jwf7je44feL8n1H7Jp82f9u/Uf9gxRKt/lj3uA5LzA9lQNz+VPP+TVbO3ZJIFpdqRwIQo0ESC/MWp+PHE5QGAuEYu76z8DEL/y2XROID1ZhLhBU/aQXQlq8w3+eI2PgcNmfH/sX/JrKTVqPDP/oo15BJKQ6IF9dsWtC6byRQ2uu+odT5L73XpfmZuw2hjuBdy3t4y8tIu9Gt5+0a/KFwOFEktSOXZ+IyMEReSW27kwLgcM7DPwSBwAAAAAAYASwiQMAAAAAADAC2MQBAAAAAAAYAWziAAAAAAAAjICU895j6vLye6xMu1SdsL9QJCQ09wMBs5LUDQRx1v1Aq9zzv19qzNk16gRCCpuRIGn/GigP/ZKBH9KWSn5YdR7GwuDSZCCwtBcIzo1cu8h9nT7st7Pqh2mmiUW/nXYgVGzoX7fImKCKHwwZlSoNuyY0Pg4CAe6Bfhp6HkqxIOm08O2hbMUXab7Py3aN9MNIgP7YtF2Si15I6yD7Y2Vv6N/jcsEPtCyu+8HGvWk/1L4yDHQjc12Tr5+1m0gzR+2amzV29eT3m3LPj+ntl71w785g126j0Q4c7DDwQ0Xb4/76uR5ZyiT/33J7FW/cKEcCd/t+0GlkrktT/ronciCIO9ZKUhqafS0ybwSuc6TP5LULdk3/4K12TXnbX/vah/BExs3ANZMkFd+6P+uWzvv9TPpAuPxW3rZrJGmyGwi/D64FI32+VYsdEFIfBA8Wqb59T/edX+IAAAAAAACMADZxAAAAAAAARgCbOAAAAAAAACOATRwAAAAAAIAR4CXumAGNkXDVUMiWG0r1gkhg7JYfnhgJtsztLb+dSBhp0Q9dypuBaxANoDLlSDBtJAxOUm77QdKRoLBQ+Gukz+2u+e1EnqFWoG/XJ/2ayP0JPA+KPKtSaNzK64F7VBv3a25W4HDoegfu6z7KzzzsFYzV/UbqfoC1rj7t15jjUbHqB8cXW4Fg7abf/9Ir3mjXlK+fs2tyJKAy8jyZ8sozflFkvdX3g31L0wt2TeQ6F68/b31+LDJPzs7aNVpesUtqk/64nWuBsabr981SxZu78k4g7DSyVpryDwTJFx/02wmNAX64c66Y421kLXuT/hb1/e9Wyk/ZNTnyHuTOUTV/HoyGyKelt4bqPl8kLD+yRp3YCKxRJeXxwDo1MkZI0pVrdkltcT7UVI48K5LSLW/f0+f4JQ4AAAAAAMAIYBMHAAAAAABgBLCJAwAAAAAAMALYxAEAAAAAABgBZrCxGRgVCD0NhRRH2pGkaiBIKRAcliqBkMqIQHBWXrtg16T5k3ZN6B4FAspyd9euSbPH7RpJUiBEOe+u2jWduSW7pjYMBMZGAu4iYbaRkM9IWF1tyi4J9Z9yIExSUu53/LZmjgba8YNl0/Rhv52Ny3aN1q77NYHwyv2Ujt1ufb4z7odyVzb8INS0cItdExmP0sSi18aVx+w2Ivf4Rt8PK5wf9wN3I2NeKHR46qD1cfe+SFJefc6uSZFx9fqzfju33G/XKDIWm+umfPFRv41j/rMZO9AgME82pv2akrlOr08rNbxnOrf8g1FSoWzXaMwfa4YN/xlIT/2xX7Nwwq6x19qFwFoxILc37Zr1Sf99cLrijZuSVNj258HBhBdwXtyOBf7ul9D6YNtfn6VDgbFOkiYO2CWtYeyQi/p84DsOAocASPGDl/aIX+IAAIC/VGSjAAHmBg6CgifGwONu4CAo+g/asLgbOMBLhU0cAAAAAACAEcAmDgAAAAAAwAhgEwcAAAAAAGAEeIlWHTPIdcsPCVUtEJQqScPs1xT84DTV/eC8fMkPDwwZCwQojfmhy/nph/x2KoHAuYjI39PckLa2Am0FrnfFr6k8/rBdkxcC/89u5Bnq9vya/sCvmZ7wa8qBPtfyQ4CzJM0GskK2/DC9XEh+O8t+oF6e9YMbQ2pVv2Y7MG5L0j7FjAwf8gIqKxOBAP3Afc7Ty347VTPUtbnhh4wv3/A+L0l9P6B19lF/vZEXA1kdR2/1a4ZmVsX6FXuuyLuB4MydbbskF/xxS9N+gHR+7gG7pvMLn7Brql95xis4fMhuIz/0oF2TbvFD7ENr7sjazFz75O1l/xlo+sHO+YB/b3TtebskTfpjer4cCPEP3Jt81QvFT5G14kZg3Nj2Q2hn7r7brtGFj9gl7sq3ICm3vP45jKzdJBXe+O5Q3efL5x/xazb9+6xq8B2+73+/WuS9QFKuxQ4jCbV1KXDgh6TCG962t8+F/uv/rYm8fAIviGzgAC+IbOAAoyhyShx8gc1++OwNHMS4GzgIcTdwEONu4AAvlZfHJg4AAAAAAMDLHJs4AAAAAAAAI4BNHAAAAAAAgBHAJg4AAAAAAMAI8E6nanfN/3rR+7wUS8qPnJATbStweo1u1kkvBe92SpKagdMMZgNp9n2z70ixMLxux69ZPOLXSNJ24NSVtt9/0l33+u10zZPkJKkUTJV3RU5RSYH95kgIa2RMaPunLkiK9e9y4DSn+ZlAO4HrELnegVMuQieV7aPCfa+1Pp9z4DS2lUt+zeQBuyRVvNP8citwMtjxU35NZL44HzgFshAYVyLPu7l2Skt3+m30Aqf5rAXG4sgY2dzwayb8cav2D77J+nxuBvrz9at2Sbo7EKBcDKznIoHYgYMd3JOW0qlb7DZC65FW4JCKQyfsklSNnJYZuDer3hozNbzTBiVJjcDfEli7pOMn/HYCJ1Gme77Yb8e8N2ngz0/54p/YNftq3j+eMy2d8Nu5fsGvkaSpwEmRkWdKUpo7btfka2djbR1+aQ8+4Zc4AAAAAAAAI4BNHAAAAAAAgBHAJg4AAAAAAMAIYBMHAAAAAABgBHjJaW6Y1ZgfOuQGpkmSdgIhrpLSXaftmry64rcTCRuLhJ4GgqTz1cDfc/yQXRMKkGz6QY2a98Ox8pOP+e1Iypf9YOPCiSW7pv+7n7FrSm+4w67J65t2TZq6OSGzg8fO2zXFrwgE3HX80NL80KN+O1Io8DzNBELSI8Hvk+N+TWQcXvSfh0iA637KF72Au3x91W4jcp/zY39g18gcy5s/9Um7ifo3vdquiYwrobFoK/C8n/1ju8YNN2z/6H+w2yiM+UG4pdsX7BpNemHYkvxDMSSp7welD657QenDTX8sKS75Y2MhcM1Ca7MF/9CJyLyvG2bg7GKgjYof4p+v+O8P/U/+oV1TujMQIh8Y04cXr5lt+GPg8LPP2DWFmUCgbOV5u2R4xe83g3V/fC6/9qT1+RQ5+CHwPryf+r/+YbsmlQPvlM3YQUP95/1DLgpTgYM+JJXv9sOGN9/3bKitqb8dOJhGUtpj9jK/xAEAAAAAABgBbOIAAAAAAACMADZxAAAAAAAARgCbOAAAAAAAACPgJQ02jgQ6apj9mulJv0ZSvnjZrknHDvvtPH7OrgmF+k4HwiAjAZrnLtg1kdDlUF+47ocNqz/waySlKT+wOm/64V1pouK3c+mqXRMJ086R+7q8ZpcUput+O5GQ4gtX7Jp0+oRdI0n56nKgxg+7zMuBwOrJQAhfJLw8EqpZi4XZpaOhshfJ27ve51e3/EaaLbskEig4+JQ3N1VffdBuQ8v+OiBHQrIjNYXk1wTkjhfsW5z1n7/CfGCMrPgBnZFQz8j9zGvecyZJxfu9QP/CufN2G+moH8aeL3kBtZLU+aAXoC5J1fv8wN1Q4Ko5BueHn7SbcMPAJSlf8K9zYSFw8EjJDxF35w1JSkVvfBqc99e/xUP+O0Dnk5fsmuoXH7NrCkdm/Jo7/PuZ6t4zkFc37Db6n44F41ZeESp7keKBQCB9YK2Vb/jvOJJUCcxfqREMiw6860y9zX/3l6TBI/6zIkmFN+zxc6H/OgAAAAAAAG4qNnEAAAAAAABGAJs4AAAAAAAAI4BNHAAAAAAAgBHgpXO1vXC+UABev2/XpKof/CpJiny/QBhyJHg5FDg88AN6c88Pw0yL83ZNKNg4EDgcuQZuqNmfaXf8mkAgXnFh1q7JLT8YO/K8RgLCdDwWEObKzz1v10T6Qo6EaUtKjUAw9q4fEFq43Q8UDI0LkXE4EF4e+W77ynzu0+HAeBkIiU6hYdkblyPja2iMCAThpntvt2siIf0pMh6veGHupftP220osHaKzBORcP5IQGbhlsAayA1Kj4SkB8asNOGHila/ygtplmSv0SWFwq1daWnBLwr8LWlpzq+xK2JjQGgcNPtN2g6seQJ/S/Xt/rOZZvz3oMhBDpFDYbL7rhF4Zspve7Vds68O+M9GaB4+4QeSSwo9H2nKP8xHkvLl637R5HioreLBwMLMwC9xAAAAAAAARgCbOAAAAAAAACOATRwAAAAAAIARwCYOAAAAAADACPBSVt3goXE/uDMdCIQAlYPBtINACNxjZ+2Swhu/wq5ZLfjhgXPZDxsbPvj7dk3vVX6eOukAABtfSURBVG/0a4Z+2Fgh+WHIvaEfNjzZiUTbScsl/x6lFGvL9XuX/ZDZd91yyK7JD37YrtGrvswuWe9esWumH3zYrkn3frFdM/z1X7FrJCm94T675tNV/xm/bcrvCzNdvya3zVBRSboWCGIcC473++Q9B19rff7dt/htVIt+EOrXfr8/Xr7uG80xrCAt1L1g1z/4ET/M/cv/iR/S+5uf8EMOX3/XEbvm8JgfbFs7ZZfot7/PW2/9xC/7z8W3fof/t3znv/Tn8s+s+n3gu+70v9v3vDsQHmv6oZ9t2TUHx5bsmrf/gL9WuPvrt+yaf/Ea/99yv/Z/9kNqf+L7/7/27jzKjrJO4/jzu327093ZOgtZCBECIQmLkICJBkFQIzOOelRcB4+7jo4zgxsuM46OOu7bzLjNcTmiqDgObgeXUUAhoiwBYkiAsASSkI2Q0J3O0kkv977zx60+aZp00s9rB7jx+zmHQ7q6fvXerd6qervu826x1v/sG/xj0DM/7T+Xb33HD/Z94Su9oHJJWvkh/7O580KzT2uRnn/aHqvk8pv8c4r2TX5fM/1+//yg3OcHvJ97sT/xwSkT/ONNo/lRayz5/ZkkLc6qerRXbVtk1/T82J/IYMOpGQHKkr70mgfsmrU78yY1Wjlxjl2z6RMZkzpI6hqb9xgvP2t463EnDgAAOCR3AAd53AEc4InMHcBBHncAB3ncARzgcOGjCAAAAAAAUAcYxAEAAAAAAKgDDOIAAAAAAADUAQZxAAAAAAAA6kCkNPygwnTT+71Uw5I/RpT2dNk16vHTyCX5s21Jiln+bBbpvg1+zS5/BoSYMMauyZlBTO0Zs9BkvEdpX14auCvG5c12k/ZmzG7W63/m+rbstmsaZ/kzDlTa/VlHGjJmd0mbtvk13f5noeHMuX47G/wAxpg7y66RpPSn1X5Ntz+LQtrj73ultlF2TXVXRjsnTLVrVM0L1y094wsjMjVc9cb3WQ8g3Z/xmRrjz2hQ7ciYAXCS1/9Xd/jHpdJsf9Y77fBn2cn5XFQ2ZxzLKn47YSZhVrb650HR7M9OVl54nF2jPr8vTls77JpqxvGofIbXF6dev8/qvWmtXdO4OGOKuozz05z3Jkdl9eEPKm6Y7p/P9tyx3a5pWnysXaPd/v7Z94DfpzUu9Ka1S5u22m2kjP4sp38qH+PPHljJOPdVyT/MN8yd5hXs8GemTV1516kNF146MuctP3uj/UZXtvpB2Q0z/JnrpMxzi7aWrLYqD/rvXylzlqmc829JKr/hsmG979yJAwAAAAAAUAcYxAEAAAAAAKgDDOIAAAAAAADUAQZxAAAAAAAA6kDZWTlVvNC0aPUDY2PKRLsmbfGDUiXlBcdliCfPs2v6jp5j1zR2PGjXaNN6uyTOPM9vp+yHQqVdfkjbzlbrIy1JGt/nh7hKUtpyj1/U5ge5jmqb4bfT4L8Opb1+8F71j1f77Zx2ol2jo2baJen6P9g1McN/f3L7n1hwkl2zcZIfGjetxQ+fLnf5gXaxOWN/yJERKjiS0lOXWOuXJt3iNzLGDyavTjnOrrlv12Zr/Y17Gu02Tp7gB9R2V/wsxxu3jrZrjl7gB4WfMdkPXC2XvOPfqI6MPuVhP2y2d/YCu+bW7Q/ZNaWT/fDUhWvutGs6F5xu17jGn+q/ZlmTaIyfbtfcu3udXTNr7JPsmg1nPGCtf3zjFLuNyzf7obYveIF/fIxOP3R7z1i/rxkTfr/RJS/sdfdp7XYbRzUfZ9dcsc47bkjS06f5E4GMLrf5Nd0ZEx8kb/98uNEP4U3Km5BhclbVo5XOf75dc/s+//O0fFveRDEvP8Hfp0J5mc83b9th15zTmBca3zvBDM02cScOAAAAAABAHWAQBwAAAAAAoA4wiAMAAAAAAFAHGMQBAAAAAACoA5HS8MOWqte+w0pmitGt9gPKkbq78wr7/KCinOeU8/high9sqS4/QDI97Ie6xaQJdo2a/EBM9fTaJW74tiRF5IVjqW2sX7PPD3fLCuCuZoSojfODxbI+cx2ddk2Mz3itS/4YddrrP5/cz0/q9j8LMckP+svp59KeLr+djP01S8bzkaTSki9l7uiPlO79pLdz5TzePj9wVxP9SQHUY37em/1wTjVlnAf0ZHz+HvSDfZVzLMvov+xjxe6M59/sB/TH1Nl2TWr3J0NQpx+ar/F+SK0d6J9zbG33z5k0JmMfyJgEIWu/yQiqtj9rOed/kfE35owJHTQqo09bf79fMznj2D3Gi7WNJv8cLj24xq7J2m9yjoPjMvqAvpxzbPN8uZRxKpFz7JQU0982Muctd3/cvyjY6U9woZl+ULokqWO7X7Mv89p/esZkMRkTv0jKug6RpDj24mG979yJAwAAAAAAUAcYxAEAAAAAAKgDDOIAAAAAAADUAQZxAAAAAAAA6oCXAmYGV6ZyRqBvTtBgRhipJKnc4Ne0Nvs1OcHLbRkhaNrhl1QygsOam/yacRmhXhnhaZETZrv6PrtGkmL2PL9ozV1+zeSMfWJ0RjD2Hj+wM933gF0TM6fbNTmhnTmhy1nhvBMy9iFJOWl1cewpdk3q3uW3054REpsRMpc6MsLimjP67RGUNm31CjKOM1kB+nfc7bdz4rFeQZf/WcoKDs2pmTXXr9nph7pmhX5va7dWj/kL7CbSPavtGh2/0K9Zu8ouSZ0ZfVBOEKrbf2ecl+SE36d7/DDo0rkZx/32jHDQnOPkTC8QO624xW8j43gSJ59o16Q//NGuyQr27fXPLUpnmsf7fbvtNuzjmaRo8a+D+lb5+0D5hc+2a/TAOrskbTX3m4xJTWJ2zjXdyEnmMUiSotEPJK9OPcGukaRYc69dU7ljU1Zb5b/OuD4aNyWrrZxzDAd34gAAAAAAANQBBnEAAAAAAADqAIM4AAAAAAAAdYBBHAAAAAAAgDpgBRunzj3e1jv8kK3HVLsf5Npz01q7punsOXaNImN8LSeUdXNGsOOUiXZNVJNdo8mT/JomPxBUZS/fu1+M8sPN0sxj/IY2bfZrmlv8moz9obJtr11TPj0jVCwjqLnv99faNeWz/LDqh79wg10jSRNfe5JftCkjGDtDVuDw9o6RfyAHMiZjHx9BPed4YYujNmcEp5f98PicINjOGTOt9Zsb/CDY3b1+oGJv1Q9bnfaw3xf1zvD396a2GXZN2rLGWr86yXtfJKk0dZtdo8232yU9J/thyJ09fnjqUSvvsWvSoiXW+rt6/SDg8SU/cLja9Uu75sEx/jnT5MmL7JrGjIk30g4vUDSmTrbbeOi4aXbNUbf64d57n/c8u6Z1+4N2Tc45fdriPZ+Y6Qei7z7rLLtmzO+vs2vKz/YfW9+PrrJrGuZOtWtKZz/TWr96e0ZQd3dGIP4Iijl+8H/lSv99bmhbZtdIUpz1N37N9Iz3QVIcc7pd097nH8MkacLuw3uvDHfiAAAAAAAA1AEGcQAAAAAAAOoAgzgAAAAAAAB1gEEcAAAAAACAOhApDT88refLL89Ip/VEQ2TVNczICJnd3WPXxLQ2u6Zytx+I1DDdD5Ds27jLrslR2e4HdJUznk/q7rNrcl43SapsfWxCx0qtfohyzmMrtY3ya8b7Nd0r/TDNng4/gLV1nr9/lxf44aC9y9bbNZJUPnac39aaHY9JO6r43fa+m/3gxsYZ/r5XzeiD+zoyAtwljf3qVXkHl0GqP3m99YJW2v2Q3lJro11T2WpOPCCpYepou8YV45r9ool+kHnPNX4QbtMpfuBq31p/v815natd3ue8stmfSKJpUUbQfs4ECm1+v9W3cqNdUz7en3RB1aq1eu8qfzKI0kR/H4hm/1yhNNcP3c6Z0CBH72rvdSsf4x/zu5dtsWuan+sHvubsA9Vt/v5ZOsEP6a2u817ntMd/Ljnn5pFxTGvI2G967/YnWMg5X67u8s5dyrMyJvSQ1PCyb4/IeUvlstfYJ4I517sq5d0bUlnt77u517yNJ/mT5uRcu0l553+S1PimHwzrfT8i7sTJGcAB+j1WAzg4MmUNrADAENwBHGQyB3CQxx3AQR53AAd53AEc4HA5IgZxAAAAAAAAjnQM4gAAAAAAANQBBnEAAAAAAADqAIM4AAAAAAAAdcCanWrfR19gpVuXJrbYD6g0tsmuUSUvnC4nkTxnNp6mOf6MCaVTZtk1aYOf7t17p/98Guf5yd5xrD9rQtrlp/lnJaPv6/ZrJMV4P9A2dfgzQcT8BXaNtjxgl6R7/ZmZUq+/78X4Vr9mQsZsNUvvtmtyZsTITZ/P0bh4jl/UljFz32132TXVLn/Gir71O+2a3FkCmt754xGZ5SHt+J43y8M+vx9La+61a2KqP9OS3fcdfZzfRiVjFsiWCX47Y3Oef8Z707HhsLfTdcxsu4nWnf6+FKMy+obOTXaNRmfMGtXlzwKmRm82m5h8gt1EWnmtXZMj5i2ya9KK6w7DIzmAJnOmoWnT/TYy9k2NyficPbTZrxmXMZlBlz97oHZ4s+8413P9SmeeZ9ek5UvtGo3xz/s0MaNPj4xrgF7zHM7sZyRJe/3+WZJi1ntH5rxl2zf9D0eDfz2+9+Pfs2skqeU9F/hFzXmzEWcd96f6x2RJiq1r8uqmvfUvZ3YqAAAAAACAIx2DOAAAAAAAAHWAQRwAAAAAAIA6wCAOAAAAAABAHbASIlPVy0XKCaAsTfcDTFNHRgCa8kJZRz3laL+hjGDRLOP8kKfS2Iyg3ZwAzQwxOiMILSPYOGUGG2uK/zpkJZTt2u7XtPqh4nH6PLsmLV/ttzNjql2jst+XNJ42za6JSRmBqusyghElyexPJUnT/eekbX54uR1cKUk7/P0omhvsmtJ4P5B+RHWY77cZTilJ0eLvv9rd5de0mgGNOSHFOeG5vXv9dnY+5LfT1WHXaEe7X2Mey1o7/D4/Kwx7/ll2jXoygtwr/nujzowgUHeygZ0P2k2kvRnPv88PfVdOgPTRGceHHRkB0tOf5K2/yZ80Qc0Zk5zsyDjW5YQU78645shpp9k71sUo/xog3b/cr+nOOA5knMMpZUxa057Rd27z+vSY1Ga3oYkZodsjaXNGIP+MY+2Sln+50G9Hkjav82v2+Z9DSdKUjMmGdmb0LZKSG5pt4k4cAAAAAACAOsAgDgAAAAAAQB1gEAcAAAAAAKAOMIgDAAAAAABQBzKSpoav5w4/YKp3mR8elFJWXKyaJvihmltu81+yWW/329nxnVV2zfiXnWjX3HmpH7r05OkZAU+T/SCw3pvW2jWlVj+QtborLxxr7cf8sL7Zr/VDPrtv9cMXq33+PtEwzv9sd9zvtzP9Qn9/WPcNPxhz1OiKXXPU2f5n+wfvygiGlPS080bbNTOXXZHVlqthsh8qXtnuB+s2zvaDpKudmUHkIyTd6+331Y1+eG73Cv/z3tvt/01m9BlewN+mX/mBnlNP8wO8o9F/Lk2n+kHzlW1+gHJlu19TbfdqHlrjh7pOmeuH5zY/6J+jrb/E7yNnvsLfzysb/EDw8mzvPGPdt66x25j5PD90vPE5p9s16fqb/Jpu/zPQeaV/fjF28d3W+g1H+xOWqKfXr8mYRKSyZqtdE43+OUy10z/XbpjuBRX3rMw4N2/wz+H2rPffm/GL/GuAaF1n15RmTbFrVPbez+qq++0m+tavsGskadQ/X5RVN1jvdf4kJO3XL7NrxpmZ5/023OIf8ybNPLyhwQN17/H3eUma/FRz8ojCqHe/Y1jrcScOAAAAAABAHWAQBwAAAAAAoA4wiAMAAAAAAFAHGMQBAAAAAACoA5HS8EMHd73tOVZCYaXiB2a1nOCHffZt9APwJKk0MS9wyNW9xQ/OHXPOVLumb/1OuyYnJLQ0xg8PbjjKD0otTfYDBFNGSHHPXe12jSQ1n+8HSVfu9QNLc16HaPFDwvrW+QG9fWvzQn1dTQv8/aHndj+0M3X5gX2REaYtSdHsB6WNOm+231CT//h6b7jPrqm2+yFz+x7ygzibWqp2jSSN/uLVeQn4g1Que42X1JsR0pu6/Nclh/sZjKl+QK0bGllrxw8pVjXjc5Gxb1T+6E86UJrsBZTGicfZbVRvudOuKc3z20kdnXZNHO+nXaYV/vOpPLTHWr98zql+Gzf6AaE9qx+2a1re9hy7Jt3ifzYrGf12efFJ1vpp/Wa7jcoG/9yiYWZGeO70o+yavmX32jU51xulp833CpLfB6Z7/AlEtNufxGDv0gfsmtY3nWPX5HzW+u43rwEyjjWNT59n10hSnPmxETlv6fvWhfYMA6UnTbLbiZPn2jWSlG68NasuRyzw+i9JUtm/ppKkdPNtWXWlCy4Z1vvOnTgAAAAAAAB1gEEcAAAAAACAOsAgDgAAAAAAQB1gEAcAAAAAAKAOMIgDAAAAAABQB6zZqQAAAAAAAPD44E4cAAAAAACAOsAgDgAAAAAAQB1gEAcAAAAAAKAOMIgDjICISBHx+QE/XxwRH36MH8O3I+Klj2WbAAAcThFRiYgVEXF7RPw8Itoe78f0lyYiPhwRFz/ejwPAyIiI3Y9hWx+NiCUH+T3XLxkYxAFGRrekCyJick5xRJRH+PEAAHAk2JtSmp9SOlVSu6R/OFwNRUTDoJ8jIo7Ic2XOOwAcbhHRkFL6UErp6hHc5mHpuwZv94neRx6RBybgcdAn6euS3jn4FxFxbET8NiJWFv9/UrH82xHxhYi4RtKni790fSciroyIdRFxQUR8JiJWRcSvI6KxqPtQRNxc/FXy6xERj+kzBQDg8XGDpBn9P0TEe4rj4cqI+MiA5a8plt0WEd8tlj3ir739f4mOiPMi4pqIuEzSqog4LiJWR8RXJS2XNDMizo+IGyJieURcHhFjitp1EfGRYvmqiJhXLB8TEZcUy1ZGxEuK5QfczkARMT8ibizqfhoRE4rlsyPi6uI5LY+IE4rl7y3auS0iPlUsuzYinlL8e3JErCv+/bqi3Z9LuvIQr+EHIuLuiLha0tw/830D8AR3oOuViGiIiPuLAe22iKhGxDOK9a+LiNkH2M664lrlD5JeNrDvjYhPRcSdRRufO0DtvxfrlwYtvzYiPhERSyW9PSKOiogfF33XzRHx9GK9ofre3QO29dKI+Hbx7wNdi309Iq6UdGnx/D87oI98S1F3XvGYfhQRd0XE9/uvxyJiYURcX/TJyyJi7FDb+XMwiAOMnK9IelVEjB+0/MuSLk0pnSbp+5K+OOB3cyQtSSm9u/j5BEnPk/RCSd+TdE1K6cmS9hbLJenLKaWFxV8lWyQ9/7A8GwAAniCidpfMsyVdUfx8vqQTJS2SNF/SmRHxjIg4RdIHJD0rpXS6pLcPY/OLJH0gpXRy8fNc1Y7bCyTtkfSvqh2rz5B0i6R3DajdXiz/b0n9Xzn6oKTOlNKTi2P/76J2p+7BttPvUknvK+pWSfq3Yvn3JX2leE5nSdoSEc+V9CJJTy2Wf2YYz3WxpNemlJ51kNfwTEmvlLRA0gWSFg5juwDq26OuV1JKFUn3SDpZ0tmSbpV0TkSMknRMSmnNENval1I6O6X0P/0LImKipBdLOqVo42MDCyLiM5KmSHp9Sql6gG22pZTOTSl9XtJ/SfqPlNJCSS+R9M1inUf1vcN43oOvxc6U9MKU0oWS3lhsb6Fq/eCbI2JWsd4CSe8oXpvjJT09Ipok/VDS24s+eYlq13AH206WJ/RtQkA9SSntjIhLJV2k2g7bb7FqJ0GS9F098iTr8qKD7Pd/KaXeiFglqUHSr4vlqyQdV/z7mRHxXkmtkiZKukPSz0fyuQAA8ATREhErVDsG3irpqmL5+cV/fyp+HqPagMTpkn6UUtouSSml9mG0sSyltHbAz+tTSjcW/36aaifpfyz+0Nqk2h1B/X5S/P9W7T/WL1FtEETFY+iIiOcfYjsq/gjUllJaWiz6jqTLI2KspBkppZ8W29tXrL9E0iUppS7juV41YL2hXsOxkn7av92IuGIY2wVQ34a6XrlO0jMkzZL0SUlvlrRU0s0H2dYPD7Bsp6R9kr4ZEb+U9IsBv/ugpJtSSn83zG0ukXRy7P8ywriin3xU33uQ7fUbfC12RUqp/zrufEmnxf67OMer1kf2qHbc2ChJA45RnZK2pJRuLtrfWfx+qO0MPO5YGMQBRtZ/qnb79SUHWScN+PeeQb/rlqSUUjUielNK/etWJZUjolnSVyU9JaW0IWrhyc0j8sgBAHji2ZtSml8McPxCtUycL0oKSZ9MKX1t4MoRcZEeeZzt16fiDvTitvemAb8bfCwe+HOoNvDxt0M8vu7i/xXtP6+OAzyGQ23nYIb62vSB2pEGPFc9+hxh8HM70Gv4jiG2C+AvR38fcJ2kt0o6WtKHJL1H0nmSfi9JEfEbSVMl3ZJSelNRM7hPVUqpLyIWqXZH5Ssl/aOkZxW/vlm1OwEnHmQweuA2S5IWDxhsUfFYhuoTBy47WJ84+OeQ9E8ppd8Mauc87e/7pf39/1DtH3A7fw6+TgWMoKLj+V/Vbpvrd732jwq/StIf/owm+jue7VH7Lj1p7gCAI15KqVO1O10vjlpG3G8kvSH259PMiIgpkn4r6eURMalYPrHYxDrVbpOXal9Zbhxm0zeqdpv87GJ7rREx5xA1V6p2gaKiZsJwtlM8x46IOKdY9GpJS4u/5m6MiBcVtaMiorVo5w3Fv4d6rgc7TxjqNfy9pBdHREvx1+0XHOL5Aqh/Q12v3KTaVzirxV2AKyS9RbXBHaWU/qoIn3+TDqLoZ8anlH6l2teQ5g/49a8lfUrSL4s+51AG97Hzh1g+ofjn1og4qcjaefEwtt/vN5L+Pvbnks6JiNEHWf8uSUdHxMJi/bFRC0h2t3NIDOIAI+/zkgbOUnWRpNdHxErVTsiG8/38A0op7ZD0DdW+XvUzHfxWRgAAjhgppT9Juk3SK1NKV0q6TNINxVeQfyRpbErpDkkfl7Q0Im6T9IWi/BuSzo2IZZKeqgP8pXiINrdJep2kHxTH8RslzTtE2cckTYjaBAS3SXqmsZ3XSvpssc58SR8tlr9a0kXF8uslTUsp/Vq1jKBbitv5+zN5PqfaBcP1euT5yODnNtRruFy1ry6skPRjFRdrAI4YrRGxccB/79IQ1ysppW5JG1Trs6RafzBWtWsRx1hJvyi2v1SDJoNJKV2uWj99RUS0HGJbF0l6ShESfKdqdwpJB+h7i+XvV+1Ozt9J2mI85m9KulPS8oi4XdLXdJBvMqWUeiS9QtKXivavUu0P8NZ2hiP2f1sDAAAAAAAAT1TciQMAAAAAAFAHGMQBAAAAAACoAwziAAAAAAAA1AEGcQAAAAAAAOoAgzgAAAAAAAB1gEEcAAAAAACAOsAgDgAAAAAAQB1gEAcAAAAAAKAO/D+5TfVam5dvUQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x720 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "grid_kws = {\"width_ratios\": (0.4, 0.5, 0.1), \"wspace\": 0.05}\n",
    "fig, axes = plt.subplots(1, 3, figsize=(20,10), gridspec_kw=grid_kws)\n",
    "\n",
    "for label, ax in zip(['Normal', 'Recurrence occured', 'Low-risk recurrence'], axes):\n",
    "    sns.heatmap(data=expression_matrix.loc[:, labels_aliquots[label]], \n",
    "                cmap=\"Spectral_r\", vmin=0, vmax=18, \n",
    "                ax=ax, cbar=False, \n",
    "                xticklabels=False, yticklabels=False)\n",
    "    ax.set_ylabel('')    \n",
    "    ax.set_xlabel(label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig.savefig(\"./plots/expressiondata.hormone.recurrence_status.png\", bbox_inches = 'tight', transparent=True)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}