exploring.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.backends.backend_pdf\n",
    "import seaborn as sns\n",
    "from statsmodels.formula.api import ols\n",
    "from statsmodels.stats.anova import anova_lm\n",
    "import statsmodels.api as sm\n",
    "pd.options.display.max_rows = 999\n",
    "from statsmodels.stats.multicomp import pairwise_tukeyhsd\n",
    "import numpy as np\n",
    "import networkx as nx\n",
    "from  scipy.stats import spearmanr "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option('display.max_colwidth', -1)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read table with RPKG (generated by parser_bbmap.py)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "rpkg=pd.read_csv(\"TSV/RPKG.tsv\",sep=\"\\t\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read table with deepARG + plasflow + kaiju results (generated by parser_kaiju.py)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "args=pd.read_table(\"TSV/deepARG_plasflow_kaiju.tsv\",sep=\"\\t\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ORF_ID</th>\n",
       "      <th>ERR594345</th>\n",
       "      <th>ERR599133</th>\n",
       "      <th>ERR594405</th>\n",
       "      <th>ERR599037</th>\n",
       "      <th>ERR594291</th>\n",
       "      <th>ERR599168</th>\n",
       "      <th>ERR599087</th>\n",
       "      <th>ERR599024</th>\n",
       "      <th>ERR594412</th>\n",
       "      <th>...</th>\n",
       "      <th>ERR598997</th>\n",
       "      <th>ERR599015</th>\n",
       "      <th>ERR594406</th>\n",
       "      <th>ERR599007</th>\n",
       "      <th>ERR599146</th>\n",
       "      <th>ERR599159</th>\n",
       "      <th>ERR594353</th>\n",
       "      <th>ERR599096</th>\n",
       "      <th>ERR594342</th>\n",
       "      <th>ERR594354</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>gene_756282|GeneMark.hmm|1215_nt|+|1342|2556|TARA_PSE_RAW</td>\n",
       "      <td>0.003365</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000284</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0022</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000501</td>\n",
       "      <td>0.005935</td>\n",
       "      <td>...</td>\n",
       "      <td>0.003321</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001699</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000136</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000292</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>gene_4591261|GeneMark.hmm|1452_nt|-|2|1453|TARA_PSE_RAW</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000748</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.010127</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002459</td>\n",
       "      <td>0.000462</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000498</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>gene_3406871|GeneMark.hmm|783_nt|+|1951|2733|TARA_PSE_RAW</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000412</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000499</td>\n",
       "      <td>0.007779</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.001131</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000355</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000255</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>gene_6732935|GeneMark.hmm|792_nt|+|360|1151|TARA_PSE_RAW</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.001338</td>\n",
       "      <td>0.000632</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.002792</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001270</td>\n",
       "      <td>0.001615</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001817</td>\n",
       "      <td>0.024539</td>\n",
       "      <td>0.003357</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>gene_3501111|GeneMark.hmm|1065_nt|+|119|1183|TARA_PSE_RAW</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000585</td>\n",
       "      <td>0.010766</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 379 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                      ORF_ID  ERR594345  \\\n",
       "0  gene_756282|GeneMark.hmm|1215_nt|+|1342|2556|TARA_PSE_RAW  0.003365    \n",
       "1  gene_4591261|GeneMark.hmm|1452_nt|-|2|1453|TARA_PSE_RAW    0.000000    \n",
       "2  gene_3406871|GeneMark.hmm|783_nt|+|1951|2733|TARA_PSE_RAW  0.000000    \n",
       "3  gene_6732935|GeneMark.hmm|792_nt|+|360|1151|TARA_PSE_RAW   0.000000    \n",
       "4  gene_3501111|GeneMark.hmm|1065_nt|+|119|1183|TARA_PSE_RAW  0.000000    \n",
       "\n",
       "   ERR599133  ERR594405  ERR599037  ERR594291  ERR599168  ERR599087  \\\n",
       "0  0.000000   0.000000   0.000284   0.0        0.0022     0.000000    \n",
       "1  0.000748   0.000000   0.000000   0.0        0.0000     0.000000    \n",
       "2  0.000412   0.000000   0.000000   0.0        0.0000     0.000499    \n",
       "3  0.000000   0.000000   0.000000   0.0        0.0000     0.001338    \n",
       "4  0.000585   0.010766   0.000000   0.0        0.0000     0.000000    \n",
       "\n",
       "   ERR599024  ERR594412  ...  ERR598997  ERR599015  ERR594406  ERR599007  \\\n",
       "0  0.000501   0.005935   ...  0.003321   0.000000   0.0        0.001699    \n",
       "1  0.000000   0.010127   ...  0.000000   0.000000   0.0        0.000000    \n",
       "2  0.007779   0.000000   ...  0.000000   0.001131   0.0        0.000355    \n",
       "3  0.000632   0.000000   ...  0.002792   0.000000   0.0        0.001270    \n",
       "4  0.000000   0.000000   ...  0.000000   0.000000   0.0        0.000000    \n",
       "\n",
       "   ERR599146  ERR599159  ERR594353  ERR599096  ERR594342  ERR594354  \n",
       "0  0.000000   0.000136   0.0        0.000000   0.000292   0.000000   \n",
       "1  0.002459   0.000462   0.0        0.000000   0.000498   0.000000   \n",
       "2  0.000000   0.000255   0.0        0.000000   0.000000   0.000000   \n",
       "3  0.001615   0.000000   0.0        0.001817   0.024539   0.003357   \n",
       "4  0.000000   0.000000   0.0        0.000000   0.000000   0.000000   \n",
       "\n",
       "[5 rows x 379 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rpkg.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Merge the tables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_results=pd.merge(args,rpkg,on=[\"ORF_ID\"])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>#ARG</th>\n",
       "      <th>query-start</th>\n",
       "      <th>query-end</th>\n",
       "      <th>read_id</th>\n",
       "      <th>predicted_ARG-class</th>\n",
       "      <th>best-hit</th>\n",
       "      <th>probability</th>\n",
       "      <th>identity</th>\n",
       "      <th>alignment-length</th>\n",
       "      <th>alignment-bitscore</th>\n",
       "      <th>...</th>\n",
       "      <th>ERR598997</th>\n",
       "      <th>ERR599015</th>\n",
       "      <th>ERR594406</th>\n",
       "      <th>ERR599007</th>\n",
       "      <th>ERR599146</th>\n",
       "      <th>ERR599159</th>\n",
       "      <th>ERR594353</th>\n",
       "      <th>ERR599096</th>\n",
       "      <th>ERR594342</th>\n",
       "      <th>ERR594354</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>PATA</td>\n",
       "      <td>78</td>\n",
       "      <td>452</td>\n",
       "      <td>gene_756282|GeneMark.hmm|1215_nt|+|1342|2556</td>\n",
       "      <td>quinolone</td>\n",
       "      <td>NP_417544.5|FEATURES|CARD|quinolone|patA</td>\n",
       "      <td>0.893879</td>\n",
       "      <td>33.8</td>\n",
       "      <td>382</td>\n",
       "      <td>192.6</td>\n",
       "      <td>...</td>\n",
       "      <td>0.003321</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001699</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000136</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000292</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>UPPP</td>\n",
       "      <td>20</td>\n",
       "      <td>185</td>\n",
       "      <td>gene_4591261|GeneMark.hmm|1452_nt|-|2|1453</td>\n",
       "      <td>bacitracin</td>\n",
       "      <td>U2Q591|FEATURES|UNIPROT|bacitracin|uppP</td>\n",
       "      <td>0.999933</td>\n",
       "      <td>30.1</td>\n",
       "      <td>166</td>\n",
       "      <td>77.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002459</td>\n",
       "      <td>0.000462</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000498</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ERMC</td>\n",
       "      <td>10</td>\n",
       "      <td>239</td>\n",
       "      <td>gene_3406871|GeneMark.hmm|783_nt|+|1951|2733</td>\n",
       "      <td>macrolide-lincosamide-streptogramin</td>\n",
       "      <td>YP_492693|FEATURES|ARDB|macrolide-lincosamide-streptogramin|ErmC</td>\n",
       "      <td>0.899664</td>\n",
       "      <td>30.2</td>\n",
       "      <td>258</td>\n",
       "      <td>65.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.001131</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000355</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000255</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>KSGA</td>\n",
       "      <td>11</td>\n",
       "      <td>262</td>\n",
       "      <td>gene_6732935|GeneMark.hmm|792_nt|+|360|1151</td>\n",
       "      <td>kasugamycin</td>\n",
       "      <td>YP_002847275|FEATURES|ARDB|kasugamycin|KsgA</td>\n",
       "      <td>0.989528</td>\n",
       "      <td>31.2</td>\n",
       "      <td>260</td>\n",
       "      <td>108.6</td>\n",
       "      <td>...</td>\n",
       "      <td>0.002792</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001270</td>\n",
       "      <td>0.001615</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001817</td>\n",
       "      <td>0.024539</td>\n",
       "      <td>0.003357</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CFR</td>\n",
       "      <td>7</td>\n",
       "      <td>349</td>\n",
       "      <td>gene_3501111|GeneMark.hmm|1065_nt|+|119|1183</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>A9KK15|FEATURES|UNIPROT|multidrug|cfr</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>31.2</td>\n",
       "      <td>356</td>\n",
       "      <td>161.8</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 410 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   #ARG  query-start  query-end                                       read_id  \\\n",
       "0  PATA  78           452        gene_756282|GeneMark.hmm|1215_nt|+|1342|2556   \n",
       "1  UPPP  20           185        gene_4591261|GeneMark.hmm|1452_nt|-|2|1453     \n",
       "2  ERMC  10           239        gene_3406871|GeneMark.hmm|783_nt|+|1951|2733   \n",
       "3  KSGA  11           262        gene_6732935|GeneMark.hmm|792_nt|+|360|1151    \n",
       "4  CFR   7            349        gene_3501111|GeneMark.hmm|1065_nt|+|119|1183   \n",
       "\n",
       "                   predicted_ARG-class  \\\n",
       "0  quinolone                             \n",
       "1  bacitracin                            \n",
       "2  macrolide-lincosamide-streptogramin   \n",
       "3  kasugamycin                           \n",
       "4  multidrug                             \n",
       "\n",
       "                                                           best-hit  \\\n",
       "0  NP_417544.5|FEATURES|CARD|quinolone|patA                           \n",
       "1  U2Q591|FEATURES|UNIPROT|bacitracin|uppP                            \n",
       "2  YP_492693|FEATURES|ARDB|macrolide-lincosamide-streptogramin|ErmC   \n",
       "3  YP_002847275|FEATURES|ARDB|kasugamycin|KsgA                        \n",
       "4  A9KK15|FEATURES|UNIPROT|multidrug|cfr                              \n",
       "\n",
       "   probability  identity  alignment-length  alignment-bitscore  ...  \\\n",
       "0  0.893879     33.8      382               192.6               ...   \n",
       "1  0.999933     30.1      166               77.4                ...   \n",
       "2  0.899664     30.2      258               65.1                ...   \n",
       "3  0.989528     31.2      260               108.6               ...   \n",
       "4  1.000000     31.2      356               161.8               ...   \n",
       "\n",
       "   ERR598997  ERR599015 ERR594406 ERR599007 ERR599146 ERR599159  ERR594353  \\\n",
       "0  0.003321   0.000000   0.0       0.001699  0.000000  0.000136  0.0         \n",
       "1  0.000000   0.000000   0.0       0.000000  0.002459  0.000462  0.0         \n",
       "2  0.000000   0.001131   0.0       0.000355  0.000000  0.000255  0.0         \n",
       "3  0.002792   0.000000   0.0       0.001270  0.001615  0.000000  0.0         \n",
       "4  0.000000   0.000000   0.0       0.000000  0.000000  0.000000  0.0         \n",
       "\n",
       "  ERR599096 ERR594342 ERR594354  \n",
       "0  0.000000  0.000292  0.000000  \n",
       "1  0.000000  0.000498  0.000000  \n",
       "2  0.000000  0.000000  0.000000  \n",
       "3  0.001817  0.024539  0.003357  \n",
       "4  0.000000  0.000000  0.000000  \n",
       "\n",
       "[5 rows x 410 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_results.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read the table with the manual curation of ARGs and prepare a table \"curation_yes\" with ARGs that are not in any scenario and marked for quantification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "curation=pd.read_csv(\"TSV/curation_ARGs.tsv\",sep=\"\\t\")\n",
    "curation[\"Quantification\"]=curation[\"Quantification\"].str.replace(\" \",\"\").str.replace(\"Yes\",\"yes\")\n",
    "curation[\"Scenario\"]=curation[\"Scenario\"].str.replace(\" \",\"\")\n",
    "curation.columns=[\"#ARG\",\"Type\",\"quantification\",\"scenario\"]\n",
    "curation_yes=curation[curation[\"quantification\"]==\"yes\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "yes    313\n",
       "no     281\n",
       "Name: quantification, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "curation[\"quantification\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "v      187\n",
       "iv     41 \n",
       "i      34 \n",
       "ii     10 \n",
       "iii    9  \n",
       "Name: scenario, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "curation[\"scenario\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Remove from all_results the scenario i, sequnces miss annotated that will not be used to any analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "results_quantification=pd.merge(all_results,curation_yes,on=\"#ARG\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "curation=curation[curation[\"scenario\"]!=\"i\"]\n",
    "all_results=pd.merge(all_results,curation,on=\"#ARG\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Select from all_results only ARGs marked for quantification, creating a new dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>#ARG</th>\n",
       "      <th>query-start</th>\n",
       "      <th>query-end</th>\n",
       "      <th>read_id</th>\n",
       "      <th>predicted_ARG-class</th>\n",
       "      <th>best-hit</th>\n",
       "      <th>probability</th>\n",
       "      <th>identity</th>\n",
       "      <th>alignment-length</th>\n",
       "      <th>alignment-bitscore</th>\n",
       "      <th>...</th>\n",
       "      <th>ERR599007</th>\n",
       "      <th>ERR599146</th>\n",
       "      <th>ERR599159</th>\n",
       "      <th>ERR594353</th>\n",
       "      <th>ERR599096</th>\n",
       "      <th>ERR594342</th>\n",
       "      <th>ERR594354</th>\n",
       "      <th>Type</th>\n",
       "      <th>quantification</th>\n",
       "      <th>scenario</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BCR</td>\n",
       "      <td>11</td>\n",
       "      <td>370</td>\n",
       "      <td>gene_3196191|GeneMark.hmm|1182_nt|+|3178|4359</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>ZP_03283366|FEATURES|ARDB|multidrug|Bcr</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>30.6</td>\n",
       "      <td>366</td>\n",
       "      <td>205.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000313</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001247</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>efflux</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BCR</td>\n",
       "      <td>11</td>\n",
       "      <td>363</td>\n",
       "      <td>gene_4124281|GeneMark.hmm|1215_nt|+|21074|22288</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>ZP_03283366|FEATURES|ARDB|multidrug|Bcr</td>\n",
       "      <td>0.948012</td>\n",
       "      <td>30.1</td>\n",
       "      <td>366</td>\n",
       "      <td>153.7</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>efflux</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BCR</td>\n",
       "      <td>26</td>\n",
       "      <td>365</td>\n",
       "      <td>gene_2148178|GeneMark.hmm|1197_nt|+|29279|30475</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>ZP_03283366|FEATURES|ARDB|multidrug|Bcr</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>30.9</td>\n",
       "      <td>349</td>\n",
       "      <td>164.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>efflux</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BCR</td>\n",
       "      <td>8</td>\n",
       "      <td>353</td>\n",
       "      <td>gene_1187778|GeneMark.hmm|1221_nt|+|66870|68090</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>ZP_03837521|FEATURES|ARDB|multidrug|Bcr</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>31.5</td>\n",
       "      <td>356</td>\n",
       "      <td>171.8</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000314</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.006264</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.004629</td>\n",
       "      <td>efflux</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BCR</td>\n",
       "      <td>23</td>\n",
       "      <td>352</td>\n",
       "      <td>gene_1359064|GeneMark.hmm|1197_nt|+|53501|54697</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>YP_001177494|FEATURES|ARDB|multidrug|Bcr</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>30.1</td>\n",
       "      <td>336</td>\n",
       "      <td>160.6</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>efflux</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 413 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  #ARG  query-start  query-end  \\\n",
       "0  BCR  11           370         \n",
       "1  BCR  11           363         \n",
       "2  BCR  26           365         \n",
       "3  BCR  8            353         \n",
       "4  BCR  23           352         \n",
       "\n",
       "                                           read_id predicted_ARG-class  \\\n",
       "0  gene_3196191|GeneMark.hmm|1182_nt|+|3178|4359    multidrug            \n",
       "1  gene_4124281|GeneMark.hmm|1215_nt|+|21074|22288  multidrug            \n",
       "2  gene_2148178|GeneMark.hmm|1197_nt|+|29279|30475  multidrug            \n",
       "3  gene_1187778|GeneMark.hmm|1221_nt|+|66870|68090  multidrug            \n",
       "4  gene_1359064|GeneMark.hmm|1197_nt|+|53501|54697  multidrug            \n",
       "\n",
       "                                   best-hit  probability  identity  \\\n",
       "0  ZP_03283366|FEATURES|ARDB|multidrug|Bcr   1.000000     30.6       \n",
       "1  ZP_03283366|FEATURES|ARDB|multidrug|Bcr   0.948012     30.1       \n",
       "2  ZP_03283366|FEATURES|ARDB|multidrug|Bcr   1.000000     30.9       \n",
       "3  ZP_03837521|FEATURES|ARDB|multidrug|Bcr   1.000000     31.5       \n",
       "4  YP_001177494|FEATURES|ARDB|multidrug|Bcr  1.000000     30.1       \n",
       "\n",
       "   alignment-length  alignment-bitscore  ...  ERR599007  ERR599146 ERR599159  \\\n",
       "0  366               205.3               ...  0.0        0.0        0.000313   \n",
       "1  366               153.7               ...  0.0        0.0        0.000000   \n",
       "2  349               164.1               ...  0.0        0.0        0.000000   \n",
       "3  356               171.8               ...  0.0        0.0        0.000314   \n",
       "4  336               160.6               ...  0.0        0.0        0.000000   \n",
       "\n",
       "  ERR594353 ERR599096 ERR594342  ERR594354     Type quantification scenario  \n",
       "0  0.0       0.001247  0.0       0.000000   efflux   yes            NaN      \n",
       "1  0.0       0.000000  0.0       0.000000   efflux   yes            NaN      \n",
       "2  0.0       0.000000  0.0       0.000000   efflux   yes            NaN      \n",
       "3  0.0       0.006264  0.0       0.004629   efflux   yes            NaN      \n",
       "4  0.0       0.000000  0.0       0.000000   efflux   yes            NaN      \n",
       "\n",
       "[5 rows x 413 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_quantification.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#table with multiple ARGs (including scenarios ii-v) in the same contig\n",
    "multiple_ARGS=all_results.groupby([\"contig_id\",\"plasmid\"])[\"#ARG\"].count()\n",
    "multiple_ARGS=pd.DataFrame(multiple_ARGS)\n",
    "multiple_ARGS=multiple_ARGS[multiple_ARGS[\"#ARG\"]>1]\n",
    "\n",
    "#formating a table with `multiple ARGs` in contig distribution\n",
    "multiple_ARGS_distribution=pd.DataFrame(multiple_ARGS.groupby([\"#ARG\",\"plasmid\"])[\"#ARG\"].count())\n",
    "multiple_ARGS_distribution.columns=[\"Number of contigs\"]\n",
    "multiple_ARGS_distribution.reset_index(inplace=True)\n",
    "multiple_ARGS_distribution.columns=[\"Number of #ARGs\",\"plasmid\",\"Number of contigs\"]\n",
    "multiple_ARGS_distribution=multiple_ARGS_distribution.set_index(\"Number of #ARGs\").pivot(columns=\"plasmid\",values=\"Number of contigs\").fillna(\"0\").reset_index()\n",
    "multiple_ARGS_distribution.index.name = None\n",
    "multiple_ARGS_distribution.columns.names = [None]\n",
    "multiple_ARGS_distribution.columns=[\"Number of ARGs\",\"in chromosome\",\"in plasmid\"]\n",
    "multiple_ARGS_distribution=multiple_ARGS_distribution.astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Number of ARGs</th>\n",
       "      <th>in chromosome</th>\n",
       "      <th>in plasmid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>3503</td>\n",
       "      <td>689</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>365</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>116</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>35</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6</td>\n",
       "      <td>22</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>7</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>9</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Number of ARGs  in chromosome  in plasmid\n",
       "0  2               3503           689       \n",
       "1  3               365            37        \n",
       "2  4               116            13        \n",
       "3  5               35             2         \n",
       "4  6               22             0         \n",
       "5  7               10             0         \n",
       "6  8               6              0         \n",
       "7  9               2              0         \n",
       "8  10              2              0         \n",
       "9  11              2              0         "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "multiple_ARGS_distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "multiple_ARGS_distribution.set_index(\"Number of ARGs\").plot(kind=\"bar\",stacked=True,figsize=(10,10))\n",
    "plt.legend([\"plasmid\",\"chr\"])\n",
    "plt.ylabel(\"Number of Contigs\")\n",
    "plt.xticks(rotation='horizontal')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "97244"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(all_results[\"taxon_name_kaiju\"].dropna())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "98.02328511667758"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(all_results[\"taxon_name_kaiju\"].dropna())/len(all_results)*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#all_results[all_results[\"class\"]==\"Alphaproteobacteria\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "#all_results.groupby([\"superkingdom\"]).count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "unclassified Prymnesiovirus                   21\n",
       "Chrysochromulina ericina virus                19\n",
       "uncultured Mediterranean phage                9 \n",
       "Cafeteria roenbergensis virus BV-PW1          8 \n",
       "Megaviridae environmental sample              8 \n",
       "Chlorovirus                                   4 \n",
       "Vibrio phage pTD1                             4 \n",
       "Mimivirus LCMiAC01                            4 \n",
       "Oryctes rhinoceros nudivirus                  4 \n",
       "Bacillus phage PBC2                           4 \n",
       "Pacmanvirus A23                               3 \n",
       "Bacillus phage Moonbeam                       3 \n",
       "Lymphocystis disease virus - isolate China    3 \n",
       "Whispovirus                                   2 \n",
       "Klosneuvirus KNV1                             2 \n",
       "Pseudomonas phage 201phi2-1                   2 \n",
       "Salmonella phage SPAsTU                       2 \n",
       "Catovirus CTV1                                1 \n",
       "Asteriusvirus                                 1 \n",
       "Vibrio phage 1.121.O._10N.286.46.C4           1 \n",
       "Ranavirus                                     1 \n",
       "unclassified Myoviridae                       1 \n",
       "unclassified Chlorovirus                      1 \n",
       "Mimivirus AB-566-O17                          1 \n",
       "unclassified Tsarbombavirus                   1 \n",
       "Bacillus phage pW4                            1 \n",
       "Vibrio phage PWH3a-P1                         1 \n",
       "Pseudomonas phage PA1C                        1 \n",
       "Enterobacteria phage vB_KleM-RaK2             1 \n",
       "Tomelloso virus                               1 \n",
       "Ralstonia phage RSP15                         1 \n",
       "Marseillevirus LCMAC101                       1 \n",
       "Kaumoebavirus                                 1 \n",
       "Colobine gammaherpesvirus 1                   1 \n",
       "Myotis gammaherpesvirus 8                     1 \n",
       "Heliothis zea nudivirus                       1 \n",
       "Edafosvirus sp.                               1 \n",
       "unclassified Coccolithovirus                  1 \n",
       "Bacillus phage vB_BcoS-136                    1 \n",
       "Name: taxon_name_kaiju, dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_results[all_results[\"superkingdom\"]==\"Viruses\"][\"taxon_name_kaiju\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'v'}"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(all_results[all_results[\"superkingdom\"]==\"Viruses\"][\"scenario\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# generate a list with contigs (ids) and how many ARGs found\n",
    "full_Table_multiARGS=all_results[all_results[\"contig_id\"].isin(list(multiple_ARGS.reset_index()[\"contig_id\"]))]\n",
    "#full_Table_multiARGS[[\"#ARG\",\"predicted_ARG-class\",\"contig_id\",\"plasmid\",\"taxon_name_kaiju\"]].head()\n",
    "multi_args_contigs=full_Table_multiARGS.groupby(\"contig_id\")[\"#ARG\"].count().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "contig_id\n",
       "TARA_ANW-k99_1343221     11\n",
       "TARA_ANE-k99_4428305     11\n",
       "TARA_ION-k99_13063158    10\n",
       "TARA_PSW-k99_8257882     10\n",
       "TARA_PON-k99_11447815    9 \n",
       "                        .. \n",
       "TARA_PSE-k99_11048376    2 \n",
       "TARA_PSE-k99_11025044    2 \n",
       "TARA_PSE-k99_10958269    2 \n",
       "TARA_PSE-k99_10938553    2 \n",
       "TARA_ANE-k99_10006132    2 \n",
       "Name: #ARG, Length: 4804, dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "multi_args_contigs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>#ARG</th>\n",
       "      <th>taxon_name_kaiju</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>26447</th>\n",
       "      <td>VANRI</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26448</th>\n",
       "      <td>VANRI</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28333</th>\n",
       "      <td>GOLS</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30137</th>\n",
       "      <td>ILES1</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31726</th>\n",
       "      <td>MTRA</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>65008</th>\n",
       "      <td>VANRA</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67249</th>\n",
       "      <td>ADEL</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67250</th>\n",
       "      <td>ADEL</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86370</th>\n",
       "      <td>MEXL</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>89801</th>\n",
       "      <td>K678_12262</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97519</th>\n",
       "      <td>RLMN2</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             #ARG      taxon_name_kaiju\n",
       "26447  VANRI       Micavibrio sp. TMED2\n",
       "26448  VANRI       Micavibrio sp. TMED2\n",
       "28333  GOLS        Micavibrio sp. TMED2\n",
       "30137  ILES1       Micavibrio sp. TMED2\n",
       "31726  MTRA        Micavibrio sp. TMED2\n",
       "65008  VANRA       Micavibrio sp. TMED2\n",
       "67249  ADEL        Micavibrio sp. TMED2\n",
       "67250  ADEL        Micavibrio sp. TMED2\n",
       "86370  MEXL        Micavibrio sp. TMED2\n",
       "89801  K678_12262  Micavibrio sp. TMED2\n",
       "97519  RLMN2       Micavibrio sp. TMED2"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_results[all_results[\"contig_id\"]==\"TARA_ANE-k99_4428305\"][[\"#ARG\",\"taxon_name_kaiju\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>#ARG</th>\n",
       "      <th>predicted_ARG-class</th>\n",
       "      <th>taxon_name_kaiju_contig</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>26447</th>\n",
       "      <td>VANRI</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26448</th>\n",
       "      <td>VANRI</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28333</th>\n",
       "      <td>GOLS</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30137</th>\n",
       "      <td>ILES1</td>\n",
       "      <td>mupirocin</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31726</th>\n",
       "      <td>MTRA</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>65008</th>\n",
       "      <td>VANRA</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67249</th>\n",
       "      <td>ADEL</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67250</th>\n",
       "      <td>ADEL</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86370</th>\n",
       "      <td>MEXL</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>89801</th>\n",
       "      <td>K678_12262</td>\n",
       "      <td>beta_lactam</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97519</th>\n",
       "      <td>RLMN2</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Micavibrio sp. TMED2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             #ARG predicted_ARG-class taxon_name_kaiju_contig\n",
       "26447  VANRI       glycopeptide        Micavibrio sp. TMED2  \n",
       "26448  VANRI       glycopeptide        Micavibrio sp. TMED2  \n",
       "28333  GOLS        multidrug           Micavibrio sp. TMED2  \n",
       "30137  ILES1       mupirocin           Micavibrio sp. TMED2  \n",
       "31726  MTRA        multidrug           Micavibrio sp. TMED2  \n",
       "65008  VANRA       glycopeptide        Micavibrio sp. TMED2  \n",
       "67249  ADEL        multidrug           Micavibrio sp. TMED2  \n",
       "67250  ADEL        multidrug           Micavibrio sp. TMED2  \n",
       "86370  MEXL        multidrug           Micavibrio sp. TMED2  \n",
       "89801  K678_12262  beta_lactam         Micavibrio sp. TMED2  \n",
       "97519  RLMN2       multidrug           Micavibrio sp. TMED2  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_results[all_results[\"contig_id\"]==\"TARA_ANE-k99_4428305\"][[\"#ARG\",\"predicted_ARG-class\",\"taxon_name_kaiju_contig\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "22766    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "24653    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "27336    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "28927    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "46319    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "73372    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "76477    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "86485    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "94930    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "96668    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "98746    Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12\n",
       "Name: taxon_name_kaiju_contig, dtype: object"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_results[all_results[\"contig_id\"]==\"TARA_ANW-k99_1343221\"][\"taxon_name_kaiju_contig\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>#ARG</th>\n",
       "      <th>predicted_ARG-class</th>\n",
       "      <th>taxon_name_kaiju</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>22766</th>\n",
       "      <td>ARLR</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24653</th>\n",
       "      <td>VANRI</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27336</th>\n",
       "      <td>GOLS</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28927</th>\n",
       "      <td>ILES1</td>\n",
       "      <td>mupirocin</td>\n",
       "      <td>Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46319</th>\n",
       "      <td>VANRF</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Alphaproteobacteria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73372</th>\n",
       "      <td>VANRO</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76477</th>\n",
       "      <td>TETB(60)</td>\n",
       "      <td>tetracycline</td>\n",
       "      <td>Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86485</th>\n",
       "      <td>MEXW</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Parvibaculum lavamentivorans</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>94930</th>\n",
       "      <td>LMRD</td>\n",
       "      <td>macrolide-lincosamide-streptogramin</td>\n",
       "      <td>Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96668</th>\n",
       "      <td>CEOB</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>uncultured Alphaproteobacteria bacterium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98746</th>\n",
       "      <td>ADEF</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Parvibaculum lavamentivorans</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           #ARG                  predicted_ARG-class  \\\n",
       "22766  ARLR      multidrug                             \n",
       "24653  VANRI     glycopeptide                          \n",
       "27336  GOLS      multidrug                             \n",
       "28927  ILES1     mupirocin                             \n",
       "46319  VANRF     glycopeptide                          \n",
       "73372  VANRO     glycopeptide                          \n",
       "76477  TETB(60)  tetracycline                          \n",
       "86485  MEXW      multidrug                             \n",
       "94930  LMRD      macrolide-lincosamide-streptogramin   \n",
       "96668  CEOB      multidrug                             \n",
       "98746  ADEF      multidrug                             \n",
       "\n",
       "                                               taxon_name_kaiju  \n",
       "22766  Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3   \n",
       "24653  Alphaproteobacteria bacterium HGW-Alphaproteobacteria-12  \n",
       "27336  Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3   \n",
       "28927  Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3   \n",
       "46319  Alphaproteobacteria                                       \n",
       "73372  Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3   \n",
       "76477  Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3   \n",
       "86485  Parvibaculum lavamentivorans                              \n",
       "94930  Alphaproteobacteria bacterium HGW-Alphaproteobacteria-3   \n",
       "96668  uncultured Alphaproteobacteria bacterium                  \n",
       "98746  Parvibaculum lavamentivorans                              "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_results[all_results[\"contig_id\"]==\"TARA_ANW-k99_1343221\"][[\"#ARG\",\"predicted_ARG-class\",\"taxon_name_kaiju\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>#ARG</th>\n",
       "      <th>predicted_ARG-class</th>\n",
       "      <th>taxon_name_kaiju</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>23890</th>\n",
       "      <td>VANRI</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Tistrella mobilis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41811</th>\n",
       "      <td>BCR</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Tistrella mobilis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79157</th>\n",
       "      <td>VANXA</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Tistrella mobilis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81175</th>\n",
       "      <td>VANSO</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Tistrella mobilis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87587</th>\n",
       "      <td>MEXH</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Tistrella mobilis</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        #ARG predicted_ARG-class   taxon_name_kaiju\n",
       "23890  VANRI  glycopeptide        Tistrella mobilis\n",
       "41811  BCR    multidrug           Tistrella mobilis\n",
       "79157  VANXA  glycopeptide        Tistrella mobilis\n",
       "81175  VANSO  glycopeptide        Tistrella mobilis\n",
       "87587  MEXH   multidrug           Tistrella mobilis"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_results[all_results[\"contig_id\"]==\"TARA_PSE-k99_4996023\"][[\"#ARG\",\"predicted_ARG-class\",\"taxon_name_kaiju\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>#ARG</th>\n",
       "      <th>predicted_ARG-class</th>\n",
       "      <th>taxon_name_kaiju_contig</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>23890</th>\n",
       "      <td>VANRI</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Tistrella mobilis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41811</th>\n",
       "      <td>BCR</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Tistrella mobilis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79157</th>\n",
       "      <td>VANXA</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Tistrella mobilis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81175</th>\n",
       "      <td>VANSO</td>\n",
       "      <td>glycopeptide</td>\n",
       "      <td>Tistrella mobilis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87587</th>\n",
       "      <td>MEXH</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>Tistrella mobilis</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        #ARG predicted_ARG-class taxon_name_kaiju_contig\n",
       "23890  VANRI  glycopeptide        Tistrella mobilis     \n",
       "41811  BCR    multidrug           Tistrella mobilis     \n",
       "79157  VANXA  glycopeptide        Tistrella mobilis     \n",
       "81175  VANSO  glycopeptide        Tistrella mobilis     \n",
       "87587  MEXH   multidrug           Tistrella mobilis     "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_results[all_results[\"contig_id\"]==\"TARA_PSE-k99_4996023\"][[\"#ARG\",\"predicted_ARG-class\",\"taxon_name_kaiju_contig\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "contigs_all_args=pd.DataFrame(results_quantification[[\"contig_id\",\"#ARG\"]].groupby(\"contig_id\")[\"#ARG\"].apply(lambda x: (\";\".join(x))))\n",
    "contigs_all_args.columns=[\"All ARGs in contig\"]\n",
    "contigs_all_args=pd.merge(results_quantification,contigs_all_args,on=\"contig_id\")\n",
    "count_arg=pd.DataFrame(results_quantification[[\"contig_id\",\"#ARG\"]].groupby(\"contig_id\")[\"#ARG\"].count())\n",
    "count_arg.columns=[\"# ARGs in contig\"]\n",
    "contigs_all_args=pd.merge(contigs_all_args,count_arg,on=\"contig_id\")\n",
    "contigs_all_args[\"ORF_ID\"]=contigs_all_args[\"ORF_ID\"].str.split(\"|\",expand=True)[0]\n",
    "desc=pd.read_csv(\"TSV/desc.tsv\",sep=\"\\t\")\n",
    "desc.rename(columns={\"ARG\":\"#ARG\"},inplace=True)\n",
    "contigs_all_args=pd.merge(contigs_all_args,desc,on=\"#ARG\")\n",
    "#contigs_all_args.to_csv(\"../resistome_app/data/deep_with_tax_levels.tsv\",sep=\"\\t\",index=None)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>#ARG</th>\n",
       "      <th>query-start</th>\n",
       "      <th>query-end</th>\n",
       "      <th>read_id</th>\n",
       "      <th>predicted_ARG-class</th>\n",
       "      <th>best-hit</th>\n",
       "      <th>probability</th>\n",
       "      <th>identity</th>\n",
       "      <th>alignment-length</th>\n",
       "      <th>alignment-bitscore</th>\n",
       "      <th>...</th>\n",
       "      <th>ERR594353</th>\n",
       "      <th>ERR599096</th>\n",
       "      <th>ERR594342</th>\n",
       "      <th>ERR594354</th>\n",
       "      <th>Type</th>\n",
       "      <th>quantification</th>\n",
       "      <th>scenario</th>\n",
       "      <th>All ARGs in contig</th>\n",
       "      <th># ARGs in contig</th>\n",
       "      <th>description</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BCR</td>\n",
       "      <td>11</td>\n",
       "      <td>370</td>\n",
       "      <td>gene_3196191|GeneMark.hmm|1182_nt|+|3178|4359</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>ZP_03283366|FEATURES|ARDB|multidrug|Bcr</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>30.6</td>\n",
       "      <td>366</td>\n",
       "      <td>205.3</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001247</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>efflux</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "      <td>BCR</td>\n",
       "      <td>1</td>\n",
       "      <td>bicyclomycin/multidrug efflux system\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BCR</td>\n",
       "      <td>11</td>\n",
       "      <td>363</td>\n",
       "      <td>gene_4124281|GeneMark.hmm|1215_nt|+|21074|22288</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>ZP_03283366|FEATURES|ARDB|multidrug|Bcr</td>\n",
       "      <td>0.948012</td>\n",
       "      <td>30.1</td>\n",
       "      <td>366</td>\n",
       "      <td>153.7</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>efflux</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "      <td>BCR</td>\n",
       "      <td>1</td>\n",
       "      <td>bicyclomycin/multidrug efflux system\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BCR</td>\n",
       "      <td>26</td>\n",
       "      <td>365</td>\n",
       "      <td>gene_2148178|GeneMark.hmm|1197_nt|+|29279|30475</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>ZP_03283366|FEATURES|ARDB|multidrug|Bcr</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>30.9</td>\n",
       "      <td>349</td>\n",
       "      <td>164.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>efflux</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "      <td>BCR</td>\n",
       "      <td>1</td>\n",
       "      <td>bicyclomycin/multidrug efflux system\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BCR</td>\n",
       "      <td>8</td>\n",
       "      <td>353</td>\n",
       "      <td>gene_1187778|GeneMark.hmm|1221_nt|+|66870|68090</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>ZP_03837521|FEATURES|ARDB|multidrug|Bcr</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>31.5</td>\n",
       "      <td>356</td>\n",
       "      <td>171.8</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.006264</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.004629</td>\n",
       "      <td>efflux</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "      <td>BCR;MEXH</td>\n",
       "      <td>2</td>\n",
       "      <td>bicyclomycin/multidrug efflux system\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BCR</td>\n",
       "      <td>23</td>\n",
       "      <td>352</td>\n",
       "      <td>gene_1359064|GeneMark.hmm|1197_nt|+|53501|54697</td>\n",
       "      <td>multidrug</td>\n",
       "      <td>YP_001177494|FEATURES|ARDB|multidrug|Bcr</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>30.1</td>\n",
       "      <td>336</td>\n",
       "      <td>160.6</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>efflux</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "      <td>BCR</td>\n",
       "      <td>1</td>\n",
       "      <td>bicyclomycin/multidrug efflux system\\n</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 416 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  #ARG  query-start  query-end  \\\n",
       "0  BCR  11           370         \n",
       "1  BCR  11           363         \n",
       "2  BCR  26           365         \n",
       "3  BCR  8            353         \n",
       "4  BCR  23           352         \n",
       "\n",
       "                                           read_id predicted_ARG-class  \\\n",
       "0  gene_3196191|GeneMark.hmm|1182_nt|+|3178|4359    multidrug            \n",
       "1  gene_4124281|GeneMark.hmm|1215_nt|+|21074|22288  multidrug            \n",
       "2  gene_2148178|GeneMark.hmm|1197_nt|+|29279|30475  multidrug            \n",
       "3  gene_1187778|GeneMark.hmm|1221_nt|+|66870|68090  multidrug            \n",
       "4  gene_1359064|GeneMark.hmm|1197_nt|+|53501|54697  multidrug            \n",
       "\n",
       "                                   best-hit  probability  identity  \\\n",
       "0  ZP_03283366|FEATURES|ARDB|multidrug|Bcr   1.000000     30.6       \n",
       "1  ZP_03283366|FEATURES|ARDB|multidrug|Bcr   0.948012     30.1       \n",
       "2  ZP_03283366|FEATURES|ARDB|multidrug|Bcr   1.000000     30.9       \n",
       "3  ZP_03837521|FEATURES|ARDB|multidrug|Bcr   1.000000     31.5       \n",
       "4  YP_001177494|FEATURES|ARDB|multidrug|Bcr  1.000000     30.1       \n",
       "\n",
       "   alignment-length  alignment-bitscore  ...  ERR594353  ERR599096 ERR594342  \\\n",
       "0  366               205.3               ...  0.0        0.001247   0.0        \n",
       "1  366               153.7               ...  0.0        0.000000   0.0        \n",
       "2  349               164.1               ...  0.0        0.000000   0.0        \n",
       "3  356               171.8               ...  0.0        0.006264   0.0        \n",
       "4  336               160.6               ...  0.0        0.000000   0.0        \n",
       "\n",
       "  ERR594354     Type quantification  scenario All ARGs in contig  \\\n",
       "0  0.000000  efflux   yes            NaN       BCR                 \n",
       "1  0.000000  efflux   yes            NaN       BCR                 \n",
       "2  0.000000  efflux   yes            NaN       BCR                 \n",
       "3  0.004629  efflux   yes            NaN       BCR;MEXH            \n",
       "4  0.000000  efflux   yes            NaN       BCR                 \n",
       "\n",
       "  # ARGs in contig                             description  \n",
       "0  1                bicyclomycin/multidrug efflux system\\n  \n",
       "1  1                bicyclomycin/multidrug efflux system\\n  \n",
       "2  1                bicyclomycin/multidrug efflux system\\n  \n",
       "3  2                bicyclomycin/multidrug efflux system\\n  \n",
       "4  1                bicyclomycin/multidrug efflux system\\n  \n",
       "\n",
       "[5 rows x 416 columns]"
      ]
     },
     "execution_count": 271,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "contigs_all_args.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "metadata": {},
   "outputs": [],
   "source": [
    "argcount=pd.DataFrame(results_quantification.groupby([\"#ARG\"])[\"#ARG\"].count().sort_values(ascending=False))\n",
    "argcount.columns=[\"counts\"]\n",
    "argcount.reset_index(inplace=True)\n",
    "argcount=pd.merge(argcount,results_quantification[[\"#ARG\",\"predicted_ARG-class\"]])\n",
    "argcount.drop_duplicates(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(style=\"white\")\n",
    "plt.figure(figsize=(12,10))\n",
    "\n",
    "sns.barplot(data=argcount[0:20], y=\"counts\",x=\"#ARG\",hue=\"predicted_ARG-class\",dodge=False,palette=[\"C0\", \"C1\", \"C2\",\"C3\",\"C4\",\"C5\",\"C6\",\"C7\",\"blue\",\"yellow\",\"red\",\"k\",\"green\"])\n",
    "plt.xticks(rotation=90)\n",
    "\n",
    "plt.ylabel(\"Number of ORFs\")\n",
    "plt.xlabel(\"ARG\")\n",
    "plt.savefig(\"Figure_2_.png\",format=\"png\",resolution=\"300dpi\")\n",
    "plt.savefig(\"Figure_2_.svg\",format=\"svg\")\n",
    "plt.savefig(\"Figure_2_.eps\",format=\"eps\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sum of RPKG of all ORFs for each ARG\n",
    "ARGS_RPKG=results_quantification.groupby([\"#ARG\"])[[x for x in all_results.columns if \"ERR\" in x]].sum()\n",
    "ARGS_RPKG=ARGS_RPKG.T\n",
    "ARGS_RPKG.reset_index(inplace=True)\n",
    "ARGS_RPKG[\"run_id\"]=ARGS_RPKG[\"index\"]\n",
    "ARGS_RPKG.drop(\"index\",axis=1,inplace=True)\n",
    "\n",
    "# sum of RPKG of all ARGs for each class\n",
    "class_RPKG=results_quantification.groupby([\"predicted_ARG-class\"])[[x for x in all_results.columns if \"ERR\" in x]].sum()\n",
    "class_RPKG=class_RPKG.T\n",
    "class_RPKG.reset_index(inplace=True)\n",
    "class_RPKG[\"run_id\"]=class_RPKG[\"index\"]\n",
    "class_RPKG.drop(\"index\",axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>#ARG</th>\n",
       "      <th>AAC(2')-IB</th>\n",
       "      <th>AAC(3)-IA</th>\n",
       "      <th>AAC(3)-IC</th>\n",
       "      <th>AAC(3)-IIA</th>\n",
       "      <th>AAC(3)-IIB</th>\n",
       "      <th>AAC(3)-IIC</th>\n",
       "      <th>AAC(3)-IIIA</th>\n",
       "      <th>AAC(3)-IIIB</th>\n",
       "      <th>AAC(3)-IV</th>\n",
       "      <th>AAC(3)-IXA</th>\n",
       "      <th>...</th>\n",
       "      <th>WJ63_05905</th>\n",
       "      <th>WM16_03815</th>\n",
       "      <th>WS68_16125</th>\n",
       "      <th>WS71_01720</th>\n",
       "      <th>WS86_20025</th>\n",
       "      <th>WT08_17490</th>\n",
       "      <th>XM57_27510</th>\n",
       "      <th>YFBG</th>\n",
       "      <th>YH67_08905</th>\n",
       "      <th>run_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.003300</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.033254</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.005777</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.114394</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.013866</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.008997</td>\n",
       "      <td>ERR594345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.013541</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.011959</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.042889</td>\n",
       "      <td>0.004296</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000677</td>\n",
       "      <td>0.002032</td>\n",
       "      <td>0.009349</td>\n",
       "      <td>0.001705</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000913</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ERR599133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.015116</td>\n",
       "      <td>0.028709</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.523718</td>\n",
       "      <td>0.013767</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.013767</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ERR594405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.000669</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.004180</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000456</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.068754</td>\n",
       "      <td>0.000928</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000439</td>\n",
       "      <td>0.000439</td>\n",
       "      <td>0.079927</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.055201</td>\n",
       "      <td>ERR599037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.009093</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.041236</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.035821</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.011940</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002683</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ERR594291</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 314 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "#ARG  AAC(2')-IB  AAC(3)-IA  AAC(3)-IC  AAC(3)-IIA  AAC(3)-IIB  AAC(3)-IIC  \\\n",
       "0     0.000000    0.0        0.003300   0.0         0.0         0.000000     \n",
       "1     0.000000    0.0        0.013541   0.0         0.0         0.011959     \n",
       "2     0.000000    0.0        0.000000   0.0         0.0         0.000000     \n",
       "3     0.000669    0.0        0.004180   0.0         0.0         0.000456     \n",
       "4     0.009093    0.0        0.000000   0.0         0.0         0.000000     \n",
       "\n",
       "#ARG  AAC(3)-IIIA  AAC(3)-IIIB  AAC(3)-IV  AAC(3)-IXA  ...  WJ63_05905  \\\n",
       "0     0.0          0.000000     0.033254   0.000000    ...  0.005777     \n",
       "1     0.0          0.000000     0.042889   0.004296    ...  0.000000     \n",
       "2     0.0          0.015116     0.028709   0.000000    ...  0.000000     \n",
       "3     0.0          0.000000     0.068754   0.000928    ...  0.000439     \n",
       "4     0.0          0.000000     0.041236   0.000000    ...  0.035821     \n",
       "\n",
       "#ARG  WM16_03815  WS68_16125  WS71_01720  WS86_20025  WT08_17490  XM57_27510  \\\n",
       "0     0.000000    0.114394    0.000000    0.000000    0.000000    0.013866     \n",
       "1     0.000677    0.002032    0.009349    0.001705    0.000000    0.000000     \n",
       "2     0.000000    0.523718    0.013767    0.000000    0.013767    0.000000     \n",
       "3     0.000439    0.079927    0.000000    0.000000    0.000000    0.000000     \n",
       "4     0.000000    0.011940    0.000000    0.000000    0.000000    0.000000     \n",
       "\n",
       "#ARG      YFBG  YH67_08905     run_id  \n",
       "0     0.000000  0.008997    ERR594345  \n",
       "1     0.000913  0.000000    ERR599133  \n",
       "2     0.000000  0.000000    ERR594405  \n",
       "3     0.000000  0.055201    ERR599037  \n",
       "4     0.002683  0.000000    ERR594291  \n",
       "\n",
       "[5 rows x 314 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ARGS_RPKG.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>predicted_ARG-class</th>\n",
       "      <th>aminoglycoside</th>\n",
       "      <th>bacitracin</th>\n",
       "      <th>beta_lactam</th>\n",
       "      <th>bleomycin</th>\n",
       "      <th>chloramphenicol</th>\n",
       "      <th>fosfomycin</th>\n",
       "      <th>fosmidomycin</th>\n",
       "      <th>macrolide-lincosamide-streptogramin</th>\n",
       "      <th>multidrug</th>\n",
       "      <th>peptide</th>\n",
       "      <th>polymyxin</th>\n",
       "      <th>qa_compound</th>\n",
       "      <th>quinolone</th>\n",
       "      <th>rifampin</th>\n",
       "      <th>sulfonamide</th>\n",
       "      <th>tetracycline</th>\n",
       "      <th>triclosan</th>\n",
       "      <th>run_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.113089</td>\n",
       "      <td>3.626901</td>\n",
       "      <td>6.774687</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.684291</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.001272</td>\n",
       "      <td>8.063829</td>\n",
       "      <td>37.256312</td>\n",
       "      <td>0.027218</td>\n",
       "      <td>1.206318</td>\n",
       "      <td>3.406160</td>\n",
       "      <td>0.001021</td>\n",
       "      <td>0.003793</td>\n",
       "      <td>0.238331</td>\n",
       "      <td>8.162053</td>\n",
       "      <td>0.492593</td>\n",
       "      <td>ERR594345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.169852</td>\n",
       "      <td>1.956439</td>\n",
       "      <td>4.673196</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.255936</td>\n",
       "      <td>0.033900</td>\n",
       "      <td>0.210209</td>\n",
       "      <td>0.836814</td>\n",
       "      <td>15.729232</td>\n",
       "      <td>0.104756</td>\n",
       "      <td>0.114979</td>\n",
       "      <td>21.075838</td>\n",
       "      <td>0.013198</td>\n",
       "      <td>0.000703</td>\n",
       "      <td>0.238539</td>\n",
       "      <td>27.631654</td>\n",
       "      <td>0.266815</td>\n",
       "      <td>ERR599133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7.724537</td>\n",
       "      <td>3.141729</td>\n",
       "      <td>16.546842</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.880136</td>\n",
       "      <td>9.062190</td>\n",
       "      <td>0.559059</td>\n",
       "      <td>13.464074</td>\n",
       "      <td>223.633203</td>\n",
       "      <td>0.280950</td>\n",
       "      <td>6.081802</td>\n",
       "      <td>13.048547</td>\n",
       "      <td>0.014410</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.772657</td>\n",
       "      <td>21.898546</td>\n",
       "      <td>21.144209</td>\n",
       "      <td>ERR594405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.460290</td>\n",
       "      <td>0.936811</td>\n",
       "      <td>3.807054</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.141872</td>\n",
       "      <td>0.027517</td>\n",
       "      <td>0.001995</td>\n",
       "      <td>0.808983</td>\n",
       "      <td>6.523131</td>\n",
       "      <td>0.026986</td>\n",
       "      <td>1.087789</td>\n",
       "      <td>4.768618</td>\n",
       "      <td>0.001553</td>\n",
       "      <td>0.000870</td>\n",
       "      <td>0.194552</td>\n",
       "      <td>15.902825</td>\n",
       "      <td>0.114893</td>\n",
       "      <td>ERR599037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.203552</td>\n",
       "      <td>1.062067</td>\n",
       "      <td>4.048345</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.472469</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.117158</td>\n",
       "      <td>1.136423</td>\n",
       "      <td>14.998656</td>\n",
       "      <td>0.048914</td>\n",
       "      <td>0.096175</td>\n",
       "      <td>16.720561</td>\n",
       "      <td>0.154478</td>\n",
       "      <td>0.031108</td>\n",
       "      <td>0.209008</td>\n",
       "      <td>31.764691</td>\n",
       "      <td>0.422134</td>\n",
       "      <td>ERR594291</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "predicted_ARG-class  aminoglycoside  bacitracin  beta_lactam  bleomycin  \\\n",
       "0                    0.113089        3.626901    6.774687     0.0         \n",
       "1                    0.169852        1.956439    4.673196     0.0         \n",
       "2                    7.724537        3.141729    16.546842    0.0         \n",
       "3                    0.460290        0.936811    3.807054     0.0         \n",
       "4                    0.203552        1.062067    4.048345     0.0         \n",
       "\n",
       "predicted_ARG-class  chloramphenicol  fosfomycin  fosmidomycin  \\\n",
       "0                    3.684291         0.000000    0.001272       \n",
       "1                    0.255936         0.033900    0.210209       \n",
       "2                    25.880136        9.062190    0.559059       \n",
       "3                    0.141872         0.027517    0.001995       \n",
       "4                    0.472469         0.000000    0.117158       \n",
       "\n",
       "predicted_ARG-class  macrolide-lincosamide-streptogramin   multidrug  \\\n",
       "0                    8.063829                             37.256312    \n",
       "1                    0.836814                             15.729232    \n",
       "2                    13.464074                            223.633203   \n",
       "3                    0.808983                             6.523131     \n",
       "4                    1.136423                             14.998656    \n",
       "\n",
       "predicted_ARG-class   peptide  polymyxin  qa_compound  quinolone  rifampin  \\\n",
       "0                    0.027218  1.206318   3.406160     0.001021   0.003793   \n",
       "1                    0.104756  0.114979   21.075838    0.013198   0.000703   \n",
       "2                    0.280950  6.081802   13.048547    0.014410   0.000000   \n",
       "3                    0.026986  1.087789   4.768618     0.001553   0.000870   \n",
       "4                    0.048914  0.096175   16.720561    0.154478   0.031108   \n",
       "\n",
       "predicted_ARG-class  sulfonamide  tetracycline  triclosan     run_id  \n",
       "0                    0.238331     8.162053      0.492593   ERR594345  \n",
       "1                    0.238539     27.631654     0.266815   ERR599133  \n",
       "2                    0.772657     21.898546     21.144209  ERR594405  \n",
       "3                    0.194552     15.902825     0.114893   ERR599037  \n",
       "4                    0.209008     31.764691     0.422134   ERR594291  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class_RPKG.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## reading metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# metadata from http://ocean-microbiome.embl.de/companion.html - http://ocean-microbiome.embl.de/data/OM.CompanionTables.xlsx\n",
    "# meta -> tab W1 ; meta2 -> tab w8\n",
    "\n",
    "meta=pd.read_csv(\"TSV/meta_runs.tsv\",sep=\"\\t\")\n",
    "meta.rename(columns={\"value\":\"run_id\"},inplace=True)\n",
    "meta[\"Environmental_Feature\"]=meta[\"Environmental Feature\"].str.split(\")\",expand=True)[1].str.split(\"(\",expand=True)[0]\n",
    "meta[\"fraction\"]=meta[\"Size fraction lower threshold [micrometre]\"].astype(str)+\" - \"+ meta[\"Size fraction upper threshold [micrometre]\"].astype(str)\n",
    "meta.rename(columns={\"Biogeographic biomes (Longhurst 2007)\":\"Biogeographic_biomes\"},inplace=True)\n",
    "meta[\"Ocean_sea_regions\"]=meta[\"Ocean and sea regions (IHO General Sea Areas 1953) [MRGID registered at www.marineregions.com]\"].str.split(\")\",expand=True)[1].str.split(\"[\",expand=True)[0]\n",
    "meta[\"Marine_provinces\"]=meta[\"Marine provinces  (Longhurst 2007)\"].str.split(\")\",expand=True)[1].str.split(\"[\",expand=True)[0]\n",
    "meta2=pd.read_csv(\"TSV/meta2.tsv\",sep=\"\\t\")\n",
    "meta2.rename(columns={\"PANGAEA Sample ID\":\"PANGAEA sample identifier\"},inplace=True)\n",
    "meta=pd.merge(meta,meta2,on=\"PANGAEA sample identifier\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "list_for_meta=[\"run_id\",\"Marine_provinces\",\"Environmental_Feature\",\"Ocean_sea_regions\",\n",
    "                                                 \"fraction\",\"Biogeographic_biomes\",'Mean_Lat*',\n",
    "                                                 'Date/Time [yyyy-mm-ddThh:mm]','Latitude [degrees North]', \n",
    "                                                 'Longitude [degrees East]','Sampling depth [m]',\n",
    "                                                   'Mean_Long*', 'Mean_Depth [m]*', 'Mean_Temperature [deg C]*',\n",
    "                                                   'Mean_Salinity [PSU]*', 'Mean_Oxygen [umol/kg]*',\n",
    "                                                   'Mean_Nitrates[umol/L]*', 'NO2 [umol/L]**', 'PO4 [umol/L]**',\n",
    "                                                   'NO2NO3 [umol/L]**', 'SI [umol/L]**',\n",
    "                                                   'AMODIS:PAR8d,Einsteins/m-2/d-1', 'Okubo-Weiss', 'Lyapunov_exp.',\n",
    "                                                   'grad_SST_adv', u'retention', 'Mean Depth MLD Sigma [m]*',\n",
    "                                                   'Mean Depth Max Fluo [m]*', 'Mean Depth Max N2 [m]*',\n",
    "                                                   'Mean Depth Max O2 [m]*', 'Mean Depth Min O2 [m]*',\n",
    "                                                   'Mean Depth Nitracline [m]*', 'miTAG.SILVA.Taxo.Richness',\n",
    "                                                   'miTAG.SILVA.Phylo.Diversity', 'miTAG.SILVA.Chao', 'miTAG.SILVA.ace',\n",
    "                                                   'miTAG.SILVA.Shannon', 'OG.Shannon', 'OG.Richness', 'OG.Evenness',\n",
    "                                                   'FC - heterotrophs [cells/mL]', 'FC - autotrophs [cells/mL]',\n",
    "                                                   'FC - bacteria [cells/mL]', 'FC - picoeukaryotes [cells/mL]',\n",
    "                                                   'minimum generation time [h]']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "class_RPKG=pd.merge(class_RPKG,meta[list_for_meta],on=\"run_id\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "ARGS_RPKG=pd.merge(ARGS_RPKG,meta[list_for_meta],on=\"run_id\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AAC(2')-IB</th>\n",
       "      <th>AAC(3)-IA</th>\n",
       "      <th>AAC(3)-IC</th>\n",
       "      <th>AAC(3)-IIA</th>\n",
       "      <th>AAC(3)-IIB</th>\n",
       "      <th>AAC(3)-IIC</th>\n",
       "      <th>AAC(3)-IIIA</th>\n",
       "      <th>AAC(3)-IIIB</th>\n",
       "      <th>AAC(3)-IV</th>\n",
       "      <th>AAC(3)-IXA</th>\n",
       "      <th>...</th>\n",
       "      <th>miTAG.SILVA.ace</th>\n",
       "      <th>miTAG.SILVA.Shannon</th>\n",
       "      <th>OG.Shannon</th>\n",
       "      <th>OG.Richness</th>\n",
       "      <th>OG.Evenness</th>\n",
       "      <th>FC - heterotrophs [cells/mL]</th>\n",
       "      <th>FC - autotrophs [cells/mL]</th>\n",
       "      <th>FC - bacteria [cells/mL]</th>\n",
       "      <th>FC - picoeukaryotes [cells/mL]</th>\n",
       "      <th>minimum generation time [h]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.003300</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.033254</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>174925.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>174925.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.013541</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.011959</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.042889</td>\n",
       "      <td>0.004296</td>\n",
       "      <td>...</td>\n",
       "      <td>4263.0</td>\n",
       "      <td>6.70</td>\n",
       "      <td>7.17</td>\n",
       "      <td>17195.0</td>\n",
       "      <td>0.735</td>\n",
       "      <td>302791.0</td>\n",
       "      <td>123447.0</td>\n",
       "      <td>426238.0</td>\n",
       "      <td>3263.0</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.015116</td>\n",
       "      <td>0.028709</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>598682.0</td>\n",
       "      <td>48034.0</td>\n",
       "      <td>646716.0</td>\n",
       "      <td>669.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.000669</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.004180</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000456</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.068754</td>\n",
       "      <td>0.000928</td>\n",
       "      <td>...</td>\n",
       "      <td>4988.0</td>\n",
       "      <td>5.49</td>\n",
       "      <td>7.22</td>\n",
       "      <td>17308.0</td>\n",
       "      <td>0.739</td>\n",
       "      <td>972004.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>972004.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>46.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.009093</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.041236</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>521106.0</td>\n",
       "      <td>64001.0</td>\n",
       "      <td>585107.0</td>\n",
       "      <td>6166.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 358 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   AAC(2')-IB  AAC(3)-IA  AAC(3)-IC  AAC(3)-IIA  AAC(3)-IIB  AAC(3)-IIC  \\\n",
       "0  0.000000    0.0        0.003300   0.0         0.0         0.000000     \n",
       "1  0.000000    0.0        0.013541   0.0         0.0         0.011959     \n",
       "2  0.000000    0.0        0.000000   0.0         0.0         0.000000     \n",
       "3  0.000669    0.0        0.004180   0.0         0.0         0.000456     \n",
       "4  0.009093    0.0        0.000000   0.0         0.0         0.000000     \n",
       "\n",
       "   AAC(3)-IIIA  AAC(3)-IIIB  AAC(3)-IV  AAC(3)-IXA  ...  miTAG.SILVA.ace  \\\n",
       "0  0.0          0.000000     0.033254   0.000000    ... NaN                \n",
       "1  0.0          0.000000     0.042889   0.004296    ...  4263.0            \n",
       "2  0.0          0.015116     0.028709   0.000000    ... NaN                \n",
       "3  0.0          0.000000     0.068754   0.000928    ...  4988.0            \n",
       "4  0.0          0.000000     0.041236   0.000000    ... NaN                \n",
       "\n",
       "   miTAG.SILVA.Shannon  OG.Shannon  OG.Richness  OG.Evenness  \\\n",
       "0 NaN                  NaN         NaN          NaN            \n",
       "1  6.70                 7.17        17195.0      0.735         \n",
       "2 NaN                  NaN         NaN          NaN            \n",
       "3  5.49                 7.22        17308.0      0.739         \n",
       "4 NaN                  NaN         NaN          NaN            \n",
       "\n",
       "   FC - heterotrophs [cells/mL]  FC - autotrophs [cells/mL]  \\\n",
       "0  174925.0                      0.0                          \n",
       "1  302791.0                      123447.0                     \n",
       "2  598682.0                      48034.0                      \n",
       "3  972004.0                      0.0                          \n",
       "4  521106.0                      64001.0                      \n",
       "\n",
       "   FC - bacteria [cells/mL]  FC - picoeukaryotes [cells/mL]  \\\n",
       "0  174925.0                  0.0                              \n",
       "1  426238.0                  3263.0                           \n",
       "2  646716.0                  669.0                            \n",
       "3  972004.0                  0.0                              \n",
       "4  585107.0                  6166.0                           \n",
       "\n",
       "   minimum generation time [h]  \n",
       "0 NaN                           \n",
       "1  14.0                         \n",
       "2 NaN                           \n",
       "3  46.0                         \n",
       "4 NaN                           \n",
       "\n",
       "[5 rows x 358 columns]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ARGS_RPKG.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aminoglycoside</th>\n",
       "      <th>bacitracin</th>\n",
       "      <th>beta_lactam</th>\n",
       "      <th>bleomycin</th>\n",
       "      <th>chloramphenicol</th>\n",
       "      <th>fosfomycin</th>\n",
       "      <th>fosmidomycin</th>\n",
       "      <th>macrolide-lincosamide-streptogramin</th>\n",
       "      <th>multidrug</th>\n",
       "      <th>peptide</th>\n",
       "      <th>...</th>\n",
       "      <th>miTAG.SILVA.ace</th>\n",
       "      <th>miTAG.SILVA.Shannon</th>\n",
       "      <th>OG.Shannon</th>\n",
       "      <th>OG.Richness</th>\n",
       "      <th>OG.Evenness</th>\n",
       "      <th>FC - heterotrophs [cells/mL]</th>\n",
       "      <th>FC - autotrophs [cells/mL]</th>\n",
       "      <th>FC - bacteria [cells/mL]</th>\n",
       "      <th>FC - picoeukaryotes [cells/mL]</th>\n",
       "      <th>minimum generation time [h]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.113089</td>\n",
       "      <td>3.626901</td>\n",
       "      <td>6.774687</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.684291</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.001272</td>\n",
       "      <td>8.063829</td>\n",
       "      <td>37.256312</td>\n",
       "      <td>0.027218</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>174925.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>174925.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.169852</td>\n",
       "      <td>1.956439</td>\n",
       "      <td>4.673196</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.255936</td>\n",
       "      <td>0.033900</td>\n",
       "      <td>0.210209</td>\n",
       "      <td>0.836814</td>\n",
       "      <td>15.729232</td>\n",
       "      <td>0.104756</td>\n",
       "      <td>...</td>\n",
       "      <td>4263.0</td>\n",
       "      <td>6.70</td>\n",
       "      <td>7.17</td>\n",
       "      <td>17195.0</td>\n",
       "      <td>0.735</td>\n",
       "      <td>302791.0</td>\n",
       "      <td>123447.0</td>\n",
       "      <td>426238.0</td>\n",
       "      <td>3263.0</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7.724537</td>\n",
       "      <td>3.141729</td>\n",
       "      <td>16.546842</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.880136</td>\n",
       "      <td>9.062190</td>\n",
       "      <td>0.559059</td>\n",
       "      <td>13.464074</td>\n",
       "      <td>223.633203</td>\n",
       "      <td>0.280950</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>598682.0</td>\n",
       "      <td>48034.0</td>\n",
       "      <td>646716.0</td>\n",
       "      <td>669.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.460290</td>\n",
       "      <td>0.936811</td>\n",
       "      <td>3.807054</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.141872</td>\n",
       "      <td>0.027517</td>\n",
       "      <td>0.001995</td>\n",
       "      <td>0.808983</td>\n",
       "      <td>6.523131</td>\n",
       "      <td>0.026986</td>\n",
       "      <td>...</td>\n",
       "      <td>4988.0</td>\n",
       "      <td>5.49</td>\n",
       "      <td>7.22</td>\n",
       "      <td>17308.0</td>\n",
       "      <td>0.739</td>\n",
       "      <td>972004.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>972004.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>46.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.203552</td>\n",
       "      <td>1.062067</td>\n",
       "      <td>4.048345</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.472469</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.117158</td>\n",
       "      <td>1.136423</td>\n",
       "      <td>14.998656</td>\n",
       "      <td>0.048914</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>521106.0</td>\n",
       "      <td>64001.0</td>\n",
       "      <td>585107.0</td>\n",
       "      <td>6166.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 62 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   aminoglycoside  bacitracin  beta_lactam  bleomycin  chloramphenicol  \\\n",
       "0  0.113089        3.626901    6.774687     0.0        3.684291          \n",
       "1  0.169852        1.956439    4.673196     0.0        0.255936          \n",
       "2  7.724537        3.141729    16.546842    0.0        25.880136         \n",
       "3  0.460290        0.936811    3.807054     0.0        0.141872          \n",
       "4  0.203552        1.062067    4.048345     0.0        0.472469          \n",
       "\n",
       "   fosfomycin  fosmidomycin  macrolide-lincosamide-streptogramin   multidrug  \\\n",
       "0  0.000000    0.001272      8.063829                             37.256312    \n",
       "1  0.033900    0.210209      0.836814                             15.729232    \n",
       "2  9.062190    0.559059      13.464074                            223.633203   \n",
       "3  0.027517    0.001995      0.808983                             6.523131     \n",
       "4  0.000000    0.117158      1.136423                             14.998656    \n",
       "\n",
       "    peptide  ...  miTAG.SILVA.ace  miTAG.SILVA.Shannon  OG.Shannon  \\\n",
       "0  0.027218  ... NaN              NaN                  NaN           \n",
       "1  0.104756  ...  4263.0           6.70                 7.17         \n",
       "2  0.280950  ... NaN              NaN                  NaN           \n",
       "3  0.026986  ...  4988.0           5.49                 7.22         \n",
       "4  0.048914  ... NaN              NaN                  NaN           \n",
       "\n",
       "   OG.Richness  OG.Evenness  FC - heterotrophs [cells/mL]  \\\n",
       "0 NaN          NaN           174925.0                       \n",
       "1  17195.0      0.735        302791.0                       \n",
       "2 NaN          NaN           598682.0                       \n",
       "3  17308.0      0.739        972004.0                       \n",
       "4 NaN          NaN           521106.0                       \n",
       "\n",
       "   FC - autotrophs [cells/mL] FC - bacteria [cells/mL]  \\\n",
       "0  0.0                         174925.0                  \n",
       "1  123447.0                    426238.0                  \n",
       "2  48034.0                     646716.0                  \n",
       "3  0.0                         972004.0                  \n",
       "4  64001.0                     585107.0                  \n",
       "\n",
       "  FC - picoeukaryotes [cells/mL] minimum generation time [h]  \n",
       "0  0.0                           NaN                          \n",
       "1  3263.0                         14.0                        \n",
       "2  669.0                         NaN                          \n",
       "3  0.0                            46.0                        \n",
       "4  6166.0                        NaN                          \n",
       "\n",
       "[5 rows x 62 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class_RPKG.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Removing samples from fraction virus and girus\n",
    "ARGS_RPKG=ARGS_RPKG[(ARGS_RPKG[\"fraction\"]!=\"< - 0.22\")&(ARGS_RPKG[\"fraction\"]!=\"0.1 - 0.22\")]\n",
    "class_RPKG=class_RPKG[(class_RPKG[\"fraction\"]!=\"< - 0.22\")&(class_RPKG[\"fraction\"]!=\"0.1 - 0.22\")]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "ARGS_RPKG.to_csv(\"TSV/table_env.tsv\",index=None,sep=\"\\t\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "class_RPKG.to_csv(\"TSV/table_env_class.tsv\",index=None,sep=\"\\t\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "classes=[x for x in list(set(all_results[\"predicted_ARG-class\"])) if x in class_RPKG.columns]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "env=['Marine_provinces', 'Environmental_Feature', 'Ocean_sea_regions', 'fraction', 'Biogeographic_biomes']\n",
    "for e in env:\n",
    "    pdf = matplotlib.backends.backend_pdf.PdfPages(\"boxplots_RPKG_\"+e +\".pdf\")\n",
    "    for c in classes:\n",
    "        fig=plt.figure(figsize=(12,10))\n",
    "        sns.boxplot(class_RPKG[e],class_RPKG[c])\n",
    "        plt.xticks(rotation=90)\n",
    "        plt.ylabel(\"RPKG\")\n",
    "        plt.title(c)\n",
    "        pdf.savefig( fig )\n",
    "        plt.show()\n",
    "    pdf.close()\n",
    "    plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "matplotlib.rc('xtick', labelsize=20) \n",
    "matplotlib.rc('ytick', labelsize=20) \n",
    "\n",
    "for e in env:\n",
    "    #pdf = matplotlib.backends.backend_pdf.PdfPages(out+\"tukey_class_RPKG_\"+e +\".pdf\")\n",
    "\n",
    "    all_c=pd.DataFrame()\n",
    "    for c in classes:\n",
    "        #print c\n",
    "\n",
    "        tukey = pairwise_tukeyhsd(endog=class_RPKG[c],     # Data\n",
    "                                  groups=class_RPKG[e],   # Groups\n",
    "                                  alpha=0.05)          # Significance level\n",
    "\n",
    "        fig=tukey.plot_simultaneous(comparison_name=class_RPKG[e][1],xlabel=\"RPKG\")    # Plot group confidence intervals\n",
    "        plt.title(c,size=22)\n",
    "        plt.xlabel(\"RPKG\",fontsize=18)\n",
    "\n",
    "        plt.tight_layout()\n",
    "        #pdf.savefig( fig )\n",
    "        #plt.show()\n",
    "        df=tukey.summary()\n",
    "        res=pd.DataFrame(df.data[1:],columns=df.data[0])\n",
    "        res[\"class\"]=str(c)\n",
    "        res[\"env\"]=str(e)\n",
    "        all_c=pd.concat([all_c,res])\n",
    "        #all_c.to_excel(out+\"/tukey_class_tests_FPKM_\" +e +\".xlsx\",index=None)\n",
    "        \n",
    "    #pdf.close()\n",
    "    plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>group1</th>\n",
       "      <th>group2</th>\n",
       "      <th>meandiff</th>\n",
       "      <th>p-adj</th>\n",
       "      <th>lower</th>\n",
       "      <th>upper</th>\n",
       "      <th>reject</th>\n",
       "      <th>class</th>\n",
       "      <th>env</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Coastal Biome</td>\n",
       "      <td>Polar Biome</td>\n",
       "      <td>-0.0943</td>\n",
       "      <td>0.1235</td>\n",
       "      <td>-0.2046</td>\n",
       "      <td>0.0160</td>\n",
       "      <td>False</td>\n",
       "      <td>sulfonamide</td>\n",
       "      <td>Biogeographic_biomes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Coastal Biome</td>\n",
       "      <td>Trades Biome</td>\n",
       "      <td>0.0169</td>\n",
       "      <td>0.7117</td>\n",
       "      <td>-0.0259</td>\n",
       "      <td>0.0597</td>\n",
       "      <td>False</td>\n",
       "      <td>sulfonamide</td>\n",
       "      <td>Biogeographic_biomes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Coastal Biome</td>\n",
       "      <td>Westerlies Biome</td>\n",
       "      <td>-0.0127</td>\n",
       "      <td>0.9000</td>\n",
       "      <td>-0.0680</td>\n",
       "      <td>0.0425</td>\n",
       "      <td>False</td>\n",
       "      <td>sulfonamide</td>\n",
       "      <td>Biogeographic_biomes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Polar Biome</td>\n",
       "      <td>Trades Biome</td>\n",
       "      <td>0.1112</td>\n",
       "      <td>0.0372</td>\n",
       "      <td>0.0046</td>\n",
       "      <td>0.2179</td>\n",
       "      <td>True</td>\n",
       "      <td>sulfonamide</td>\n",
       "      <td>Biogeographic_biomes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Polar Biome</td>\n",
       "      <td>Westerlies Biome</td>\n",
       "      <td>0.0816</td>\n",
       "      <td>0.2398</td>\n",
       "      <td>-0.0307</td>\n",
       "      <td>0.1938</td>\n",
       "      <td>False</td>\n",
       "      <td>sulfonamide</td>\n",
       "      <td>Biogeographic_biomes</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           group1            group2  meandiff   p-adj   lower   upper  reject  \\\n",
       "0  Coastal Biome   Polar Biome      -0.0943    0.1235 -0.2046  0.0160  False    \n",
       "1  Coastal Biome   Trades Biome      0.0169    0.7117 -0.0259  0.0597  False    \n",
       "2  Coastal Biome   Westerlies Biome -0.0127    0.9000 -0.0680  0.0425  False    \n",
       "3  Polar Biome     Trades Biome      0.1112    0.0372  0.0046  0.2179  True     \n",
       "4  Polar Biome     Westerlies Biome  0.0816    0.2398 -0.0307  0.1938  False    \n",
       "\n",
       "         class                   env  \n",
       "0  sulfonamide  Biogeographic_biomes  \n",
       "1  sulfonamide  Biogeographic_biomes  \n",
       "2  sulfonamide  Biogeographic_biomes  \n",
       "3  sulfonamide  Biogeographic_biomes  \n",
       "4  sulfonamide  Biogeographic_biomes  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_c.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax =plt.subplots(2,2,figsize=(25, 25),sharey=True)\n",
    "\n",
    "sns.set(style=\"white\")\n",
    "\n",
    "\n",
    "tukey = pairwise_tukeyhsd(endog=class_RPKG[\"quinolone\"],     # Data\n",
    "                                  groups=class_RPKG[\"Biogeographic_biomes\"],   # Groups\n",
    "                                  alpha=0.05)          # Significance level\n",
    "\n",
    "fig=tukey.plot_simultaneous(comparison_name=class_RPKG[\"Biogeographic_biomes\"][9],ax=ax[0,0])    # Plot group confidence intervals\n",
    "ax[0,0].set_title(\"Quinolone\")\n",
    "#ax[0,0].text(0.01,0.1,\"A\",ha='center', va='bottom')\n",
    "\n",
    "\n",
    "ax[0,0].legend('A',numpoints=None, loc=2,handlelength=0, handletextpad=0) \n",
    "\n",
    "tukey = pairwise_tukeyhsd(endog=class_RPKG[\"bacitracin\"],     # Data\n",
    "                                  groups=class_RPKG[\"Biogeographic_biomes\"],   # Groups\n",
    "                                  alpha=0.05)          # Significance level\n",
    "\n",
    "fig=tukey.plot_simultaneous(comparison_name=class_RPKG[\"Biogeographic_biomes\"][9],ax=ax[0,1],xlabel=None)    # Plot group confidence intervals\n",
    "ax[0,1].set_title(\"Bacitracin\")\n",
    "ax[0,1].legend('B',numpoints=None, loc=2,handlelength=0, handletextpad=0) \n",
    "\n",
    "\n",
    "tukey = pairwise_tukeyhsd(endog=class_RPKG[\"fosmidomycin\"],     # Data\n",
    "                                  groups=class_RPKG[\"Biogeographic_biomes\"],   # Groups\n",
    "                                  alpha=0.05)          # Significance level\n",
    "\n",
    "fig=tukey.plot_simultaneous(comparison_name=class_RPKG[\"Biogeographic_biomes\"][9],ax=ax[1,0],xlabel=None)    # Plot group confidence intervals\n",
    "ax[1,0].set_title(\"Fosmidomycin\")\n",
    "\n",
    "ax[1,0].legend('C',numpoints=None, loc=2,handlelength=0, handletextpad=0) \n",
    "\n",
    "\n",
    "tukey = pairwise_tukeyhsd(endog=class_RPKG[\"polymyxin\"],     # Data\n",
    "                                  groups=class_RPKG[\"Biogeographic_biomes\"],   # Groups\n",
    "                                  alpha=0.05)          # Significance level\n",
    "\n",
    "fig=tukey.plot_simultaneous(comparison_name=class_RPKG[\"Biogeographic_biomes\"][9],ax=ax[1,1],xlabel=None)    # Plot group confidence intervals\n",
    "ax[1,1].set_title(\"Polymyxin\")\n",
    "\n",
    "ax[1,1].legend('D',numpoints=None, loc=2,handlelength=0, handletextpad=0) \n",
    "\n",
    "#plt.xlabel(\"RPKG\",fontsize=18)\n",
    "\n",
    "fig.text(0.57, 0.0001, 'RPKG', ha='center')\n",
    "\n",
    "\n",
    "plt.tight_layout()\n",
    "\n",
    "\n",
    "\n",
    "plt.savefig(\"Figure_3.png\",format=\"png\",resolution=\"300dpi\")\n",
    "plt.savefig(\"Figure_3.svg\",format=\"svg\")\n",
    "plt.savefig(\"Figure_3.eps\",format=\"eps\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tukey = pairwise_tukeyhsd(endog=class_RPKG[\"bleomycin\"],     # Data\n",
    "                          groups=class_RPKG[\"Marine_provinces\"],   # Groups\n",
    "                          alpha=0.05)          # Significance level\n",
    "\n",
    "fig=tukey.plot_simultaneous(comparison_name=class_RPKG[\"Marine_provinces\"][1],xlabel=\"RPKG\")    # Plot group confidence intervals\n",
    "plt.title(\"Bleomycin\")\n",
    "plt.xlabel(\"RPKG\")\n",
    "plt.savefig(\"Figure_4.png\",format=\"png\",resolution=\"300dpi\")\n",
    "plt.savefig(\"Figure_4.svg\",format=\"svg\")\n",
    "plt.savefig(\"Figure_4.eps\",format=\"eps\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Runing OLS generic model with all covariates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "def eta_squared(aov):\n",
    "    aov['eta_sq'] = 'NaN'\n",
    "    aov['eta_sq'] = aov[:-1]['sum_sq']/sum(aov['sum_sq'])\n",
    "    return aov\n",
    " \n",
    "def omega_squared(aov):\n",
    "    mse = aov['sum_sq'][-1]/aov['df'][-1]\n",
    "    aov['omega_sq'] = 'NaN'\n",
    "    aov['omega_sq'] = (aov[:-1]['sum_sq']-(aov[:-1]['df']*mse))/(sum(aov['sum_sq'])+mse)\n",
    "    return aov"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "formula1=\"Marine_provinces + Environmental_Feature + Ocean_sea_regions + fraction + Biogeographic_biomes + Latitude + Longitude + NO2 +  PO4 + NO2NO3 + SI + miTAG_SILVA_Taxo_Richness + miTAG_SILVA_Phylo_Diversity + miTAG_SILVA_Chao + miTAG_SILVA_ace + miTAG_SILVA_Shannon + OG_Shannon + OG_Richness + OG_Evenness +         FC_heterotrophs_cells_mL + FC_autotrophs_cells_mL + FC_bacteria_cells_mL + FC_picoeukaryotes_cells_mL\" "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/ipykernel_launcher.py:42: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n",
      "/home/rcuadrat/.local/lib/python3.6/site-packages/statsmodels/base/model.py:1752: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 3, but rank is 2\n",
      "  'rank is %d' % (J, J_), ValueWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "out=\"TSV/\"\n",
    "ols_df=class_RPKG\n",
    "\n",
    "ols_df.rename(columns={\"Latitude [degrees North]\":\"Latitude\",\"Longitude [degrees East]\":\"Longitude\",\n",
    "                       \"Date/Time [yyyy-mm-ddThh:mm]\":\"Data_Time\",\"NO2NO3 [umol/L]**\":\"NO2NO3\",\n",
    "                       \"NO2 [umol/L]**\":\"NO2\",\"PO4 [umol/L]**\":\"PO4\",\n",
    "                       \"SI [umol/L]**\":\"SI\",\"FC - heterotrophs [cells/mL]\":\"FC_heterotrophs_cells.mL\",\n",
    "                      \"FC - autotrophs [cells/mL]\":\"FC_autotrophs_cells.mL\",\"FC - bacteria [cells/mL]\":\"FC_bacteria_cells.mL\",\n",
    "                      \"FC - picoeukaryotes [cells/mL]\":\"FC_picoeukaryotes_cells.mL\"},inplace=True)\n",
    "#ols_df.drop(\"run_id\",axis=1,inplace=True)\n",
    "\n",
    "#'Date/Time [yyyy-mm-ddThh:mm]'\n",
    "ols_df.columns=ols_df.columns.str.replace(r\".\", \"_\").str.replace(r\"-\", \"_\").str.replace(r\" \", \"_\").str.replace(r\":\", \"_\").str.replace(r\"*\", \"_\").str.replace(r\"[\",\"_\").str.replace(r\"]\",\"_\")\n",
    "par=list(ols_df.columns)\n",
    "parameters=[x for x in par if x not in classes]\n",
    "parameters.remove(\"macrolide_lincosamide_streptogramin\")\n",
    "#parameters.remove(\"multidrug_mutation\")\n",
    "\n",
    "# formula1=\"+\".join(parameters)\n",
    "# OLS with all the metadata \n",
    "reg3 = pd.DataFrame()\n",
    "reg3_ = pd.DataFrame()\n",
    "aov=pd.DataFrame()\n",
    "# pdf = matplotlib.backends.backend_pdf.PdfPages(out+\"models_aov.pdf\")\n",
    "# writer = pd.ExcelWriter(out+'aov_results.xlsx', engine='xlsxwriter')\n",
    "# writer2 = pd.ExcelWriter(out+'aov_results_significant.xlsx', engine='xlsxwriter')\n",
    "for c in classes:\n",
    "    formula=\"%s ~ \"%(c.replace(r\"-\", \"_\")) + formula1\n",
    "    res = ols(formula, ols_df).fit()\n",
    "    dfres={\"coeficient\":res.params,\"p values\":res.pvalues}\n",
    "    ddd=pd.DataFrame(dfres)\n",
    "    ddd[\"coeficient\"]=ddd[\"coeficient\"].apply(lambda x: \"%.4f\" % (x))\n",
    "    ddd[\"variable\"]=c.replace(r\"-\", \"_\")\n",
    "    ddd[\"Formula\"] = formula\n",
    "    ddd[\"model p value\"]=res.f_pvalue\n",
    "    ddd[\"model Adjusted r squared\"]=res.rsquared_adj\n",
    "    ddd[\"r squared\"]=res.rsquared\n",
    "    #final result of OLS\n",
    "    reg3=pd.concat([reg3,ddd],axis=0)\n",
    "    \n",
    "    reg3_ = reg3[(reg3[\"p values\"] < 0.05) | (reg3[\"model p value\"]< 0.05)] \n",
    "    reg3_[\"coeficient\"]=reg3_[\"coeficient\"].astype(float)\n",
    "    reg3_ = reg3_.sort_values(by=[\"coeficient\"],ascending=False)\n",
    "    aov_table = anova_lm(res,typ=2)\n",
    "    aov_table[\"classe\"]=c\n",
    "    \n",
    "    eta_squared(aov_table)\n",
    "    omega_squared(aov_table)\n",
    "    #print(aov_table)\n",
    "    aov_sig=aov_table[aov_table[\"PR(>F)\"]<0.05]\n",
    "    \n",
    "    #aov_table.to_excel(writer, sheet_name=c[0:30])\n",
    "    #aov_sig.to_excel(writer2, sheet_name=c[0:30])\n",
    "\n",
    "    aov=pd.concat([aov,aov_table],axis=0)\n",
    "\n",
    "    res_ = res.resid \n",
    "    fig = sm.qqplot(res_, line='s')\n",
    "    plt.title(c)\n",
    "    #pdf.savefig( fig )\n",
    "    \n",
    "plt.close()\n",
    "# pdf.close()\n",
    "# writer.save()\n",
    "# writer2.save()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "ARGS_RPKG_fornet=ARGS_RPKG[ARGS_RPKG.columns[0:-45]]\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/rcuadrat/.local/lib/python3.6/site-packages/ipykernel_launcher.py:12: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  if sys.path[0] == '':\n"
     ]
    }
   ],
   "source": [
    "corr = ARGS_RPKG_fornet.corr(method='spearman')\n",
    "\n",
    " \n",
    "# Transform it in a links data frame (3 columns only):\n",
    "links = corr.stack().reset_index()\n",
    "links.columns = ['var1', 'var2','value']\n",
    "\n",
    " \n",
    "# Keep only correlation over a threshold and remove self correlation (cor(A,A)=1)\n",
    "links_filtered=links.loc[ ((links['value'] > 0.8 ) | (links['value'] < -0.8 )) & (links['var1'] != links['var2']) ]\n",
    "\n",
    "links_filtered['color'] = np.where(links_filtered['value']>0, 'red', 'blue')\n",
    "# Build your graph\n",
    "G=nx.from_pandas_edgelist(links_filtered, 'var1', 'var2')\n",
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "node_att=all_results[[\"#ARG\",\"predicted_ARG-class\"]].drop_duplicates()\n",
    "node_att.reset_index(drop=True,inplace=True)\n",
    "node_att=node_att[node_att[\"#ARG\"].isin(G.nodes())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "# df = pd.DataFrame(index=G.nodes(), columns=G.nodes())\n",
    "# for row, data in nx.shortest_path_length(G):\n",
    "#     for col, dist in data.items():\n",
    "#         df.loc[row,col] = dist\n",
    "\n",
    "# df = df.fillna(df.max().max())\n",
    "\n",
    "#layout = nx.kamada_kawai_layout(G, dist=df.to_dict())\n",
    "pos = nx.spring_layout(G,k=0.25,iterations=20)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "carac= node_att.set_index('#ARG')\n",
    "carac=carac.reindex(G.nodes())\n",
    " \n",
    "# And I need to transform my categorical column in a numerical value: group1->1, group2->2...\n",
    "carac['predicted_ARG-class']=pd.Categorical(carac['predicted_ARG-class'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "No handles with labels found to put in legend.\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAt4AAALeCAYAAACdlBZfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzs3XecFdX9//HXzG3bFxZYlo6ySweRLiBirMGu2DX6tXyTmPxS7NGowdgRW76xRVFjiRFbNKJiwYZSpShIld7L9nLbzO+PkYUFdtly79wt7+fjsQ9n587MORdh9z3nfuYcw7ZtGxERERERiSsz0R0QEREREWkJFLxFRERERFyg4C0iIiIi4gIFbxERERERFyh4i4iIiIi4QMFbRERERMQFCt4iIiIiIi5Q8BYRERERcYGCt4iIiIiICxS8RURERERcoOAtIiIiIuICb6I7ICIiIiLS2FiWxcaNGyktLT3o66mpqXTu3BnTrP04tmHbth2rDoqIiIiINAfbt28nGAzSqVOnA8K1ZVls2rSJQCBAdnZ2ra+pUhMRERERkf0UFBTQvn37g45om6ZJ+/btKSwsrNM1FbxFRERERPYTjUbx+XzVvu7z+YhEInW6poK3iIiIiMhBGIZRr9eqo+AtIiIiIuICBW8RERERERcoeIuIiIiIHERNk//VZ2JABW8RERERkf0kJSWxa9eugwZs27bZtWsXSUlJdbqm5vEWEREREdlPOBxm48aNVFRUHPT1pKQkOnfuXOPMJ/tT8BYRERERcYFKTUREREREXKDgLSIiIiLiAgVvEREREREXKHiLiIiIiLhAwVtERERExAUK3iIiIiIiLlDwFhERERFxgYK3iIiIiIgLFLxFRERERFzgTXQHRKT5sYuDWJuLsTYVYW0sgmAEbMBnYmanYnbJxOiQjtE2BcMwEt1dERERV2jJeBGJCduysJbvIjJzPfa2EvCaEIo6gXt/fg/YNgS8eEZ2xntkB4xkn+t9bijbsrBW52OtL8RaV4C9oxTClvPeTANS/Zgd0zG7tcLs0RqzbWqiuywiIgmk4C0iDRZdvpPwO8sgajthuy58Jtg4AfyY7hiexl8BZxcHiczbTHTuJrBsCFdzg7GH13lPRrsUvKO7YvZq2yTep4iIxJaCt4jUm10eJvzucqzVu52R3obwmRhpfnzn9sPMSY9NB2PMtm0iszYS/WyNE7Qj9XjPfg9Gqq9Rv08REYkPBW8RqReroJzQlAVQHnZGumPFa+I7uw+e3u1id80YsPLLCU9dgr2rrOE3GQBe0xnlH3cYhqk6dxGRlkDBW0TqzC6oIPjMfCd0x+MnSCML39a2EkIvLNz7kGis+EzM7q3wnddfpSciIi2AftKLSJ3YoSjB5xfEL3QDRCzCb/6AtakoTg3UnrWzlNDzC6AixqEbIGxhrSkg/Nr32JbGQEREmjsF74TZBfwALAKWABuJX4oRiZ3I9FVQFsfQXdmQRej1JdiROj6sGUN2KEron4sgGMc+RCystQVEZqyJXxsiItIoaB5v1xQB7wGfASuACsAP7KntjOAkmcOAkcBZQCfXeylSE2tdAdHF2+r3UGF9lIWJfLoG34m57rS3n8j0Vc5Id7yFLaKzN+Lp2w6zgx64FBFprlTjHXdrgWeBT3A+YKioxTnen47tD1wFDI9X50TqJPj4HOydZe426jUJ/HYERkbA1WatdQWEXl7s3k0GYLROwn/NcNV7i4g0U/rpHjcRnMB9MfAhEKJ2oXvPuSHgW+Ba4E9AcRz6KFJ71qYi7MLa/h2OIdsmMn+T682GP17taugGsEtDWMt2utqmiIi4R8E7LrYCFwHPAUGgIb+8K4DPgTNwgrhIYkRmbYh5EN1VXsQJ/76eikio+oOiNtG5m7GjNbc9ceJEXnjhhZj0y9pZhr2tNCbXqpOQRWTmevfbFRERV6jUJOY2AFfg1HTH+oGsJOA+YEyMr9v47flrahia7ziW8vLyKrfLy8sJBAKYpnM/fv/997NmzRoee+wx/H5/5YqUHtPDtAn3ctJrN1aeWxYJkuwNVD6xMGX8jQzr0BuAR+e9wWPz3+SNMycyqH3VWu17vnmJrKQMfnXk6ZX7Zm78nvtn/4sfC7aQGUjhlqMu4ZQ+o/FfOIClxRu5/vrrWblyJXl5eTz44IP0798fgG3btnHKKafw9ddfO/1tgPB7K4gu2Nywe+b97Cov4oJ37uTdc+4hyVtD/7wm/quGYGZXv7z8xIkT6d69O5dddlnsOigiInGn4B1T23FKSwqI35QPScAjwNA4XT/x7KiFtWo31roCrHUFTk3xngVLDCDFh9khHaNbKzw9sjBz0hLa3+ZixIgRTJo0ibFjx1bumzx5MmvXruXRv9xP6Ol51S4c0+Opi/nkgsl0z8ypst+2bY791x8pDpVzao+RTDz6fypfC0bDjH7pt7x7zj10SGsDwMr8jVz0zt1MOvaXjOk8gOJQGUXBMrpl5WCN7cK46y7gqquu4rLLLuOll17iqaee4quvvqoM2hdccAGXXHIJp5566gF9rNNNRtgC2671TcZryz7j3VVf4zO9GIZB98wcbj3qEkZ07FN5Xq1vMnoehfe4w1mWVujKTYaIiLhHs5rEjAVcjzPSHc97mYqf2nkbaBXHdtxnF1UQmbOJ6PzNzh9h6CCfGNhAaRhr1W5Yk0/0i7UYrZLwju6K2Tcbw6vqqXiwtxRDPT5tmLtlGdvLCrj3mKu5c+Y/uXXUpfg9zo+dRdtWke5PrQzdAH//9m0u7PszxnUdBEDrpHRaJ6VD1ObrGV8QjUa5+uqrMQyDK6+8kieffJKZM2dy7LHHAnDUUUfxySefHDR4r1y5snK7upuM008/nccmP0zwgZmwz7za3105pXK7x1MX898J91S5yXht2WdcfcSpXDf8PGzb5rVln3HN9EeY84sn8JgmwWiYN1d8ybvn3LO3P/kb+cMnfz/gJoOoTcXqHVzx6K+q3GRcccUVlTcZ7du3Jzc3l+nTpx/0vYqISOOklBIzrwFriH15ycEEgb+60I47bMsmPHM9wf+bQ3T2RmfO5IOF7v1FbQhb2DvKCE9bQfD/ZmNtTPyCK82RtbOsdv9P9vPmii85rttgxh8+AoBP1+19TmH57g0c3qpDleMXblsFwM+n3sTIF3/DtZ88TkFFCQArli+nT/fuGJs3w9atsHMnfXJzWb54MZSWQjBI3uGHs3Tp0vq+TQDsLSXgq/+PRsMwOD13FAXBEnaWFwKHvsnwmh5aJ6XTLbM9AN/MmV15kxEIBLjyyiuxbZuZM2dWnr/nJkNERJoOjXjHxDbg/6j9rCUNFQZmA18CR7vUZnxYu8udVfvyyxv24F7IglCQ0D8X4hnSAe/xPTQlW4y8++67fPT+h86NDtC3bTdeOe3PhzyvPBzk/R9nM+nYX+PzeDn58OG8teJLTj7cmR6zKFRGqi+pyjlbS3fz9oqveOGUm8lObc0NM55k4swXePi431C6cwfpc2dA586Vx2e0akXpe+/BH/4AQFogQFFmJng8e7+83gO3DQMuvRR8vr2vhUJgWdgbgjDwbPBW7VttRS2Lt1Z+SZf0drRNzgSqv8nomtGen0+9ifyKEkZ17Mfto39Bq6Q0Vm5eS58+fao809CnTx+WL19eObqfl5fHtGnT6tVHERFJDAXvmHgVd0a691UBPElTDt7W1mJCLyx0RlJjVZ0TsYjO34K1vRT/hQMwvJ4YXbjlOu2003jo1N87n0bUwfS18/CYnsqykTNyR/GL9+5lV3kRbZIzyAykUhquerMa8PqZ0Gssh/0UUn995Bn84r9OeUaqz0/JfuUuxaZJajhc+X2JYZBhWbDna5/XqsjOhp07nbC9R3o676al8fG897EXfQoYtb7JAHhm8TReXPIRoWgYG7jvmKvx/FRDXuebjFAF6W2rLqSTkZFBaenemVbS0tIoKtInPCIiTYmCd4OFgTd/+q/b1gKrgMSs6tcQ1o5SJ3THYynuiIW9oZDQv77Df9FAjXzHQlLdb2DeXPEFZeEKjn75dwDY2IStKO+s+pr/GXAyvbO6MmXx+1XO6Z3Vtcoo774xu2fHw3m2bVvs667DiEYhGuWHV17h8pEjIScHolFWfv89fYuKYMIEiEYhEnH+u//2ggXQqxekpOx9bccOTguFeHjA8UT6ngG+ui3Yc9XA8ZU13ivyN3L5e/eRGUhlXNdB9bjJSKKkZHeV44uLi0lN3TvTSUlJCRkZGXXqo4iIJJaCd4N9TnwfpqxJBKe2/JYEtV8/dihC6MVF8Qnde0Rs7A1FRD5eje+kvEMfLzUys9OI+j21rvPeWrqbrzctYcrPb6R3m66V+5/77gPeWvEl/zPgZAZm96A4VMbW0t3kpGYBMKHXWP7v27c5I28M7ZIzeWrhu/ys25EAjPzZ8XgWvsaz2dlceumlvPLKK9CqFaOfeQZ+mtlj1oUXcuFvfgOnn35gp/Y1YgTcdRfs83AlkyfD2rUYv70D3v6h3n8/DcOgV1YXhuT05LP1CxnXdVCdbzLysrvw7OyPsW278pgffviByy+/vPKYlStX0rdv33r1UUREEkNDgQ02GzhwCe1//7uM447bQY8eWxg0aBt/+lMhRUXWAcd06rSF//yn/IDzi4stbr+9kGHDtpGXt5VRo7Zz++2F7N697zWiwJzYvh0XhD9cDRWR+De0p+xED1w2yDvvvEPv00Yx4InLGPDsFQx49orKhwar8/aKr+jTphtHdxlIu5RWlV+X9T+J5bs3sHz3BvweL2f3HMt/Vu59YPDc3uM4K28M57x1O2Nf+T1+j5fbR18GXoOkHu2YMmUKr7/+On379uXVV19lypQpldPpbdu2jRUrVnDyySc36P2a7dMq69nra3X+ZuZtXUFea6cefd+bjD0m9BrL68u/YH3RdsrDwao3GQOH4vF4ePbZZwkGgzz33HMAjB49uvL8WbNmVdZ7i4hI06B5vBvsPODHKnuefLKEJ54o5ZFHMhkzJsDWrVFuuaWI/HyLt95qg8/njGBNmLCLH34IM3iwnxdfzKo8PxSyOfPMXWRkGNx5Zwa5uV5277Z48cUyBg70cdxx+9aKeoEvgKYxl6+1roDQy4vdXYo7I0DgtyM01WAD2LZN8P6v6jWzSU1qvaiM34P/kiMwO1dfWjFx4kS6detWZVS4OtVNJ7h3Hu+9zx3MuOjhyock4eBzlt8w48nKebxtoFVSGmfmjuba4ediGs7fu3u/eYWs5HR+Oei0yvMemfs6Ly/9GICxXQZy++jLyAyk4jmqM8s7VFTO452bm8vkyZOrzOM9fvx4vvnmG83jLSLShCh4N4gFjGbf+u7iYovBg7czeXImp5+eXLm/tNRi5Mgd3HZbOuedl8LGjRFGjtzBU0+14te/LmDevGyys5062ldeKeP++4v5+ut2pKYeKiymAk8ATeMj5+Cz87E3FbvbqN/EN74nnoE5hz5WqhV6dxnWwq2JqaxK9RG4dpRrK5cGn1+Avb7mUf26qstNhu+sPnh6ta32kLrcZIiISOOhIcAGKWf/FDJvXohg0Gb8+KozGKSmmvzsZwG++CIIwNSp5RxxhI9TTkkmL8/Lm2/uLTf58ssg48YFahG6wakMzW/g+3CHtbMMe1vpoQ+MtZBFZOYG99ttZrwju0AiHlT1mnhGdnYtdAN4R3QGf2xnxGmTnMFH5z9Yc+gGMMDMzarxkDvuuEOhW0SkCdLDlQ0SZv97l927LbKyTLzeA0NC+/Ym333njI6//no5l1/uzFBw5pnJvP56Ob/6lbP0eX6+xcCBvtp1obgY7r4Ovm8LrVtDVlbNX61aOXMWJ0B0zkZnircEsAvKsbYWY+akH/pgOSizXSpGu1RnFUuXeY/s6Gp7Zs82YLoX9Ct5DDxDOmomHhGRZkrBu0F8OOUme2VlmezebRGJ2AeE723bnFA+d26IDRuinHGGMyp+1llJ3H9/Md9/H6Z/fx+tW5ts21bLgBqNwqJV8EEdHiDMzKw+mNcU3gN1m15tX3l5eRC2wLYpj4Twe7x4fqp9vWvslawt3MrjC/6D39z7V9Jjepg24V5Oeu3Gyn1lkSDJ3kDlDBBTxt/Ile9PYsHl/6icM/lPn/+D91bPqrrvk6fxLstk0gt/r/d7EPCd1pPQlAXu1ej7TDxHd8NIqeWNaIwYHhPPiM5EZ65393kEw8A7rJN77YmIiKsUvBskmf1HvIcM8eP3G0ybVnFAjfeMGUFuvDGdqVPLsG048cSdVc6dOrWc/v19HH10gAceKKaszCIl5RAjX6YBW+s4h3hhofO1Zk3dzktJOXQ4P1iAT0tjxZJlBB/4CiybsS//nnuPuZrRnftXXvrReW9wyuEjeei4aw5o9rsrp1Ru93jqYv474Z7KB9vC0QiWbbNk5xoGZvcAYN7W5eSkZlXZN3fzMn5/5KV1e79yADMnHc/Izs5iOuE4B1IDjFZJeEd1PfSxceAd3ZXowq1Q6NKKtHtuMjLrt2KmiIg0fgreDWICXXEWsXFkZJhce20at91WRHq6UWVWk6wsk7PPTuLuu4t44IFMjjtu7wjytGkVPPxwCbfdls455yTz0ktlXH11ARMnZnD44R4KCmxeeqmMfv28VWc1SQ/Ap2thdzHs3n3wr/z8g++vblW/6pSVOV8b67aCIV4vdt+j4LQ/gT/50MfXgc/jZVB2LnO2LGNgdg92lhcSjkY4M29MlX1rCrcwNPWwmLbdUnmP6Y61fBf2rjKw4vikpdfEd24/jESUfACG18R/bl9Czy90ZdTbaJ2csJsMERFxh4J3gx3JvsEb4Jpr0mjd2uSvfy1m7dp8gkE46ig/r76axUcfBUlKMpgwIblyWkGA889PYdKkYmbMCHLCCUm8+moWDz5YzAUX7KKw0KZdO5MTT0ziyCP3ezDL6Aits52vHj1q323bhtLSmoN5deG9tI4PSEYi2LY/bvXdwzv0Yu6WZVx1xCnM3bKMITm9GNqhFy9+P71yX5f0dnTwaJW/WDA8Jv7LBhF6Zj52UTA+4dtn4r9oIGbb1EMfG0dmxww8Y7sR/XJdfEf4A56E3mSIiIg7FLwbbATwHvsvonPhhSlceGEK4CyUM2lSMdEonHFGMmecceCob3KywZIle6e7y8gwufPOTO68M/OAY/fyAiPr123DgLQ056trHUfZgsGDh/UaArydnglGzWUz036cxafrF1R+37dtN1457c+H7M7wjn144fvp2LbN3C3LGdahF0e2z+PaTx+v3De8Qx+wqbISoNSfkeLDf9VgQs8vxC6oiN2IsGmA96fQ3bWmv/vu8Y3pBhURonM3xSd8+z34fzEIs01K7K8tIiKNioJ3g43hUH+M55+fgsfjTDXYqVMsSy08wPkxvF4tBQKQk+N81ZKxeCtMWwGh6oPL+GpqvA/lyOxcysIVrNi9gblblnFx3+NJ9SXRIbVN5b7L+p9U5+tKzYwUP/7/HULk0zVE521uePj2mRg5afjP7tvo6px9x/eAJC/RL9bF7ibDa4Dfi/8XR2Bmp8XmmiIi0qgpeDeYF2f1yn8CoWqPmjAhHqNZPYFucbhuHCT5nFH2OAh4/QzIPpxP1i1ge1kBPVo7U88N69CLT9YtYNnu9Qzr0NsJdhrtjinD68F3Yi6eftmE316GXfTT6Hddqk/8HjDAe3wPPIM7NNr/R74x3fAc1prQ60ugLNyw0W+fidk3G9/JuRgB/RgWEWkp9BM/Js4FXna5zSTgVy63WX9mThpE4/cg3vCc3jz//QcMad+zct/QnF7c/tVzZKe0oltme4x2ia0Xbs7MThn4rxmGvamYyDcbsFbsdBbbse0DA6rHKSchbGG0S8U7pitm77ZNYu5qs1MGgd+MIDJzHdFZm5z3F4rW7mQD8JgYbVOcm4zDW8e1ryIi0vgoeMdEG+AGYBLOapbx5gfG4dSXNxHpfidwRao/5L0fZ/HR2nlV9s246GHaJh+61nd4xz48sfAdhnboVblvaE4vdpUXcWqPo5zVABtJzXBzZRgGRucM/Of2w45Y2NtLsbYUY20ugvKI8xCm34PRPg2zYzpmhzSMJHfn544Fw2viO+YwvGO6YS3fSWT2RuytJT+9aDhh3MYJ2qbh3Hgk+zB7ZuEd0VllJSIiLZhh23Yc5wNrSWycEehF1JguY6IV8DbQtH6BB19ciL2mIDGN+z34zuyNp3e7xLQvzZpt29gFFdhbS7ArIhC1wGtiZCY12RsMERGJPY14x4wB3AdcAuwEavnxc50lA4/Q1EI3gHd4Z8Kbimv/0XwsGWDmtXG/XWkRDMPAaJ0MrWM7T72IiDQvjb+osklpDUwB2hKfe5pk4FGg/6EObJTMvDZOuYnbPAaeIR2bRA2xiIiINF9KIjHXHngJ6IMTlGMhgBPm/wEMjtE13WeYBp5RXcHn8l87w8A7rJO7bYqIiIjsR8E7LvaMfP8OZ/aR+o5+Gzih+zScmu7eMeldInlHdsZID7jXoM/Ec0z3RjcvtIiIiLQ8ergy7rbijID/56fvazPrSQDnYc1RwBVA3/h0LUGsbSWEnv02dguRVMcAo10q/v8dqqW4RUREJOEUvF1TAXwMfAksAXbgTAtoAhYQxnlgshfOMvCnAFkJ6akbIrM2EJmxJj5LcO8R8OC/eghmlpbiFhERkcRT8E6YcmAXzmqXPiADaFnzTIc/W0P0mw3xCd8BD/7LBmHmpMf+2iIiIiL1oOAtCRWZvZHIJz/GruzEa0LAi/8XR2BqpUoRERFpRBS8JeGsbSWEpy7BLg42bPTba2IOaI/vpFwMvyd2HRQRERGJAQVvaRTsqEVk1kan9CRi1X6RHRMwTYz2qfiO74HZrVVc+ykiIiJSXwre0qjYlo21cheRORuxNxdD1HYW3bFtZ6IXw3BmWQxbkObHk5eFZ0QXzLZ6gFJEREQaNwVvadTs4iDWlhKoCGNHbQyvidEqCaN9mspJREREpElR8BYRERERcYFWrhQRERERcYGCt4iIiIiICxS8RURERERcoOAtIiIiIuICBW8RERERERcoeIuIiIiIuEDBW0RERETEBQreIiIiIiIuUPAWEREREXGBgreIiIiIiAsUvEVEREREXKDgLSIiIiLiAgVvEREREREXKHiLiIiIiLhAwVtERERExAUK3iIiIiIiLlDwFhERERFxgYK3iIiIiIgLFLxFRERERFyg4C0iIiIi4gIFbxERERERFyh4i4iIiIi4QMFbRERERMQFCt4iIiIiIi5Q8BYRERERcYGCt4iIiIiIC7yJ7oCIiIg0QLAENs6CzfNg3eeQvwaiQTC9kNwGuo6GTiOg80ho1TXRvRVp0Qzbtu1Ed0JERETqaNv38M1D8P2r4PFBpAKioYMcaIA/HawQtD8CxtwMPU8Fj8beRNym4B0jtm2zOb+cZVuK2JJfTihi4TEN0pN99MxJJzcnnSSfJ9HdFBGRpq5kG7x9Oaz93AnadrRu5/vTwZ8G57wCh42LRw9FpBoK3g1g2zaL1hfw0sw1zPtxFwCmaRAKW0QtG8MAn9fEaxpUhC06tErmvBFdGT+oI2lJvgT3XkREmpzv/gXv/tIZ3bbCDbuWNxkGXAQnPwKBtNj0T0RqpOBdT58u3crfPlxOQVmYilCU2v4hJvk82LbNSQM78P9O7EV6sgK4iIgcgm3D9Btg3hMQLovddb1J0Ko7/M+XkNo2dtcVkYNS8K6j/NIQd739PfPX7KIibNX7On6vSZLPw1/OHsConu1i2EMREWlWbBs++CN8+4/Yhu49TB9kdoGr50JKVuyvLyKVFLzrYMnGAn7/z/kEI1HC0dj8sSX5TMYf0YnrT+mDaRoxuaaIiDQjcx6Hj26IT+jew+OH9gPhqtlgaqZhkXjRv65aWrQ+n98+P4+SYCRmoRugImwxbdEm7nhjMZaleyAREdlH/pr4h25wHtLc8QPMeiS+7Yi0cAretfDj9hL++OJ8ysN1fHK8lirCFl8u38HD7y+Ly/VFRKQJsiyYej5Egu60Fy6FT2+DXSvdaU+kBVLwPoRI1OKmVxdQHopP6N6jIhzlnQUbmbN6V1zbERGRJmLVB7Dzh7pPF9gQ0Qr46Cb32hNpYRS8D2HK56vZUVRR61lLGiIYtrj99UWUBiMutCYiIo3azAcgVOJum7YFq96H0h3utivSQih412BLQTkvz1zboNlL6qosFOWpT/Qxn4hIi5a/BjbNTlDjBsx/OkFtizRvWi+2BlNnr8NyedKXUMTinW83cc3xPUnya6VLEZEWafm7AIz4VzY7yz2Yxt7fRV+et52cVIvikMGRL7dndMcQL5y0u8rpQ15pT0GFiWnYpPpsftYlyF2jCknxVf2ddv/cdB5bmM77Z+5gYLufFuSJlMPil2HsrfF9jyItkIJ3NUIRi7fnbYzpDCa1ZRjw8ZKtnHpkJ9fbFhGRRmDdF87qlGTw3Em7GNspdMAh//0xmYAHPtsQYGe5Sdvkqp/OvnjyLkZ1DLG11OSi99vw+OI0rh9SXPm6bcMbq5JpFbB4fWXy3uANkL8aohHwKCaIxJL+RVVj2PDh5BeVkHPeA5i+AAAly76gbNU3ZJ96ExueuQLD6wcMTH8yyYcPp9Xw8zBMk+DWFRTMmUo4fzOGaeLN7ECroy4k0O6wg7ZVOO9NytctIFywhYxBp8KQM3lt1joFbxGRlmrTnEMeMnVlMv/Tr5QP1ibx1qpkrh5QetDjclItxnYKsmRX1V/5Mzf72V1hct+YQibOyuC2kUX49hSgegKwYynkDGzoOxGRfajGuxrhiIVtWZQs+ajaY9qfNZHOlz9Bu/E3ULZ6FqXLP8cKlbNj+qOk9TuOTpf+jQ4XTiZz8OkYZvX3ON6M9mQOP5ekLnt/wK3ZUUJU83qLiLRMxZtrfHldkYc5W/2c1aOMs3qUM3VlSrXHbiox+WxjgO4ZVWdHmboyhZO6VXDa4eVELINP1yftfdEwYNeKBr0FETmQgnc1QhGL9IEnU/zdh1jBmhcu8LXqQCCnJ+H8TUQKtwKQ2mMkhmliev0kde6Pv02Xas9P7Tma5C4DMX17f+h5PSbrdx189EJERJoxK+rMLvKTK6dn0eeFHPq8kMMV01sD8Pp9W1f5AAAgAElEQVTKFAa0DdOjVZQzc8tZusvLD7urDvBc9mEWPZ/PYfi/cuiQGuWPg/eWmZSGDaatSeLMHuUEPDC+ezmvr0zee7JtxX/RHpEWSMG7GmHLxt+2O4GcXhR/90HNx+ZvIrh1Bb42XfFm5mAYJrs+f4byDYuxgvULzwawamvxIY8TEZHmaO8nns+euJsfLtvKD5dtZcqJ+dg2vL4ymbNyywHolBZleE6IqSuqjnq/cNJuVly+lX+P38mKAh8FFXt/5b+3JomAx2ZcF2dxnrNyy/l4fRL5FcbeC5h6wF8k1lTjXZ2fZjPJGHIm29+9l7R+xx9wyLa3J4JhYgZSSes1ltSeYzAMk+xT/0TR4vfJ//IFouWFJHUZSNaYy/CkZNa6+ahtU1Kh+bxFRFoc0wOmD6zwQV+evdXP+mIvjy5I5/FFaQCUhA3WFHq5dXgRnv2G1MZ0CnF2jzL+OieDfxyfD8DUFSmUhE2GvdIecGJ+yDL4z+pkLu9XBoYJ/rS4vUWRlkrBuxp7xhr8WZ1J7noERYum4WvVscox7c+8A19m+wPO9bXuSJtjrgQgXLCFXZ89TcGsf9HmZ79iy+t/JlrirE7Z7uQ/EsjpefD2bVTjLSLSUmX1gJ3LDvrS1BXJHNu5goeOKajcVx4xOO6Ndny+KcDPuhy4xPzVA0o56tVslu32kuaz+WaLn3+N30Wv1nsHeJ5cnMbrK1Oc4G1FIHtA7N+XSAun4F2NfT5sI2PwGWx7eyLpA06q83V8rTqQmjeG0mWfAdBhwl21Os80IOBTJZCISIvUZdRBg3d5xOC/a5J54rh8slOqTh94dm45U1ekHDR4Z6dYnJ1bziML0undOsygdmGO3m+Kwiv7l/DM9+1ZVeAltx3Q+uAzcYlI/Sl4V8Nj7o3evsz2pBw+nJIlH+Nr3bnG88IFWyhfv4iUHsPxpmYRKdlN2erZ+LN7VHuObUXAsgEbbAs7Esb0BeicVf1T6iIi0ox1PRqWvMbsC7dX2Z3stVl++daDnvLA0YWV2/Mv2nbA65PG7n39D4MPXIq+U5rF+qu2ON9kD3FmNhGRmFLwrobPW3W0OePI0yld9fUhzzN8SYR2/EjJ99OxQmWY/hSSuh5Bq+HnVXvO7i9foGzlzMrvixb+lzbHXEnPDsfW/w2IiEjT1fsMeO/XiWnbnwbDEtS2SDNn2LbLa6I3EW/N3cCjHy6jImwd+uA4aJse4L/Xj0tI2yIi0gi8cQl8/yrY0UMfG0v+NLhhB+wzxa2IxIaKiKsxMq8tiXq20WsaHNM7OzGNi4hI4zDqOvAG3G3TE4DBVyl0i8SJgnc1OrRKZkDn2k//F0umaXD+Ud0S0raIiDQSHY6EvueC18UQ7E+DcX9xrz2RFkbBuwaXjDmMZL/7Cwj06pBB1zaprrcrIiKNzPi/gT/dnbZ8KXDOy5CUmEEnkZZAwbsGI3q0pWubVEwXH+wOeE1+f3Iv9xoUEZHGK5AO500Fb/Khj20IXwoMuAhy6z5trojUnoJ3DUzT4O7zjsDvdeePKeA1OXNoF/p3buVKeyIi0gR0PwbOeiF+4duXAocfD6c+GZ/ri0glBe9D6JyVwq+PyyPJF9+SE9OA1ml+rjk+L67tiIhIE9TvXJjwqhOSjRj+6valQu+z4Lw3nKXqRSSuFLxr4byR3ThpYAeS4rSSpGFAepKPxy8fTiDOAV9ERJqo3qfDL78FMwdChz68RqbPqR0/7Wk4+0XwaFkPETdoHu9asiybB/67lA8WbaYiEru5vT2mQUayj6evHE4XPVApIiI1yc+HvB7QMx/GmZCWAqEDV6Gs1p4ZUnJ/Dqc+AWnt49NPETkoBe86sG2bN393D39L60fY4yPawBGCJJ+H/p0zmXjOQNqkuzxXq4iIND3XXQcPPeRs/+dNOCwIX90PO5aALxmiIYhU7D3e8IA/1dnvT4Ohv4ahv4KMjonpv0gLp+BdF++/D+PHszmzPbdffjer23ajIhylrn+AST4PpgHXn9KHnx/REcNwcdoUERFpmlatgr59IRyG446Djz5yahUBQqWwdRFsngf5q53vPQFIbQcdh0LHIZDeIbH9FxEF71orLIT+/WHjRkhOxl68mEW+Nrw8cw2zV+/CaxqUh6NU96fp95qVZSWXjOrO+EGdSE1STZ2IiNTS2WfDW285YXvhQhg4MNE9EpE6UvKrreuvd0I3wH33YeTmMggY1K01u0qCfLt2N0s2FLJwXT5bC8sJR208pkFqwEPvDhkM6p5Fv86Z9OuUqRFuERGpm88/d0I3wJVXKnSLNFEa8a6N6dPhpJ8WFTj6aPjsMzA1IYyIiLjAsmDYMPj2W0hNdUpOcnIS3SsRqQelx0MpKoKrrnK2k5Ph2WcVukVExD0vvuiEboA//UmhW6QJU4I8lBtugA0bnO2774Y8LXAjIiIuKS2FW25xtrt0gWuvTWx/RKRBFLxr8vHH8PTTzvaoUfC73yW2PyIi0rI8+CBs3uxs33ef88mriDRZqvGuTnExDBgA69ZBUpLzBHmvXonulYiItBSbNkHPnlBWBsOHwzffqNRRpInTv+Dq3HSTE7oB7rpLoVtERNx1661O6AZ4+GGFbpFmQCPeB/Ppp87iBAAjR8JXX4HHk9g+iYhIyzF/Pgwd6myfdx78+9+J7Y+IxISC9/5KSpwSk7VrIRBwSkx69050r0REpKWwbTj2WGfubr8fli2Dww5LdK9EJAb0udX+br7ZCd0Ad96p0C0iIu56+20ndAP88Y8K3SLNiEa89/X55zBunLM9fDjMnAleLe4pIiIuCYWgb19YvRratYOVKyEzM9G9EpEY0Yj3HqWlcMUVzrbfD889p9AtIiLu+vvfndANzqeuCt0izYqC9x633AI//uhsT5zojDiIiIi4ZdcuJ2wD9Ou3d9VkEWk2VGoC8MUXcMwxzvbQoc5cqRrtFhERN/3ud/C3vznbH3wAJ52U2P6ISMwpeJeVwcCBzkd7fr8zhVP//onulYiItCTLljm/e6JROPlkeP/9RPdIROJApSa33rq3nu722xW6RUTEfTfc4IRuj8dZJl5EmqWWPeI9cyYcfbQzZ+rgwTBrFvh8ie6ViIi0JB9/DCec4Gz/+tfw+OOJ7Y+IxE3LDd7l5XDEEc5UTT4fzJvnlJyIiIi4JRp1Bn4WL4aMDOd3UnZ2onslInHScktNbrvN+QG3Z1uhW0RE3Pb8807oBqf0UaFbpFlrmSPeX38NY8Y4JSaDBsGcOSoxERERdxUXQ14ebNvmrE65dCkkJSW6VyISRy1vxLu83Fkox7adKQOff16hW0RE3Hf//U7o3rOt0C3S7LW84H3HHbB8ubN9661OnbeIiIib1q+HyZOd7dGjYcKExPZHRFzRskpNZs+GUaPAspya7rlznbm7RURE3HTxxfDKK8727NkwfHhi+yMirmg5wbuiwnly/IcfnHlS586FI49MdK9ERKSlmT0bRo50ti++GF56KbH9ERHXtJxSk4kTndANcMstCt0iIuI+24Zrr3W2k5LgnnsS2x8RcZU30R2oLzsShbDlfOPzYHhruIeYOxceeMDZHjAA/vzn+HdQRERkf1OnOjNrAVx3HXTtmtj+iIirmkypibWrjOgPO7DWFWBvKYHyMJjGTy/akOLD7JiO0a0Vnr7tMFslO68Fg06JydKlTonJ7NkwZEji3oiIiLRMFRXQpw+sXQs5Oc5aEmlpie6ViLioUY9427aNtWIXkZnrsbeWOAHb2uc+IbrPdmkYa+Vu+DGf6Iw1GJ0z8I7uivn8wxhLlzrH3HyzQreIiCTGY485oRvgrrsUukVaoEY74m3llxN+Yyn2jtK9JSV15QFj1UL8b9yF0TUH5s+HQCC2HRURETmU7dudxXKKipxpbOfPdz6FFZEWpVGOeEfmbyYyfRVELGjIbUEU7K79CP6/f+Id1gqvQreIiCTCHXc4oRuc+bsVukVapEY14m3bNpFPfiQ6d1P9R7mr4zXxHNMd32g9yCIiIi5assRZO8Ky4LTT4J13Et0jEUmQRjWdYOTztfEJ3QARi+gXawnP2hD7a4uIiFTn+uud0O31wqRJie6NiCRQowne0RU7iX6zIT6he4+wRfTTNVjrCuLXhoiIyB4ffOB8AVxzDfTqldj+iEhCNYpSE7s8TPBvs6Ei4k6DaX4Cvx2B4VeNnYiIxEkk4jxIuXQptGoFq1ZBmzaJ7pWIJFCjGPEOv7cCwlH3GqyIEPlolXvtiYhIy/PMM07oBrj9doVuEUn8iLddUEHw8dkQcbkbHoPAH4/CSPG7266IiDR/hYXO9IE7dkBurvOApV+/b0RauoSPeEfmbWrYlIH1ZRhEFmxJQMMiItLs3XOPE7rBeaBSoVtESHDwtqMW0fmbq65A6ZaIRXTWRhpBibuIiDQna9bAI48428ccA2eckdj+iEijkZAFdPLy8pwNG8rLy/F7vHgM5x7grrFXsrZwK48v+A9+c2/3PKaHaRPu5aTXbqzcVxYJkuwNYPz0/ZTxN/Lass94d9XX+EwvPo+X/m0P447Rl9GjdcfK815f/jk3ffY0j574O84uHIzRKjnu71lERFqIm2+GUAgMAx56yPmviAgJrvGOfLuZURf9nHvHXs3ozv0r9z867w3WFW7joeOuqfH8Hk9dzCcXTKZ7Zk7lvhtmPElOahbXDT+P8nCQW754hvVF23jjrDsrj7no3btYvmsDg3Ly+OczU/D0zY79mxMRkZZn5kwYM8bZvvxyeO65hHZHRBqXhJaaWBsK41rfnewLcHruKFbs3li5b1PxDuZsXsbdY6/ky/WL2LpkTfw6ICIiLYdlwbXXOtspKXDXXYntj4g0Oomt8c6viOv1S8MV/GfVTPq27V65780VXzGg3WGcfPhwerTuyNufvh/XPoiISAvx6qswZ46zfeON0KlTYvsjIo1OQmq8K0WqX6Vy2o+z+HT9gsrv+7btxiun/blWl31m8TReXPIRAY+PI7J78MC4X1a+9taKL7m03wkAnJ47ijfnfsxv+Ev9+i8iIgJQVubUdoMTuK+/PrH9EZFGKbHB21P9AyfjDx95yBrv6lw1cDzXDT/vgP3zti5nY/EOTs09CnCC9+S5U/n+++/p37//AceLiIjUysMPw4YNzvY990BqamL7IyKNUkKDt5Hq7rymby7/EhubU1+/pcr+qVOnKniLiEj9bN0K997rbA8eDJdcktj+iEijldDgbXbNBJdmWQpGQkz7cTZ3j72KY7sOcnZ6TaYnreGxqVO47bbb8HoT+wGAiIg0QX/+M5SWOtsPPwxmwtemE5FGKrHBu0N6ta+99+MsPlo7r8q+GRc9TNvkzHq1NX3tfJK8fs7KG4PP89Pb9nu44MJRPPSvp5kxYwYnnHBCva4tIiIt1KJFMGWKs3322TB2bGL7IyKNWkLn8bZDEYKTZiZm5UoAn0ngpjEYGp0QEZG6sm04/nj49FPw+WDpUsjNTXSvRKQRS2jiNPxezF5tXSs3qcIEzxE5Ct0iIlI/773nhG6A//f/FLpF5JASnjq9R3UBbwK6YZp4RnR2v10REWn6wuG9Uwa2aQO33ZbY/ohIk5Dw4G12ysDISnF31Ns0MDpnYLZJcbFRERFpNp58EpYvd7b/8hdo1Sqh3RGRpiGhNd57WDtLCT09v8YFdWLKZxK4ZjhGZpI77YmISPORn++UlezeDb17w+LFTo23iMghJHzEG8Bsm4pnbDfwudAdn4n3hFyFbhERqZ+//tUJ3QAPPqjQLSK11ihGvAFsyyb08iLsDUXxG/n2mZg9svCd2w/DSMQTnSIi0qStXAn9+jk13scfD9Ong36fiEgtNYoRbwDDNPBfMACjQ1p8Hrb0mphdM/Gd01ehW0RE6uemm5zQbRgwebJCt4jUSaMJ3gCGz4P/0kGYuVmxLTvxmpj92uG7cACGp1G9ZRERaSo++wzeesvZvvJKGDgwod0Rkaan0ZSa7C/6ww7C7yyDsAVWPbvoMcDvwXdWHzy5bWLbQRERaTksC4YOhQULIC3NKTnJyUl0r0SkiUnokvE18fRph9ktk8jna4ku3OpMNxiqZe233wMGeIZ0xDumG0ZSo32bIiLSFLz4ohO6Af70J4VuEamXRjvivS87HCW6ZDvRBVuwt5c6S8x79quri1jg82C0T8M7uANmn3YYiViYR0REmpfSUujZEzZvhq5dYdkySE5OdK9EpAlqEkPBhs+Dd1AHvIM6YNs2FAexdpTtnf3E58Fsl4KRHkhsR0VEpPmZNMkJ3QD33afQLSL11iRGvEVERBJi0ybIy4PychgxAr75RjOZiEi9qRZDRESkOrfe6oRugIceUugWkQbRiLeIiMjBzJ/vzGQCcP758Oqrie2PiDR5Ct4iIiL7s20YNw6++AICAeeByu7dE90rEWniVGoiIiKyv7ffdkI3wB/+oNAtIjGhEW8REZF9BYPQrx+sXg3t2sGqVZCRkeheiUgzoBFvERGRff39707oBvjrXxW6RSRmNOItIiKyx86dkJsLhYXOqPfCheBtEkteiEgToBFvERGRPSZOdEI3wOTJCt0iElMa8RYREQFn5pL+/SEahZ//HKZNS3SPRKSZ0Yi3iIgIwA03OKHb44EHH0x0b0SkGVLwFhER+fhj+O9/ne3//V/o2zex/RGRZkmlJiIi0rJFozB4MCxe7MxgsmqVM42giEiMacRbRERatueec0I3wJ//rNAtInGjEW8REWm5ioshLw+2bYPDDoOlSyEpKdG9EpFmSiPeIiLSct13nxO6Ae6/X6FbROJKI94iItIyrVsHvXo5S8SPHg1ffgmGkeheiUgzphFvERFpmW65xQndAA89pNAtInGn4C0iIi3P7NnwyivO9sUXw/Dhie2PiLQIKjUREZGWxbad0pJvvnFqulesgC5dEt0rEWkBNOItIiIty9SpTugGuP56hW4RcY1GvEVEpOWoqIA+fWDtWsjJgZUrIS0t0b0SkRZCI94iItJyPPqoE7oB7r5boVtEXKURbxERaRm2b4fcXGfRnCOOgPnzweNJdK9EpAXRiLeIiLQMd9zhhG6AyZMVukXEdRrxFhGR5u/7751RbsuC006Dd95JdI9EpAVS8BYRkebv5JPhww/B63VCeK9eie6RiLRAKjUREZHm7YMPnNANcM01Ct0ikjAa8RYRkeYrEnFKTJYuhdatnekD27RJdK9EpIXSiLeIiDRf//iHE7oBbr9doVtEEkoj3iIi0rTYNoRKIVIOhgf8aeD1H3hcYaEzfeDOnc5/lywB/0GOExFxiTfRHRAREamRbcOmObD8HVj7OWz/DsJlTujGBisK6R2h03A47Fjodx6ktoN77nFCN8CkSQrdIpJwGvEWEZHGKRKExS/DV/dB8WaIVIAdrfkcbzLYFnQaB3/9BNZGYNw4+PRTMAw3ei0iUi0FbxERaXw2zoGp50HZTgiX1v18GwgDPwA3fw4jxsa4gyIidafgLSIijYcVhY9vhjl/d2q4G3w9D6S1hnOnwmHjGn49EZEGUPAWEZHGIRqGf58Naz51arhjyZsMZ78Ifc+J7XVFROpA0wmKiEjiWVF47bz4hG5wRs/fvBRWvBf7a4uI1JKCt4iIJN5X98GP0+MTuveIlMPU8yF/bfzaEBGpgYK3iIgk1val8MXd8Q3de0Qq4PULwLLi35aIyH4UvEVEJHEsy5m9JFLhTnt2FLZ/D/Ofcqc9EZF9KHiLiEjirPkECtfhzP/nknApzLjdqSsXEXGRgreIiCTOzAcgVOJ+u5EgrHzf/XZFpEVT8BYRkcQo2gTrv0pM26FiJ/SLiLjIm+gOiIhIC7X2czB9/HuFwdOL01hb5CHdb3Ny9wr+NKyIzIDN5PnpPLYgDb/HxmtCXqsIt48sZGj7MP9ekcy1n7fmqv4lTDyqqPKyH65N4oqPsjg3r4xHxhWwodjDyFfbk+Kt+kDlg+MWcsblUTA9br9zEWmhFLxFRCQxNn7Dk/NsnlicwSPHFDCmU5CtpR5umZnJhdPa8PbpOwE4vUc5fzu2gLAF98/N4OqPsvj24m0AdM+I8N8fk7ltRBHenz7DnboymcMzIwc098NlWyuPAcCfBjuXQ3bfeL9TERFApSYiIpIgxSs+Z/L8NP56VCHHdgniM6FLepQnj8tnQ4mHN1clVzneZ8K5eWVsL/eQH3R+fbVLjtI7K8xnGwMA5FcYzNvm58RutZwlZcv8mL4nEZGaKHiLiEhCzFu5nWDUYPxhVUNyqs/mZ12CfLEpUGV/MAqvrUyhY2qErKS9ZSMT8sp5fWUKAP/5MZkTu1XgN2sxS0q4HIo2NvyNiIjUkkpNREQkIXaXRslKomr5x0/ap0RZvNNPj8wo7/6YzMfrk/CZNr1aR3j2hPwqx57cvZy/zMqgKGTw+soU7hhRxKcbAgdcc8CLOVW+f+f0neRFgjF9TyIiNVHwFhGRhMhKMdldYRKxDgzf28o8ZCU582yfdrhT412dZC8c1yXIowvSya8wGZYTOmjw/u7S/Wq8McHjj8E7EZF4sSMW9vYS7MIgRCwwDUjyYuakYaQ2vX+/Ct4iIpIQQw7LxO8pZtqaJE7vsbfcpDRsMGNDgJuHFbOltHYzjkzIK+O899pw7eDi2nfAlwQpberabRGJM6ugnOicTUSX74TCoHNnbuCss2X8dFDEAr8Hs3MGnmGdMXu0xjCMGq7aOCh4i4hIQmT0GMm1g9/itm8ySffbVWY16ZAa5Zy8Mv5vYXqtrnVUhxD/Gr+L/m3Cte+A6YWcI+vZexGJNWtjEeFPVmNvKgbbhuhPz2qEqllltjyCtXI31rpC8HnwjO6Cd3gnDLPxPsKo4C0iIonRdQzXDH2D1oFi/jo7g3VFXtL8Fid1q+Bvx+YTqMP02oYBR3cK1XhMnxeq1nhfP6yUX14/sD49F5EYssNRIp/8SPTbLc5Idl2FohCKEp2xBuvbLfjO7YfZLjX2HY0Bw7btWjz6LSIiEmPbl8A/hkO4LDHtt+0Nv/0hMW2LCADW7jJC/1wEZeH6he6D8Zp4Tzgc77DOsbleDGnEW6SZsC2bwi3F7Fy9i/wNRYTKQmAYBFJ9ZHVtRdvcNqRnpzaJGjhpIbL7QavDYMcS99v2p8FR17nfrohUsnaUEnpuAQQjTv12rEQsIh//iF0RwXd09xheuOEUvEWauO0rd7Lo7aWsm7MJwzAwTAiX77NqnwG+ZB921ALDoMeYbgw8ow9ZXVslrtMie4y5Cd67BkIl7rZrWzDgQnfbFJFKdkEFoecXQMWBq8zGRNgi+tV6jIAX7/DGM/KtUhORJmrzkm18+ffZlOwsJRq2sK3a/VM2TAPTa9K6SybH/HYkbQ/PinNPRWoQroCHOkP5Lvfa9CbD0F/CyQ+716aIVLJtm9Az87G3lsR2pPtgvCb+qwZjZqfFuaHaUfAWaWLCFRG+mTKPFTPWEK3uSe9a8vg9DDi9N0MvGIjHV4cn2URiaeX78NoE92q90zrA71aBP8Wd9kSkisjsDUQ+XQPhGNV0H4LRJhn/r4ZheBI/20nieyAitVa8vYR//+admIRugGgoyvfvLuONP06jvKji0CeIxEPez6HX6eA5cNGbmPMmw7mvKXSLJIhdHHQ1dAPYRUEisze61l5NFLxFmojibSW8ed37lO0uj0no3iMSjFK4uYg3r3uf8kKFb0mQU5+CzK5g+uLXhi/FqSnvNiZ+bYhIjSLzNse/vGR/YYvo1xtqXZIZTwreIk1AsCTI2zd9SLAkFJcfHFbUpmxXGe/cMp1IDEO9SK0lZcAVX0JG5/gs4+5LgaG/gmNuj/21RaRW7KhFdO6mBk0b+MWGxfzyw4fqfmLEwlpV9VmSU045heXLl9e7L/WhGm+RJmD6/V+wbu5GrDh/NOcJeOh7Uh6jrhwa13ZEqlW2C176OexYCuHSGFzQAG+SE7iPvjkG1xORmuTl5VVul5eXEwgEMH9aSfLeP9zG6o/n8fi8t/GbeyfW85gepk24l5Neu7FyX1kkSLI3ULlC/JTxNzKsQ2/OeOPP/GXMZRzZ3mnnonfvYuXujYSiYTqnZ/OHYedwQve9v8PeWTmTSXP+TX5FCWN6HsnDrz1D69atndfeeYd3332Xf/zjH/H64ziAgrdII7d2zkY+efBLIkF3RqI9fg+n3XUC7Xu1daU9kQNYFsz5G3xyC0SCYNfz774vBdI7wrlTocOg2PZRRA5pxIgRTJo0ibFjxwIQ/ng1Dz32COsKt/HQcdfUeG6Ppy7mkwsm0z1z74qzi7ev5g+f/J1PL9w74r1s13pyW3fCa3pYuG0Vv3jvXj4+/0GyU1uzYvdGJrx9B8+cfD392h3GrV89i9G7LU888QQAFRUVDBkyhBkzZpCdnR2HP4EDqdREpBGzohafPfaNa6EbnAcuZzwyE92TS8KYJoz8Pfz6O+h/gTNi7a3Dw5D+dEjLgWPvhGuWKHSLNBLWusIGnf/5hkWM6Ninyr7ebbriNZ1ZuQzDIGxF2VK6G3BGu3/W7UiGd+xDqi+JPw4/j/fff5+SEmfdgKSkJAb8f/buOzzKKnvg+Hdqek8gQOiETkAUEZFiVwQsYPnZxd4VC/ZV17XiCmJFEFFXV0FAFhUFBYMIKL2GDoEAgfSemXnn/f1xmUwCIXVKBs7nefJwZ+aduXfWTXLm5txzevVi8eLFjVpXfUgDHSGasD0r9uN0+D7nuji7lMPbsmjeJcHncwtRIbYDjPoSSifBmmmw8b/uLpdGC+qE1tE/RNtLICQOkvpDv3uhw4UqgBdCNBl6VuPSx7bm7CMloeNx99/x01sszdiETbMzKCmFXgntAdieu5++iZ0rrmsbm4jFZGbXrl2kpKQAKjVm8+bNjVpXfUjgLUQTtnbWpqpdKH1EszlYN3sLFz0lgbdoAkJi4Oyx6kvXIXc35OwARxkYTRAUCc16QYh0YxWiSTt6eArfiKUAACAASURBVP/HXcv5LX1Nxd3d49vy1Yjnan16QXkJ4dbg4+6fcukT2DUHSzM2sjPvAEaD+tBdbC8nwhrivtCpExEWXrHjDRAWFsbhw4cb+o7qTQJvIZqo4pwSnp/1BAW2AsamPEuoJaziscmbJnKo9AAP9RpHdJDqPLk4YwGpBxcypuv9JIW3qfJahbYCFh34mR35W7Fp5URYo+gRk8LZiUOxmqy8vHIcFqMFMGA2mukQmcxwfRTn2wdKYx3RtBgMaic8toO/VyKEqAdd1yvKCA7rcFatOd7ViQoKo8hWfdlbi8nM0DZ9+GzDz7SNbM4F7U4nzBJEoa20ynVFJcWEh7u7WBYXFxMZGVnvtTSU/B1OiCbqyHZV9ijGGsPGnLUV92eWHMTutFW5Vtd11mevJsQUyvrs1VUeK3WU8GnaBzicDsZ0vY+n+v6TGzvfQZlWRm65u7TS3d0f4em+/+ShXuMoc5SSemABOXvyvPgOhRBCnCoMBgMYDbVfWIMuca3ZnX+wxms0XSO9QO1gJ8ckkZadXvFYekEmNruNDh3cH9y3b99O9+7dG7Wu+pDAW4gm6vD2bHQdesX1rRJMr8teRUrc6VWuTS/aTZG9gEvajGRTzjo0pzs9ZVnmEoJMQVzZ/tqK3fEoazSXtBlJ89AWx80bZAqmc3R3DpdkcmRn9nGPCyGEEA0S0rhEi6Gt+/DXwbSK2ztzD7A4fS1lDht2zcGcbX/w98E0zmzRFYCRyQP5be9q/j6YRom9jAkrZnLJkAsrdrzLysrYsGFDRdUVX5BUEyGaqMNbswBICm/DhuzVHCnNJC44gU0567it630sOvBzxbXrslfRObob3WNSmJ8+l235W+gW0wuA3QXb6RrdA4Ohbp+zSx0lbM3bRKuwNhzekY3v9gGEEEKczIwtIwD4YddyFuxZWeWxRde/Q3xIVI3P75nQnghrKGszd9CneSd0dN5dNYuHFk7CaDDSLiqRiRc8SM+jhys7xybxz0FjePS3D8grK2JgUk/eeev9itdbsGABAwYMIDEx8URTepwE3kI0UbYSdzqJa9e7bUQH4oObEWl156PZNRubczZwRftrMRlNdIvpybrs1RWBd6mjhHBL7flrkze/i8FgwKaVExscz8h2V1NeZKv1eUIIIUR1VqxYUeW2oU00D/cfzcNnjKr1uTvv/k+19z/W72o+XDuXjy8eS6eYVsy68uUaX2dk8kBGJg9UN8KtBDd396j4+OOPGT9+fK1r8SQJvIVooiq3hk+J68v0rR+RV55D7/iqaSZpeZswGowkR3UBoFfcaXyxbQrF9iLCLOGEmEMpshfUOt9d3R8iNjgezamx8sgyPkv7iNfPetuzb0oIIcQpy9QxBi11D2gN7xMxqHUKg1qn1P+JBjAmx1a5a968eQ1eR0NJjrcQTZTJ6v5cHB0UQ7Q1lu35W+ka3bPKdeuyV2Fz2piw/jXeXvtPZu78D05dqziQ2T4ymbS8Teh63drNm4wmTos/kzxbDoeKaj7EIoQQQtSVMTECQ/Tx5QB9wmTE3D/JP3NXIjveQjRRce2q1iQe0W40ZVopVpMV59EW2gW2AnYX7OD65DFVDkouz/yD9dmr6d/8HAY0H8SG7DXM2f0t57a6mOigGAps+SzLTKVP3BnHHbB06k7WZq/EbLTQpVey99+oEEKIU4Z5YBvsP24DW902gzzFEB+KsVl47Rd6mQTeQjRRzbsmYKhUeSk2OO64a9KLdpMY2pKOUZ2r3N+/2UCWZ6ZyuPQQzUISGdP1PhZl/MzULe9hd9qIsEbRM7Y3sUHu1/x48wTAgAED8cEJ3ND9Njr0bu+ttyeEEOIUZOzeDH7bDbZy301qMWI+v2nU/jfout7wRBshhNfkpucx6/H5OMp937kSwGg2cvP0UQSFB/llfiGEECcn5758bF+sA4cPdr1NBozdm2G9spv356oDyfEWoomKTorCGmrx2/wxbaIk6BZCCOFxxtZRmPq2ALMPwtAgM5ZhTSdtUgJvIZoog9FAyhXdMFl937LdEmKmz1U9fD6vEEKIU4P5wo4YWkV6N/i2mrDe2BtDUNPJrJbAW4gmrOsFHf0yr8FgoP1Zrf0ytxBCiJOfwWTEen0v7wXfVhPWm3pjTPT/gcrKJPAWogkLCg+i9xXdMAf5btfbHGSm/y19MVl8v9MuhBDi1GGwmLDemIKpT6Lngm+LEUNcCNbb+2JsVXvzOF+TwFuIJq7vtSmExoaCofZrG8tgMhDfIYZuF3fy/mRCCCFOeQaTEcuwzlhvTIFwK1gaGJoaDWA2YhrQGus9/TAmhHl2oR4iVU2ECADZu3OZM24+jnLNq/NYQsxcPWkEEU30B5YQQoiTl27X0DYexrE0HQrLQXNCbYVPrCbQdUx9WmDqn4QxNsQna20oCbyFCBB7Vuxj4fg/0GzeCb7NQSYue+l8Ers188rrCyGEEHWh6zr6gUK0HTk49+ahZxZBiR10JxgMYDRgSAjH2DoKY9sojF3iMQRIeqQE3kIEkN3L9/Hr20vQyh1g8EymmMEApiAzw144jxY9JOgWQgjR9OgPPQQffgxOB4Y5c2DECH8vqUEkx1uIANK+fxIjtd8IL83G7Gh81y9zkJno1lFc9falEnQLIYRosgxGIwaHDYPTCU7ftpv3pKZT2FAIUbv//pdms6dyrdHC30PuZ1NsPwA0e/1+CJmDTOg69L2mJ72v7I7RJJ/BhRBCNGGmSqkkmnfPO3mTBN5CBIqDB+H++wEwmw0MmHQHKYnt2TR/G5t+2IbudOJ0OE8YhJtMYLSaMQeZSbm8G10v6ERwpHSmFEIIEQCMlTaIZMdbCOFVug533QW5uer2P/8JPXoQBpx5Qx9OvzaFw9uyyNqZw4GNmeSm56kKKA475v17ic/fS4tze5Jw19UkdIrDYPRBbUIhhBDCU2THWwjhM9Onw7x5ajxgADz2WJWHTWYjLbo3o0X3ZvQa0dX9QEkJhB0tDXj2A9A53kcLFkIIITyo8o53AAfektgpRFO3bx88/LAah4SoINxUx7JJoaGQmKjGu3Z5Z31CCCGEt1X+vRfAqSYSeAvRlOk63H47FBSo26+/DsnJ9XuNjh3Vvzt3enZtQgghhK/IjrcQwus+/hgWLFDjoUPhgQfq/xodOqh/d+8O6F0CIYQQpzDZ8RZCeNWuXfD442ocHg6fflr1E39duQJvmw0OHPDc+oQQQghfkR1vIYTXOJ1w221QXKxuv/02tG/fsNdyBd4g6SZCCCECk+x4CyG85t13ITVVjS++GO68s+Gv5crxBjlgKYQQIjDJjrcQwiu2boWnn1bjqCiYMgUMjai7XXnHWwJvIYQQgUh2vIUQHudwwK23QlmZuv3uu5CU1LjXTExUZQhBAm8hhBCBSXa8hRAeN348LF+uxiNHwk03Nf41DQb3rrfkeAshhAhEsuMthPCojRvhH/9Q49hYVUqwMSkmlbkCb9nxFkIIEYhOkpbxEngL0RTY7XDzzarkH8CHH7o7TnqCK/A+cgQKCz33ukIIIYQvVE41kR1vIUSjvPoqrFmjxtdco748qfIBy927PfvaQgghhLfJjrcQwiNWr4ZXXlHjZs3g/fc9P0flkoKS5y2EECLQyI63EKLRystVionDoW5Pngzx8Z6fR0oKCiGECGSy4y2EaLQXX4RNm9T45pvh8su9M0+7du6xBN5CCCECjex4CyEaZflyePNNNW7VCiZO9N5cISHQsqUaS+AthBAi0MiOtxCiwUpK4JZb3J/ap0yB6GjvzunK85YcbyGEEIFGdryFEA327LOwbZsa33knXHKJ9+d05Xnv2RPQuwVCCCFOQbLjLYRokNRUd1pJu3bw9tu+mdcVeNvtkJHhmzmFEEIIT5AdbyFEvRUVwa23gq6r259+ChERvpm7cmUTSTcRQggRSGTHWwhRb0884W5g8+CDcO65vpu7ci1vOWAphBAikMiOtxCiXhYsgI8+UuNOneC113w7v9TyFkIIEahkx1sIUWf5+TBmjBobjTB9OoSF+XYNzZq555TAWwghRCCRHW8hRJ09+ijs36/Gjz0GZ5/t+zUYDO5db8nxFkIIEUhkx1sIUSfz5sG0aWrcvTu8/LL/1uIKvGXHWwghRCCRHW8hRK2ys1WdblCf1qdPh+Bg/63HFXhnZ6v0FyGEECIQyI63EKJWDz4Ihw6p8TPPwBln+Hc9lQ9YuqqrCCGEEE1d5cBbdryFEMeZORO+/lqN+/SB557z73qgaklByfMWQggRKCqnmsiOtxCiisOH4d571dhiUSkmVqt/1wRSUlAIIURgkh1vIUS1dB3uuQeystTtf/wDUlL8uyaXtm1VdROQwFsIIUTgkB1vIUS1vvoKZs9W4379YNw4/66nsuBgaNVKjSXwFkIIESjkcKUQ4jgHDsADD6hxUJBKMTGb/bumY7nyvCXHWwghRKCQcoJCiCp0He64A/Ly1O1//Qu6dfPvmqrjyvPeuxccDv+uRQghhKgL2fEWQlTx6afw009qPHAgPPKIf9dzIq7A2+Fwd9MUQgghmjLZ8RZCVNi7V7WFBwgNhc8+q/rpvCmpXNlE0k2EEEIEApMJHdANxoDe8W5iyadCBCCnE8aMgcJCdfvNN6FTJ/+uqSaVa3nv2gXnn++/tQghhBDV0J06zl25OPfm4dybh36oEF78FThameutPzA0D8fYNhpT+xgMrSMxuKp2NWESeAvRWB9+CL/9psbnneeu391USS1vIYQQTZRebMOx+gDaigxwOMGugX70QUOlRI1SB/qePLT0PLQ/0yHUgvnsNph6N8dgbbrhrUHXdb32y4QQ1dqxA3r3hpISiIiADRtUreymTNchMhKKiuCaa+Cbb/y9IiGEEKc4XdfR1h7CMX+7+j3laEB4ajGC2Yjlim6YkuM8v0gPkBxvIRpK0+C221TQDfDOO00/6AbVQEdKCgohhGgi9CIbts/XqqDb7mxY0A3quaUO7DM3YftuM7qt6VXuksBbiIaaOBH++EONhw1Ted6BwpVuIqkmQggh/MiZV0r55JXo+/JV4OwJdifOtCPYPl2DXmr3zGt6iATeQjTEli3wzDNqHB0Nn3zibsUeCFyBd26u+hJCCCF8TC8sxzZ1NRTbwNMVAjUdPbsE2/S1TWrnWwJvIerL4YBbboHycnV70iRo2dK/a6qvygcsd+/23zqEEEKcknSnju3LdVDqcB+e9LSjwbd99hYvTVB/EngLUV9vvgl//63GV14JN9zg3/U0ROWSgpLnLYQQwsccS9PR88rA6eUaH5oqS6ilHfHuPHUkgbcQ9bF+Pbz4ohrHx8NHHwVWiomLlBQUQgjhJ86sYrQlez2X010buxP792lNIt9bAm8h6spmg5tvBvvRb9wPP4Rmzfy7poZq29b9gUECbyGEED7k+H0PaD5u+645cfyd4ds5qyGBtxB19corsG6dGl93HYwe7d/1NIbVCq1bq7EE3kIIIXxEL7Xj3JrtvbzuE3HoaCv2o3s7taUWTbe1jxBNycqV8OqrapyYCO+959/1eELHjpCeLjneQviIpmnk5uaSlZVFaWkpmqZhMpkICQkhISGBmJgYjEbZDxMnN8eag/6bXNNxbs/G1CXeb0uQwFuI2pSVqSommqZuf/IJxDXNjlj10qEDLFqkgm+7HSwWf69IiJOOpmns3r2bdevWkZubi8lkQtd1HA53eTOz2YzBYEDTNOLi4ujduzft2rWTIFycVJKTk9XA4aTUVo7VZMZ0tAX8K4NvZ0/+IT5Y8z1Wozs0NRlN/Dj6NS7+9smK+0oc5YSYg3Cdrvp02JNsOLKbzzf+TG5ZIaGWYC7reBZPnXU9ZqOpyhpW7NnI9eddx0MPPcS4ceO8+n5PRAJvIWrzwguwebMa33orDB/u1+V4TIcOEAQ002D5l2on3xIKcZ0hooW/VydEQNM0jTVr1rBhwwYA7EfPhjidx+e1Vg7Cjxw5wu+//05qaip9+vQhJSVFAnBxUti+fTu6rlP+xh8MnvYArw25k4FJPSsen7jyOy7rcBb/Pv++45674fZPK8YdP76BeaNfpV1UYsV9zUJjGN1lMJFBYeSVFXH/golM3/gzt6cMq7jGrjn455+f06dVZy+9w7qRwFuImvz5J4wfr8ZJSTBhgn/X01hODbb/CKsmg7YEngDswOIHwXz0x4GjDCwh0Lw39LkFel6nbgsh6iQrK4uFCxdSUlJSJaiuK1eQvnr1arZt28aFF15ITEyMp5cphO8VlHulfGDbqOYVYx0do8HA3vzMKtdMXf8j5yT1IrusEHT/5XlL4C3EiZSUqB1u1zfo1KkQFeXXJTWYvRSW/RuWTwBHOdgK1f2mo19aMWiVrtfKYe/vcHAV/Pgg9LkNhjwP4QFaxUUIH9m8eTPLli1D07TaL66Fw+EgLy+PWbNmMWjQIDp39u9OnRCNpeeVgck7JXjnbl/K80s+pcheRmxwBM8McPfYyCg8woytvzN31L948c/pvitjWA0JvIU4kaefhu3b1fiee+Cii/y7noZK/xNmXAOlOeAord9zbUXq39WTYf3nMOIT6HmN59coxElg3bp1rFq1yiNBd2WaprFkyRIcDgfdu3f36GsL4Ut6LSUEf9y1nN/S11Tc7h7flq9GPFen1x6ZPJCRyQPZnX+I2duWEB/i3ih7eennPHrGaMIswWodfqxsIoG3ENVZtAjefVeN27eHt97y73oaQtdhwTj46736B9zH0mzq6/sxsP5LuPobST8RopKtW7eycuVKjwfdLpqmsWzZMoKDg+lQuQGWEIHEWPNu97AT5HjXR/uoRJJjknhhyTQ+vPhRft2zmiJ7GcM7Dai4xuDHxncSeAtxrMJCGDPGfXvaNAgP9996GsLphDm3wJZZjQ+6K7MXw66FMG0w3LoYrGGee20hAlRhYSFLly71WtDtomkaixcvJjExkdDQUK/OJYQ3GEIsPqnfrTk10gsOA/BnxiY2HtlF/89VQF9oK8GUvpK03duZNm2a9xdzDAm8hTjW44/Dnj1q/MgjMGSIX5fTID8+oIJue4nnX9tRCoc3whcXwy2/gdnq+TmECBC6rrNw4UKvB90umqbx22+/cdlll/l1106IhjDEh4LD8/nV32xZxPnt+hIfEsX23P18tHYug5JSAHi032juOW1ExbUvr/iSloN78Mgjj3h8HXUhgbcQlf38M0yerMadO7ub5gSSLbNh3XTvBN0ujjI4tAYWvwQX/Mt78wjRxO3cuZPc3Fx0H1VJ0HWdw4cPk56eTtu2bX0ypxCeYjAZMUQHn/DxH3YtZ8GelVXuW3T9O1Xytauz6tA23v77W0rs5cQGR3Bph/6M7ae6S4dbQwi3ulMjQyLDCA0N9VulIIPuq58WQjR1eXnQsydkZIDRCEuXwlln+XtV9VOcBe92gvJ838xnDoHb/4QWfXwznxBNzIwZM8jNzfX5vM2aNeOKK67w+bxCNJZ93la0NQd93zIewGzEfEEHzGcm+WFyRaryC+Hy8MMq6AZ44onAC7oBfrzfszndtXGUwsxr/VoTVQh/ycrKorCw0C9zZ2dnk5/vow/YQniQ6YyWYPJT+KnrmHr6tyyupJoIATB3Lnz+uRr36AEvveTf9TRE4SFI+15VH/HpvAdgz2Jof65v5xXCz9LS0ioa5LxaKS3NbrdXtIEHGD58ODk5OSxZsgSTyYTRaCQhIYGLLrqI1q1bA7Bnzx5mzZrF2LFjAfjss8/Yv38/RqMRo9FI8+bNGTZsGM2bq0YhTqeTrVu3cuaZZ/ryLQvRaMbECAwxIehHin07sQGMneMxhPr3XJIE3kJkZcFdd6mx2awC8KAg/66pIVZ+BP44bGUrhqVvSeAtTjkHDx6sGD/zzDMV4wkTJjBy5MgqZf8WL15Mjx49uOqqq3A6nSxatIgZM2ZUBNrVGTZsGH379sXpdPL7778ze/Zs7rnnHkDleh84cMAL70oI7zMPaYv9+zTfNrIxGTEPbOO7+U5AUk2EuP9+yDzaWvbZZ6FvX/+upyF0nf73TqbdxzHklFX9tr5oVgKtPmnJvkITjyyOpt3UFiRPS6z4uuC7BAA2Zpnp8lkiu/NNFc9df8RCt+mJ7CtU9/X/uhmpGVV3C77ZFswV76yE4iNefpNCNB1Op7PBqR5Go5GUlBQKCwspLq59189oNNKzZ0+OHKn6PZaTk+OzQ51CeJKxWwKGlhHgq70isxFT7+YYW0b4aMITk8BbnNq+/VZ9AZx2mgq8A1FBBuCkdYTGnB3u09tbcsyUOqr+ZLs3pYjttx2q+Fo4Sv0y7xnv4NYexTyxJBpdVxsRY1Ojefz0QlpH1FIqzWCEjL88/a6EaLIKCgowGhv2K1TTNNatW0dISAghIbU3otI0jfXr15OUVPVAmK7rlJR4sXqREF5iMBiwXtkdzD4KQ4PNmC/q5Ju5aiGpJuLUlZkJ9x3tkGW1wvTpYLH4d00NdXAVYGBUp2Jmbg9hTE+1izZjWyijk0t4c2VknV5mbN9CLvwugS/TQjlSYiTMonNbjzrk4Tk1yPgbOl/WiDchROCw2Wz1rqO9adMmtm3bhs1mIzg4mGuuuabG4P2nn37il19+weFwYDabueaaa6o8bjQasdvtDVq/EP5miAzCfEU3HLO3eKW2dwWLEeu1PTFYTLVf6wMSeItTk66rvO7sbHX7pZegVy//rqkxMlaCrtO3uY3vdoSyPddMhygH3+8K4fsRWXUOvINM8PbgfG7+ORanDj9ckVVbh9+jdNib2qi3IEQg0XW93oG3K8e7pKSEb7/9lgMHDtCuXbsTXn/ppZfSt29fdF0nPT2d//73v9x6660VByxBpbwIEajM3RKgzIHjp+3eCb4tRizX9sLYqm6/A31BAm9xavriC1XJBKB/f9WtMpAVZuAqijqqUwkzt4dwVgsbydEOEsOqpol8vCGczza7W71f1LaMiUPzKm53ibVjMuh0jtHoFO04bqrbf4mt8tdBmwa94u1QfNiz70mIJsxsNjc4vzo0NJThw4fzySef0KtXLyIias47NRgMtG3bltjYWHbu3FkReOu6jtksv8ZFYDOf1gKsJhxz01Tw7YljCyYDmE1Y/68XxjY1N9/xNfmOFaee/fvhoYfUODhYpZgE+i+vSiUERyeXctW8ONILzYxOPj7/8+5eRYzrd+Lawy8vj2RACxvrsix8vzOYyzuWVXl86kU5DG7lnu+bbSF8nRYKTvmTtzh1RERENKpNfHx8PB07dmTp0qVccskltV6/b98+jhw5QkJCQsV9TqeTsLCwGp4lRGAw92iGsWUE9u82qzKDjal2YjFi7BCDZURXDKFNL300wKMNIepJ1+GOO8BVjeC116BLF/+uyRMs7gNaSREabSI0ftsXxNuD82p40vFSM6z8sjeYxVcfYc1hC2NToxncqpyY4DpsQZgCsASjEA1ktVoJDg5u1OHGs88+m88//5xBgwZV+/iPP/7I/PnzAQgPD+fcc88lOTm54vGIiAhMpqaRtypEYxljQrDe3hdtzUEcqXuhzKH+pFoXBsCs2tGbz+uAqUu8V9faGBJ4i1PLlCnw889qPHiwe+c70EW3p3JdpvGD88gvNxJq0eucNldiNzBuSTQvDiggNtjJ+W3KGdyqnBeXR1VJRTmhKP/XRxXClxISEti7d+9x9z/yyCPH3Td06NDj7ktKSqqo/x0WFlalpvett95a6/zNmvm3A58QnmYwGDD3bYnptBY49+ahrdiPMz1fBeBmo0pD0XXVs8KA2hkPtWDsEIO5fxLGFv4vF1gbCbzFqWPPHnD9YgsLg2nToIHlwJqcVv2qNM9pF6kB1e8UfLg+nCkb3X+eDjLpbLw5k9f/jqBTtIOrOrlbzr80oIChMxJI3R/E4KTyE89vMEK7IY1+G0IEko4dO3LgwAG/VBaxWCxVGvQIcTIxGAyY2sVgahcDgF5sw3moCErs6JoTg8kIEVbVBTM4sEJZgy7V98WpwOmE88+HxYvV7Q8+gHvv9euSPKokG95u6ft28S5BkXDdHOleKU4pmqbx+eef+yXwDg4O5sYbb2xwLXEhhH/Id6w4Nbz/vjvovvBCONp2+aQRGgeRSbVf5y2aHVr289/8QviByWSiW7duPg9+TSYTvXr1kqBbiAAk37Xi5Ld9O4wbp8aRkTB1apW0jJPG2Y+DxQ8VDgwm6HEtBIX7fm4h/Kx3794+P+BoNpvp3r27T+cUQniGBN7i5KZpcMstUHo0b3nCBGjd2r9r8paUG0H3QzMNkxUGPOr7eYVoAkJCQhgyZIjP6mmbzWbOPfdcgoKkipAQgSiwMtKFqK9//xuWLVPj4cOhDpUCPKrggGrnfmAVFB0ER7kq/RfTAVqeAS36QrCHivuv3QybQqF9KVg985K1Mpgh6SxITPHRhEI0PR06dGDbtm1kZGQ0qrZ3bUwmE23btqVNG6kgJESgksOV4uS1eTP07Qvl5RATA5s2QYsW3p+3KBNWfgx/vQflBWAOAlsx6JV+IZusYA4GeynEtIeBT0LP/wNraP3nKyyE556DSZPAqMMDQBSVqwt6jzUcHkiDyFY+mEyIpstutzN37lxyc3O90sbdZDIRHx/P8OHDpXa3EAFMAm9xcrLb4eyzYeVKdfurr+D//s+7c5Zkww/3Q9oclUPuKKv9OS6WMECHs8bCkOfBXMct6x9+UNVZ9u1z33f7JdBucf3mbwhLGAybBKfd5t15hAgQNpuNefPmkZub69Gdb5PJREJCApdeeikWS9PrxCeEqDvJ8RYnp9dfdwfdo0bBddd5d74ts2FiR0ibDVp5/YNeezHYS2D5v+GDHnBwbc3XZ2aqDxLDh7uD7nbtYP58mPITXDIRzCE1vkSjWEIh5Qboc6v35hAiwFitVkaOHEmXLl08tittMpno0aMHl112mQTdQpwEZMdbnHzWroV+/cDhgIQElWKSRd+32AAAIABJREFUkOCduZxOtcu9/nMVOHuEQaWhDP8Q+txS9SFdh88+g8ceg9xcdZ/RCI88Ai+/rBoDuSx/FxY+BY5SPMoSCj2ugZFTT54GREJ42KFDh1i4cCG2khIcDXi+xWIhKCiICy+8kARv/fwSQvicBN7i5GKzqaB7/Xp1+7vv4KqrvDOXrsPsm2HLLA8G3ZWYQ+CSd+CMu9XtnTvh7rvh11/d1/TuDVOmwBlnVP8ay6bC/+4Ekw6N3oA7+oFg0DMw+NmTsySjEB7kdDrZ8+CDrI2KIrdlSwwWC44avm/MZjO6rhMfH0/v3r1p06aN1OoW4iQjVU3EyeXll91B9w03eC/oBljwpEox8UbQDWqnev6jENoM5m2Hf/wDyo6msAQHw4svwtixUNOfn99dCP/T4UogORicDcz7toRBVGu4egY079mw1xDiFGPMy6PDlCl0sNnIb9mSQ3PncqioiMzMTMrKynA6nRiNRkJCQmjevDmJiYkkJiYSGRnp76ULIbxEdrzFyeOvv2DAAJX+0aKFSjGJifHOXHv/gC8u8nwaR3XsRpjgBFd8f+65MHkydOpU8/PmzIErr1Tjm2+C52+AGY9AUZqqeFLbx26DSVVkiekAA8dBz+vAJJ/Vhaizjz5Sh59BfS/OmuXf9Qgh/E5+i4qTQ2mpapTjKuM1ZYr3gm5bCcy81jdBN4DBCSOBn2Pg7bdVLfLa0jxyc92/8BMT4Z0JEBsLqe1gWRr0BG4dAjmboSwXjGbAAE6HyuFu3gvaDoHuo6DFaV59e0KctD7/3D2++Wb/rUMI0WRI4C1ODs8/D2lpanz77TBsmPfm+nM8lOZ67/WPZQaSTfDEJ9B3VN2e8+ijcOiQGn/4oQq6s7Nh4UJwAMZBcMdi9bitGGxFquulJdRzDX2EOJXt2OFu3hUb692fSUKIgCGnNkTg++MP1aESoE0b99gbNAesmOi73W4XowabptTt2p9+gunT1fi66+CKK9R4zhxV6QXgmmvc11vDILw5RLSQoFsIT/niC/f4uuvA6qt2skKIpkwCbxHYiotV6oXrqMKnn4I3DyZtmwea3XuvX5O9i6Ego+Zr8vPhrrvUOCEB3n3X/di336p/DQZV21wI4R1Op6SZCCGqJakmIrCNG6fK7AHcdx+cf75Xp+t/9SNk5oez+oZiYoPdbaEvmpXApmwLy6/L5O1VEczZGYLVpD4MJIVrXNimjPv7FBFp1Xl3TTiT1oYDoOkG7E4IrnTtoquP0OqTlnSNsbNg1BGMR9O531gRwsF7b2PCf3458QKffBL271fj995z1y/PynKXIRw8WB0+FUJ4x9KlsGePGnfuDGee6dflCCGaDgm8ReD69Vd4/3017tgR3njD+3M6bLSO0JizI4QxPYsB2JJjptRR9bDjvSlFjOtXSJkD0nItvLIikivmxjPv8iweOq2Ih04rAuCbbSF8nRbKnJHZx011qMTE9ztDuLLT0bQWpwMKD514bb/+qqqdgKqgcPXV7sfmzAFXC+vKaSZCCM+rnGZy881S814IUUFSTURgKiiAMWPU2GBQ3RzDw707Z0k26E5GdSph5nZ3O/YZ20IZnVx9Le9gM/RJsPPZRTnklhn5Zlvd27jfl1LE+FUROJyV7izNqv7ioiK44w41jomBDz6o+svelWZiNHq3trkQp7rSUvf3G6h+AkIIcZQE3iIwjR0L6elq/OijcM453p/z8EYwGOjb3Eah3cj2XDOaE77fFcKoTjUftgy36gxqVc6KQ0F1nu7S9qVEWJ18uy3UfaejDBzlx1/8zDPuP21PnKhKCLocOQK//abGgwdXfUwI4Vn/+586awEwZAi0a+fX5QghmhYJvEXg+fFHmDpVjbt2hVde8c285YUVQ9eud2pGEMnRDhLDtFqfnhimkVde9285A/DE6YVMWBOOTat0r6246oVLlsCkSWo8bBjceGPVx2fPljQTIXxFDlUKIWoggbcILLm5cOedamw0qrJ5IXVP32gcd5PX0cmlzN4Zwrc1pJkc62CxieggZ+0XVnJ+m3JahDn5cktYpWVUeo2SEnfKTWQkfPzx8fmkkmYihG8cPgzz56txcDCMHu3f9QghmhwJvEVgeeghOHBAjZ96yrfVAszuAD8pQqNNhMZv+4IY1q6s1qcW2w38kRFE/8Rq0kRqMe6MAiatDT96gFNXTW5cXnhBNeoA1dUyKanqkw8fhkWL1HjoUGjevN7zCyHq6Ouv3X9duuIK75Y2FUIEJKlqIgLH7Nnw5Zdq3KuXCjp9Ka4zlXe9xw/OI7/cSKhFr3oAspJyDbbmWPjXX5FEBTm5tnP9G++c3dJGl1gHM7aHcmF7J1iPBt7Ll8M776jxBReojp3HmjVL1RQGSTMRwtskzUQIUQsJvEVgOHIE7r5bjc1m9QsuqO4HFT0iqjUq81ppF6kB1ed2f7g+nKmbwtB1VZv7gjZlPNCniFCLXu31tXnyjAJGfJ8AIdHqjrIylWLidEJYGHzySfUlyyTNRAjf2LQJVq9W4+bN4cIL/bseIUSTJIG3aPp0XTXHOXJE3X7hBejTx/frMBhY8UQH2Hd8LW2zETLuVCkwE4bmMWFoXp1e8trOpdXugrtey6VvMzsZdx+GQUfz2//5T9iyRY3feKP6ygmHDsHvv6vxeee5m+kIITyvcu3uG25QGwRCCHEMyfEWTd8338DMmWp8+ukqt9tfOl8DWP0zt8kK3UfBqlXuZkGDB8O991Z/vaSZCOEbmuZOgwO46Sb/rUUI0aRJ4C2atoMH1W43gNWqqphYLL5fx/79qh37lc+D3eb7+QFikyGmi0ox0TRVzWXqVJVGUh1XmonJpDpZCiG8Y9EiyMhQ4169oHdv/65HCNFkyd/CRNOl63DXXaqEIKj0ih49fLuG9eth/HhVrcDhOHof0Acw+XAd1nAY+CS8/rpaE8C//gWdOlV//cGDkJqqxuefD/HxvlmnEKeiYw9VSot4IcQJSOAtmq7p02HePDUeMAAee8w38+o6/PqrCrh//tl9v9GoOmRu2Qo9M30YeBsgqg3o3eCVW9VdAwao0oon8t136n2ApJkI4U1FRSqtC9TPiOuv9+96hBBNmgTeomnatw8efliNQ0JUEG7ycqRrt6v0jPHjYe1a9/0hIeqXaXY2zJmj7vsRGGkCU+0dKxvNHAxXfQ0jxqg1BgXBp5/W/L/HjBlHn2tW9YSFEN4xezYUH+0me8EF0LKlf9cjhGjSJPAWTY+uq5rUBQXq9uuvQ3Ky9+YrLFTl+CZMUAG/S7Nm8MADEBenKqlkZ6v7ExPhuXeh9BNIXwKO2hvoNJglFAY9C9N/VIcqAV58Ebp2PfFzDhxQbeRBpZnExXlvfUKc6qR2txCiHiTwFk3Pxx/DggVqPHSoCn69ISMD3n1XzZef776/c2d4/HE4+2y16/7rr+7H7r5bfRCIjgbbMJg6ELLSQKt/R8paWUKhxzUQfwW82Ffdd/rpam01kTQTIXxj/373z4fwcPnrkhCiVlLVRDQtu3a5A8vwcJVScaKqHQ21cSPceiu0bw9vvukOus85R6WSrF+vdrfPOMP9S7VbN7WL/NFHKugGsIbBbanQvFeVdvIeYQmFlBth+GS1+19erqq5TJtWe31gVzUTSTMRwru++sr9IXf0aNXMSgghaiA73sLzdB3y0+HAKti/HHJ2gKNU1aEObwFJZ0HL0yG+G5gq/V/Q6YTbbnPnS779tgqOPbWmRYvgrbdg/nz3/QaD6uj4+ONw1lmwYgX06wcbNqjHrVZ49lkYN676TpnBkTDmD/j1Gfj7Q/U+G8NoBlMQXDoRThuj0l+WL1ePPfusKlVWk4wM+OMPNb7wQoiNbdx6hBDV03V19sRFancLIerAoOt6w3pYC3GskhxYPRWW/xvK8sFkgfIiwFn1OksYGIyga9D7FjjrEYjvrILMRx9V11x8Mfz0U+PLcjkc6qDh+PHuds6gDkzedpuar1Mnlef97LPw3nvuHaxBg1QaSrdudZtr/18w4xo4shcsVO4uXzuDCcxB0OpMuPILiEqCHTsgJQVKS9W/f/+tPgjUZOJEeOQRNZ42Te3sCyE8b/VqlfoF0Lo17Nnj+b/OCSFOOrLjLRrPVgILx8HqKYDBvet7ot1fe7F7vHoKrJ0GsSnw2jp1X1QUTJnSuKC7sFA1l3nnHUhPd9+fkKByxu+7z13beu5cuP9+la/pmv+tt1SKR31+kSadCdbH4ZsHYSDQ0QLWELAVge48/nqjRaWUaOUql3vAWEg82njD6VTzl5aq6iXTptUedIM7zcRigcsvr/vahRD1U/lQ5Y03StAthKgTCbxF4+z9A2ZeC6W5Davu4bSrr0N/we3AL8CDEyEpqWHrOXhQHZj86CPIy3Pfn5ys6oDffLPa7XZd++CD6jCiyzXXqF3jxMT6z63r8OFHsAsobg6b/4LDa9ROeHoqlGTDvt1QWAamGLjmSWjVT+1yB0VUfa2PPnI3wBk3Dvr2rX3+ffvgzz/V+KKLICam/u9BCFE7u1011XKRNBMhRB1J4C0absUkWDCu8XnNoI75WoFLTWCdDdr1KlWlrjZvVukkX36pfim6nH02PPEEjBjhrnvtdMLkyfDUU+6Dla1bwwcfwPDhDX8PqamwaZMa33knxLZRX10r7TyPHq0C/V5JMOmp6l9nzx7Vnh5Umsvzz9dt/pkz3WOpZiKE9/zyCxw+rMZnnFH3dDQhxClP/jYmGubPt2HhU54JuiszabDzF/hqBGiOmq/VdVi8GC67TLWSnzZNBd0GA1x5JSxdqr6uuMIddG/eDIMHw733qqDbaFQ50Zs3Ny7oBnj//aPvwaTKDlbHtQudk3Pi93TnneqAqdGo3lNwcN3md6WZWK0wcmTd1y2EqB+p3S2EaCAJvEX9bfwGfnsB7CXeeX1HqWpMM/f2EzzugG++UdVHzj0XfvxR3R8cDPfcA1u3qhbOZ5/tfk5ZGfzjH9CnjwrGAXr3VhVD3nlHlS5sjAMHVAc7ULnVJ0qVcVUZyc2t/vGpU2HhQjV+9FHo379u8+/d665+cvHF7pKHQgjPysuD779XY7MZrrvOv+sRQgSUUybVxKk5ycsoIGtHDofSjlBwqBDNrmE0GwmNDiGxWwIJneKIbReD2erl1uSBrPAgzL0THF4Kul3sJbB5JvS4FjoPU/cVFam63u+8o9IxXOLi1IHJ++9XhyeP9fvvagd661Z1OyQEXnpJ7XRb6pHOUpNPPlEfCEAd3DwR1453SYmqzV25ROH+/SoPHVSllZdfrvv8kmYihG/MnKm+dwGGDav+Z44QQpzASR94F2eXsGn+Njb9sA2npipLOMqOT2HYvXwfRpMRp+ak4zltSbm8G3Ht5HBaFboOs272bov0yuwlMPsmuO5PmPw5fPhh1Z3iTp1g7Fi45RYIDT3++bm5Kld6yhT3fRdeqA4udujgwXXaVdlBUK3czzvvxNdWrqudm+s+xKnr6sNBQYG6/emn1b+nE5kxQ/0bFCRpJkJ4U+U0EzlUKYSop5M28C4vtrF08t/sWroXAM1eTTm3SjSbhoYGwPbFu9n1x15i28dw3qMDiWoRUeNzTxm7foX9y1QVEl8pzoN7usP8Sv/9zjpLHZi8/HJ37nZluq7ynR9+GDIz1X3x8apO+PXXN742+LHmzFEVUkDtdtf0+pUrjeTkuAPvL790p8w88ICqIV5Xe/aoxj8Al1wCkZF1f64Qou527VIdbEGlczX2XIgQ4pRzUuZ4p68+wNd3f8+upXvR7M5ag+5j6U4dh03jyPZsZj48j/Vzt6A7pc8QS9+oWoPbF4xO6ONUHxEvv1x1ZVy2THWbrC7o3rtX/TK87jp30H3LLbBlC9xwg+eDbnAfqgwLq/2g1bE73qCC9ocfVuN27eC11+o3f+U0k6uvrt9zhRB19+WX7vG119b94LMQQhx10u14r/pmA2u/24ijXGv0a+lOHUe5xt9friNj7UEuenoIJsspmv+dvw/S//DP3BYz/O9VuOSJE1/jcMCkSfDccyp/GqBjR5UCcv753lvbpk0qhxxUE42oqJqvP3bHW9fVLrkrCJ8ypf4HPV3VTIKCVNlEIYTn6Tp88YX7tlQzEUI0wEm14/3Xf9Z6LOiuzFHuIGNDJj++9Buao3675yeNzd/R/4so2k1tQU5Z1f/bXDQrgVaftGRfoYlHFkfTbmoLkqclVnxd8J06fLQxy0yXzxLZne/+8LL+iIVu0xPZV6ju6/91M1Iz3B0av98ZTPcv4lm29NsTr23NGpV+MnasCrrNZnj6adiwwbtBN6ja3y7331/79ZUD79xclZs9Z466fddd9V/v7t2qlTzApZdKmokQ3rJ8OezYocYdO8KAAf5djxAiIJ00gfeWX7az4fstHg+6XTSbRubWLBa/+6dXXr/J27MI0GkdoTFnR0jF3VtyzJQ6qqZv3JtSxPbbDlV8LRx1BICe8Q5u7VHME0ui0XWwO2FsajSPn15I64jj/7t9uy2EZ5ZGM/2SbAaEblM7TpUVF6tc7379YNUqdV///mr86qvuDpXeUlDgPmg1aBD06lX7cyqnmuzb5w7Wk5LgzTfrvwbXoUqQaiZCeNOxhyq9kbYmhDjpnRSBd2FmEX9OWem1oNtFs2nsWb6PvSv3e3WeJunASgBGdSph5nZ3QDtjWyijk+teWnBs30IOlxj5Mi2USWvCCbPo3Nbj+LzxL7aE8vLySL66NJt+ze2g2aDwgPuCn39Wge748aBpKj1j0iRVozslpeHvsz6++EKVOIS67XYDRESoxjigWk5nZanx5Mm1p6lUx5VmEhwsB72E8JbyctU7wEWqmQghGijgA29d11k4/o96H6BsKEe5xqJ3/qS8qNwn8zUJmh2KDgHQt7mNQruR7blmNCd8vyuEUZ3q3r0yyARvD87n1b8i+XhDOG8PzsN4zMbRF5vDGL8qgm8uy6Z3wtEKKqYgOLxJtWm+8UZVvWP3bvXYyJGq8+QDD1R/4NIbdN19qDIxUXXKrAuj0Z1u4movf8stKk2kvnbudO/0DxumgnohhOf98IP7HMY553i2HKkQ4pQS8IH33r/2k5Oe59OqI/ZyB6u+2eCz+fzOXgoGd0Dr2vVOzQgiOdpBYljVvzR8vCGcbtMTK74eXly1i2KXWDsmg07XWAedoo+vqZ6aEUTfZja6xVZ+TIef/wfdusF//qPuatECvvtO5Ui3bu2xt1snv/+uKqWAavFutdZ8fWWV87ATE+Hf/27YGiTNRAjfkNrdQggPCfiqJmtnbaq2IY43Oe1O0n7ZwZk3nRa4XS7LyyE7W31lZR3/b+VxwRG42l27e3RyKVfNiyO90FxtmsndvYoY16/whFO/vDySAS1srMuy8P3OYC7vWLUhz2vn5DNxTTiPp0bz9uA8lUpZXAyfvwc5Ry+6915Vdq8h6Rme4NrtNplU45v6cDXJAdXMp3Led3240kxCQuCyyxr2GkKImmVlqR1vUJWDpGSnEKIRAjrwzj9QQNau3Nov9AYD7PpzL52HNoE/OZaV1Rw4V/dv4YkD4+MYgUrpIEkRGm0iNH7bF8Tbg/PqtdTUDCu/7A1m8dVHWHPYwtjUaAa3Kicm2P0Xi/gQjW+GZTNqXjxPL43i9XPywaGBHejeXeVDDxxYr3k9KiMDZs9W4yuugFat6v7cH39U//uDanV/+eUNW8P27aqaC6igu74lCIUQdfPf/6pypaDS2mKko7EQouECOvAeetFQsvKyGJvyLKGWsIr7J2+ayKHSAzzUaxzRQWo3cXHGAlIPLmRM1/tJCm9T5XUKbQUsOvAzO/K3YtPKibBG0SMmhbMTh2I1qRQCXdeZtOFNzEYz9/V8DHupg7QFOzwfeJeW1j2Ado2LPdjUJjpadXmMi1P/usbGyVUuGz84j/xyI6EWnbpWWCyxGxi3JJoXBxQQG+zk/DblDG5VzovLo5g4tGoAnxjm5NvLshn1vzj+sSySl84ogFseg3Gv1i+twxsmT1YHOqHuhyoB8vNVyUCX6OgTX1sbSTMRwjekdrcQwoMCOvDW7Box1hg25qzlzOZqBzSz5CB2p63Kdbqusz57NSGmUNZnr64SeJc6Svg07QNah7dlTNf7iA6KJd+Wx7JDqeSWZ9M8tAUAe4t2U+wowqk7ySjeR6uw1mTvrmG3XddVTen6BNBZWSrw9pTYWHcAfWwgXd2/sbGqBnZ1/pMGrKy42S5SA6qvIvPh+nCmbHR/EAoy6Wy8OZPX/46gU7SDqyodxnxpQAFDZySQuj+IwUlVD6y2Ctf49rJsrpoXT7DFzNP/e8v/JbzsdhV4g8o3Hzq07s994gm1W+5SOeWkvlxpJqGh6mClEMLz0tLgr7/UOCEBLr7Yv+sRQgS8wA68HRp94vqxPnt1ReC9LnsVKXGns+jAzxXXpRftpshewIh2o5mfPpeLWw/HZFRvfVnmEoJMQVzZ/loMBnXWNMoazSVtRlaZa33WKrpEd8fhtLM+axWtwlqjldgofvolwvIzqw+qy6rmLjeYwaCC4uqC5RPdFxPj2Qof7Yay4saFqqzfMcxGyLhTlfqbMDSPCUOrTz95+ezjA83YYCfrb8qsuL3i/w5XebxNpMbK6zMh6Sz/B92gUkwOqQov3Hdf3de0cCF88okad+kCW7eqKgm6Xv/3tW0brFunxsOHq1b1QgjPq7zb/X//BxaL/9YihDgpBGzg7Sh3oDshKbwNG7JXc6Q0k7jgBDblrOO2rvdVCbzXZa+ic3Q3usekMD99Ltvyt9AtRjU72V2wna7RPSqC7urYNRubczcwqsP1OHQ78/bO4qLWw7Fq5eRN+ZawrM11X7jRqALk+uxER0f7rkzeiXS7Cha94J+5reFw2u3+mftYrkOV4eF1/7NzUZGqfALqA9F118FLL6m80aKi+pcBrJxmIge9hPAOpxO+/NJ9W9JMhBAeELCBt73cUbFR2CuuL+uzV9M2ogPxwc2ItLrLtdk1G5tzNnBF+2sxGU10i+nJuuzVFYF3qaOEcEvNbba35G3EbDTTMSoZp+7EqWtsz0+jV3gH7LHNoBl1D6Sjo90NVAJJbEdocTrsW+r7uXUn9Po/3897rA0bIDVVjW+6qe7t2Z9+GvbsUeOJE9354aB2vesbeEuaiRDel5oK6elq3L079O3r3/UIIU4KARt4GyqV2UiJ68v0rR+RV55D7/jTq1yXlrcJo8FIclQXAHrFncYX26ZQbC8izBJOiDmUInvNubbrslbRPSYFo8GE0WCiW3Qv1mWvoler0zB8+QX0S/L8G2yKzhkHM/8P7B48zFkb3QhdrwVrE0in+PBD9/i+++r2nNRUeO89Nb7sMtX8Z+5c9+M5OdCmTfXPrU5aGqxfr8YjRqjgWwjhedIiXgjhBQEbeJuDTOhHK9BFB8UQbY1le/5WRrSr+qf3ddmrsDltTFj/WsV9Tl1jY85a+jc/h/aRyaTlbWJIywuqTTcpsOWxp3AnB0r2sSVXNc2xO+04dAcl9iLMQQH7P2H9dR4OzVMg4y/Qqz9Y6XHlTnjiBzDOrnt3SG8oKHDnew4eDD171v6ckhK4/WiKTGSkqtntytd3ya1nOUypZiKE95WUuL/XDAa44Qb/rkcIcdII2KjRHGSmcpw8ot1oyrRSrCYrzqNBYYGtgN0FO7g+eUxFdRKA5Zl/sD57Nf2bn8OA5oPYkL2GObu/5dxWFxMdFEOBLZ9lman0iTuD7flpxAXHc3OXqk1SPt3yAWsPrmJM6+t98n69wenU2ZdTwpYD+ezPLqHEpmEyQHiwhU6JEXRtEUlcRJD7CQYDjP4a3u8O9uMb53icwwj/c8Luw3DVVepr0iRo2dL7cx/r889VPjbUvYTg88/Djh1q/O9/Q9LRv4xUrgOck3P882riSjMJC2tYm3khRO2+/979/X7eeb7vjCuEOGkFbOANYDK7DxzGBscd93h60W4SQ1vSMapzlfv7NxvI8sxUDpceollIImO63seijJ+ZuuU97E4bEdYoesb2JjYojnXZq+iXMIBwS9U83NMT+rMu8y9CX3gSLrlElZULgOoSuq6zLj2P/yzdzYqd2ZiMKmmnxObewTYZIdhixu5wEmw1Mfy0llzdvy0tokMguDns7gLN1oA3y2mbQ6DLhdDiInjqKfVLcNYs+PVXePNNuOMO3+XK6zp88IEaJybWbed9+XJ45x01vuACGDPG/VhDd7w3b4aNG9V45EjVsVII4XmV00zkUKUQwoMMuq7rtV/WNK2ZuZGVX6/HWdcOLh7W8shmRiw7msJitaoUhEsuUV/duze5nMA/tx9h/A9byC2yUebQqOt/eYvJgMFgIKVlOE/NfJOk+XPgXOAsvBN8m4MhsQ/csggswbBvn8qpnjfPfc3gwaqedpcuXljAMX77Dc4/X41feEFVJKlJWZk6iLVli/owtnEjtGvnfry01J2b/cYb8OSTdVvHSy/Biy+q8ezZqmumEMKzDh5Uf51yOtX3aWamdIYVQnhMAJbXcOswsC0GPwW3FqOTriEH3WX+bDZVq/nxx1X+b5s2qoTcd99BXv3aqntaYamd52as45lv1nIgt5RSe92DbgC7pmNzOFm9J5cb+tzKf8+8HOff4bAUOL6sd+NYQlXN7lt+U0E3qD/zzp2rWjc3a6buS02F3r3h1VdVUxtvcu12m0xVO0+eyMsvq6Ab1O585aAb1E518NH3Vp8db1eaSXi4+nAnhPC8r75SQTeo9DYJuoUQHhTQgXdUiwjiO8XWfqE3BFnpMH+KapYzc6ZKfUiqVN1k/36YMgVGj1alBM85B155BVaudP9Q94Hdh4u4+t0lpKYdpszeuHmdRhPl1mA+Ov827r/qBUqXBYHjcgiOAb2xdcYNKr1k0DNw0wKwHJNGYTDAtdeqgPa229R95eXw7LNw+unu7nKetn8/zJmjxldeCa1a1Xz9qlUq2AYYMgRJP3FoAAAgAElEQVTuuaf661x53nXN8d60SaWaAFx+uTtwF0J4lrSIF0J4UUAH3gB9ruqBOdi3qeomi5FuFydjsphUXe5Ro1RXwvR0lVYwfrzK67UezcPQNFi6VB2269dP5QnfeKNqznD4cM2TNcKOzELumLKC/BI7Ng+m45RZgtnUqgt3PzKF0pdnwPXLIQ2wA456vpgTtWse0xXuWgmDnwVTDf89Y2Ph009hwQLo0EHdt2EDDBgAjz7qPhDlKZMnu+tu13ao0mZTHwo0Te1qT5ly4jx0V553XXe8XbvdINVMhPCWdevcXWFbtlQHK4UQwoMCPvBu268Vce2iMZh8l3JiDjLT9+pexz9gMECPHvDYYyowzMmBH36ABx+E5GT3dUeOwH/+o2rDNm8OZ5wBzz0Hf/yhuhl6QGZ+Kfd++hfF5Q68kcRvswSxJ6IZY79ag3PCR/CtBu8By4EyoJwTp6HoJgiKAmMQbAGmAwdHQLPudV/ABReogPuJJ1Rw63TChAkqzWf+/Ea+u6NsNhV4g8rZHzKk5utfe02tCeBf/4JOnU58rWvHuy6Bt667A+/ISLjootqfI4Sov8q73Tfc4P+OwUKIk07AB94Gg4HzHx9UpcKJN5ktRs4bO5Cg8DqcKgwLU50F330Xtm1TpeXef181PqlcAWXVKhWoDRqk0lJcO+j79jVojbqu88LM9VUqlXiDzaGzJT2Hb5fuVncUAL8CpnHwXQTMB8q6QptBENsT0oHNQG5fuP5/MC4LykfAAWDKVHXosD5CQ1Vax19/QZ8+6r69e1WZvZtuUmlAjTF7tjpYBepwZ03nCdavV6lEoHbfH3qo5teuT6rJpk2qcQ5ImokQ3uJwqA0RF0kzEUJ4QcAH3gARCWEMvOsMzEHeDb7NjnLaZ6ygjfNAw16gY0cVwM2dC9nZqjTeE09ASor7mvx8VTbvrrvUAc3KO+hlZXWaZu6q/Ww9WIjm9H7BmjLdwIfn3cr+mKN10vv3hzvuhu2FsAboOg7GpMLNS2AaMAPYFgttB0FQODzwgHpedrY6PNkQrhzvN95wB6VffgnduqlfpA0t3PP+++rf8HAVyJ+Iw6FSTBwOCApSqTC17ZTVJ9VE0kyE8L5ff4VDh9T4tNPq1iRLCCHq6aQIvAG6XtCJ3ld291rwbTY6SczZxpAV76ua3Y1NZ/h/9s47PIpy++Of7ZveCxB6771JsQGCKBZEQWxYwKteuXb9WUGvBQHBckVQAUVQUBEVFVBBkR5Aeu+Ekt6TrfP747BsyqZBNii+n+fZJzOzszPvbJLdM+f9nu+xWEQ/OGGCaAqPH5eA7eabRTfuYedOab4yYIAEa4MHSxOZfft8BpR5hU6m/LSHQkcNdZYEHEYTr10zVlb+8x8pIPXQubP8DAvzBsUnT3qf79fPawn4zjvnHiSbTGLLt20bXH65bEtNFS391VdLJrwqbNsGK1fK8h13iMSjLN58EzZtkuVx46BFi4qPX9mMd0mZSf/+FR9boVBUHeXdrVAoaoCLJvAG6DKiPR2Htan24NtoMVCnSz0G3t8Gg0En7YSvvba4HvB8qVNHsqZffCEa8NWrxTO6WzevxKGgAH74QWQMzZqJhtiTQc/JAeDHLUloflF1l41bb2BbQktONG4l9lsbN8oTAQGSdQa5hlpnsuJFA2+93pv13rxZrvt8aNJEMlcffeS9gfnpJ5k5mDrVWyhZEZ5sN8h7XBa7dnm9tTt3ltmJyuDJeGdllT+mbdtgzx5Zvv56uWFTKBTVS3a2SMtAZqtGjLiw41EoFBctF1XgDdBpWFuu+r/LsIZaxHXkPNDpdRgtBrrf0ZGrnrkUw523w3ffiT7b6ZSsyIQJ556lLQujUXTC48bBunXifDJ3Ltx5pxRjejh4EN5/X3S/UVFoV1zBnB+3nrdt4Lng1ulZMOIRcXLxZLw7dJBr8eAJvFNSintv33knhJzpDPrOO+c/GJ1OOkXu2uWVZuTlSTb+kku8BZBlkZUlUhWQgsrWrX3v53LJeex2ybjPnFn8esujaNv48nzelcxEofA/X33lrTG56qrin7MKhUJRjVx0gTdAQodajPjgOhr3qY/BpMdgqtpl6vQ6DGYDcc2jGfbOtbS5pgU6/Zms88CBsHw5xMTI+lNPwaOP+tebOzpaMjCzZsGJE5IZfu01CQo9gZ7Dwb6dR8gqLJ49PfH5ExyfOYbjs/7F8U8eImXJFJy5Im9I++0jshK/LvO0mqZx4osnOfnls6WeS/7+DXJ3/3523Wk08b21vtyEeDLeHpmJB0/gDd6iRZCg+667ZPmrr+Qaq4P4eJlBWLTI67+9fr10lXzuubI18598IoE6lG8hOHWqtIYHOV5bH043ZVGZtvFFZSZhYUpmolD4C+XdrVAoaoiLMvAGMAeauXzsJdw64wY6DG2NJcSM0WrEFGAEH+YURosBU4ARg9lA836NuXHSIK57/SpC43x0LevaVXy5GzaU9SlTxHrKZvPvRYFIMzp0gKefhhUrpChx4UIYM4adbS9B83Ft0QMeJuGu96lz61sYAkLJXPNZ6Z18YDu1F3dBDs6cFGwphyre3w0p2/d6M7hduhTfoWjgXVRuAt4A1+mEDz6o1PgqzZAhopX3SEacTnGRad9eOmAWRdO8nSpr1y67Lfu+fdK8B6Q49umnqzamohnvsnTeW7bIeUCa95gr4aSjUCiqxpEjkkwBqaMYMuTCjkehUFzU1GznmQtAYEQAXUa0p/Mt7cg6kU3KgXRO70kh+1QuLpsLvUlPYEQAcS1iiGkSSWS98MpJVJo2FT3yoEHw55/iyJGSIo4k5RXiVTehoRIcXn89W77aSuHWk2XuqjOaCGjYhcw18yp16Px9qwio3xHNaSd/3yosMQ3L3d9k0LF7/Q5iPBvKy3h73AM8NG8uBaRLl0rg/eyz1RtohoaKbvvWW+G++0SGsnevzBqMGSOOKGFh8OuvXuu+0aNFQlISt1s6lRYWih505syqj7UyGW8lM1Eo/E9RC8Fhw6Q2RaFQKPzERR94e9DpdYQnhBGeEEbTS8sPICtNfDz89psUFP7yizwuu0wKIOPjq+ccVWDXyexyn3c7beQfXI85tlGFx3I7beQfSiTqivvRnHYy/viE8O7D0ZXTVbLA7mL/gVP0AfHYLunuUV7GG6TR0NKlIkNZsEBmEaqbXr28Up1XXxWt+QcfSIHqe+95td1GowTovnj/fW+m/KmnRLpSVSrKeBeVmUREwJVXVv0cCoWifDRNuZkoFIoa5aKVmtQYoaHSnXL4cFnfvFkK+DwSgRok3+bbHSN12bsc/+RBkmY/RGHSDkLaDarwWAWHNqEzGLHWaU1A3fZobhcFx7aU+xq3Bjmn02SlZGElFL8Z8RV4DxrkbQNfHUWWZWGxiBPJ5s3Qo4d3PDfe6HU2uOEGkZqU5NAhCbZBHFteeOHcxlBRxnvzZjhwwDsWJTNRKKqfxESva1CDBtC79wUdjkKhuPhRgXd1YLHIdOV//iPrhw5J8L1hQ40Ooywbwej+D5Fwx3skjPqAiJ63kbL4DVz5WeUeK2/fKgIbdkWnN6Azmghs2Jn8fRVb/blSUmShpL4bKs54Gwxerfe6df5//1q3hj/+kCA/+IyW3+NQU69e6YJZTZMseF6eaO1nzjx3e7+iXu2+Mt5KZqJQ+J+i2e7bb5f/a4VCofAj6lOmutDrpdHNhAmynpoqjVyWLKmxIZgN5f86dXo9gQ07g06H7XTZGXlnXjq2k7vI27+WpM/+Q9Jn/yH/UCIFx7biKswp9xzW3DNyl5L6bqg48AbxMg8MlOV33y33XNWCwSA+4n/+WTyInjRJGhzt3evd9uGHIicCeOQR6dJ5rhiN3lqAkhnvojKTyEgZh0KhqF7sdphXpN6lvO60CoVCUU2owLs60emkBfwnn0hglZcH11zj1Q37mXpRQeU+r2kaBUc247blYwqvdWabG83p8D5cTvL3rcEYGk+tYa8Sf8M44m8YR61hr2EIiiD/wLoyjx9gy6dB6nFZOXJE/LKdTu8OMTHeVuplBd4REdJtEqRgNTm5Utd+3qxf73Wl8WS/f/tNHEtefVU80z3NcZo2hZdfPq/TuTQXx1rEsbVNJBuNJ9mSsoWj2UdwuV1ix3jojIvMjTf6LvBUKBTnx48/iisUiOSsadMLOx6FQvGP4B9TXFmj3H67BJlDh0qXy9tvFxePxx7zdqH0A50aRLDhUBpOV3HJSerSt0GnBx0YgqOJvPQeTBHia52z5Qdytvxwdl9zXBPchbkEt7oCQ2BYseMEt7icvH2rCGndr4wR6Ghx8kwm/YUX5BEYCB07igVjly4QFSXBdFmBN0gGevp0yUjNmOG17fMnHgvBkBDYulWaF82aJcH4s8/C669Ld1CdTrpinoPzwem80/x4aDEbkzeSlJuE8dHG6J0NwJgF617BrWk4NQe1cg10HNWSq386Qm0lM1Eo/IPy7lYoFBcAnaZVd9tFxVnWr4fBg0V2AiJPmDjRbzrCDQfTePrzzeSVUWTpb0wuBytevQ6DVolmQp7Zga5d5VGvXvGbkssvF5/yOnUk++vPrO/WreLpDaIx90hcfv5ZLAUPFfEw79BBdOFB5c8uFGVH6nY+2z2HPem70TQNp+as8DVGhxu9Bo3iWnFry9voENuxKlekUCjKIz1dpG+errOnThUveFYoFAo/oQJvf7Nvn7Qg9gRvng6UfnCpKHS4GDRhOQX2mg+8dUCPQ5t469P/kwx///5SHJmYKA4dZXWJ9BATIxlxTyB+6pTXzm/BArjpJv8NfswYybAD7NgBrVp5nztwQIowizZHatAApk2T32s5FDgLmLF1Or8nrcDusp/z8CwGC93je3B/+wcINvto6KRQKKrGtGnwr3/J8g03SP8FhUKhqAFU4F0TnDrlbbQD0K+ftEb3Q6OdyQsS+XrLaZzl+G37gwCjjokfPUHnI1tF4160UMnhkIA2MVGKFj0NairCYACXSwLd6dOlYLO6s1KZmZJVz8+XLPuvv3qf0zTRWH/zjaw3buy1+AO5xsmTITq61GEPZB5g3JoXyHfkY3efe9DtwaQ3YTVYeb7ni7SIbHnex1Mo/tFccgmsWSPLCxeW3aFWoVAoqhlVXFkTeBrteNwpfv5ZGu2cPl2959m7l5vH/wu9u+Yz3qE46XRkq6yUdDQxmUSice+9xa3xpk4VWcedd0qWuaT+3XXmOg4flq6WUVHQpIl4pk+aJE1scsp3WamQTz6RoBu8LeU9zJ/vDbpHj5Zul2+8AVarbPv0U/Hy/uwzrw0hsDt9F8+sfJJMW2a1BN0ADreDHEcOz696li0pf1bLMRWKfyT793uD7shIuPrqCzsehULxj0JlvGsSm02CzC++kPVGjcRusEmT8z92YqJk1VNTGXfd4/zS7nLsOsP5H7cSWPUa4/O30HfC06J9zsryupeU5P33vQHurFnyfnjIzYVNm7wSlbVrJeguD51Ogl9P8WbXrqLX9gTH5aFp8to9e6RZzuHDXi15SorcDKSmQkICbN8uLeVBvrhHj4bly73HGjgQpk3jULiLp35/kkJXQcXnP0csBgsv93qVFpEtKt5ZoVAU58UXYfx4WS5a06FQKBQ1gMp41yQWC8ydC2PHyvrBgzLlmZh4fsddtkxkEqmpoNPxeKdwAu0VaKqrCbPTTs/tf9D38zOuIB07lh10Q/le3sHB0LevuL/Mmye6+BEj5DmTSQpVExKKv0bTYOdOmD1bWs736CHOJJ06iXZ7xgyR+Dgcpcfyyy/ernVjxhQv4Pz3v71FsdOne4NukBulX34RdxNPI5yffsLWoS0vL3vcr0E3gM1l45W148l35Pv1PArFRYfbXbppjkKhUNQgKvCuafR6eOstsacDyaxedhksXXpux/v8cwlIc3MlcHzhBYLee5vx81/B4u/gW9MIsBfy9PdT4ehR2da8efmvqUwTnaJ4vLMdDrj0Ujh2TF733XdiV3j11VKYWRSnUwo6p0+XzHTHjqKn79kTHn5Yvnh37fJmuoxGbyEniObTMytx550yk1ASnQ7uvluOM2wYALOH1iXbmVfxNVUDBc4Cpm+dViPnUiguGlat8s6iNWsG3bpd0OEoFIp/HkpqciGZPRvuuUe0zEajSC9Gjqz8699+25s9Dw6WIPGDD842rfly9Au8m9CbQmcl7P3OgUCzgemt3TS5e7i3+2JQkNwMXHON7xcdPQr168vysGHFW6OXhacQqkEDkXmUzKhrmhw3MdErU0lMFMlLZWjeXHy7u3aVzHbr1qK/j4+XbHpERIWHOPDNRzxl/xK7uebuZS0GCy/2HE+b6DY1dk6F4m/N6NEyCwbwyis10yNAoVAoiqAy3heSO++UzG1goATLt90mRYMVoWnyheEJumNiJGv+3ntyHL0e3niDm6a9xENXNcdiqt5fs0GvI9hqZNrd3Why/QD5AvOQlwfXXgtPPulb3hEf712uTMYbRPYBkqlavLj08zqdBPNDh8pMws8/i0/vnj1S+Pif/0CvXmU3vdmzRwo2GzcWrben6HXUKG/hZQUsqH0Kh7lmNPUebC4bX+yZV/GOCoUCCgqK3+h7OuQqFApFDaIy3n8FSjbaefRRePNN3412nE64/37RFwPUrStZ2u3bZT0iQjLOAwacfcm6A6m8+OVW8u0u7OeZ/baa9DSJC+HlYe2pFX4mkH3sMbHVM5ulIYWHnj1lLPXqFT9IdLS0am7cWDLYFWG3S2B96pRYMS5bdm6DdzpFGrJmjQTjBZXUYteqVdxjvEuXYhaC2bYsRi25E4fbx42GnzHpTUzrN4OYwJiKd1Yo/snMnw+33CLLl14qDboUCoWihlEZ778C3bqJ9rBBA1mfPFmKfuwlrOgKCqSRjCfobtxYMsyeoLtdO5FYFAm6Abo3juarsX25olUcFqMeo77qbeut9kKsDhsP9WvKjHu7e4NugI0bz5youwTFcXGyvmaN2Ah+913xg3l03idPFrPhKxOzWW42QLLZu3ZVefyAyHnathVZjifofucdkadMniwzD74oqikfNEhmGBo2FGvECRP49dfp+HpH0zdksOuNPWx+ZCtbntnOvvcOkLs/lxOLT7LxwT85vTyl2P6nl6ew8cE/ObFYZgIyt2Wx5entOPO8nS4zt2Sx9f924CoQq0UNjaVHlpzb+6FQ/JMoWlSpWsQrFIoLhAq8/yo0awarV3tbl8+dK1lwj091RoYE1IsWyXqTJuL6kZ4u68OHy+sbNfJ5+CCrkZeGtuPTBy7h+i51CXDZCSrMw+gsO0tr1ZwEOG3EZqXwr19m8v3kEdyUtBFdUb9tt1ssAEEywf36iYuIx7M8IwOGDJGsuOdGwhN45+dX3od79GgJnOH87b/+d8aBJSRE5D5duoiW2yMr+eQTaQv/1luiuW/WrPQxDh+WjppPPcWmtQuwl8h2n/4lmWNfJhF/VRztXm9Nu1daE9M3msytoju3xFpIX5de7DVp69KxxFrOroe3DSOkeTDHvkoCwJnv5OgXx6g3PAFDgMhanG4nm5M3nd/7oVBc7CQnw08/ybLV6t9OuAqFQlEOKvD+K1GrljTaufxyWfc02tmyRWz2/vhDtterJxINt1vkKG++KYF6UFCFp6gXFcTjg1vy4/9G8eqXr3Dvyrl0bxxFrXArUXoXMdmp1Es9zoCdv/Pvxkb+N/NxFk29nVs2LCLYll9ag75vnzd49jTOiY8Xl5aXXvI2xZk8Wa7h8OHizianTlX+vTnjHsLs2ZUvnCzJli0yuwASdIeESJb+ww9l2403ivazVy+Ro8yZIxrwzEyxEHz9dfnS9hSIAgcbFe9A6ipwceL7U9S7JYGIDuEYLAZ0Bh3hbcNIuLEOAEH1A3Hb3RSckMx7wYkCNIeboPrFs+71hiWQvSOHrJ3ZHP/qBMFNgglvF1ZsnyPZh1GKMYWiHObN8zbkuuEGv3QNVigUisqgNN5/RWw2mQr1FAIZjWedSoiO9mrBIyPF9q5fv6od/9gxr+66TRvYtk2W77vPG4AOGyZB6ccfi/94vXoSZIMErpdcIstz53qdWHbvLm0nuHw53HqrN8AOD5cbi4ULZX3FCtFbVoY1a7znnTpVrAGrSlFXg507xRe8bVs4ckTezx07iheAlkdyMrmJq7ij8COceu+/UdaObPZPO0inKe3RGUqLUE4sPoktxY413oqr0EXC9bU5vvAEhkADhScLscSYqT3Ye3OSnpjB0fnH0emg1XMtMYUYix3PbLDwvyunERsYW/X3Q6H4J9C5s3dm7ocffFuEKhQKRQ2gMt5/RSwWydB4CoE8QbfF4g2627cXPXdVg26QjLEHzxdQYaE3GAbJ+M6dK8vDh8NTT3mfmzjRu+xp/hMSAk2blj7X5ZeL9MQzzszM4ueprLMJSHMcT1b93Xcl418VMjPF5QRECtOyJTzzjATdIMF8ZYNugNhYsvp2w2Qu3iXTmefEGGT0GXQXJapbBOmJGWgujfSNGUR1821bGNQgEFeBi9AWoaWCbgCjzkC2/RxnABSKi50dO7xBd1wc9O9/YcejUCj+0ajA+6/Kzz/D998X32azyc8RI0TP3bDhuR27aODraRwzY4Y4jXhYtUqCcZAW7yNHeosmv/nGm/32FFZ27OjbhQXkdT/9JG2aS+7jKQytDDqd11pw376qNx2aNcur437gAZH1vPeerA8eXDUP9TO43M5S24xBRpx5TjRX+ZNJ5kgz1hgLSYtOYI2xYI4w+9zvyLxjRHWPJGtHNrkHfTfocfoYh0KhAD791Ls8cqS3VkShUCguACrw/isyb54Egnl5pQNVvV66NZblwFERmuYNdgMDJUtdWOjtpOlh+XL52amTWOhZrd6gV9Ok8LBkYWV5GAzw/PPw668QFeXdPnFi8RuBirjlFq+V3zvveLfnpcL+pfDnbNg4Q37u+wlyk+V5t9tbVFmnjmS97rlH1kNDpfGQrupuL0a9qZS+OrhREHqjnswtFWehI7tFcPrXFKK6R/p8PnV1GvYMB/VuSaDOkFoc+ewYbh+WkCa976BdofhH43JJnYYH5WaiUCguMOrW/6/G1KlS1AcSZHvkFKGhkvG22cRqMDlZ/L6ryubNXneRtm3l54cfwokTxffzSFoeeMAbkN5/P/z3v2LFN2uWjCM3V57zSEAq4tJLJZPfs6es22xS0PjwwzBhgshpysNqlSz9a6/B9h/g44GQkgj2HDBaz7xfbkAHeiM4C8EcBAEtwH4mSz9mjHSqPHBA1idPlmC8KmgaHDpExPpVOMyFxW5hDQEGag+OF122AUJbhqIz6MjenUPO3lz0Zm+AH9k5AnOEmeBGpQtj7ZkOji88QeMxDdGb9ET3iSI9MYNTP52m9jVeDbjD7SDS6jtwrw7cTgdupwO9yYzeoD4yFH8jli+HJHEFom1br2uUQqFQXCDUt+j54iiEgjRw2cFghoBIMJXRIbE8PN0oX3tN1nU6b9DdoYNkhU+flkx4WprY8504IcFqWRIPXxT1sr3hBgl8Pdluq1X05B5NeViY6Ls9REXB3XeLPKOgoLjDSUUZ76K08dHi/O23Rd4yf36ZloiAvE/9a0EeEAQcWcJZE22X3fdrCmyQvxpGAIVA3TR4cao817+/XFNFpKaK3/f69d5HaioBQPiMy0mLLv47j+sXizHUyMmfTnNo1lH0Fj2B9QKpNTCO7F3ZZ/fTm/WEtgjxecqjXxwjsnM4IU2CAdDpdNQfWZddE/YS0SmcgNpyTovBQoS14rb2lcHtcnLyz5Wc3rGOlF2JZB0/gMtuQ6fXo7ndGK0BhNVtSmyrbsS36UFcmx7oqvL3p1DUJMq7W6FQ/MVQriZVxZ4POxeIjCFpLWQfl4BbpwfNLcFfSB1I6AFNBkLrYZJxLQ+nU7KwH39c+rmRI2H6dK+0ZM8eGDhQbPk8z3/8sTSZqQzNm8PevbJ8+jR8+SU8+KCsP/ywFB96tN5jx8KUKcVff+CAyFM0TVqwFxRIYWVmZtVuAEJCJFtev760lvdk3ENDpUGQL5/d7BPw9W2QtB4cvrXOlcKhg5Ma/BQIa3YWswYE5Jo2by4eZHuy4z549alOrO1RhaLMaqZNdFte7f16xTuWQ0FmKvuWzmPfT3Nwu104C/PLb26k02O0WDFaA2lxzSgaXz4Uc3BY2fsrFDVNbq4US3ske8eOQe3aF3pUCoXiH44KvCtLxiFYPVG0w+jAkVvxa0xBgAbt7oBej0Nk49L75OdLVrlkd0eDQfTPY8eW1h6fPCluJFu2yPqAARJAhxTJnGoa5J4CW47cEJgCwBINwaGSSQ8Lk8C7SRM4flwkHgcPSvdLT+C9axe0aFF6zDfdBF995V2/7DKvJryyNGsmBZIREXIzcfvtsKRIB8aHHpLr90hP9i+F+UNFOlIdhYQu5Ibp5i9Aa1o8yN661Zv194XBINPW3btD7dosK9jCjPa5FFprfgLJYrAwsuVtXN/kxnN6vaZpHPhlAZs+eR3N7cLtKGPWoBwMZit6o4nu979C3e4DKn6BQlETfPqpN8t91VXeBjoKhUJxAVGBd0W4XbDmLVj+ggR87rI7PZaJ3gh6E1z6PPR6EvTSdZCMDLj2Wm9DFw/R0SK58DTS8UVWFlx/vfhgg2isP50Mx3+AwysgeTtoLjkvSPDttMFpJxwDrJ3gintFww0S6E6eLE14HA7JPJfVpKaonzaI1rxkY52K6NsXVq6U5YICydhPmADPPedtdNGpk/iUO3fAlyPAWVC1c1QGB7AI2FHOPg0bQrduEmh36yY2hIsXw7RpsHo1hRYDt8+6EtsFCLxNejOzB35CsNm3XKU8CjKSWTXlUdIP7cRlO//31mAJIL5ND3o89AbmwKqPR6GoVvr3F3cokALLc3AtUigUiupGBd7lkXUM5g2BtH3nJ23wYAqCiEYw4lvIN4lkpKSdXqdO8HVQPGkAACAASURBVPXXpeUPvrDZ4PbbYPuX0BuI1YFJLwF3eWiA3gK5Tljpgm0m2HMQ1q3zSjxq1SpdcFmUjh3Fnxvg8cele2ZVuOUWb4Ogw4e91/vHHzID4CmIahUAN7tAq3omttI4gPnAfqSJTrdu3kC7a1eIiZH99u4V95NZsyC9eLv3/41uzbL+dXEZa07vrEdP7zq9ebzrUxXvXILc5OMse34EtuwMNHcFfy9VGZPRTFB0LfqN/wxrWFTFL1Ao/MHx49L0S9MgOFgaeFWis6+ietE0jdNZhaTl2nC4NEwGHVHBFuLCrOjOwUVKobgYUMWVZZG6Fz7uBQUZFQeylcWRByk74P2O8HkAbC/RPOa220TPHVDJ4syCUzAgCVqaQOcAtMqNVQdoNilOvALobwLXIa/dHkjWuzz69vUG3r/9VrnxFqVo2/iTJ72Bd+/ectw774Rff4DBBXKj4E9MwO0BMHQFtOpaXNrjcIiMZ9o0aRlfBkMXHmT5FQm4avA/yqg3cnPzEVV+XX76aZY9N5zC7AyZCalG3E47uSlJLHvhVq56dQHmINWaW3EBmDvXW6Nw003+D7o1DY6vhWOr4fBvcHIT2LLPzDoaITge6l4C9XpDwysg4hx7MPwNSM2x8e3G46zel8KB5Fzcbg2jwZuQcLrc6HU6GsUGc0nTaIZ0TiAm1FrOERWKiwsVePsi86gE3flpVHvUp7mhMBMGZorkIxMp/HnrLfHJrmwW4M/ZsPhBcNlAdx6aZzNAPnwyACyFZ4JypFhS0yo3ng0b5NG1a+XPWzLwLkp0tGje/9sRCrdW/pjng84Ff46H1meaFh05Ik2FPvrI2+7eg6eItEjnzLg+g7itTj/mpC7BxjnIkaqIxWDhxiZDqRdar0qvc7tdrHj1Pmw5mdUedHvQXE7yU0/yx+SxXP7cxyqzpahZNK14d15/upkUZstn8eo3oTADXA75TC6JLRvS9sKOBRKM1+4qssOmg7zSw785O5Oy+HD5ARIPpoEO7EX6Ddh89B7YkZTFvtM5zF55iC6NIrnnssa0TgivySErFBcEFXiXxO2CuYMl0+2vVKsesAK3AguiYP6XUqBYWf54A1aMq17Ns6sQOgARwOdIcWF+ftmZIk/HSg+TJsHnn1f+fOUF3gBHfgP9fslG1wQuu2jj57wIn2+EH34o7eoRHi4OCUVnAwYMgFdega5dufbzeaw4kcbhesF+lZzo0RMdEMPNzYdXvHMJdi36kNzk49UqL/GF2+kgdd8WDq/8loZ9r/PruRSKYmzeDDt3ynLdutI7oLrRNNg2D76/X25gKytF9Ox3dCWc2gyhCTBsAcT5sFj9m1DocPHe0j18uykJu9NdpW9NT3C+Zl8qGw+lc03HOvx7QHOs5ovjZkSh8IUy4C3JH2+Ig0l1yUvKQg9E6GHGHVULute9A7+N90+hoRloqINhSOY7I8P3fi6XfLmBeH+DyDE8FoeVoazAOz9fdN5zxoh1Y03iyIPl46Vw0hN06/XQqpXcgGRmeoPuXr2ksHXJEvEwHz8e/Yhbef6VDYTmOND76Z5Nh44gczAv9RyPoYqZsuwTh9jx9bRqKaSsDC5bAYkfjacgM7VGzqdQAMW9u2+7rWo2p5UhLwXmDILvRkvjrnOt/7HnQuoemNENlo+TpM/fjAOnc7j57ZV8uykJWxWD7qJoSFb8+81JDHt7JftO5VTnMBWKvxQq8C5K6l74/ZXqKaSsDEY3bJsGKbsqt//xdbDsKXD4MSA1atAI6EnZgfeePRIgg9gaggTjU6dW/jxFA+8VK8THvGNHcVMZ3AcK93kb49Qk8UAk4vc7cKAUW+7cKZlukDH+8IM4slx6KRQWihXiiy8CEOm2MqH2/YRZI9Drqjdro9fpCTaH8EafCcQFxVX59du/fO+c7ALPB7fTyd4fP63Rcyr+wTgcMG+ed/3226v3+JlHYFoHOPRrNX1PaJJEWT0BvrgRnDX7/3k+7EzK4r4P15GcbfMpJTkXbE43KTk2xny0ju3HMqvlmArFXw3lalKUb0bBlk/9n+0uis4AbYbD0Dnl7+cohHebQ9bRmhmXA+gzBwb6sOD65BMpfgSxFrzsMnFYCQqSJhURProoapo4DaxfL+4pq1bB6tW+z90Xup+K5XS+gU0jTxNp9X6oD/g6hh1pJtYOP82kjSF8cyAAU5H0cv1QFz8PTWF7qpGh30fz0w0pNAyT3+fWFBO3/BDF0htTqBviovu8WFILDOh13tff3LSA/3ZqCJ+nybV4aN4cXn4Zhg71ZtBSUqT7p8cOsmFDyZa3bEl6YTqvrnuFI9lHsLkKK3q3K8RisFIrqBbP9XiB2MDYKr/enpvFN/dfisvhQ3/qZ0yBIdw4YxV6Y03phhT/WBYvhmuukeWuXeXzprrIToIPOkN+in/qI4wB0OhKGP7NX173fTA5l3s/XEu+zX/flYFmAx/c052m8cqaVHFxoTTeHgqzYfsXNRt0g5xv11dQ8C4ElFNYsvwFyKvBKXsDsOV5uOrW0gWWHn13WJhY7t1xhxQi5uWJK8tTT4ksIzHRG2ivX1+6SLEk0dFi49d1O8wrpG6Ii2/2B3B3G8ks7Uo3UuAsPpZ/tcvlqa6lpyXbRDu5q3UeT6wMZ8HgNJwaPPp7OI93zqFuiPd3PPOqNPrWKZFlOvGnFL6CuK289JJMWRuL/Lvs3Clf8IcOyXqvXrBw4VnrwUhrJBP6TuTHQ4uZtWMmLrcLp1b1Ilijzoheb2BEi1u5rsn1GM4xi37wt4WgvzBFjprbzfENv1Cv58ALcn7FP4iiMpPqzHa7HDD7SihI81tRMs4CyaT//AwMmOCfc1QDNoeLR+Zs9GvQDZBvd/HYZxuZ/3AfrKa/9o2IQlEVVOB9hu49enA6LZxNI20VZljNBsmQJgS76F+vkAc75BJqlm12F7y2IZRvDwaQbdMRaXVzVYNCxvfMPnvMhfsDmL4tiP2ZRoJNGq1j3Dxc51W6jSrjw9aWC+vf9Y+uuyz0QOEJOLoK6vcu/lxiovzs3FmC8oceksAbYNw4aWHvaUtfFlaruILY7RLAb94MDRrI8V4LBywMbZLPl/u8gfeCvYHc1DSfCYmVs6h7tFMO/b+KYc7uQFLy9QSZNEa1rsT0cAwQFwvPvwD33uvtnulh2TIYNszbYOi22+DDD0vtp9fpGdzoWrrX6sm3Bxax5PCPaEBhJX6PAcYANE3jynr9ua7J9cQHnV9L+uPrf8Zlk8z7fzfkMKxpAM3Cvf/++zOdfLA9H5NefgWhZj2XJ5jpFmc+u8/jf2TL82evD17p6f1dpBa4eX1jLr1qmbihsdcS01mYR9LG5SrwVviXzExYtEiWjUbpB1Bd/PYyZB+rnq655eHIl8/61jdDnS7+Pdc58v4v+8jMrxlJTFaBg/8t28ujV7eskfMpFDWBCrw9OPOrlGEtdMLuDBOvrAvl+m+j+f66VAJNGu/+GczWFBOLr0shLtDN8VwDa096g5cPtgbx3pZgXu+dxWUJNkwGjeXHLCxZ8lPZgffWOaC7AHJ8zS42WUUDb4cDNm2S5cxM6NHDW2gJ0oWyZNCt00Hr1t7GNN26QZs24tm9fj3k5EizC50Ock9LW3gsdIqz89X+QPZlGGkU5mTRwQAWXZta6cDbYoBJfbO4Y0kkbg0WX59auaSvyQQbf4U6rUs/N22a3Gh4umuOHy/dNsuxzIsOiObuNvdwe8s7WHNyNZuTN7E7fRen8k5j0OnR6XRomoZLcxEbGEuLyJZ0iO3IJbV7YTFYyjxuVcg8uqfCfULNOp7vFoKmaezOcDJzZwENQgzEBnqzTY91DCY6wPffYmKynUCjjj9TnAxpqGEo8man7vvz/C9CoSiPL78UyRvA1Vd7G1+dL6e3weqJNZf4cBbAgmHw0G4wVs//f3Wx52Q2CzccqzZNd0XYHG4WbTzOoPa1aVknrEbOqVD4GxV4e3DaGdqs8hlWqxE6xDiYNSCdPvNj+WJvAKNa5/NnipmBDQqJD5IPprohLuqGyAd2tl3HxI0hTL40k6sbenW/A+rbGBDow1LvDN3veIXTuaFsGlng12x8Uc2zSQ+d4+y8XrCMOtbPYNNuCZLXrJGCQvAG4CUxGqXBTvPm0KyZZLKNRsluOxzSrXPTJu9x3G549lnJgttPgsmrufZkvXvUstM03El8UPHpzQ+2BTNrp9fycED9QqZe5i3KaR7pwKDTaBbhokl46WzVPUsjKer891z3bEa2N4GuhBba5ZIOnVOmyLrFIl7Bt9zi+z3wgclgom/CpfRNEHszp9tJniMPh9uBSW8i0BSISV/9Ouj89GRcVSiq1Ol0tIw0EWAq5GS+u1jgXRaaprEx2cHVDSz8cNjGrgwnbaK815KXcgKXw47BZC7nKArFeVBUZlKd3t0/P30mGVCD5KfCjvnQvpqLQ8+TWb8fxO6qmaDbg93lZubvB5gwolONnleh8Bcq8AYokGYi55JhDTZr9KljY90pC6Na59Mp1s70bUGYDBrd4+20iHCeTYZuPG3G5tIxqIGPD3FbNqQeBWOoBKieR146uJ3UDdH8no0Hr+a50An/tyqc51dY+fib26AC5UgxnE749Vd5VJY33pCfMcDd3s03NS3gxu+jOJpj5Kampd1cxrT1rfH2MH5tKD1r2dmSamLRASvXNS7+3n80IL20xpswcBfx6s7JgVtvhe/PNNeJjZUp7R49Kn99PjDqjYRZ/J/FyU87icFkrrSjiVvT2JXuJN+hEW2t3EzLgSwXOQ6NDtEmknJdJJ52FAu89UYThVlpBEXXKucoCsU5cvCgOA2B+O17CizPl+wTcPAX/N8+twT2XLG2/QsF3pl5dlbtSSnV3sDfaBqs3Z9Geq6NyOC/1gyAQnEuKDtBgNyTZ6UCngzr70kWnxlWX8QHuci0yVv57w65PNg+l4X7A7l6YQyd58Yxf6/oXTNseiKtbnz2VsmzQ+v64ggSFyeNHxo3hmu7gqadHZcHTzbeF0Wz8RmFer44c/6i2XidTrLxw5r5nj61GmFwwwL2ZpugdoVvQfXhopiNYEKIi3ohLn49ZuFqXzcs5fB7kpmlR6y80SeL13pl8cKaMDIKK6M10cB4xp/82DHo08cbdLdpI8Wi5xl01yRuh43KeDNm2zWeW5PNM6tzmLWrgGsbWqgTXDzb/dafuTy3Jpvn1mTzzQHv7yMx2UHLCCNWo46OMSZ2ZTjJc3gzYzqd7sw4FAo/MKeIK9Qtt5SuyzhXEqdVvptwdZN5CE5urni/GmLxn0kX7K0A+H5z0oU7uUJRjaiMN4DTGxBUlGH1xck8A+EWCTIMerirdT53tc6nwAlf7Anksd/D6RjjIMLiJr1Qj9ON7+Db14x+LSAdv2fjS1Lg1PHtwQA6xdqhfqXehvMnNBQiwsFU3DJxYt9Msmx6Ak0alZUW5jt0PLUynJd6ZhNpdXNlPRt969h4aW1YMSmKTxwFEFYPNmyAIUO8biwDB8IXX8g4/0boDCYqk7HzaLydbo3Fh23sz3LRt07xfR7pUFrjbXNpbEtzMLyp3OA1DDUQYtaxOcVJ79oyo6Jp2plxKBTVjKbBp0W84qtTZrLzy7Myk5u+j2JnmonNt53Ccuaz+j8rwiuU+H2xN4DHfw/Haij+P7jy5mTig9x0nxfLm30zz868LTpg5Zk/wvnoqix67v0BanWs0pDfeecd1q1bx5wiNyO9evWiYcOGpbY9+eSTWK1WJk6cyNGjRzGZTLRs2ZJJkyZRr149AI4cOcLzzz/PbytX4dYZCWrWm/DuNwPgzEklY9Wn2JIPoDMYCWzYhfAeI9BVYIeYvnIWtlN7cGYlE9l3FEHNihfwO7OTyVgzF9upPej0JoKa9WZNnTHc0acRM2fOZP78+ezevZvrrruOKR75H7Bx40befPNNtm3bhl6vp2fPnrz88svExUnfg0mTJvH2229jNntnen/++Wfq16+pLzmFQgXegsH7T1g0wzqpb8UG/nkOHX8kWXi4Y2m5Q4BRgvCJG0PYm2mkTx0bZoPGT4etXNPIR/bWV3I9DEiXxYr0zr6ID3KxNVWu798dcgm3uFm4P5Bxa8KIsLp5ums2NxfJens0z/kOHVEBbj4blFa1WVadztv1MTpaujpGRvp+7NoFTzwh+7700tkmNLxVH/C+Pw1CXWW8OfD+1mA+3O7VeFsMGtvvOM3rG0JoEu7kxibeaxvXM5vLFsTw+3ELfRPkZmvUkqhiPt5969j46AYT/PCr2JEVnHn9Qw/BW28VtxT8m2AJDkNzV16XadTrGNzAwhsbc9meVlwy4ottqQ5sLvhyfyFfn8mCFzg1EpPtZwNvt9OBOfjvdcOi+Juwdi3s3y/LjRtDz57Vc1yXE9LluMdyDKw7ZSbUrLH0iJVri3x+VyTxA+gca+ebIWkVnnL+3gDGrQ1j9sA0usY54MgK4NkqDbtHjx689957uFwuDAYDp0+fxul0sn379mLbDh8+TFxcHHfddRczZsygd+/e5OXl8dtvv2EwSOBst9sZMWIEd955JyeaDien0I0jy2sLm7HqU/QBodS59S3c9nxSfpxI7s5fCWnTv9wxmqLqEtioG5kbFpR6TnM5Sf5xEiGtriDqin+h0+lxZJ1i32n5jo2Li2Ps2LGsWLGCwsLi36NZWVmMHDmSyy67DKPRyLPPPsujjz7KZ599dnafIUOG8M4771TpPVUoqpO/XxThD4JiKCpcq0yG1eaCPekm/rs+lDCLm1vOBK8ztgXROspBx1g7Jj18vT+APIeeNlEOQs0aj3fO4dnVYRj1cGmCDaNeY2WShdVJFp5r1hAibaIp9jyM3nH5MxvfNEIKDz2aZ5cblhyxctP30Sy/JpPYa/tLUP3tt3Lge+6Bfv1KB9M6nXhfZ2WJ68mMGZCQ4Htw8UUs8pKTvct1urFuxJc+X2LUQ9J9JwCYclkmU8rIXo+/JLvUtkirm623nz67vm5Ecql9AMhvBjfdJMt6vXTkfOgh3/v+DQiuVR+301Fsm8ut4XB7/7ZK/pkb9TourWNh2VFbhYF3YrKD7nEmrqrvnd7PtGm8syWP0/ku4gINmIPDMAeqRhgKP1DSu7u69BApO0VyZnfw5b4AOsU66BhjZ8HewGKBt4eyCu7LwqbTczAglAKDka9jG/F+MqzfkMMNo8I4GhmApSCLxsnbCCrzCL5p3749DoeDHTt20K5dO9atW8cll1zC0aNHi21r0KABqamp1K1blz59+gAQHBzM4MGDzx5r/vz5xMXFcctto5gzaQU6o4Y5qu7Z5505qYS3vhKd0YTBGIY1oS2OzBMVjjGk1ZUA6DaV/mzJ2/cHhsBwQtpedXabOaoudoebtBwbV199NQBbtmzh5MnipgRXXHFFsfVRo0YxdOjQCsejUNQkKvAGCbyL2PVVlGH9aEcQmibTiv3qFfJQh9yzmY0Ao8b4daEczjKi00GjMCfT+6VTP1SOd3+7PGID3UzdHMxDy8MJNmm0jXbwcDcN3i5huaZp8N0D8MTXgH+z8Z7A24NBD1c3LOSpPzTWO2txzbffwqxZ3sD7//4PGjXyfeIxY2DCBCmyfOcdb+FkSYq2jS/6AdryRtj/kxQY1TRuI3x/ppI0JESkJYMG1fw4qhG93kBIrQZkHdt3dttHO4tr+xuElp4a7hZnYulRGzvSHLQuI/jOKHRzIMvF9Y2thJq9/0OhZmgabiAx2cHgBgYiG7aqpqtRKIpgs8n/qIfqbJqTcfBsEP/lvkDua5tLpxgH1y6KJiVfT0yg76xMSYlfUTRga3AUX8c2YmtwFBbNRa5hL8v+1JF7II+mY5uxNyGAg24XZrcbu95Aw4+HMHRvAN1PmDC4ke8Ft1t+lny43Zg1jY4mE2vvu492cXGsO3qU7lYrcTYba+++m3bR0axLSqK7203bCRM4cPAgL7Zpw4CgIDpYLATpdGePvyklhQRN4+6O7TnidGGMSCD8kpGYIyWZEtKmP/kH1mGp1Ry3LZ+C49sI63z9eb3t9uSDGEOiSflpMvaUw5gi6hB+yUiCatUnPc9OVEjl9ftr166lWbNmxbYtW7aM1q1bExsby1133cWdni7MCkUNoQLvM6x7ugUcXl5qe2UzrB5ua5nPbS3Lz0bf2KSgmAQCgHq9S++o00FkA4oWxvkrG18STYOlR6xk2fQ0rRslGz2NcyIipD16WTz8MEyeLIH3Bx+Iz3WIj2xnbKxXmlIy8P7+/rKP70+cTtiGZO2//16KKS8C4tr2IPvEITSXk2e7Vi7zbDboGN/Du+/E3qWlIhFWPW/62A4wuo3k6gxmC/Htep3DqBWKCli8GDIyZLl377KTAeeCsxA0jfWnzBzPMTCkUSGRVjf1Q50sPBDA6LZlN+MqKvED2JRsptkntbDpDGg6MASl0HZcDPno0YDs3TmENA0moLYUdTv1BpxndNL7Il1MbZeBobXGfR/u5LLfT1RYKt0zJIR1ycmMXr+edTEx3JOeTpzRyJzAQEZv3Mi6mBhG5+ZSf+dOFphMTM/P536zmTy9niEFBfw3K4sgTeNkZCSrLRZe1ixkjnyDUwdWk7rsbWrd9Co6gxFLfDNyd/9G0uwHQXMT2LQXAfXPz/bPmZeB7cRuogf8G2vtVuTsWEbqsrcJHzkBZxWsDHfu3MmUKVP4+OOPz2679tprGTlyJDExMWzatInRo0cTFhbG9def382CQlEVlKuJh6aDwBhQ8X7+wAnsdBRvROOhVudiU6cNQl20jykdKINk45vNiqfNJ/GMXRFOu2g7i4aklsrGd5wTT9tP45m9M6hYNh5E89x0ZjzNZ8fzRmIIUy7LonmXfvKkp1V8587lT+fWqSP2eyCSk48+8r2f0ehtclE08DZaoPNofFeb+hEXsAVo0ER0oxdJ0A3QtN9w9IYL03ZZ0zQa9hlyQc6tuMjxl3c3gN4I6FiwN4C+Cd6Oxjc0KWDB3sByX1pU4mfX6Ymva6L9xLa0n9SODhPb0XZc8RmgerckYEu2ceSzY2g+/PoKAk3khpj537/a8OJL3ciIDgKzWdxbrFYICICgIAgOhpAQuhuNrLdayQgPJ91opFF4OF1CQki0WsmIjWWP0UiP8HCoXZvOsbF8EBTENrOZr41G1gUG8nZCAjRqhDUwkK4GA33q1AWzhZC2A3EX5uHIPIGmuUlZ8haBDTqRcNf71L7tbdy2PLLWl9ZtVwWdwYQlvikBdduhMxjPntOWcQKTT1eC0hw6dIjbb7+dcePG0b1797PbmzVrRnx8PAaDga5du3LPPfewePHi8xqvQlFVVMbbQ/s7YfkLF+7809bBpE7QqRPcdx+MGCGt1Gt3Zt2tacV9pc9Q3dl4n5pnmw4cMZIJ/vOMFKZz54qv57HHvF+Kb70lGmlfhYm1aom++9QpyXzrdLBjBzw8D25wQU3atrqAP4Ds/aLxnjxZumxeBITWaURY3aakH9hesyfW6and6TIsoRE1e17FxU9qqmS8QQLQYcOq9/jWMApcOr47GIBLgw5zxBnD7tKRZdezI83312dRiV+KycqcWs3JPpqNTV/2160p1ETTh5uwd8o+jn5+nPoj6vrcz2Y1sr1DPA/Mqs8rvV6jcXhjn/t1Liggp2VL5j77LF22bIHp0wkB4gYMYO511xE3cyb1PDOYRegADBo/nt0HDsDs2bScMIENGzYQ9clc7BOWQ5GMs9uWhys3jeDWV6IzmDAYxH0ka+PCs64n54I5MgHb6f2ltjtcLqIq4eN9/Phxhg8fztixY7nJU6tTBp6uwQpFTaIy3h6CY6HJQCrjd1ytuIGDgGfWctMm+Ne/JCC96y7YuANCatJIuwR6De5+AWbO9Haa7NKl4te1awcDBsjy0aPSztkXHp23zQZpaXD33dC2LexJgsVA5Rsunh92YEM4eGoyV62C7t1h5EgZ/0VAm6EPYLDU7KyOwWSm1XX31eg5Ff8QPv9cEgIgtp/h4dV7/Lj2LNnnxqCHFcNSWHqjPFYMS6Z7vI0v9xXPettcsDXFxN1LIwmzuLmytZtHm/UiT2/EXYnvFXO4iWYPNyF7Vw7Hvizbs9qpScfb//vjKQ5klg5QAQICAmjXrh3Tp08vlvHt1q0b06dPp8eZPgTr16/ns88+IzU1FYD9+/ezdOlSOnUSuciNN97Ipk2b+HPDGoLMenK3L0NvDcYUXhuDNQRDSDS5u5ajuV24bfnk71t1Vv9dHprLieZ0gAaa24XmdKBpEtQHNumJPfkghUk70Nzus+cMi61HRJAZp9NJYWEhbrcbl8tFYWEhzjN/BydPnuTmm29m1KhR3OFjBmTJkiVkZmaiaRqbN2/m448/ZoDne0qhqCFU4F2Uvs95G6fUFC7gdx/bCwqkJXmfPvBrPnABPJA14ChwOk9uBjxUJuMNkvX2MHEiPlueFS2wbNxYAnzPfqeiIag5uPx8M+QEQpvAkjS5QSiqE507F5o3l5b2OWV3yPw7UKfz5cS26orOWDN/SwazhQZ9hxDV+OKR7Cj+QvjLu9tDcCwL9gVyc7N86gS7iA10n33c1SqPhfslE+5L4vfF9Rm81KonOQYTbp2OvEN5bH5ka7FH3pHSs4/mSDPNHm5MxuZMkhaV7w5S4Czg2T+eISU/xefzPXv2JDU1lW5FZu26detGamrq2WA8NDSUpUuXcuWVV9K0aVNGjhzJoEGDeOCBBwBo0qQJ77zzDs888wy7Z9xPwZHNxAx4GJ1BsvfR/R6i8Nh2kuaM5eT8p0FvILz78Arf2pQfJ3F81hjsyfvJ+GM2x2eNwXZSitpN4bWIvOw+MlZ9StKnD509Z4sEmTWbOnUqjRs35t133+Xrr7+mcePGTJ06FYB58+Zx5MgRJk2aRNOmTc8+PCxatIhevXrRrFkzxo4dywMPPMDNN597dl6hOBd0mppnKc4P/4ZNH4HTd0fHasUYAM2GQWIcvPcewqdgDwAAIABJREFU5JchA7EAj1HzsbcNsNwBExZ4/awDAyE3t3KWXZoG7dvDtm2yvmIFXHpp8X2eeEKC8qLodDJtnJEBvy+De4Fw/COMcgFB8fDwdgg8U0Rqs8G778LLL4tG3UNcnGy7+264QHrp86UwK43vxg7EWeB/xxhreAzXTv0Jo7V8PaxCUWV274aWLWU5JgaSksDkhw/IOYPEYamKvF23Lb9H1MZeQSOZ88WgM9A8ojmv9ZmAzs9tJResO8J7y/ZS6Kh8gWN1YTHpuf+Kpoy4pEGNn1uhqG5Uxrsk/SdAQGTNnMsaDtf/T6z3Dh+Gp5+W4pii6HSgmeFPwHdNpV9wa5ButPB6GytTvh7LN9c1YnvrSPI1u4yzMvdrOh08/rh3vWSAPXMmFOk6BoibSL9+MH8+LFsmwf9MIAPJTFcnLiCsLjy4yRt0g+hFH3tMmnI89JA3yD59GkaPho4dZWx/Q6xhUfT+z1sYzP6d2TFYAuj7xHsq6Fb4h6LZ7hEj/BN0A3R9EMxV85/fGhzJyvBafg+6AVyaiwNZB1h6ZInfzzWwXW3cFyhNp2kwuGOdindUKP4GqIy3L05tgY97+9dH2hQEo36H2iWsl9LSpBjx7beLSxvMwFg9lOEdW93YdHqeaHYJhwPEKs5kd2FyuLFZDDQ6lM3Q9Lp0e2kmxoo0w3a7WA+eODNtunOnaDEHDYItW4rvW6eO7Of5k9TrRYqSlCTX/3RLMB8BR+WaB5WLSw9N+sOwuRBYwY3W7t3w5JPw3XfFt199Nbz5JrT6+3lUH1nzI+v+9wwuu63aj22wWOn75PvEt+lR7cdWKHC75TPFU3uRmFh5+VuVz+WCifGQn1qp3TXgvpaXkWyp2RtOq8HKp4M+w+JnqeRLX21l6baTFQbgefvXkPHHJ6W2G4KjqHXTK1U6p14HV7aO5+Vh7av0OoXir4oKvMvi2Br4dIB/gm9TENz2E9T34d3tIT1dOiZOneqVO9QHbkWCUD9SqNPzTWxD5tZqXuY+AflO9HoDd3e8n37Nri1/mvONNyRLDtChg0hPXOW0u9fpYPhwkbd8841su+su+PhjmPM8bP+vOA2ey/tgB4xmuGk2tKtYi1iMX36RTHjRGwaDQbLg48Z5rRH/JpzY/BurpjyKy2FHc53/dILeaMZoCeDSp98nulnHahihQuGDFSvg8stluVUr2L69+rpV+uKPN+C38ZW64d8eFMH4Rl0pNNSsYZjVYOXetqMZ0OCqinc+Dw6n5HLntDXYymoi4QcsRj0zx/SkUWxwxTsrFH8DlNSkLOr2hFErITi++vy9HYA1Gkb9Vn7QDdJ+fdw4kaCMGydZ4iPAKvzq9GHT6dkXGM4X8U3L3a8g0EieVceMLdN4dvljpBWUkxEaM0Z8ZkEsCT1Bt8lUWis9ZIjs07ixN+geMACmT5dM+OTFMAVYBhijwBwM+gqmmV2IZCUbCLkOnjlV9aAb4Morxcv8o4+87e5dLnj/fWjSRCRDhaVbSf9Vqd3xUga/9SMxzTudt9uJwWylTpfLufbtJSroVviXkt7dftY20+MR+R6oBAtjG2GrAYlJSQpdhXy170u/W+M1iAnm9t4NsZpqJnSwmvTcekkDFXQrLipUxrsi7Hmw5FHY8un5FVzage2AfhAsXFz1L4usLCn4mzwJOmZAN6o9823T6TkQEMoLTbpXSZ9ocLoxmwMZ1/u/tIhsUfxJpxPuvVccWoq9yFA6692xo9gpTp8uwTpIhvz336Xz5ddfw9Chsr15c8mcn1wPh1bAkRVwcjPYssDtlOYXTgMctskNy2krTPhCAvvqIDdXAu2JE72FpwANGsDrr8PNN/s/IKgmNE3j6Oof2Dr/HQoyknEVFlTKVVOnN6A3mgip3ZD2w/9D7Y59/T9YxT+b/HwpcvYUeB89CgkV29edNyc2ifywnO+AbvPiOFFgpN2rrTEGezPeO1/bQ8HxAtqMb8nJxadIT8xEZ/D+g1lizLT6vxbkH8tnz5T9tHyqOdZY8avOO5rPvrf30/KZ5liiLGx7fgeOLGeZ51j0yzd0adHVD2+AF6fLzcj/reZoWl6lSn3OFZ0OEiICmfdQL4wGlSNUXDyowLuyHF0NK16EIysBDVyVSDsbzKDTQ91LYLkGs8+0pP/6a7jhhnMbR06OOKAsexl65Ivkoho+kwp1etaHxTGlXruzrYqrilVn4qXer9Iq6ozmeckSuOWW4s4gJenRA7ZvBUs+NK0HY8fCI49Dqga160kHyVq1RNfZvr1MK4PY/I0Y4fuY6ekS+P7yi6wnJIg+u0OHc7qucjl+XKwGPymhZ+zZUxrw9Pj76Jw1TSNtz2b23ncDyeEmCgNMGAOD0dBEvKrToQOc9kICImKIb3sJza++g/B6zS700BX/FObOFW99kBmon3+uuXOvfB1+f7lMyUmnz+PJDrASfWkMsZeJ7KwgqYADHx7Glmw7G3ibIszUubaWz2MkLTpB7qF8mo1tDG7Y9cYeontGEXu5HG/b8zvQGfXEXhrt8xxvL5zK0G7lN42pDk5k5HPXB2vJKXDgrwAixGrk49E9qBsV5KczKBQXBhV4V5WsY5D4Aez/EVJ2yW15UamD2wmaC6JbQpOroMv9EF5fWqK3aAHZ2VJEuHMnhIae+zhyc+G9lyBpCgS7zjn7bdfpsekNvFWvHYlhcec+njNYMfFWzwnUGfmAN/D1RcdouLsTcBRO7/Y6lnj+Gk1ASC1o0Bva3gpb8mHEmS/c1q1FZ+3L0u/AARg8GPbskfUuXeDbb4v7hfuDxETRf/9ewpR9+HB47TXJhP8d2LDhbLdOx0svkDHsGgqz03E7HRhMZgIiYgiv36LiolqFwh8MGgQ/nbH3mz3bP/7d5bH0CdjwP5/Bd7svahPYJ5a0bTm0fErqY45/nYQh0MCJ705VKvB2O9zsfG0PcZfH4Mh2kr07h+aPNEGnlwz5tud3EH1JFJlbs3yeY+ysh3my/1N+uvjiHErJZcxH68ktdFSr24lOB8EWI9Pu7kbjuKo5yigUfwdUy/iqElYXrnxFHm43ZByE3FPgsoHBAsFxENFYHDmKUquWBGAPPiguHc8/L4WT50pwMDw1EXKeh3fvg5SvwOqW32glMuD5egNunY4fouqzMLYRedXUVMXmsvHGvHt4a/kflAqLdUB7oA8QkgZpP4Pmlqy9ryR7/knYuUB8dHPyoC+wFnjpJd9B98qVMpOQlibrN94otmOBNeAw0KWLFH198414kx84INs//xwWLoRHHoFnnjm/m62aYPnys4umfgOIbeXfaWuFotKcPAlLl8pyYKD8f9c0/SeAJQxWvlpKduLU67E2CsG1IYuCU4VYYy2kb8yk+aNNOfHdqUodXm/S02BkXfa/fxBNg5ZPNDsbdHsIahhE2voMn+c4nH2o2i61IhrGBDNrTA8e+2wTJzMLqsXf22rSExcWwKSRnUiIVFakiosTFXifD3o9RDWRR2UYM0YkCevWiV779tsr1369PELC4Jn5on1sFQLt3FAP+H/2zjM8qnJrw/f09A6EXkMXEJEm0pFjARHEXhD1nKPid+z9WLF3UQ9WxE6xIFZUEBAE6R1C7wmQ3jNlfz9WNnsmmfRJg/e+rlyZmczs2ZNkZj97vc96VjRSRdbAbTFT4DBj97hJt9nZGRLFkuhmrIhsgtsUWO+cZjFztEkI88a0Zfw8r4NA93i43AwFSWDxyI5VdLGlMEuGCA0CBlqgp59Gm08+gRtvBGdR2PkDD8DTT5c8AapJTCYR/hdeKHagJ5+E9HQZyPPcc9KU+dRTsp/WevrWW7hQvoeGwtlKdCtqGY8Hco6JqDWZJUNbj/v8/HP5OYjoLj7zoDYwmWDII9BuBMy5DHJTTgpwXXbG9o0mdWUqYR3CCIp3YI/yLWok/3aM44uNaZORPSJpe13rk9eDmwWD2URwkyCC4v3HA5b2HAWu2m3ubhYdwie3DOTjP/cyc8keXG4P7ipUvy1mE1aziWvOacukwe2Up1txSlNPj/6nKBYLvPOOZM663SLEV64MjAhLTYX9HmkktNshyAb2HHDA27d1ZXerSI4GhZJXCzFXBcFWPr+yI6N+P0R4SDTcdQ44fwZXPliqsSZpA3DD7AnQdSKMfQ9MFnj0URHZIEkp77wDN9wQiJdSNex2qXBfd50I7bfekibT48fh3/+GadPg5ZdhdM1Gf1WawkL480+5PGiQvA6FoibRNNi/BLbMgQN/wontIm5NRStaHifYQqBJT1i5XSbYplP7FpPitBwAtyfCH0/Aqrd8fhTTN4bEV3dSkFJIbN+SMwKajGxcqtUE4ODXhwlPCCN3fy6pq9OI6RNd4j6lPUdd+EatFjOTh7RnZLd4Plm2lwVrDmB2FpJnL9+OFmQzo2kw6ox4rh3UjtZxys+tOPVRp5W1Tc+ecNddcnntWql8B4LnnzcuDxoEQaFwAg46Yljcrjl7QiNrRXTrmDSNX0e0gMtjwPVLUVUoQIcFZy5smQ0zR8FVlxmiOyZGJkrWpej2JjZWJnNu2QIXX2zcvmUL/OMf8qU3i9YHVq2CnBy5PHx43e6L4tSmMBdWTIPX2sDnF4lvOnmDWPZc+eDMkS93IeSnw/7F0DEZbgX+aYdWBRVfMaspbMEw6jm49zhc+DbmotVDR6wde5ydjC2ZRPWKrNQmM7dnkbExk9ZXtqTVFS05OPcwrpySGfulPYfdUncny63iQnl4bDd+/PJu/vPLOwzJ2EujcAdWzUNwQR4hBbkE28xYzSbiwh0M7tyI/4zuzI/3DuORcWco0a04bVAV77rgscdkJPr+/fDIIxKR17Jl9bb59dfG5YEDT1oGvnvxOlymndR2LaQgyMq8S9sybutCzDUx6t6VB3uXgFUT73iHBPjhB0goO3+8TujYUbzfixbJSdf69XL7L7/IicLNN0tWe5PqN7dWCy9/98kBJQpFoDmwTGwa+emVm0KrH62aFsLcK6DVOTBuJoRXLGO7xrAFQc9rsTpexFpUqW9zdStcuW4sDgtaBb0X7gI3+z8/SIsJzbCGWYnsHkFE53AOfnXYx4qi4+85Wke0CdjLqhIbNxK6czvj2M64O6+Ca4eSe/O/SJ87j0J7EPZtW4gMdRDqUNJDcfqiKt51QWioUenOyYH/+7/qbU/TpPEIZJm2KFlDa9GCxbYDeEwaqavS2Pb8DtbduZEND25m51u7yd6VzZEfjrLmtvUkLzrus8nkRcdZc9t6jvwg281KzGLjw1tKPPXRn5JIfH1XidudmU5+f3A7Cz1R1XttZWHRxM9+TVuJHayPotubYcMk/WTGDCNlxeMRa0xCgvjA63IAjy68IyMlU12hCCRuF/x4O3w8CrKOVE50F8eZA3sXwbSOsPWrwO1jZcnMlNSk22/HmpKGJa8AAEcjB6Gt/TcHJv96jHV3bjz5tf6+TQAc+e4oQU0cPtaRlpc2J3NLFpnbskpsx99ztI9sH6hXVjX0ApDVChddBEBIVgbN0pNpU5BOs5hQJboVpz3qHVBXXHQRXHopzJ0r1dBvv4Vx46q2rYULjWXXuLiTwvvYlOvxsJXk34+RtOAYra5sQUSXcMxWMxlbM0nfmIHZYcbR2EHqylSaDDNGnqesTMVRNMShLGL6xnDkxyQK0wqxRxvLnKmr0whtGUxhcyukpFftdVUEO9AxCVxJQEk/Zb3DYoFJk2DiRHjxRfnKzZV89gcfhOnTRYBffnntDuDJz4dly+Ty4MH1t/lT0TBxFcKX48QyUp1BZN54nFDohK+vlYbMs28JzHbLwuUSS9avv8rXihVyG/CX3cwV3btTbCwYJouJs96SGQJtrmtNGz/Va4CWE0sOArKGWen5fPeT1894qpvfx5osJgZO78eArgOq8KICyDffyPdhwyC6yJuenS3f66IZVqGoh6iKd13y+usykRHg9ttFfFUFb3+3PsrcamX3Rf3R8j0c+T6JVpe3ILpXFBaHBZPFRNQZkbQY3xyA0NYheAo95B2RA2LekTw0p6fUio03jlg74R3CSFmZ5nN7yt9pRPeLYUtYLYhhVz7MvbzuPZ+VITRUYhETE0WI6yJ7/34ZDDRwIPz1V+3tz4oVkr4CymaiCCweN8yZCPv+qF6VuzRcebDgHlj3UeC3rWmwcye8/bYkFsXFyXvzscekEdll+K8dzVvTLT24bjocgQh7BC3CqmlZrA67dsk0YfAdEKcL73CVya1QgBLedUuzZvDMM3L50CFJ56gK3gJtb1GE3/jx7LdkkpKYisflIapn2U0+MX1jSPlbxHPKyjRi/HTjl0ZsvxhS/049eT3vSB75R/OJ6RPNnuDayK3WIG2vpCI0NJo3F+vJ6tUwdKhx+4oVcoC//HLjb1qT6DGCoBorFYFl2Quw57fAVbr94cyFH26FpI3V31ZKivTg3HwztG0rPRq33Sarkt5TeKOiJNbwf/8T0blnD+PHPEyQzX8EYE3isDi4pMMETLW5SlYcvdptMvmu3uoFJVXxVigAJbzrnltuMfKS33hDkk4qQ06OUVGwWIzLt9xCTmE2zhwn1lArJkvZH8ixfaNJXZ2G5tZIXZNGbN+SEValEdUrksI0Jzn7JBEj5e80IrtHYA21UlDF8fOVxpkLy16sneeqCXr3FvH77be+XvXZs2Xi6f33+x70A43u746NhTPOqLnnUZxeHN8Gi0sfsx5QXPlSWXdXspu7oED+/x98UOYqNGokJ7zvvy8rUDpWq9iwnnpKToxPnICvvpKI0Pbire7V+EyCLbU/1VXTNIa3GlHrz+uDLrz79/edFKysJgqFD8rIWddYLPDuu/KB73bDP/8p2d7ekxnz0uHoWjiyGlJ3gjNPxtSHxcPqwxAJZCAZ1m43dOkCQ4agbdqBNdSKK8eF5tbKFN/2GDtBjRwcnneEoEYOH792uS8hyELUmZGkrEwjpFUIqavSaHVZC3a8tpMNh/IouBIcXi9n3TEbr6wNZ3WyHbMJ2kS4uK5LDpd3ymP5ETuX/RBLsFXDZIImIR6m9Mzi8k5Gtaz5e80ItnowAQ6LxuAWBTx7TgaRe36FnBMQGlfx3399wmSS2MHzzxev9+OPQ1qa5Gu/8AJ8+KEM5bn55sB6sHNy5H8OpOpem0OHFKcumgZzLhdBXDtPCJmHYOlzMPS/Ze/X5s2GT3vJEumz8EfXrjBqlHwNGVKueDSbzNzaawovrX6BAndBNV5LxXFYgriy81WE2Opw0uORI8bKq7fNBJTwViiKoYR3faBXL7jjDhmqsmaNDFy55WbY/KUs06bulsxYV77k2p7EBIUmuA1wA2vy4W+kim4yEWILIaxdKGarmfQNGUT3LjthJKZvNPs/O0iba1pV+iXE9othzwf7iOwegebyENTUQfauHGxBZhbsD2JMOzn4rk62ceWPsdxxZjavD00n2uFh0wkbb20IOymum4R6WHNVMpoGCw86uGFBDGc1KaRDlNG29Ov447SNdJNVaOLfv0fz8tpwnhxqgsMroeOFld7/eoXdLkk311wDU6dKAo7TKRW2W2+VATwvvSQCPRBLy8uWGRM/lb9bESj2LYb0vdSq6dmZC8tfhEH3gdWrOfzIEfjtNxHav/0GSaWMcG/cGEaOhPPOk+/Nm1d6F/o17U/vJn1YnfQ3Tk9NZKkamDDRLLQpF3eoYmN+oJg3z7hcmvBWHm+FAlDCu/7w+OMwZw4cPADf3QOZD0rlsbDoQ6vA3we4Bvaig5oN6Ff0FT4fMifQIqwlYWFhNLswngOzD2GyQESXCEwWE5nbs8hKzMZsN4RbzFnR2KPthLUrfZCBx+nxuW6ymjCZTIR3DMNiN3Pgy4PE9IkmdXU6oW1D6dBUY05iyEnhPXVlBBM75nFbr+yT2+jRyMk7I32bM0E05YhWBUQ5PGxLtfkIb51wu8ao1vn8si8ICjPh8KqGL7x1YmLglVfkROr++42l3G3bZCz9qFFyslZda4jydytqgmUvQGFOHTyxBms/hRNNjar2lpJRqAAEBYl9RK9qn3FGQFZ8pvS6nVt+mozLU4hmrjnftcPi4L6+D2Ix1ZKlrzT0GMEzzoAOHYzbNU1VvBWKYqg15fpCWBi8eC/cAvR1givXEN0VxVr0dWARvNmJMw6tRdM8NBnZmBbjm3H052Q23L+ZjQ9v4djiEyUaLs12MxGdwzHb/f9bONOdrLtjo89XwXGpwJtMJmL6xVCY6iSmXwypK1NpdFYk53csZPEhB8dzzeS5TKw5ZufCthVrsvJosGC/g9R8M20jSk5vA0gvMPHLviB6N3aC5pK4slONhAQ5sP3xh3jBdX79VVZL/vnP0it4FUH3d8fHi59coagm/fr2oc2D60nN9xWd533diObvNeNgloVLv4/l8+2+9ojlR+yc9bkMkjqaY6brzHj+TjJsb4ez5ba1x2wAXPp9LO0+bErCjHg6fxTP+PmxbEvKhxk3ycmpPjlWx2SS99D990vlOy1NBlndc49MFQ6QzSp8zWaeu28JIbkuTJ6aqfg7LA4eHfA4zcMqX5UPKGlp8tkE0mzqTV6ezCoAJbwViiJUxbu+sHk2JN4HccgkxurgcUFhNtELH+U/oZG82LoHsX1jfAYz6JRV3W47ycibDe8YfjKLtjSaj2lK8zFNyd6VTUFqIU3ODGfYsd3MibDyze5gxrTLw6OZaBLiKXM7yTlmusyMJ99lwqXBY/0z6R7nK7z/8U0jzCbIdppoG+HilSFFWeHZ1RCg9Z0hQyRD+LPPpBHs8GE5qL33Hnzxhdx2550QXInmrowMSVQBsZnUZSqC4tTBVUDLCI1vdwUzubtUvbelWslzVfz/q2moh4f6ZnLPkkgWjD9OkBUe+DOKyzrmyol2EVMHZnBV51zcHnhlbTj/tyiaX8cdl6ObC2jVyqhojxghkYA1yYoVMHo0LbKyeOHB5TzwwmDyrG5ctsBUpc2YsVvsPDrgcbrH1YNG6O+/N2IVS7OZgBLeCkURquJdH9jwKXw7SeK2Aqh7TM4czs48ziN71mDWyha7gSRlZSoRXSJwhFrpnJPOJR3ymJMYQpRDw2zSSM4t+9+uSaiHbdcnsX3SUSZ3y2HZkZKDfH6+5Djbrk9iz+SjXNc1l/Hz48h3UcwDfwpiNsO110r+95NPQkhRxTA7Gx5+GDp1EmHuqeDfe+lS477K360IFK4CJrTPYe5O4yRwTmIIlyZULt3k6s65NA7x8OracGYnBrM73cr9Z/ufd2Axw8Xt89iZXrT09+I9sGMH7NsnCSWXX147ovu8805G6LW8ZDJv3fI7PTem4Mj3v2pXGRwWB20i2/LasGn1Q3SDYTNp2xZ69PD9mfdsCuXxVigAJbzrnt2/wfx/1ljGrd3jpHt2KlMObqqR7RfHU+ghdW062TuzWfvQVvp82pj3NoWxNdXGngwLZzUu5Me9FavIOizwcN9Mtqda+Xmf/2xcmxmu6pTLgSwrO9JsYCl/2uYpQUgI/Pe/Mtxj8mSjUn3woDRl9u8vAz7KQ/m7FTWBO5/eTQrIcprZmWbF7YF5e4KZ0KFyn3MmE7w0OJ2ZW0N5/K9IXjg3nWCrf+tGoRu+3hVM78aF4LDBwDaSwV1bqzjFRDfPPw9z5xKVUcijU1fznyMJhNpCCbJUPuc7yBKEw+Lg6i7X8srQ12gW1izAO19FcnLEqgNiMyn+u1YVb4WiBEp41yX5GTJxsSYHSwBBmptBaUmclXmsRp8HIH1jBiaziZ4PJzBvYioLxh/nj4nH6BdfwNydITzcL5PZicH8b0PoSf/nlhQrt/zuPzfcboF/9cjm1bX+P7TdHpiVGEyQxUOrcBeEN/V7v1OWZs3ggw8k/91bOK9aBeeeK6Pp9+wp/fG6v7tlS2jXrmb3VXH64JFG6Akdcpm7M5glhx0kRLmIDy3ZIF0eLcLcNAl1E2b30L9pyRWt/y6PoMvMeDp91JSPtoRyZ+8s+UzNPFztl1FhiovuadPktqNHATD168eg21/l4/M/Y8qZ/0fbyHbYzDZCLMGY3CVPJEyYCC7QcOS7aJKUyw3R5/Px+Z8xrsMlmE316LD9yy/i44aSNhNQwluh8IPyeNcl399Sa13/QZqbO/dv4F9dhpJjtdXY86SsSKVxvygu9iTR2ZYnaSvApK45PPpXJA/3zWT2hSm8vCac19eFYzFD2wgX13ct/fdwRac8Xl4TzoL9Ds5rLdm4o75uhAkwm6B9pIsPRqURbddgez78+KOIztNpabNXL2kW+/57uPdeWWIHmDsXvvtO4gkfflim7emkpMD69XJ5+HDl71YEEBGTlybkMf77WA5kWUvYTCwmKBaShMtjwmbyFaJvrg8j2uEh1Gpi+sYwpvTybTp/amAmV3XOxaPBqiQ7NyyIYe5FJ+haG0N7oKTofvNNiQTVU4jCwsT+ZbNhAwa3GMLgFkNIyTvBzvSdJL75OIn2dPKjQvH07EGQNYj2ke3plGKh/ZW3E5+Ui+mCEPjhptp5PZVBf43x8TBgQMmfewvv0+nzWKEoAyW864rDq2HHPKilIQsAQR43VyYl8n6LbjX2HAlT2hNXmMf123zTRca2z2dse4kUPLOxk0/PT/X3cAY2K2TNVck+twVbNTZfZ9x2+OYj/p88H/hqOTx3oQyY6dtXBOXw4XJQCKr9Uc61iskEY8bAP/4B77wDjz0GqakygOell2Q0/RNPSAqKzQaLvf5Gyt+tCChyEtci3E2rcDcLDzp4eXC6zz2ah7k5lOXbcHggy0KLcKMqnphmZfrGML4fd4JCN1wyP44L2ubRLrJk5dxsgn5NC2kT4WLJYQddrbUwQbKokdJHdI8c6Zs+NG3aycmW3sQGxxEbHEf/VuNl+iXAmpd8Hzv4F5le++OPYgurT3awwkI50QcZ/OUvEcbb460q3goFoKwmdcfyF2txoptg1zyMSj2E3VP55d4KP4fbzf371mKvxWZO48nNkFZHAqTJAAAgAElEQVTUbOhywfLlMoBm+HCp9I4YAU8/LRPWXNVvdKq32GwwZQrs2gV33y3XQSrcU6ZI1u733/v6u5XwVgQSLzvES4PTmX1hCiE230r22HZ5zEoMYd0xG5oGu9MtvLc5lLHtxLrg0eDuJVHc0jObDlEuusa6mNwth/uWRqGVktC3OtlGYrqVjrEmCGtSYy8PMER3ZqZcf/NNmSp79dXGJMyJE+H668veziWXGKJ1zhzfnz3zjPH+veeeijdN1wZ//AHpRSdT/mwmoKwmCoUflPCuC3JTYMd3UBfiFBPnph2tkS3bXXBn9Jl0ctbstDa/mMzQbRwcSZOmwieflPHn9qIM4IICEZqPPAIDB8pwmosukgE169fXrwNaoIiOlkr3tm0wYYJx+44dUhn/6CO53r69RK4pFIHCGoRe9W4T4aZno5KfCUNbFvBQ30zuWhxF55nxXPdLLBMT8rimi4jW9zeHkucycWtPQ7zd0TuL43lmPt9h5H8/sjyShBnxJMyI5z9/RHNfnyyGt7NAs7Nq7vX5E9233QaPPirThwFatJCVp/IsXI0by2cViPD2Pqto316m1QKsWweffx7Ql1EtdJtJZGTpJ+5KeCsUJTBpWmm1A0VN0e/MbiSfSGPt1cnEBBmC77yvG7ElxcaKK5J5eU043+4Oxm6RP0+LMDejWuVzW69sIoqmVRa64dlVEXy3J5jMAhMxQR5Gt8nnyQGZPs936fexbE2xse6aJBwW2B0Uyb0dBuCyBui8y6PhcMOdZ9/LwPje8FJ8rVfzsYXCdb9Cy2I+w7w8qXwvXAi//y5Nh/5EdkyMHDx0a0qnTqee53npUrjrLiO3W6djR6leNT3NGlMVgSU/X95n8+fD1tkwMBXqKmTIbIP7U8FRA2Jv5UrxdBcX3YsWyaqapslnx8KFhqAuj+nTZUItSKP0mWcaP0tJEQGekSEnyDt21L1tzuOB5s1lcNfVV8Onn/q/33PPyXwBkFjH1q3930+hOI1QFe+6wJVPy3A33+4yPIj+hkvc0iObxElJbLwmiVeGpLPmmJ1x38WR65T7vbk+jI3Hbfxw8XESJyUx96IUzoj1rSwdzLKwMsmOyQQL9suHdbusDM5anRyYXNl8F/FpTp47+0kGthkGQZHQ+2aoDX+ljskMsR2hRf+SPwsONiwmK1aI53n+fBk007Oncb/UVPjqKzmAduki1aprrxVf9P79tfdaapJzzxXR8OmncqKhk5go0zGnTjWWyBWKipCcDB9+KFaDuDiZFjl9OqxPrdujS2Tr2hXdqanyeaHXse6/v+KiGySKrzS7SWwsPPSQXD5wQDzjdc2KFca03NJsJqA83gqFH5TwrgtcBSdjtnTKGi4RZIVejZx8dF4qaflmZiXK49Yft/OPNvnEh3owmaBluJuJHX2jCefuDKZ3YycTE3KZkyjLsyYNHv4tgzs3xBDmsRFsspd4zvKwF7ixFbq5aNFx3hryFu3b9DF+OPI5CPIfD1gjWINg4qyKVagjI30tJsePy4Hulluk8qtz5IgI1MmToU0bqTjdfLNMiKzOePa6xmyWCtW4cb635+RILninTvDJJ6em9UZRfTQNNm6UE9n+/WWV5MYb4dtv5X9Ip0kCmOooa9oWCgPuDPx2SxPdmiYNy4eL4gv79JEm5srQuLFMpoWSdhOA22+XyE+Q331KStVfRyDQbSZBQdLMXRrKaqJQlEAJ77rA46J3k8JKD5cIs2uc27yAlUmyftu7cSHvbgrlo60hbEu1+m04mrszhEs65DK+Qx6LDzk4nmuGsAj49BUGPvsFM8fN4fY+d9E+sj02s50QawimUsZnBlmCCDLZicwo5IpZO5lx+wquv+kdbB06+t7RHgKXfgFaLYTm2EJg2JMQm1C1x8fFwaWXwttvyxLuwYPw8ccwaZJxoAPJwn7/fbjqKhEb3btLRN+330JaWkBeSq2yZIl8T0iAm24yqm2HDsF110kijH4fxelNQYHkNU+ZItMJe/aUXomVKw2BaDbLisoLL8D27bKKcu10sNeB2NI80PPawG6zuOieNk1EN0ivxFdfyeWQEIkOtFe+mMFll8n3XbtgwwbfnwUHi+AGsZxMnVr57QcKTTOmVY4eDaGhpd9XF95Wa9V+JwrFKYiKE6xt3C70jFu96t2/aWGFh0vEh7rZeEI+wG7vlU2Uw8M3u0J44q9IooM8PHB2JpcVVb3/TrJzKMvC2Hb5xAR5aB3h4pvdwfzzLA8UygeizWxjUPNzGdT8XI7nHmdneiKJqYnsSNtOrisXj+bBYbHTOqINXczNaP/vh2m9cqcMcfjuOzj7bD+v0Q3PfwEbXDAUqKnPW2swtBsF/QNY3dItJvqy8Z494tXUv44VDSHaskW+pk2TSnvv3oY//Nxzyz4Y1TUHD8rBHWDUKHjrLamo3X23ZIGDNIgNGSJL4M8/Dx061N3+Kmqf48clwm7+fBHd2dkl7xMRIdXOMWPg/PPFEuFNwgUQHHPys6ZWsAZD75vAEcDMaH+ie8oUubxzp7x3dF5/3XflrDKMHy9i3uORqnevXr4/v/pqePVVabJ86y3ZBz8xhTXOpk3GUK7x48u+r/5/Ex5+6vXMKBRVRAnvWscoS5c1XKI0juZYiHKIDcBihkndcpnULZc8F8zaEcLdS6I4s5GThGgXcxKDGdyi4GQD5yUd8piTGMI/z8rzm6jSKKQRjUIaMbDZOSWfOCsLBg+G9Tvl+vT/iZ+zOIWFUjGdNUuux8fAmXngDvB0TlsItB4Cl83xnx8bCEwmObDpNhNNE7Gti/A//pDqk6aJUF2zBl58Uao7/fsbQrx/f3DUo1H2+rRKMHKBe/SABQvgp59EgG/fLrd//bWIrylTxIoSXYsWIkXtoWmwdav8refPl8hNf0tobduK0B47Vk4wy6pimi1w6SyYObzGp/OeJCgKRj4buO39/XfpotvpFDGsW2zGjxfbTVXR7SaLFkl299SpvmLVbJbPl5Ej5bkffhi+/LLqz1dVdJuJxSK2vbLQPd7KZqJQnERZTWobsxV/wyUuaFN+CkiO08Sfhx30iy85dCfYKiI80u4hMd1Kngvm7wlmxVE7vT5tQq9Pm/DepjC2ptrYkmIV4VpRnE6xY+hTDh95RDyNxcnNlUEKuuju1Qumb4PzX5XnM1lKPqYqWIOh5yS48juw1NwUzhKYTL4Wk5QUSUl5/nk5OAcXefZdLt9Iw6goqSw/+6xUz+o6Q9xbeOu+UpDXd8EF4uF96y2x4YD8/V99Vareb7wh1xUNn8JCWeH4z3/k5LJ7d0mgWL7cEN0mk8RvPvssbN4Mu3dLVXfEiIpZB1r2h7NqqdnaGiy9HvYArTb9/be8b/2JbhAf96pVcrlZM3j33epXdSdOlO/+7CYgv/fzz5fLs2bJPtY2us1k6FDfJm1/6BVvJbwVipOoindtYzJJJaiIlwank1FgJsSm4Sqln63ADTtSbTz9dwSRDg+XF1lJ3tsUSrdYJ2c2LsRmhq93BZPjNNM91skv+4KxmOH38cexmY3K1b9/j2buNgvdYjtVbH/1xqEFC+T6ddeJoCxOWppUP5Yvl+vnniuVs8hIaPwvaDcS5l4Bx7eBs/Tx8GViDZHUlEu/hDaDq7aNQGKxSCNVnz5w330iZFauNCrif/0lIjU/XwSObuMIDxfBO2KEVJy7d6+5qn1xNM0YnNOzpyGuvbHZJDv4qqtkgMfrr8trS00VkfbWW1J5GzNGLR83NFJSZFVj/nz4+WdDVHoTFibe3TFj5ESsUaPqPefI5+HgckjeVHOTem0hMOgBaH1uYLZXXHS/8Yav6F6yRN4bIO+Bjz8uabWpCuPHy/OUZjcBOdH/5Re5zz33yATa2nof7t4tJ+b6vpaHt9VEoVAAKse7TujXvR0vDjjM4OaFPre7PND6g2Ylcrw1TXK8R7bKZ0qvbCId8if7dFsIn24PYV+GFZMJ2kW6uOPMLEa1LuDqn2LoGO3isf6+B9bvdgfx6F9RrN6yF6ujAlmwjz1mCO2RI+GHH0pWupKS5ECtfyBfdJEslQYXq3J5PLBlNvz5LKTuAlcBaBWYomkPA3s4DLwb+vw7cBWtmiY3F5YtMzLE16zxnxYSF+ebIZ6QUHMH0j17DF/oHXdIJbs89u6FBx6Qv6k3w4bByy/7Zg4r6heaJk3DuoVk2TL//4OtWhkWkiFDAm+Nys+Ej4bAie2Bz/i3hcDZU2DUc4F53/gT3d4+7vR0sWYdPCjX771XmkoDxbBhYmNLSJC/nb/XdOONEuMIsvp28cWBe/6yeOkleb0gjdjNm5d9/4QEqd4PHy6fgQqFQgnvOmHFG/Db/bU/ZEanUTe4bXP593v/ffE2gxxoli6Vhipv9u6Vg9Tu3XL96qsl+9pWjgUkaQOsnwn7l8DxrXJwMVnku8cN7kKIaQ8tB0L3K6DtiNqrCtcU6elSKdMr4ps2+b9f8+ZyoNIr4t7pKtXF+2/63XcitirKsmUygMd7edtkkgSYqVNluV1R9zid8rf67jsR23ojrTcmkyTXjBkjX2ecUfNV08IcmHMZ7Ftc9VUvH0wSJTp8Kgy8KwDbw/B0Z2TI9eKiW9PgyisNO92ZZ8rKViBPVN5+20hMWb/ed96AzuHDImrz8iQCdPNm6S2pac45R1Y1+/eX110eTZtKYWbsWJg3r+b3T6FoACjhXRek7YO3utSN8LYGw5DH4Nz7y77fTz/JAdntFuH3118lqxubN8tB6mjRCPrbb4fXXqu8QNY0yDwEBVnS9GkLgciWtevfrguOHZPKli7Ed+70f78OHYxq+LBh0oRVVa6+WsZOm81iHYmMrNzjPR4RHQ88IMM8dEJCxG5zzz31O9HlVCUtTawj8+fLezc9veR9QkLk/TpmjDRGN2lS+/upabLqNf+f8vnnLiz/Mf6whUJUa5g4Gxp3C8y+lSe6QSwl118vl4ODZcpk586BeX6dpCQ5idU0aaAsLTrwv/81fva//8G//x3Y/SjO0aNyDNA0sbvcd1/5jwkPF7vJVVdJzKJCoVDCu874aBjs+6P2n9caBHcdgpAy/Ih6lFxOjgizZcugW7GD24oV4v/UM6wfe0y+lOe36hw8aIjw3383BnIUp3t3oxo+eLA0b1YETZMDelKSxEBWpzErL0+838884zudrlkzue3aaxv+CkV9Z+dOw0KydKmcJBenRQuxfo0dKydtdT1qXCc7GZY+A3+9DU5XBUfLm8RmFhwHg+6X2EBLgKq8xUX3669LE7U3u3eL51r3LU+fDv/6V2CevzhDh4p3u2NHSRjy97mamSkn5cePy8n4rl0166X2HmuvT7stC49HqvCaJr+n6dNrbt8UigaEEt51xc6fYc7E2s24NVmg6wTp/C+NvXthwAAZBW23SxNP8dHHv/4qkw/18eL+DlKK6qFpciD1zhA/caLk/cxmOOssoyJ+zjmlV5y3b4cuXeTyffdJ1aq6JCfLCdd77/l6h3v3Fv93ZcZmK8rG5ZKVJ91CsmOH//v16WNYSHr1qr8nwzk50LQRtM2DC1pAdD4UZMqqnNsNWdkSAGUDYttBq4Fw1j+h1aDAvqaKiG6XSxrGV6yQ62PHire6pn63ek43SLpJjx7+7+dtS/nvf/03vgeK0aOlyb5799Jtct5kZxsnAnffLf5whUKhhHedoWnwwTlw+O+KNRgGAlsI3LoFotv4/3lKigg3/YD+xRdwxRW+95k7V5YNnU5J9ZgxQ6qbiprF4xFrjy7CFy/2n0hhs4n/Uq+I9+tnNMP+73+SVgJiSxg9OnD7t3mzHFz19BudceOk8ay86pjCPxkZcvI7f74MtElNLXmf4GBpfNYtJA3Fa//FF/JZAtL7cO65kJsCKTvh0D644kooBFLNkOesmRWUVaukR6Us0Q2+Tebx8dJIXt20l7Lwtps88gg89ZT/+zmdIoQTE8VKtHNnzfz909Kkqu5yVVzgJyWJxxvk9/f444HfL4WiAaKEd12Svh/e6hagRqNysIXCqBeg763+f56XJwegZcvk+gsvGN3rOu+9Jz5Cj0eaiebMqVxzniJwuFziL9WF+J9/yt+wOCEhMGiQiPAFC+S+VqscSGsiW/fnn0WAb91q3Ga1GgN4ysv9rQxuJxxZLV96k64zT6qQ9jBo1gdangPNz4bG3etv5bc4e/YYFpLFi/3nvjdtalhIhg+Xv3NDY+xYeY0tWsD+/b7C2rtaCnDkiCHiAkVFRfeyZWLp0ld0fvlFKuQ1zZAhckLSqRNs21b6/++338Ill8jlm26Sz+lA8+mnRoFl3Tr/MYfF2bXLOOF+8UXp/1AoFEp41zlr3oOf7wBnxSZXVgmLA5qdBTcs9V818njgssvgq6/k+pQp0ljk/UH//PPSUAdyQJw/33f4iqJuKSiQZXBdiK9YUfqgnpgYqUANHy7e/UALUpdL0lMefVT8pzrR0XLbrbdWbPhKaWQegdX/g7/flAQcj7P0RmVbke0mpBGccy/0vDaw48QDgdst+e+6hcT7pMWbM880LCS9ezdsD31qqlSOnU4RZC++6PtzTZOTe31Y08qVksISKCoqujMyRGTu2yfX77wTXnklcPtRFm++aTR3btwoyTP+0DRZLVi2TP4nNm4s2ZNTXSZMkME5bdrIiWFFPjPWrzeiRmvSD69QNDCU8K4PLLgPVr1dM5Vvix2i2sLNK2X4jD/uvFPSSECsAXPnio0E5EP9gQeMnNpGjaSq2bt34PdVETiys40M8YULYfVq//dr3Ng3Q7x9+8AJ8YwMmXj42mtyYqDToYMIrYsvrtxzOfPh9wdh9XT5v6zsMBZdhI9+WbzCdVkBz8qSFYj58yUb359/3+EQy9CYMVLdbtGi9vezpnjvPWP67Zo1/j9P4uOlhwDkM2nChMA8d3HR/dprMhjKH9dcY6Rx9OghJwC11aDqnSJSnr1jxQrpzQFpev/hh8DtR26uzBrIy6vcicfSpbJSAFIxv/rqwO2TQtGAUcK7PqBp8PsjsPK1wFa+rcEQ0wFuWAzB0f7v8+qrks0M8sH9++/G4Bu3W6wl778v11u2lMbKThWceqmoPzz7LDz0kFxu3VqW9v3RsqUhwocPD4zY27dPRpF/+aXv7UOGyEG8Iidxh1bC7ImQewJcfiw1lcEWCvE9YcIXENWqetuqDPv3GxaSP/6QaaDFadJEfNpjx4pv+1SNZhw+HBYtKttG0aOH0cRXljiuDJUR3Z99JsIbRGyvXh34SnJ56HaTzp1lJaSsk8XLLhP7H8jn+PDhgdkHbyvL0qViXasIP/0kJwH6NmpryI9CUc9pwGuVpxAmE4x8Gi75GByRUqWuLtZgOHOyVLpLE92zZxuiOyFBlrp10V1QII2Vuuju3FkqqEp0N0z0NAaHQ4TO0aPS3HbzzcYkS5BIw5kzJau4ZUv5e99yixzQvW0jlaFNG3kuffCGzuLFksAxaVLp0YkAm2dL/GbmweqLbpCVpcN/w/SecHR99bdXGh6PVEgfflhEZJs2Yh1YsMBXdPfoIfdZsUK8zB98ICLlVBXdR47IiQfIMJrSxKR3Xr0+JbI6rF5dcdG9b5/RiAySyFHbohtg4kT5vn27NDCXxbPPGoPL7r3X/4TSqvDNN/K9cWOjql4Rsr0Su2qin0ShaKAo4V2f6DoB/m8XdPiHCGdTFTJq7eEQ0RKu+w0ufBNswf7vt3Sp0SzTqJFUJ+Li5Hp2tixvz50r1886S6ougZygqKg93G4RuSAHzuBgWca/4gp4911pgtq3T0ZQX3ONbypCYqL4My+7TA68PXvKcvP8+f5TVcpiwAAR319+KSIUZLVn5kw58XvsMd+DNYjo/nZSYAS3Nx4X5KfDjMGQtDFw283JkerejTfK77F/f8k1945fs9slUebNN+X3vmGDDELp169h+7YryuzZ8ncHEd6loX8eQfWF9+rVsoJQEdHtcsn7QP//vvBCXxFem0yYYJyY6NXs0mjf3tjPtWvlZLe6OJ1SkAE5GdQtiBXBO9+/JvPFFYoGhrKa1FdO7IBXrwBtPXgAuxnMfioYJosMlXAVyHj1c+6F9qPLPoBv2yaxgWlpkobwxx8yUAWk6emCC6RSB5LDPG9eyVHxiobD6tXG3/fJJ8UvWhaaJoJb94cvWiRRk8WxWKRirdtSBg6seLpGfr408E6d6nuAbtoUnn4arrsOjq6Bj4YGXnQXJzgGpuyA0Ljy7+uPQ4cMC8nChb5+dp24OMNCMmrU6S1E+vWT7OzevcXfXRpTpkieNVR8RLk/iovuV1+FO+4o/f5PPSVNwCAnm5s2VW9abHUZPFgKJRWxm5w4IQI8MxNatZJo2Op40n/7Tf5fQYoz//hHxR/7xhvGyc2WLdC1a9X3Q6E4hQjQ2C9FwInrBHNzYQ/QDHjoeohMh7TdIrLNVgiJg1bnQPN+0LwvhMeXv92jR+H880V0m80y/lsXZUeOSEzWli1yfexY+Xl9mXanqBqLFhmXhw0r//4mk1hMdJuJxyPiQ5+ouXixVKb1NI6VK2WZ226Xqvbw4dIUePbZpaeXBAXJEJ9JkyTf95135HmOHoXJk+HN1+DSpJoX3SBDrL67Ea6cV7H7ezxSUdRTSNaXYlfp1s1IIenXr3LVwlOVXbuMiallVbvBt+J94EDVnq+yonvFCnjiCeP6Rx/VregGsZssXSp2ky1bJLe7NOLipJfjgQfkdzZtWslY2Mrw9dfyPSKi8p5xZTVRKPyiKt71FZdL/LgejwihjIzqV8mysqRZZ906ue4d8bRrl1Q29Nis668Xf7dVnZs1eM4/X5JoQkLkhKs6UX4gy89r1hgV8WXLpIJdnNBQiTnTK+K9epUuPrdsEYHw009y/TzgLCAA7Q4VwhYiPRZdS0nOyM2Vk4758+H77+UEocQ2bPL+0lNI2rWr2X1uiEydaqy4HDhQtn3NO07PbJaVhMp8Hume7vR0uV6e6M7MlPi7PXvk+v/9n8QM1jVHjkiTs6ZJJd77xMAfeXly0nzwIERGyqj72NjKP6/HI8979KgMOtLTXSrKQw/JCTnISmp0Kb1GCsVpxmlgKGygfPON0RzTvXv1RbfTKZUTXXQ/9JAhujdulE51XXTfcYf4fZXobvg4nVItA/kbV1d0gzEd86GHZCk6LU2q6o88InYT/f8mJ0cE/333iSUlLk7SEaZNkyVz73P+bt1kMuPPP0OfTtCH2hPdIGlCP07xbUg7ckQ88GPGiHAZO1Zi8LxFd0yM+IFnz5bm019/FcGmRHdJNM3wHQ8eXH7PiHfFW18NqSiVFd0gfzdddHfvLrML6gPNmok1EMr3eYP0cDz9tFzOyDAuV5a//zZ+53qqSWXwtpCdqo3CCkUVUMqqvqKniUD5S7LloWkisn/5Ra5fe61UnkCqlRdeaCzFPvWUJCw0lCl/irJZtUoEMAQuXqw4QUHSCzB0qPz/ZGXJJE29Ir5unfwPpqdL4+G338rjmjTxjS5s21aaDq0TYenzgLNm9rc0CrPh1+mw4rhUtkvzH3fubFhIBgxQJ6gVZdMmYzhQRT7T4op57g8erFiDd3HR/cor5YvuWbOkyRdkpfHzz+uXxW7iRHlPbdsmq0PlJaxcfbW87vXrZeVgypTKnwzqNhOHo3Lebh3damK3B+aEX6E4RVBHjPqKHv8G1R8c8cQTMGOGXB4xQkS9ySTVxfHjZWnSZJIP6Lrq3lfUDJX1dweC8HCxt5x/vlxPTRVf+O+/ixDftk1uT06WCqheBW3dGoYPhTZzqXXRDVCQDZ/dBp8Uu91ikQqtLrY7dKj9fTsV0P/OVitcemn59y8uvA8dKv8x/kT3nXeW/ZgDB3ynKr7wQulTIuuKCROMRsU5c8oX3mazDKkaNUpWvR56qGSOfllomhEjOHp01TzauvBW/m6Fwgfl8a6P7NljZCvHxvqfaldRPvxQos1A8oKXLBHf35dfSuXb5ZID4ccfV7+yrqh/jBghYjciQpJJ6kN19uhROSHQK+J79xo/awb9QhuTnGdh7dXJxAQZ1o/zvm7ElhQbK65I5uU14Xy7Oxib2fj4ah3h5rcJx9l8wsqE7+P4+ZLjtI10A7DxuI3Lf4xlwfjjtAx30++LxpzIs2A2adjMcFaTQp4blE7zYA88C4RHyYnD2LFS7YuKqq3fzqmJpsmKxv798nv98cfyH3PokG+F+6WX4O67S7//mjXSSFkZ0e12y2rLkiVy/R//kH2rjyt+gwbJCmXXrkYDfHno/R0gTdB9+1bscZs2yfECpGgzaVKld/fkc7dubdgYFQqF8njXS954w7g8enTVt/Pzz8ZY5hYt5IASGSlNlVddJaI7OFjiApXoPvUoKJDcbJCKbX0Q3SCRgVddJSsve/bI1/vvy22dJbayZbibb3cZGfTbUq3kuXzF0C09stl5Q9LJr98myICf7nEuJnXL4d6lUWgaOD1w15Io7jkri5bh7pOPnzE6hZ03JLH26iQaBXv47/JIMNvhm+lw7JjYDa64QonuQPDXX8a01Ip+1hRvCCwry7sqohukuq2L7kaNRGTWR9ENxjCdrVsNy055vPCCES17772+fRVloVe7LRZZ5akKusdbVbwVCh+U8K6PfP+9cfm666q2jbVrZTnX7Rax/dNP0qTzzDMSEadpcvuCBcZYX8WpxYoVRtpITfm7A0HbtrIq89lncMuFYIIJHXKZu9MQ3nMSQ7g0IbfCm7yrdxbHcs18uj2EaevCCLVp3NAtx+99g6xwYds8EtNsYLdCU5MxAVARGHSbSVAQjBtXsccEB/vmwpcmvIuL7pdfrpjoXrXKyOsGWR2Mr0Aka13hbTmsSJMliGVGr1YvWSK9CxVBF95DhlQtEQWU1UShKAUlvOsbLpex9G61SsNaZdm3T8R0To4IiG++EU/gPfdI4yRIY9vixbJ8qTg1WbjQuFxb/u7qckzGYvduUkiW08zONCtuDxYETpcAACAASURBVMzbE8yEDhXP9HZY4OXBGTzzdwTvbArj5cHpmEspZOa5THy3J5jejQsl3eR4BZfxFRXD5ZLUF5DqaWUSmsqbXulPdN91V/nbzc42Vv1Aelsuuqji+1UXtGhRuXQTnSeflJMYgPvvN15zaezda2TTVyXNREcJb4XCL0p41ze++MKINOvVSzrKK0NqqvgUk5Pl+kcfSZbyjTfK8ivIuO4//5Tx34pTF72xMibG8GvWd1xGHrhe9V5y2EFClIv4ULfPXd/ZFEaXmfEnv/7zh68lpFOME4tJo3OMiw5RJcXGjQti6DIzns4fxbP0sINbehYJhYLMwL+u05lFi8S6A5W3tJUlvKsqukEaFXftkstdukgjYkNAt5ts2VJxu0nz5sbvZft238Qsf+jVbqj46oQ/dKvJ6TylVaHwgxLe9Q09fQQkH7gy5OfDxRfLmGCQHNrx4+XDWt9u164iulUyw6lNbq6RjDN0qOHzrO94+WsvTcjjm93BzC7FZvKvM7LZdn3Sya/Xh6b7/PzJFREMaFrI0Rwz83aXjIb74LxUtl2fxJ7JR5k6MINLv4/jWK5ZpsIqAoduM4mIMJJuKoq38E5OhsJCubx2rW96SWVE99y5YisBibn7/HNfS0t9pip2E5As/UaN5PJjj/lmbBdHjxHs21eq7FVFVbwVCr80kKPxacSqVcblysQIejySUvLnn3L9ttvg3/+WjG49N7lvX/H5NW8euP1V1E+WLZMYMWg4NhMAu1EdaxHuplW4m4UHHVzQxs9kzDJYctjOgv1BPH9uBs+ek8Gjf0WSlu/fa2IxwwVt8zGbNP5OckCQaqYMGPn58NVXcnn8+MpnY3sLb02ToUZr10qlOy1Nbq+M6D50yGg4B5ms2KtX5fapLmnRQoZUQeWEd0QEPP64XD52TBJi/JGcbDRkjx9f5d3E7ZaTf1DCW6EohhLe9YkdO4wqQXx85aoN99wjlRyQqvd//ysHJ93nO2KE5ChXtVFG0bDw9nfX58bK4jT3jTt7aXA6sy9MIcRW8dTTXKeJ+5dG8fiATGKCPIxoVcDg5gU8viLS7/01DX7ZF0RGgZmECA9sPiFxat5TLBVV46efZBQ7VC05qXiW9++/+4rul16quOh2u6VZXX/sqFHlD9apj3jbTfRM/Ipw883QsaNcfuklOYkpzrx5RvJJdfzduV4rVMpqolD4oIR3fcI7RrAySSOvvipfAP36iV9x2DCjej5+PPzwg6o8nE7o/u4mTcTD2lBoOQBMxsdSmwg3PRv5H6bzv41hJMyIP/nV/eMmADy3KpwOUS7GezVjPjEgk0UHHSw5ZPRM3PBLLAkz4uk0M57nV4fz2tB0OoUXwNMfiSc+Lk6aAZ9/XqqAus1BUXF0m0njxlU7ASwuvP/zH1/RXVaud3Feftl4X8TGSv9LQ7FgeeM9fKgyVW+bDZ57Ti7n5orlpDi6zaRrV0OkVwVvK4s67igUPqgBOvWJtm2NQQN//ml0sJfF3Llw2WVSpejQAT75BC6/XKaxgTRVTp9efzKcFTVPZqY0VLrdkkOti5+GwPHt8O5Zki5SF7iAp0v5WVCQnNiee658DRigqnllkZUlgjs/X0aWT5tW+W3873/+p+lWVnSvWSN/L91+9c031WscrGsGDpRs9O7dZXWmomia/O8uWyYnHRs3GlMwMzLEB+50wiOPwFNPVX3/EhOhUye5XBkrkEJxGtAAT/dPUQoLjQETdjv071/+Y/78UxowNU0+MF9+WSbt6aL73nvhvfeU6D7dWLpURDc0LH83QFwnCG9WN8/tBjYhDZ49e0p/xNlnG++f/HyJ4Jw6VQZbRUVBnz5iV/jqKyO5QyHMm2fkyFd1QFfxijdUXnTn5Eh0oC66//nPhi26wbCbbN4sSSUVxWQy/N0ej8QL6vzwg/E7qo7NBAzLJKiKt0JRDCW86wsff2x4684+WyaGlcX27SKyCwoko/Wpp6S58rhM7+O552RqWX2dwqaoORqqvxvk/3XQA2Cvg4O1ZoIVyPtwwwYRIjt3ygCSt9+WpfkRI4wEDI9HKqmvvy7L/02aSJXvpptg5kyZyHk6LyjqKy2tW0u1uSqkpvpe79q1cqIbpNqamCiXO3UyYlUbMlW1m4AUdfTH//CDYb/RbSatW8OZZ1Zv/7yFt1oVUih8UMK7vvDxx8bl8qZVJiVJVndamiwX3nOPVN0yM0W4vPOObyVDcXqhH0hbtID27et2X6pC9yuB2j5hNEGLXrBgvfiI9Sbk9HTJPb71VhEmF10kYnzlSllhGjfOtyqbmAgffCBivX17+Rtcfjm8+aaIebfb77Ofcpw4IVNxQexOVSkArF0rMXjeVHauwbffwrvvymWbTaIDQ0Mrvy/1jZYtjVXRygpvkDQXfTrrPffIqsBPP8n1Sy6pfsFGebwVilJRwru+sGaNcbmsGMGsLFkC120p118vH6L5+fJB+uWXvnFZitOL1FRj6tzw4Q1zxcMeAue/DrZaFEjWIBg3Qywmr70Ghw+LoLngAqMBb9MmGUXepo2sJnXuLPc5dkyGmbzzjli/Wrc2tnvkiExtvP12ia2LjZX373PPic+2oKD2XmNtMneuMSGxKjaTdeskvSSz2DCj0sbG++PIEVl90Jk6FXr3rvy+1Fd0u8mmTcbshorSoQPccotcXrtWUrD0JJLq2kxAWU0UijJQzZX1gY0bjSmSLVsaHu3iOJ1iL/n5Z7k+ahT8+qtcDgmRitzo0TW/v4r6yzffGPm7M2ZI5bUhomkwczgc+BM85Yy4ri62EBhwFwwvpZns8GFZkfrwQ2PaoU7TpnLye8MNvikQBw+K1/7PP+X75s3+t+1wSL6+3rA5cKBkLjd0hgyRmQFdukjsXWVOANetE0uPnl5SnLy88vPAPR75LPztN7k+fLh8VjbEFJPSOHDAOMl76ilpiKwMJ07IqkxmpqwC5ORIr9DRo+VbHcvjgw+Mk56//xb7pEKhAFTFu1YpKCjg6NGjHDhwgAMHDnD48GGys7PRvGMEx4zx/2BNkwqFLrp79DBEd1SUHGCU6FZ4+7sbWmOlNyYTjP8UHBHUqO3EYoeYDjDk0dLv07w5PPig2EiWLJGTGd3nffSoVK87dRLhPGOGVPtatpSGvrfflopkSgp89500PPfvbzRsFhSIMH/mGZnqGB0tVdn//EeqxklJNffaa4pDh+Q1gfwOqiO6X3ih5InIoUPlb+e11wzRHR0tnvtTSXQDtGolKTtQNbtJXBw89JBczsmR7xdfXH3RDb5WE+XxVih8UBXvGsTj8XDgwAG2b9/OsWPHKCwsxFL0oWYqOhi53W7MubnEHDxIwooVJLzyCrY+fUpu7MknjdzVli2NJdf4ePFSnnFGbbwkRX2ne3epMLZrB7t31/XeVJ/kzfDhICjIBAL8UWWxQ2QruPlvCI6u3GOzssRCMmOGWEa8CQ2ViM/JkyUS1J/wzM0Vn7heEV++3BA/xenQwaiIn3uuVCnrs4Xo5ZfFNwzih+/QoWKPKya6tWeewXnhhTjHjMWVloZmMgMa5htuwH7RRdh6nIGlceOS21m/XgSpnrs+d27lpgA3JF55xWg23b7diPCrKHl5UjXXm/K//FJ6EqrL1KliXwE5VlVn9LxCcYqhhHcN4HK52LhxI5s2bcLj8eB0+h8AUhxrQQFaSAgJCQn06dOHEL2qNmOGHMRBqtvp6XK5XTuperdrVwOvQtHgSE6WEzGQZd733qvb/QkUx7bCR0OgIAvcAfJE20IgJgEmLaq86C7Ojh3yHp05s2SFOiFB3rvXXQfNyohJdLlEMC5dalhUdDFUnPh4Q4QPGiSrX4GoUgaKPn2kZ+Xss8VmUBGKRLeWlkZ+UBDZ3c6g8PhxTEFBkJPtGw5jNmMKDUUrKMAcFkbINVcTdt21WJo2lROaPn2MiY433ijNsacq1bWbgHjpf/9dLt9+u+8gt6rywAMyeArkRCoqqvrbVChOEZTwDjDJycn89ttv5Ofn465igoHZbMZisXDOOeeQsGcPposukjSEoCAjF7d7d6l0N20awL1XNGhmzZIECYDPPpNl/lOFvDSY/y/Y+UM1h+uYpJFy4D0w5L9gsQVsF3G5xAo2Y4bYSlxe3nSzWZKIJk8WO5ndXva2NE2sLboQX7oU9u71f9+ICPGG62L87LPL90DXFN6DU155RZpRy2PdOrSRI8nJzyczIhIcQWgVLFYA4pPXNBxDhxBtMmGZ8aHc3qGDCPpTvbmvf39ZPenRQ5JzKoPHIyuo+vh4m01OJNu2rd4+TZkCb70ll51ONUtCofBCCe8AoWkaq1atYtOmTVUW3MWxms3Eb93KqLfewuZ2G1FkAwZI/mp0NSt1ilOLf/9bkjVADqSn4knZjvnww22QnwaFOVTYfmKyiLWkcTcY+wHE96jR3eTYMTn5+eADsf54Excn6SeTJ1fOInb4sK8Q37zZf0643W40bA4aJHaXyMjqvZ6K8sQT8PjjYoU5dKjsKj/AunW4Ro4izWTGabOhVceHbTZjcrmIzEgnxFmIafny06Opz9vas2NH5Ua9r1xZclhbIKbdXn+9NCQHBYmdRaFQnEQJ7wCgaRqLFy9mz549uFyBTWCwFBYSmZzM2Jdewp6fD+edJ+klp0IWrSKwdOwontrOnY2l9lMRTYN9f8CyF2HPb1LB1jzg9PZIm2QIj+YBNMkGH3CnCO/a3tfVqyUR5fPPS8bj9ekjiShXXln5E+m0NPGG60J81Spj8qA3JpNUQ7194jVxUqZpkmKyYwcMHWrkyZfGunUUnHceKVY7mskUMN+6yePBkdCBmN9/w2QL4IpGfWX/fom4BPFWP/xwxR+rW0IcDjlJ0y0n1U0imTBBjlNxcaVbphSK0xQlvAPAsmXL2LFjR8BFt47Z6STm8GHG7t2LdebM8pepFacfhw7JkjFI+s3bb9ft/tQWbicc3wpH1sCJbdKEaTJDUIxUtZueBdHt6keiRW6uxD3OmGEIHB2HQ2IgJ0+W6Luq7G9enggmXYgvX+6bp+xN+/a+PvGEhOoL33XrjJzsd9+Fm28u/b7r15M/ahSpVnv1qtylERSEo19fYmd+dHqI73795G/fs6eR418emia2oJ07ZTDU009L1rymweDB8McfVf+fGD1arJBt2pRukVIoTlOU8K4m+/btY+HChTUmunWsbjddevZkwMCBNfo8igbKJ58YE0/nzPEdKa2of+zdK82YM2aUzO1v3VoiCydNMiqZVcHlkhkB3vaUY8f837dJExHguhjv2bNkw6YzD5I2QMqOIp+9CRzh0Lg7NOoKDz4ML74oPuGkJIiJ8f9c69dTOHIkJ2yOmhHdOsHBBI8+j5i33qy556gvvPSSRFWC+OwTEsp/zJYt0isEsiJzww1y4jdjhtw2b57MjagK55wjJ35nnCH/gwqF4iRKeFeD/Px8vvzySwr12KoaxmKxMGbMGBr7i9BSnN54HzCPH/cdY66ov3g8kr0+YwZ89VXJSZbDh8vfdvx4CA6u3nNpmlQ3vYX4nj3+7xseLg2b55wBzY9C+lLIPiJpMJoHPEX9JmYLYAJXPqRosMIF8aNh3s/+t7t+PdqIESRb7bgtlhqPRTQFBxP9xusEX3B+jT5PnbNvn9EQ+fTTRj53WTz1FDz6qKyuJCfLZ8bhwyLa8/LEsrZpU9UaI3v0kMcOGCACXKFQnEQJ72qwcOFC9uzZg8fjqbXnDA8P54orrjiZA65QAFIZ3b+/askGivpBWprkKH/4ofjCvYmMFB/45MniCw/U+//IESNLfOlSqU5qGsQC/wCKkuqoqFujEHDYoMeVMOpFCPMqEqxfDyNGkO5ykxMSWmv2H1NEBE2WLcVSWgX+VKFvX/H59+oltp/y6N1b7jdsmO/grUceEfEOMH06/Otfld+Xdu1kVWfUKLGcKBSKk9QD42PDJC8vj71799aq6Aapsh+qyOQ2xenD3r0iuqFhT6s83YmOFn/+qlVy8nTHHRAbKz/LyBAR1LevLN+/8krptpHK0KyZDPuZNk2E8Ynj8N71cJsV2iGCuzIWaTugOWHTFzAtATbPltuLRLcrI4Oc0NoT3QBafj4ZU5+uteerMyZOlO/r18OuXWXfd98+Q5xfconvz+67T0bHgwxt855CWVH03oJTPcpRoagCSnhXke3bt9fJ8zqdTjaoiqbCG+9q1fDhdbcfisDRowe8+qpUpOfOhQsvNMTqli0yrbB5c7GgfP+9b2Z4Vck5AbNHQvJcMLmqd3TwOKXRdd4N8O5IGDUcUlPJDg0DSy1nOhcWkjfvOzxVEZANCb2vww7MehaWPgefng9vdoXX28O0zvDBIPj1QZj9BOiaeNw43+1ERBhTkpOTxT9eWfTftRoXr1CUQFlNqsgnn3xCXl4ezzzzzMnbnE4nVqv1pA3koosuIjU1laVLl2KxWDCbzTRq1IjzzjuPlkUJFPv27ePrr7/mrrvuOrmdXbt2sXTpUpKSkrBarTRq1IgBAwbQqWgwhcVi4corrzQmWypOb665RjKjzWZISVFT4k5VjhyRbOQPPxSvtjfx8dJce8MN4s2tLNnH4L2+kHUUPAHuWXECh0D7DI62aFvh4Th70HgcjZmYMFFxa00hGlej8S4mooseZwoJIeLBBwibfENVXkHDIGkDPDIcmqQCZnBY5ASoOCYzFJrA5IaMcPi/WdB+tO8qhNMpjZeJiRASIhX0ikZQulzSYAtw663GIJ0GiqZpHE3PY/uRTLYcyuBIWh4FLjd2q5nGEUF0axFJl2aRtIgJwWxWFlBF+SjhXQVyc3P5/PPPS9hMXnvtNcaOHUs7rxHuf/zxB6mpqYwfPx6Px8OiRYvYsGHDSaFdXHhv3bqVefPmMXr0aLp27YrD4WD//v1s3LiRsUUd5jabjWHDhtGmOokHilMDTZPK59Gj4v1dtaqu90hR02gaLFsmDZmzZkFOju/PBw4UL/hll1Ws4liQDe/0hvR9/oVaOVz6fSxbU2ysuyYJR1EQyh1/RPHt7mBsZg27Bc6IcfJE6yzC/27COx43h4Ani5XU++LhK0y0LBLL9+NhBCbOK7r+KB5WAflADHAtJsYV/WwNGu+hMb1omx+jkYrGHV7PYU1IoMkfxupQ8+bN+fPPP2nrNaVx1qxZ3HXXXbz99ttcfPHFlf5d1AmZh+Gb6+DgX+AqACppf7SHQXAsTJwFLfoZt3/zjayogERDvvtuxbaXkWGc/N93nzE6voGRml3AvDWHmLViP/lON2aTiTyn22dmlQkItlvwaGAxm7jkrBZc2q8V8VHVbIRWnNIoq0kVOHHiBJbiUVsVwGw206NHD7KyssgpfrBEzqx/+eUXBg8eTO/evQkKCsJkMtGmTZuTohuksp6cnFyt16A4RUhMFNENyt99umAySfTfBx9IbN+HH8p1neXL4aabpAo+aRIsWeJ/wqXOT/8HmQerJLoPZllYmWTHZIIF+33H1N/SI5udNySx+qpk4kLc3L0rguCOmaVsyZcTaKwBhnjddj0m5mFiEWZexsR0NLaVMrl0NPADUv3Wce3bV261fc6cOURFRTF37twK7Wedommwbga82Rn2LwFXHpUW3QCF2ZCxHz4aBj/dAc58uX3cOIkFBP8TWEvDOzu+AVpNcvJdTP1mExe/soQZS/aQnusk3+kht9Bd4m2kAbmFbvKdbnIKXMxauZ/Lpv3JfZ+vJTW7wO/2FQolvKvAiRMnqpTb7Xa72bBhA8HBwQT7iQZLSUkhMzOTrl27lrutpKSkSj+/4hRE+btPb8LCxF6ydKlMjHzgAcMSkJsrWeFDhshU02eekbg4b/YshC2zJA6wCszdGUzvxk4mJuQyJ9G/9S3YqjGuQx470q1E9s/EbCt/kXUl0AlweFlM2mPCrltHir5KazNvgokIYLPXbSa7HVfizlIeAYcOHWLFihW88MILLF68mGOBaF6tKTwe+P5W+HGKCGdPADz+rjxY8y58eA7kZ8gJ3osvGs93//0V2463l76BNVeu3H2CCa8vYcHmJJxuD4Wuyp3ION0ahS4Py3edYOIbS/l181GUqUBRHCW8q0Bubm6l3kxbtmz5f/bOOzyKcu3D92xLNr2HEEoICSX0phRpIkgRVMCCCvZ+PJ9HQLCXI1bsFSsi6qGIIBZQRIpU6b2FkoSeXjbZOt8fb5LNkoS03U2A976uvTI7+868M5tk95lnfs/v4dVXX2XatGls2bKFG2+8EU0FVf0mkwkQloFVUVRUuy9KyUVGSVtunc416ym59GjVCl55RTTk+eUX0ba7RGt76JBoJd6sGQwfLgo2CwtgwW3FzXBqx/yDflyfYGJ0QiEr03w4ayr/uVZgVfjxkJH24VYUjYpv88Iq95uMWupkWJbXcNAXBzegEgEU52PphlIqMykhDjg3zLbu3VvpnPPmzaNTp06MGDGCxMREFixYUOVx1guqCovvgx2z6vS7qxBbIZzZDV9eISRIvXo5izZ/+cX5eXM+yma8L5DAW1VVvlhxiCnfbyXbZK1xwH0uNrtKgdnOtIW7efWnPTgcMviWOJGBdy2w2+01Gt+uXTumTp3KpEmTiIqK4sSJExWOKymWzKtG9b23bQwlDRCHQ7R1BmEzd4F8yUk8jE7nDK6PHxfuKB06iNccDvjtN2E91y8G8tJrPc3GUwbS8rSMii+iY6SV5kE2fkx23smbsTOAtl83os+cKAqsGt4ekI2iBW2AnWWoXInD5VGWPKCi/PkUNKxA4VMUBqBgOM/x+RXvpwTVbsdRgcSvhPnz53NdscPHdddd13DlJmteh13/c3/QXYLdDBkH4X/XiiD/lVecTXQmTxZ/Q+fjAgy8P/nzIN/8fYQiq3u/V4usdpbuPMELC3bKzLekFBl414La6LtBBNbXXHMNK1eurDC4Dg8PJygoiL3nycrU9RgkFxG7d4sulSD13ZKKiYwUfuDbt4vC24ceEs14ADrkgVJzXXcJ8w4Y6dfETJivCFauTyh0kZvc3yGfvbefYtttp5l5dSZxQc6ExbAIK8vRuDzKEghUFlZqUeiMwhlUfjjP8ZmK9wNwEw76F5no8vxzbNiwodzYf/75h9TU1NKCyuuvv559+/axa9eucmPrlbN7YcULYK38AsIt2M2QtkFoyBMShL88wObNosnT+bjANN7zN6YUF1B6JplVZHWwct9pPvzjgEf2L7nwkIF3LQgKCqpQKlIdIiIiaNmyJWvWrCn3mqIoXH311axatYqtW7diNptRVZWUlBQWL17sMtbf379W80suIsre9pX6bsn5UBThevPhh3DyJJcntSXujxgyi1w/x4YsiCT2s8ak5mkZ+3M43+1zzTuvPWGg23fRFNpgUbKR5ak+tPs6ms6zo/lsZwB7MvW0ntmIjDL7XZHqw+jF4bSa2YgO30Qz95CRs1oVKimMBEhEIaWKU7IDaefZx1EgsXh5DhpWGf3Z9vIrXH755eXGzps3D1VVGTJkCJ07d+aaa64pXd9gcNhh3o211uPXGGuBKLzNSROt5YOCxPonn4TzSR0vII13akYB7/++32NBdwlFVgfzNqSwKzXbo/NILgxk4F0LIiIi6pRx7t27N1u2bKnQ2SQpKYmxY8eybds23nzzTaZPn87y5ctLPbxBuKM0bty41vNLLhJKCisNBqHFlEiqg9EIPlqaBqksPOSUhuzN1FFoq54P8dKjRvRaeKJHHiG+KouvPcuKG84QbLDTIshGeHEW/OfDvtz/ZyhjEwvZdMtptt92mt4xFk6atGgDK5fsXQbsA8zFgXUmKr+jYkLFjso6VH4HelTi730GlRygfdmVOh26+BblxhYVFbF48WJef/11fv/999LHSy+9xMKFC2tVSO8R9vwgLB/Pc7Hhduxm+OtZiIiAJ54Q644dgw8+KB1iszs4cDKXnzan8friPTx9zIenxjzBtJGPMueMlu0pWRRaGsh7WAaHQ+Xpedux1lHPXV3MNgdPzduO2Vozqark4sPLLcQuDiIiIirUeT/66KPl1g0YMKDcuiZNmvDkk08CInNdtnkOQEJCAgkJCZXOr9VqiSxp6Su5NLHbYeVKsdyrlwimJJLqYjMzpmUB8w8auau9SADMO+DH2EQTr28KqnLzeQeN3NjKxMOd8lmR5sPsvf60DLGh18DJAi2JoTZUFV5YH8SjXfK4pY1TONI00I7NbkWvKtjzKv4KCkehOyqrgMEIB5MfUHkVEXY2Ah5DoV8lgfdSYASUuqAAqIWF6Nu3Lz926VJ8fX0ZO3Ys+pJiVOCmm27ijTfe4K+//mLw4MFVvice5+/XhIOJN3HYhJ582Lvwf/8HH30Eqamo06axZ/iNfLsrk9X7zqDXanCoanHm2B/aCSNIw54C9Ac3Y7Y6SIoNZnzfFvROjETbABrN/LnnFCnpJrxZ95htsjB3wzHGXxFf9WDJRYsMvGuBwWAgMDCQnJycepnf4XAQFRVVL3NLGgjbtkF28W1LKTOR1BS7ma7RZn44ZORglo74YBuLDhtZNDK9WoH3t8MyS5en98tm6IJINAp8elUmV8SKzpeHsnW8vz2QES1cZQkTu+WhOiB/RwBFR5xSlo3n3IC9D4UXULkKCEVhRjW7V1pQ+Qm13HhNRASaMhK948XWii1atKiwWY7RaGR3db2rPc2Z3ZBede2PR1A0sH0WXPYwvPQSu558mZdGTeTU/3Zi1uhQVbBWYjhgcYDFLF7bkZrNc/N3YNBpeGxYGwZ3iCnt8lwfzFp9hEIvZ5/NVgffrzvGrb1byC6XlzAy8K4lnTp1Yu3atV6/DakoCi1atMBgOF89v+Sip6y+WxZWSmqKKgKOMQkm5h800jPGQmKIjUb+NQ9EmgTYifa3U2hT6BnjbDefVazzjvIrv09FAxrj+W/xx6PwdQ1axZdgQGHeudvp9RhHXlPjfTUY9i4Ah5WNpwy8tCGIA1k6tBpIU8+8+gAAIABJREFUCLHxQq8c9mfpeGxlKPe0z+eFXs4mRUuP+nLXH2HckGjinQHZpOZp6fm/aPx0ru/99H7ZXNuyyKXjaAnNg+wsi5qFucsDfBTRnYV3vIFZW/z9U8Nssclix2Sx8/JPu/l1+0meub494QE+tX5basuh03mkpHu4QLUSiix2Nh7OoGdCRL3ML6l/ZOBdSxISEli7dq3X59VqtXTs2NHr80oaGCX6bqMRKigWk0jOS3HANDaxkNE/h5OSp2NsoquPiFaBc2vObA4FveIabX2wLYBQHwf+OoVPdgTwr85CDhFarPM+Y9LSLKii4NuL9/g1GgLuutN787mboyvJK7Jz+9IIXumTw8j4QiwO2HjKB0Px+xgXZOPnw0aeuTwXXfHNg3kHjcQHl08O7b39VOmYc3mwYz5Teri6bqmnCxj/0RpO55qdQXcdKLI62HQ4g5ve/5sP7+hB65iq77K4k1GD+5ObnVl8BajFJzqB0D4T0AWEkbHyC0zJ61E0zvBIFxRJo9EvYstL5+Scx2ly12comqrrvDJXz8R8aj+2nDOE9bsT/1ZXYLLYWbwlTQbelzAy8K4lOp2Ojh07smPHDq9lvRVFISIigogI+Q97SWO1ik6FIJrmyLsfkppSfIu/SaCdZoF2lqf68GY/V8eF2AA7aXmuwUVKnpYmZYoiD2Tp+GRHAD9fl47FDtcvjmB4i0Lig+20DLbR2N/Gr0d9eaBj+eyiw+a92n5Dl87omjXz2nxu59Q2DueIr+vrEkQDIqMG+jcRbcl3ZuiJNNrx16usSPPhqmZmsooUNp02MCaxkIzCur3XZpsdR9ZRzGp03c6jDDaHSn6RjQe/3MhHd/agTeNgt+27Kiw2BxFD/o1vbDtUm5Wstd+Qve5bIgY/AkBQx2EEdx9d53n04U3xi7+M7H9c3XF2p9WPTFXSMJCuJnWga9eupU1vvIFWq+VKqeeVbNrk9MqVfw+S2lAmmze9XzZzR2Tgd04r91Hxhcw54MfWM3pUFZKztXy2y59R8SLwc6gwcVUID3bKJyHERlK4jbvaFfD46hBUVcT2z/XM5Z0tgczZbyTPouBQReOdySuDsed4Ke/j60vwCy94Zy5PYC2Coizig21oFPi/FSEsT/Uh21xehjM2sZD5B8V30qLDRoY0LyrNiNfpENARq6bWeT8VYbLYeXjmJk5mV93R1B2oqoq1TEWlotNjbNEda1bFje3qQmDSIHxjk1C0epf1Z/PMDdLpReIdZMa7Dmg0GgYPHszChQtr3M2ypuh0Onr16kVAA/dFlXgBqe+W1BWdL2gNgKW4sU35z68BTc08eVkuj60M4USBlgijg3GtTdzWVkhSPt/lT6FN4aFOTqeNR7vmMWRBJN/t9+PWNiauiS/CX5/Fe1sDeHptML46ldahNu5PMmHZ7YXPMqORgHvuxtC+nefn8hS2QtDoCDRYWDgynQ+3BzB5dQhnTRqubGrmjb7OOxVD4wp5fn0QuRaF+Qf9eO7yXJanltdQd/imkcvzn0alkxgqAsEZOwOYucdZhDqkeRHT+lvwxeyhE4SiQjPPPP8dn237BkWrFZ0yS36WXa7rT52OLPSiI2cxDpsZ0+GNGKK85zTiq9dyLN1Em8beldhIGgYy8K4j4eHh9O3bl9WrV3ss+NbpdMTHx9OmTRuP7F9ygVGi7w4MhG7d6vdYJBckGxZ9Bt+N4NxYSqeB4/c6M383ty7k5tYVZyLv61DAfR1cJSQ+Wlh5w1mXdQObmhnY1HUiFS2n/gkELHgMnQ5dTAxBj/3Hc3N4BWdmOzHUxjsDRKB9KFvHI3+F8Nz6oFLJiVEHg5qaeXdrIFlFGno0slQYeO8cX7nG+/4O5TXe+aofDg/eILdrtBwyhrMgP4Axm3/x2DwAhSGNIDKC9D8+AI0G1WpBYwwgcujE0jG5O5eQt+fP0ufG5l0I73+P245BUUQ7ecmliQy83UCrVq2w2WysW7fO7cG3TqejRYsW9O/fv16tlyQNBLMZSrqe9usnsjgSSU1p3B1U7zQOqQglKgm/22+l4PMvUAs9IDHQatGEhRIxbw7KhV4DYfAXftrnkBBi48ZWJmbv9S8NvAHGJpq48ZdwHuuaV26b2qKiYMKzvQKKDEbeH/YAQ5v44G8tAptN9Ctw509ALf4ejRj8L6HxdjgoPLaVs7+8RqMxLwEQ1GFojTXeaTMfLF1uNPYldAHhlQ9WheRFcmkiv7XdRFJSEn5+fvz111/Y7XYcjrp/qWm1Wjp16kTXrl1l0C0RrF/vbNcs9d2S2qL3hS53wz8fgcPq3bkNAXDFFII63IKi0ZD/6WfuDb4NBjQREUQu/BFto0ZVj2/oaPUQ2JhDKadYluLDqPhCGgc4OJ6vYWGyH12jXO8a9Iqx8P3wDNqHu+/36oOFw8S5bX+VT+TLkiffZsxlbiiEtVrh6FE4eBAOHRKPAwcwpp0Bi/O9UTQa/Fp0I2vN15hPH6z1dE3u+LjaY1WE3ERyaSIDbzcSFxfHzTffzIoVKzh58mSt3U50Oh3+/v4MHjyYsLAwNx+l5IJG6rsl7uLyR2DzDO8H3iiQNBZFUQh6fDL6pCSyJk5CNZtFsFSXPRuN+AwcQOjrr6EJDXXT8TYAGnfH/+RPbD1j4LOdAeRYFIINKoOaFfHM5bn8etS3dKiiQN/Y80t42n7tekEyqVse9xc7z3y8I4DPdzk13j5alTXjTWQR4sYTqpgiq53Za44yukfT6iWbLJbywXXJ8tGjItN9DmEAic4OpqqqUpSyDYfZhD4khsKU7VVOq9ptUDa5ptWiKOWlOKrdJvTkKqgOO6rNClotRVY7zcL9y42XXBooqrzf4RGOHz/O9u3bOXnyJECVEhRFUdBqtQQGBtK5c2datmyJRiNNZyTn0L8/rFoFoaGQng7yb0RSF+bdBPsWgd1zhXMu6P1h4AvQe6LLantGBlmPTcLy99+oVmuFAdP5UPz9Qacj9J23MA4Z4s4jbhisfw+WTQFbUdVj3cjYn8PZk6Hns1sbMVH7KgAZK79A5x9KcPfRpb7W+vBmNLr++dLt7EV5nPjuMbR+ITS++Y0K923JTCNnwxws6cdwmPNpes+XgMgEf3lfT+KjiotvLRY4cqTi4PrYsZr9rQQHQ2IiLXMKMRflCx9vBbQBEQR1Go5/Qq8KfbwVrZ7Y8e+Vnu+5RA6biG9s+QLeMz+/hvnUftexwx+nSetO/DxpQPWPW3JRITPeHiI2NpbY2Fjy8/NJTk7m5MmTnD17lqKiotKA2uFwoNfrCQ8PJyYmhri4OOnRLakckwnWrRPLAwbIoFtSd675BJL/8E7grWggPBF6PlruJW14OBFff4V1/37yP/sc048LUfR6IUGp6M6hqqKoKmg0aNu0IfBfD2EcPhzFx/tdED2OqkJ6lAhAvfgvn5qnZcMpA4EGlQ+OtYPzmH6oNjOWzDQMYU0AMB1ajy4wQmR8K0HRaDHG9yAg6UrS/3jfud5qYe+rHxC/a7kzuK6JdDMkBBITISGh/M/wcFAU/r1kH/9bf4yK0o7h/e8mvP/dFe5aFxhReoFQHaKumVLh+rbSzeSSRgbeHiYgIIBOnTrRqVMnAGw2GzabDVVV0el06PX6KvYgkRSzdq3zVrzUd0vcgTEURn8D824Eq6nq8XVB5ws3zIXzdPzTt25N6PQ3CH7heSybN2PZvgPLunXYjqWIwmK7HeV4GgaLBR+LBb2qov9j6cV5EWqxwPffw1tvwY4dMA5IwGvB9/yDRrpGWWkXqTL3YDoR5wm8/RJ6Yzq4FsPlNwJQcGgt/gm9yd+/qtJt9CEx6ENisOacdllfiJad+08w4o8/Kp8wLEwE0pUF11VwVftG/LgprV6cRfwMWkZ0ifX6vJKGgwy8vYxOp0MnnSgktaHERhCkvlviPlqNgL5Pweppngu+dUYRdIcnVmu4xt8f33798O3XDx75l/OFjAw4967g2bMQ7b6OivVOVhZ8+im89x6cKNPUZQ3QUktFnuueYP5BP+5sX8SJqP4ULtqG3ZSD1q/i7pL+Cb048/MrBPcYiy3nJKrVLHyxzxN4n489sa1FAF02oC67XMfap6TYYKKCfEjJ8PDFZgXotBquaBXp9XklDQcZAUokFwolhZVRUZCUVL/HIrm46PckOOyw5lX3B986I4z5VgT4dSUsTEgJssu0t09NvTgC7yNH4J134IsvoKCMP7qPD9x+O/znP7B1Khz6Dewe9D9HdBdNy9PSP17Hvb53owt6EVPyBgI7VKyf1/qHogtuRNGJPZhP7MU/oVed5i9s207UsHgIRVGY0Dee6b/s9WrW26DTcMNlzdBpL8I7NJJqI3/7EsmFQF4e/POPWB44UFgXSCTuZMAzMOpzYfencYMEzqGFgBiY8Ae0vb7u+wPxd9+ypeu6VM+0MvcaGzfCjTeKTO577zmD7ogIeO45SEmBGTOgTRu49gtxIeNh5h0wckUTK9N9n6QIX/xa9qTg4JrzbuOf2BvTgb8xHd6IX2Jvl9cKDq0jbeaDpM18kLNL3qpyftUL0qGhHWOICvLBm5+kfgYtt/SO8+KMkoaIzHhLJBcCq1c7q/elzETiKTqMg7gB8OMESF1XnP2uofGVHXAAOxzwxlJo1sG9x5iQAJs3O5+npbl3/97A4YDFi2H6dPj7b9fXWrWCxx6DCRPAeE6Q7Rcu7h7MvUG0kvcAhTb46bARi6rD9u3ngLDFUy0mLBkplW5njOtO1tpvMUTEoQsIx5ZzqvQ1/4ReNcqC+3jB41qn1TDtxs7c89l6zDbPN5Py0Wt4cWwn/H1l2HWpI/8CJJILgbL6bllYKfEkgTEw/ndIWQNr34Dk30HRgrXgPBspIlOu2iG/KXy9H9JVCHwFvvvOvceXkFA8pSIcPy6kjLfJBLNmiYLJg+c0a+nXDyZOhGuuOX+xaKsRMPx9+PURjwTfv6YEoWp0RFz/CmicXT8zln9MwcG1lW6n0fsQNXwyGp/q+VOrqgp2W2lXTtVmBUVY9yVEB9btJKpJYqNAJvSN55u/j3hUcuKr1zC4fQyXtay68FNy8SMDb4nkQqBE3x0b6ww8JBJPoSjQ/ArxyD8NB38VgXjqGsg+JjTGiiKcSsJbQbO+0LQXJI6AM1nwQSJgFq4cjz8OnTu779hKpCYlXnAXQuB95gx8+KF4ZGQ412s0cMMNIuDu0aP6++t6N2gNsPj+Ym9vN7Xj0PgyZ384hpZ90Aa4FgAGJF1J1rrv8G1ceX2JIbJFtaey52e4eGKnzbwfbUA4Lca/Sedm3mt+dFf/eE5mm1i26xRFVvdnvn10Gjo1C2XKSFmXIxHIBjoSSUMnM1PoPVUVxo8XGTOJpCEzebKQUQAMHQq//ea+fa9eLbLDJfTpU16u0VDYt09kt2fNEnaIJfj7w733wv/9H8TF1X7/p3bAvBsgN61uRbG24scisB3QMuCJhdi09WN162fQ8uEdPWgbW7GDiidwOFSm/7KHX7efdGvm21evpUd8GC/f2Bm9TpbUSQTyL0EiaeisWuXM7kl9t+RCYOpUCCpuErJkCaxY4b59n3vHp6FlvFVVnO/IkdC2LXz2mTPobtwYXntN6NLffrtuQTdAo47w0C7oPVlIfQwBNdveWvzYB7wnfupUBz2yjqK4K4teQww6Da1ivNtgRqNRmHxNElNHJuFn0KLT1K3kUqMITfcDgxJ47eYuMuiWuCD/GiSSho7Ud0suNMLDYUqZrn1PPEGFbQJrQ6NG4OfnfH78eI1bzHsEq1VIa3r0EBfIP//sfK1jR5H1PnJESG9CQtw3r1YPA5+HyWdg2PsQ0Va40vgEgfacTp4OoAiR3c4HVgPvAksDoEQu/tpr3DZpHL4G7ytRfXQaburZHG0dA9/aoCgKQzs1Zt6/+9I9PgwfnabGAbhGEXrutrHBfPtQH27uFYemHs5F0rCRUhOJpKHToQPs2gUtWsDhw/V9NBJJ9SgoENnpU8XuFj/+CNdd5559d+wIO3c6n6elifqH+iA3Fz7/HN59V1j/leXqq2HSJBg0yHsWoGYzrFgKy7+DAyvAfBr0iGC7ADgBnNVCpyuEDGjoUJGZ79QJ9u+HkBDU/fu5bvZeTucUeeeYizHoNCz8Tz/CAnyqHuxhjqUXMGf9MX7dlIK20IRNo8Vs8C03zqDToNcq2OwqA5OiGdc7jtZezthLLixk4C2RNGTOnHE2B7n7bvEFL5FcKHz8MTz0kFhu21YEy1o3WMWNHi0C+RLWrYOePeu+35qQmip8tz/9VATfJej1cOutwhKwg5utFCvjyBEh6fntN3GHrKACB5omTWDYMPG48koIPkdD/ccfMKS4Qc5dd7Fmyqs8NXebRwoOK8JXL7LdD17VyivzVRfzxMkcnLOYfTGJbL/+dk5FNcVic6DXKkQE+tC5eShtGgfTOiYQYz3cJZBceMi/EomkIVNWGyv13ZILjXvugTffhORk2LtXyC3uvLPu+z1X5+1NL++tW8U5zZkDNptzfWgoPPAA/OtfQsvtSQoLYeVKZ7B94ED5MQYD9O0rAu2hQ0W32/Nl3QcPhrFjYf58+PJL+tx7L31aRbJq31msds8H3xGBvtwzoOE5NvmsW0P74/tpf3w/Y2e8AF271vchSS5wZOAtkTRkSmwEQQbekgsPvR5eegnGjRPPn3tOLPuWv2VfI7zdvVJVRZA7fbprzQUICdh//iMuKAJqWNxYk/kPHnQG2itWQFEFMpAWLZyB9sCBNT+et96CX38VfuMPP8zUVWvYcjSLbJPFbRL9ivDVa5h2Y6eGV4RoNsOWLWLZaBQSJ4mkjjSwv3KJROJCyZd869aez6JJJJ7gxhudPt6pqfDRR3Xfp7ecTcxm+PJLaN8ehg93DbovvxzmzRMB8SOPuD/oLigQBZoPPyzOt3VrYT+4ZIkz6Pb1FUH2u+8KfXZysvAKHzmydsfTtCk884xY3rKFwFlf8cldl+Hvo/NYa3UfnYaXbujUMHXR27c7HWl69ACdzFVK6o7UeEskDZXjx4UuE+DBB90TsEgk9cGSJSITC8LxJDm5vMa4Jhw75mrFN3asCILdRUYGfPIJvP8+nD7tXK8ookB04kTo3du9BZOqKuQ4v/0m3q9Vq8BiKT+uVStnVrt///Jt5euKxSK06QcOCPnM/v0cxcj9X26koMiGzeGekEFBWO5Nu7EzfVpFVjm+Xnj3XXj0UbH8+OPCClIiqSPy8k0iaahImYnkYuHqq2HAACGRyMgQko3//rf2+2vSRMhYrFbx3F0Z7+Rk4a/95ZdCR12C0SikJI8+ComJ7pkLRFHmn3+KQHvJkvKuKCCsEwcNcjqQxMe7b/6KMBjEBcfVV0NWFjzxBHGff853D/fhxQU72Z6SXecmM756DZFBvrx8Y2cSG3mnPXytWL/eudyrV/0dh+SiQma8JZKGyt13iwAAhLtJZAPNCkkk1WH9emfw4ucnrDFLHHtqQ5s2Ql4BQoZ1/Hjt97V2rSiY/PFHV7/xqCghI3ngAdE9tq6oKuzY4cxqr1njWqBZQrt2IsgeNgyuuAJ86sFeb+xY+OEHsbx+PVx+Oaqq8sfOk7z2814cqkqhpWYBuI/DhmowcFufOO7q3xKdtoGrXVu0gKNHxfKJExATU6+HI7k4kIG3RNJQKfnQ79BBfFlLJBc6ZW0AH34YPvig9vu65hr45RexrChCi6uvQZtzux0WLhQB97p1rq+1bSvkJLfeWvdC0KwsYdVXktU+ebL8mMBA4SoydKjINDdrVrc53UFKingfTCbo1g02bCi1grTYHCzfc4pZq49wPNOEVqtQaLFXWIBp1GvBYsYvP4dxmxczct6HBEeFeflkasGpU85Au3lzZwAukdQRKTWRSBoiR444P+ilzERysTBtGixaBA4HzJghvK5rK50o62yiqiIj2bx51dsVFMBXXwlJybkNqQYOFA1vhg4FTS2zsQ6HcMIocSBZv16sO5fOnZ1Z7V69anbR4A2aNYOnnhKPzZvhs89E5h/RNGZox8YM7diYE1km9p7IZVdaNrvTcigw21BVISdJjA6iQ7MQ2uQcJ37gSDSo8NtQuP32ej65aiBlJhIPIQNviaQhUlbfLdvESy4W2raFO+4QEiqbDZ59FmbPrt2+KvLyPl/gffKkyLB//LHIQpeg1cJNN4kMd209mtPTYelSEWwvXQpnz5YfExIiGtQMGyay2heCbGHiRJg5Uzi3PPmkkJ+cI7lpHOpH41A/BrVrVPl+1MbQLgl274ZvvrnwAm9vN2eSXNQ0cIGVRHKJUhJ4Kwr061e/xyKRuJPnn3dqlr/7Tli21YbqWgru2gV33SVcUF5+2Rl0BwaKwPLwYfj225oF3Xa7kKc89xxcdpnQgt92m7iIKBt0d+8u7PnWrBHr58wRFx4XQtAN4vf0/vtiubjQslYoCowfL5aXL/duw6PaUlZ+JANviRuRGm+JpKGhqsK14cQJoa3ctKm+j0giOT+qCjkp4mEzg0YHxjCIaAM6Q/nxkyYJbTUIf+wSrXZNOHBAeFuX8PrrMHmy83j+/FPMsWSJ63ZNmwp3knvugaAaeEefOiWy2b/9Br//7po1LyEiQmSzhw4V2e2oqJqfV0NkzBhYsEAE0OvXi4uNmpKWJuQrqips+R5/3P3H6S5sNvG3UVgoLj5yc4Xbi0TiBmTgLZE0NMoGFJMmwRtv1O/xSCQVYcqArV/B3h/gzC5w2EFbJjhRHWArgpA4aDEQuj8IMcWNdDIyhLY7N1c8X7my5nd2LBZh81ein37kEWFTOGeOCLjPzaR37Soy3DfcUD09tdUqsp4lRZFbt5Yfo9GIRjolvtrdutVeG96QSUkRLjKFhSKLv359aaFljRg0SGS827WDnTvd64PuTrZudd4B6dVLuN5IJG5CarwlkoZG2e54Ut8taWic3Qsrnod9i0DRgK2M33XZ5RIyD0LWYdgxG0Ljod8z0O4GkfF8+mkxZupUIceoSSBmMAhN95Ej4vny5SKYP9dWcMQIcQHbv3/V+09LcxZFLlvmvDAoS6NGTk/twYMh7AJw6KgrJYWWTz8t7sB9/jncf3/N9zNhgvg97d4N27ZBly7uP1Z3IGUmEg8iM94SSUPjpptg7lyRUcrKElpUiaS+sdtgzWuwahrYzSKjXRv0/hB7GQydAZ37OjtDLlwI115bs31dcYUI2M/FYBBB3mOPiYLOyjCbxfYlvtq7dpUfo9VCnz5OB5KOHS/OrHZVmM3Qvj0cOiQuNg4cEF1Ia0JenvBuLyyE//wH3nrLM8daVyZMEEWgID6Lb7ihfo9HclEhA2+JpCGhquKL6exZeYtT0nDIPQ7fXA3ZR8Bqqvv+NHrQ+YDvOJj4mViXlCT86qsjYdi0SchK5s51bXgTHg4PPSQ8witrznP0qDPQ/vNPYS94Lk2aOAPtQYPq1t7+YmLJEvGeANx3n7CErCm33iqKaqOjxR0GXQO88Z6YKC4wQMhsmjat3+ORXFTIwFsiaUjs2iUa5oCw75o2rX6PRyLJOgKf94TCTHBU0GWxLuiMsMof/kgXz7/6Srh+VITDIYowp0+HVasqHpOZCaGhruuKioSGvERCUtLtsix6vdCYl0hI2rVruPrj+qakCZKiiKY6PXrUbPuywfuvvzqXGwrp6c4uwbGxF4YDi+SCogFeakoklzDSv1vSkMg7BV/0AVN67aUl58NWCH3skAbsRdjz3Xyza7fIwkJx2/+tt8oHzW3bwt69zudZWcIv+9AhZ1Z7xQqxj3OJixNB37BhonFOQID7z+9i5O23xftaWCjuLtS00PKqq4RO/tQp8XttaIH3hg3OZanvlniAS1CoJpE0YEoKKw0G6N27fo9FcmmjqjDvRjCd9UzQXTqPBUZrIBhxW//jj8X6s2fhhRdEAeX99zuDbo1GaG7XrxdSk7JMnCj8vVu1gv/7PxF8lwTdPj7C6u+dd2DfPuHf/dFHMHKkDLprQvPm4m4cCMnPF1/UbHudDm65RSwvXCh03w0JWVgp8TBSaiKRNBTsdnGLMytLODCsWFHfRyS5lNn8OSx5FKwVaKDdjgbSHPAFImN9/fXw/fdCJlKCvz/cfbcIqFu0EJnun346f1OXxESn1V///uDn5/EzuSQoKhKSuNoWWpZ1NDmfvKg+uOoqof0H+PtvUVgrkbgRGXhLJA2FLVuEDzCI7n7PPVevhyO5hMk/A+/GeynoLsamgV8csO2c9TEx8O9/w7hx4n+kxFc7JaX8PvR6ZwOboUOhZUuvHPolyW+/ieZHIO5IfPJJ9bdVVeEOs2uXkPmUtVCtT+x2USOQlycy87m5witeInEjUuMtkTQUpL5b0lDYNANUu3fn1DmgP87Au00bIUnQaESg/cwzoqPgufj5ganYaeXOO2vntCGpOcOGCfvHRYvg009FJ9Du3au3bUkL+SlTxJ291NSG4RyyZ49T+tK5swy6JR5BarwlkoZCSdbHaKxdS2aJxB047LDhXdF10tsYgebFyykp8OyzomnLypXOoDswUEhRZsyAY8dEMWYJJ054+4gvbd55RxTCqqootHTUoBbg1ltFAK6q8O23njvGmrB+vXO5V6/6Ow7JRY0MvCWShoDV6rRI69NHFIJJJPVB8u9c/pWeuC9iyCxy/YoYsiCS2M8ak5qn5dEVIcR9EUPiV41KH1f9IGzYdqXraD2zEUdynG4XO87qaft1I1LznOt+PGRk2I8RJH7ViC6zo7lteRgb2xS3nTeV8Qvv1MmZHU1PhwULhI90s2aimLKE1FS3vx2S8xAX5yy0/OefmhVaxsYKj3QQ7iYNQfVaNvCWhZUSDyGlJhJJQ2DzZsjPF8tSZiKpT46uBFWlaaCDhYeM3NVe6Lz3ZuootLl6Wz/YMZ8pPcq7UrSPsHFHuwImrw5h3ogMbCo8tiqESd3yaBooJCwzdvjz4fYAXr0ihwFNzOi1Kn+l+rAgvesJAAAgAElEQVQ01ZfLsIgd9eoF8+dD48aVH29ZHfexY3U7d0nNmTwZvv4akpNFoeuYMaLgsjqMHw/LlgmJR9kal/pCOppIvIDMeEskDYGy+u6BA+vvOCSSYysBlTEJJuYfdGpc5x3wY2xi9btWPtY1jzMmDbP3+fH+1gD89Sp3thNBfK5FYfrmQKb1yWF4iyL89Cp6DQxpbuaZviYIKc56r18vMtzno2zGOzvbNVMu8Ty+vvDee2I5IwOeeqr6244e7XSaKWnRXl9kZzs94aOihHOOROIBZOAtkTQESvTdgYHVL1CSSDzBmd0AdI22kGfVcDBLh90Biw4bGZNQQSOaSvDRwpv9cnh5YxAzdgbwZr9sNMUJ882nDZjtCsPiKtCRG/zh3mvEsqo6pQyVca5ziew06H2GD4dRo8TyjBnC37s6BASI4BuEfWRFxbPeYuNG53LPnrJzqcRjSKmJROIFCnOLSD+UydlDGZzen46lwILqUNH6aAlvFkz0oSIi/KII6tsdRSf/LSX1hKqCJR9R5Uhp1rtnjIXEEBuN/F2dTmbsDGDmHv/S50OaF/HugOzS563DrGgVlVahdhJCnEFVlllDmK8DXUWpH9UB1w6BGcuEndsvv8Dq1dC3b8XHHBwsvL+zi+dNTRUNdCTe5Z134Pffhcf3v/4Fa9cKR5qqGD8eZs+GM2fE9iUWhd5GykwkXkJ+w0skHsJhd5C69QTbftjNmQMZaA1abGYbqt21iOjEjtPsazseNUmLr7+OTj/vo9WV8Rj8DPV05JJLFodrYD02sZDRP4eTkqerUGZyf4eKNd4lvLg+iF4xFran61mU7Mu1LUWGO9THQWaRBpuD8sG36gA/H6EdfuYZsW7qVNHMpLIsZMuWok4CZIFlfdGihdB4P/ecaLv+1Vei4VFVDBokvNpPnoRZs+ov8JaOJhIvIaUmEokHSN16gtl3L+DP6X9zas9ZHDYHVpO1XNBdglXvh03nQ75Zy4ZZW5l1+w9smbsTh92DrbolknPRaF2eNgm00yzQzvJUH4ZXJAs5D6uOG/j9mC+v9c3hlT45PLsumKwiETh3i7Zg0KosOepbfkNFAzofePRRobUFkT1dvLjyydq0cS7LwLv+ePxxiI8Xy1OmQGZm1dtotc4W8osWQU6O546vMhwOcbEAIksv5X4SDyIDb4nEjVhMFv58aw2/v7ySwqwirIU11yzazHbsFjtb5+9i/v/9QlZKdtUbSSTuQFHAN8Rl1fR+2cwdkYGfvvp2byarwpTVITzfK5cwXweDmpnpF2vm+fXBAAQZVCZ1y+OptcEsOepLoU3B6oDlqT68tNYHAhsL/W9JxhuE1tteSVOfsgWWFXW0lHiHcwstn366ettNmCB+FhXBDz945tjOx4EDkJUlljt0EH97EomHkIG3ROIm8tMLmPvIzxxZewybpe5d/2xmO1lpOSyY9Bspm4+74QglkmrQqLPL07ggO50irRUO/XhHgIuPd/tZ0QC8+k8gCSE2RpcpxnyhVy5/pfqwKk141D/QsYDneuby7tYAOnwTTY/vovlqtz9XNyuARl3ERvfd53SX2L278kYrZQPv5ORanLTEbYwYASNHiuVPPnFKgM5Hx47iAfXjbiJlJhIvoqhqQ3Ctl0gubAoyTPzw2K8U5ZpRHe7/l9IatAye0pfm3Zu4fd8SiQsrXoCVL4FaTw4TATEwqUwHym+/hdtuE8vNm8P+/eUbTK1f7wyYWrSAw4e9c6ySijlyBJKSRAb78surV2g5fbrQ9QMcPSp+197igQeEGwvAzJlw++3em1tyySEz3hJJHbGZbSx64neK8jwTdAPYLXaWvb6as8kZHtm/RFJKi0Ggr0B77Q0ULbQc7Lpu3DhnNvTYMZFFPZeyGe9Tpzx3fJLq0aKFKIgFoZ2eObPqbW65xRmce7uFvHQ0kXgRmfGWSOrI359uZP8fyW6Rl1RFYKMAbvpgJFq9turBEkltUFV4pwXk1EMXSL0f3LUGYlzlLvz6q5AwAERECDlJUJDzdVUVjViKigtAc3OFJ76k/igshHbtRPY7IkLcqaiqo+XVVwtLwTZtRDdLb3hp5+UJO0qHA0JDhTZdenhLPIjMeEskdeDU3rPs81LQDVCYVcim73d4ZS7JJYqiQO/JIgj2NqHx5YNugGHDoF8/sZyeDm+95fq6okBsrPO5dDapf4xGZ6FlerproWxljB8vfu7bVz1teE04swc2fgjzboK342BaAPzXF96OhEcdMA64LgrS1osLOYnEQ8jAWyKpA6s+Wo/dS0E3iILLnT/tpSBTtsWWeJDOE0DnZbmJ3g8GvVzxa4oCr77qfP7mm6LhSlnKdrCUgXfD4JprxAOERGjLlvOPv/568C9uyOSOIkubBXbNgU+6wKfd4ffJsHuuuJtjLQC7GRxmCARaAc0OwTdDxB2fDR9CUW7dj0EiOQcZeEskteRscgZ5p/O9P7ECe5cc9P68kksHn0AYPdt7WW+tD7S8GlqPrHxMr15w7bViOT8fpk1zfT0pqXRRTUn1WL2FpIa8+64ohnU44OGHxc/K8Pd3bSFvrdhNp1qkroP3WsJP98CpbWArFI/zYhedW3OOwbIp8HYz2D2/9scgkVSA1HhLJLXkz7fWkLz6aL18wfsEGJgwaywarbx2lniQ+eNg30Kw1ax5To3xDYF/HwK/8POP271bFFo6HKDXw/79mCMbc3RDGicX/s3p3afJ9YvEodUBCopGwT/cj6hW4cS0i6b5ZU0IjPQ//xwS9/P88/DCC2L5yy/hzjsrH7tsGQwuLrBdvNiZMa8uNgv8MRk2f1aNQLsa6P2hxUC47mvwq0KjfiFSmA0ZB8BqEneWDIEQ0QYM9SA1u0SQgbdEUgtUVeXLm/7Hm/9MY2TcGOKDEktfO5qbzKwDn9I6pB03JUwoXX/KdIJP97xL84B4bm9zPwAvbpqCXqMHFHy1viSFdWRwkxFoFBFQf71vBmkFKWgUDRpFIdrYmGHNrqVJeDOGPTuQmKQor5635BLDnA9f9IaM/WC3eGYOvT+M/x2a9a7e+DvvhJkzORvcnB1XPcwRpQmKRsFWdH77Q61BC6pKdNsoOl+fRJMuMSiyiM47nFtoeeCAKGSsCLsdmjWDEyfgxhthzpzqz2MpgNnD4MQm9wTdJWgNwuby7jUQFFv1+IaMzQJ7F8D2WXByCxRmgt4IFP8vqCrYTKKJVZNe0PUeaHFl1XaQkmqjq+8DkEguRHJPnV9i4qfzJy3/GCZbAX46kWHbnrGZcJ+IcmPvT3qUMN8IMovS+Xr/DCJ9o+gaeXnp68OaXUvXyMtwqA5WnljGwiNzeCj4MdKTM2XgLfEsPgFw50r4qh9kHAK7mzPfen8Y91P1g27AMuVp/t5p5HB0NxxmHaqmejUWJbUYJ3ac4syBdMLjQhg08QoCo2SXQo9jNArJyahRzkLLDz6oeKxWC7feCm+8IVrIZ2cL15GqsJmFPvvEFvf/ndotkHccPu8J922GgAvwc7cwC/5+FTZ9AqpDSGpKMFcg6clJEY+Dv4BPEPSeBD0eBp3Be8d8kSIvYSSSWpCenImiqTxbplW0tAltx+7M7QA4VAd7MnfQPrxLpduE+UbQNKA5p0wnK3xdo2hoH9aJs0VnsFsdnNh1um4nIZFUB2Mo3L0OEoa4T/OtM0JAY7jjL4i/stqbHd9xiu+nbeFwbE/sOh9UTe1sNW1FNs4czGDuv35mz1JZL+EVRo50WkJ+/DFs3Vr52BJ3E7MZ5ldTY734ATi51f1BdwkOG+SfhtlDweG9gnq3sH8xvNsS1r8L5lzXoLsqLPmQdwKWPw0fthUXNpI6IQNviaQWZKVmYy08/63tjuFd2Z4hPqSScw4QaYwmUB9U6fj0wjOk5B8lzKdinavdYWNnxlaa+DcDIPNYdi2PXiKpIT4BMG4RXP8N+ATX3vFE0YjgvfMd8O+DENuj2pseWn2E3/77F0W5Zuxu+OpS7So2s411X2xi/cwtSNWlF6huoWWHDtCpk1iujrtJ8jLYM9e98pKKcFiFHnr9u56dx13YzDBvHMy/GYqyhItLbbGaIOswfHkF/PWCtFysA1JqIpHUAoup6mr7pgFxFNlMpBedZUfGZjqFd8PqKL/dp3veQ8WB1WGlXVgnukf1cnl9SepP/JH2CzaHFZ1Gxw0thW7cbqmnlt6SS5ek0RDXHzZ+BBvfE1/slryqt9MZxRd1m1HQZwo07lqjaQ+vS2HFe56x7rSZ7ez+dT8anYbLbqvAQ1ziPlq2hClT4MUXRbfIWbPgjjsqHjthAkycCKtWiRbycXEVjzPnwQ+3iMDQG1gLRPa39SgIT6h6fH1hLYRZQ+DkZvdekNgKYe3rkH8CrvlENhuqBTLjLZF4kA7hXfnnzFqO5h2mTUi7Csfcl/Rvpnb5L2Pib+F4fipWh2sR29Cmo5jS5QWe7PoSNyfcyfzk2Zw2naS0GEYi8SZ+4TDgGZh0CsZ8C93ug8gkUYCm8xW6bZ0RNDpRoNV6FAx+HSamwQ1zahx0Z6Xm8Nfbazzql1/ij394bYrH5pAUM3WqM4h+/HHIyqp43Lhx1Wshv+ULUVTpTewWWPGcd+esCXYbfDvC/UF3CVYT7JgNSye6f9+XADLwlkhqgcHfUK24t2N4VzadWUdCcGv02sqLUhRFoV1YJ5oENGPViWWVjNHQPLAFYT7hJOceQOcj28ZL6hGNVvhuj5wBD++GJwvg0aNi+d+HYGo2TDwuJCqX/6tqq8AKcNgd/PHaKq90hrWZ7ax8fx2FuR62TrzUKSm0BDh7Fp59tuJxMTFOW8FZsyqWNqgqrH1DuHB4E9UunEEKK7loqG9WvwzHN3hWemM1weYZcOAXz81xkSIDb4mkFoQ1DUbvK5RadtWOzWEtfThw6hZDfcK4vfX9XBk7tFr77dNoAFvSN5Jvrfj2fWr+Mc4WnSbSGE14XCV2XBJJfaDVQUA0hDSHoMZgqLtf9vYf95B3Jh+8JCe1WeysfH+9dya7lBk5EoYPF8sffQTbtlU8rqTI8sAB+Oef8q8fWS6KBesFDWz9qp7mPg+ndwn3Em9Ib6wmWDC+4V6ANFCkxlsiqQURCeGljXO+P+j64ds0IM7lebPAFtXeb7RfDM0CWrD21EqGNBWNI35LWcTS1MUABOgDGdj4atpEtiOm3QVoaSWRVBObxc6Webuwmb3nIOGwOUjbdpLstBxCmgR7bd5LDkURWe9ly8BiEYWWq1eX94q+7joICBCdSr/5Bi67zPX1/YvBUsCcA0Y+3RHA0VwtgQaVoXFFPNEjl2AflTc3B/Le1gAMWhWdBhJDbDzbM4fu0VbmHDAyaVUIvloVjQJNA+1M6Z7L4OaiCDE1T0vP/0XTLtzK76PPlk6bWaSh67fBRH//Dht2P+bpd6v6qKoopPR0w6uyWE3w279hdDWKYCWAbKAjkdQKVVX58uY5VTbt8BR6o47hzw2iUdvIeplfIvE0B/46zN+fbMTq5f8xRavQdkgCfR+4vOrBkrrx7LPw3/+K5Zkz4fbby4+54w74+mvReOf4cTCUkezN6MYnS/fz8Y4A3umfzRWxZk4VaHlyTTAZhRoWjkrn/W2BHM3V8v7AbKwOeO2fIH44aGTLraeZe9DI9/v8WDgqA4cK3+7z48X1QWy65TTBPmpp4B0fbOOzqzJpEyb+Fr/Y5c/MPf5YHBo27D3RcAoMj60WDYSsXta8a33gPykXpr95PSClJhJJLVAUhZZ9mqNo6+cDV6vXEtWq5ppZieRCYduC3V4PukHYDB5YfhirWboGeZypU6F5c7H8+OOiWc65lMhN0tNhyRLnelUl7/he3twcyH975TCwqRm9RmStPxmURWq+lgWHjC670mvghkQTZwq1ZJldwx+NAmMTCzHZNBzJdRUDjEkwMe+g08N+/kEjYxNNgCqazDQU1rzhPXeXsiga2PKZ9+e9QJGBt0RSSzqMaoNG6/1/Ia1BS4eR9TO3ROINLCYLz//8BC9tfhLTOdm7T3e/y4ubprDl7EY+2jUdm8MZIJtsBUzf9iKHcvZzNDeZFzdN5ZUtz/Dqlmf4cOcbbEsvrxNWVZX3drzGR7veLF2naDWcPZThuROUCPz84J13xPKZMxUXWg4YALHFbdrLenoXZrLppBazXWF4C1dphb9e5cqmZlYd93FZb7bD3IN+NPa3Eebr6iFud8Cc/Ub0GpUmAa7ypjGJhSxKNmJ3wIEsHQVWhS5RFkCBrCO1OXP3Y8qEw38AKnMOGBk0P5KWXzai8+xopv4dTI5ZJIne3BxI889jSPyqEW2/bsSoRRFsOq0HYM4BI02LX2s9sxFX/RDJH8ec72FqnpbYzxozfkmYy9SP/OHDm+9U0olUUg6p8ZZIakl4XCghTYLIOOzlwhIV2g5J9O6cEokXSU/ORFEg1BDKrsxtXBbdB4DTppOldpvxQQnsztzOqhPLuLKJKF5emrKYxOA2JAS35mhuMoH6QP7T6SlUVeVQzn7+d+hrmgTEEeHrlGgdyz9CgS0fh+rgeEEqsf5NsVvtpB/KoHG7aO+f/KXGtdfCsGHw22/w4Ydw993O5jkgWsjfdhu89hosXuxsIW8rItOsJczXga6CHES0n50d6QZaBttZfNjIshRf9BqV1qE2vhjs/MzecsZA268bYbIq6DTw3oAsIoyuQXmMv52WwTZWH/dh7UkDYxKL3UJUFWZ8AAF/Q1CQ6yM42PW5n59nJSnHN4DWwCdbtBVKb8b9Gs7CUekAjGpZ6CK9ufePMLbcKjohd4uyuEhvHloeWiq9Kfue/XNaT4/oMn0p7GbIPSEKqyXnRQbeEkkd6P9wTxY98btHPYbLovPR0un6dhhDatk5UCK5ADibnImqQoeIruzI2FIaeG/P2EzH8G78dWIpACPjxvDpnvdICutIvjWPI3mHeLBd+WI3RVFIDGmDUWfkjOmkS+C9I30zrUOSsDms7EjfTKx/UxxWByd2naHjtUneOeFLGUWB996Ddu1cCy3LBqnjx4vA22yGefPg3ntBoyPM10FmkQabg3LB92mTljBf8bk8Ml4EmhXRtTjQLLAqTFwVwsZTBka1LF+cODbRxNyDRjadNvDjyAySc7Qi8J77Axz5oerz1GjKB+c1eZQE8gEB5YtQAY7/Q15+IW9uDufNftkMbCoKREukNz3/F1Wp9ObjHQGVSm+m/h3CkVwdnSOdQfZDHfN57Z8g5l9T5q6QohW+4TLwrhIZeEskdSAyIZz2I1qz+9f9nndfUCAgwp+uN7b37DwSST2Tc0LYaTYJaMbOjC2cLTxNuG8kuzO3c2ebh0oD7xCfMAY0HsxPR+dRZCtkeLPrMer8yu1PVR0cyNmHyWYi1NdZG2G1W9iTtZMx8bdgU638fGwBQ5peg1ajI+90vndOVgIJCULj/dJLsGaNkJRMmOB8vV076NIFtm4Vnt733gs+QXSLLMKg9efXI74uwXKBVeGvVB+m9sjjZEH1+h3461Ve6ZNN7znR3NzaRPsIV43/iBZFPL02mA4RVmID7CLwRoGQKAg2Q27u+duoOxwiW1+Rjr2mBAaWD8w7bmdTlqZK6U3LYOf3VG2lN7cnFfDlbn9WHTfQL7a44ZvdAsc3CW9/yXmRgbdEUkd63NqJYxvTyDmVh2r3nEmQzqBl8JR+UtstueixWZxBT4dwkfVuHhhPhG8UQYYgl7E9onqzI3Mr0X6NaRPq2h02z5rHa1ufE/76qoMhTUcQ4xdb+vre7F3oNDpaBifiUB04VDsHc/bRJrS91+5iSYp54gkRcB87BpMnw6hRQlJSwvjxIvD++284cgRatCAoLIrHuubwzLpgAg2qi7Qixt/OmEQTH2wLrPYhhPqqjGtt4u0tgXwxxFVC6KdXmTsig2CfMgGqAmw4CnqjCKxNJhGAl33k5JRfd75HTg7Yq/jby8sTj+PHnesSILPI6BXpja9O5ZEueby+KYh+senFa1XITa36TZbIwFsiqStavZZRLw/hh4m/UphViMMDwbfOR8vVTw0grHlI1YMlkgucsheXHcO78vX+T8g2Z9Ipolu5sYqiEOkbRaChvO92icbb5rDxZ9pvHM1Npmd039LXt6dvJim0IxpFi0bR0jakA9szNtMmtD1KRdGLxHP4+cHbb8Po0aLQ8rnnnB0uQbSQnzxZBKWzZ8Mzz0Dj7jzUaRGhPg7+uyGIY7k6AgwOrm5exPsDs6hNc9972ufTZ040ezJ0BBpcP8s7lZFbAKJ7q75YvqHRCBlIQAA0roPcQlWhqKjyoLyygD1gGWFm70lvbmlt4pPtAfxepvgSu7n2530JIQNvicQNGEN8Gf3mMBZN/Z2CDJPbsmWKRkFr0DL0qQHEdmzkln1KJA0dvxCnFjXEJ5QQQxgHc/YzMu6GWu1Pp9FxVZNhfLhrOvuydtMmtB25lmyO5iVzwpTK3qydAFgdVmyqDZO1gOgA6ZHvda67DoYOFbaBH3wgCi07dhSvNWoEQ4aIIsxvvoGnn4b4QZD8O+PamBjXpmIbvYndKu4CDHBTq0JuauXaVr1xgIMjd58sfX783hMVbtsv1sqGaVfW8ASrgaKA0Sge0TUo7v2oA910uzFoVa9Ibwxa+E+3PN7YFETr0OILEp2xgj1JzkVe0kskbsIvxMjYd0fQ+sp4tIZapFrOQeerIzw+lLHvjJBBt+SSIiIhzKW2bmTcWCa0vg+D1lD5RlWg1ejo2agfq04uA2BHxlbCfSN4uP1k7m/3KPe3e5R/dZhMkD6YXVnbiEmSzUC8TkmhpcEgpBsPP+yqmy7x9D54EDZsgPY3A/XUA1DvB90frJ+5KyKoKUEGlce65vPMumD+SvXB6hAWgA/8GVoqvakJZaU3FTE2oRCzHf5K8xVe3iHV79J8KSMDb4nEjeh9dPR98HJGvDCIwCh/dL46oQOsyT58deh8dVx2WydGvzGM4JjqaxQlkouByJauPsFhvuE09m9S5/12Ce9OjiWb/dl72J6xme6RvQjQB7o8ukVezo7MLUQlygZV9UJiopCUgNBzz57tfO3aa0VhIYist38kJAynxh+y7iAgGpr28v68lRHXH7QGHuqUz9Tuefx3QxBtZsZwzaIIYvztzBmRUWvpzfJUX/ZklBdIaDUwqVse2WYNaPQQ290NJ3LxI1vGSyQeQlVVTu4+w/Yf95C27SRagxaHzY7d4lqoomgU9L467FY7gdEBdBnTnvg+zdD5SCWY5NJEVVVm3joXS4G16sEeQKvXMG7GdfiHl3dIkXgBkwnatoWUFCG32L9f2OkB3HmnaC8fFgYnT8KZbfD1QO92bNT7w7B3oevd3puzKg7/CXNGgzm3fubXGuCx4+AfUT/zX0DIb3aJxEMoikLj9tE0bh+NtchGxpEs0pMzOL0/HXOBBdWuCjlJXAhRrSKIbBku/bklEsT/TturE9n50z4cNkfVG7iZyMRwGXTXJyWFlmPGwOnTotCypMPlhAki8M7MFHrva6+FdjfBru/BVr4I0O0oGohoA53v9PxcNSH2MmHpV18ExMigu5rIjLdEIpFIGhx5ZwuY8+Ai7FbvBt56o45BE6+geY+6S1skdUBVRUfLpUtF98qtW6FDB6H9jouD1FQRmM+fD+Z8eK8lFJzx/HHp/OChHRDW0vNz1ZQfboNd/wPVy1aYen8Y9Ar0fMS7816gSI23RCKRSBocgZH+xHZshKL1rn5Xb9TTtKvsvlfvlBRa6vXCQrCk0FKjgVtvFWMWL4asLPAJgJsWiIJHT6IzwvD3G2bQDdDrMdD5VD3O3ah26Dyh6nESQAbeEolEImmg9Hu4J1p93R2CqovWoOWqSX1lk6qGQqtWMGmSWF69Gr79ViyXuJtYLDB3rlhu1gdumOc5SzudEQa+AF3v8sz+3UHjrhDTFTReVBHrjNDtfvAt76MvqRj56SKRSCSSBol/uB9X3NfDK4XGWoOWVgPjiWknbQQbFE89BU2biuXJk0UTmaQk6FbcTOmbb5xjWw2HWxaDIcCNwacigsur34I+k920Tw8yerYodPQWxjC46hXvzXcRIANviUQikTRYWl0ZT7PujdHVxgutmmh0GoKiA+h1V/nOmJJ6xt9fFFoC/H979x5kVX0YcPx7n/tkF5a3LA8XdiEKSCoqyi6aQhJbIFJToo0pZMam42g7jk1pNQGHaRmMbaaDOkwNth1rwEfQidJgM21Bo7zEqBgSExZYQITykPe+7t679/SPiwodVFjYsxf2+/nrcjnn/M7Z4Y8vZ3/nd/btg/nzc58/uuu9di1s3/7J9lWT4S/rYeik3Nzj85Eoht418J034Jq7zu9YYek5FL7yw/O/9rMRL4KZP/nk7Z06K4a3JClvRSIRJn+3loGj+3dKfEfjUUr7lvC1hV8hUehCX3np1ltzb60EeOwx2Lw59wr52Ml/D6eu9Q3QYyDM+h+Y/iMoHwLxEjiXZ3STPaCwF0yaC3dvhv5jLshlhGb8XfCFP+rcOe+JYpi8EIbc0HljXKJc1USSlPey7Vl+8dgGGtbtIpO6MKs2xAvjVAwu5w/nT6agNMRfz+vc1dfD6NGQTsOkSfDqqzBtGrz8MgwfnnubZeQMD+IGATz7MKx4AIYCySgUleaW3guyueUBo4ncA4JBAAOvhhv+CmqmQ+wi/o9Yth1e+CbU/+zCr3GeKIZJ86Du/gt73G7C8JYkXTR2vfkBryxaRzqVIdvBpQYj0QixRJRrvjWOMdNGEYl2wZsPde4eeAB+8IPc56VLIR6H22/P/XndOrg+9ybJIAhoy2TJBgHJeIzYvLmwcGFuu3+YB9PHw4k9uXW/Y0ko7pt7KLFXVW7VlEtFNgur58KGRZBpOf/jReMQK4Cpi2Hc7PM/XjdleEuSLiqpxhRvLnuXLasaIAKZ1sxZ7RdLxiAIGHz1ICZ8+/coH9ijk89UF1RTU+6Nlrt3w4ABsGkT1NRwIhtSGiUAAAjWSURBVNXOq3fP5Z3aqWzefZS9R1sIgoBIJEI2G1De1szI99/ji7t+Rd28v2D4V+u6+krCtfct+MnM3Drn6aaOHSNRCgPGwtefgZ5DLuz5dTOGtyTpopROZdj++k7e+896Dr9/jEgEIrEoZAMCcne2CQLa01nKBpRSfePlfOGr1RSV+4bYi9bzz8PMmQBs/esHWRZcxurkQGJBlpbkZz/kF8+kiRUWMKRPCbNqL+dLV/Qn3l2WjsykYNO/w9qHoXH/yeknn5N/0XhuGk7/sVD7tzBqxpmn8+icGN6SpItekA04vr+RwzuPkG7NEAQB8YI4PQeV0bOynFi8mwTWpS4IaLl5KosS1fx87O+TThSQ5dxjsCgZo0+PAhZ+YxzVA7rRbz6CAHavh/eWw67X4eB7QJCL7CCAbAbihbkHSofdBGO/BX1GdvVZX1IMb0mSdFHYvPso9y99k8YTLaQS5/eWxgiQjEe5Y+Iw/uymEUS741z/bPbkFJTm3IOmyVIo7u2d7U5keEuSpLy3tv4g33tuE6lMxx6q/TSFiSi1NX2Z//Wx3WfqibqM/8IkSVJee2P7h50S3QCt6Sxr6g/ydz/9Nd6LVGczvCVJUt7af6yF+5/pnOj+SGs6y2u/288z63d12hgSGN6SJClPBUHAg8//irb2zovuj7Sms/xo1VZ2H+rgknvSWTC8JUlSXvqPt/ew5X9P0J4NZwpIuj3L3OXvOuVEncbwliRJeSebDfjnVVtpTbeHN2YA7x9q5u2dR0IbU92L4S1JkvLOhu0fkgoxuj/S2tbOsrU7Qh9X3YPhLUmS8s7SNTtobgs/vAPglw2H+fBEKvSxdemLd/UJSJIkbdy4kQULFlBfX080GqW1sC/lE/6EzJE9NG55jf7Tv3fa9nufnUNF3beJJAo5+PI/ctkdi4gmCk/bZt9P51NSU0ePKyefcczWPb/h6MblZI7tI5osoeeE2yiuupZ4LMJbDQfZ9PNlPPfcczQ2NjJs2DCWL19OeXl5p/0MdOkzvCVJUpc6ceIEs2fP5qGHHmL69Ols3XuEWQt+TCb6+ZlS0G84sZJetOz4JSU1tR9/33b4A9JH9lI8/Loz7pc+sodDryyh4sY7KRx0Jdm2FoK2ZgBa0u0sWfwo0SM7WLFiBYMGDWLLli0UFJzf2zIlw1uSJHWphoYGAGbMmAHAjsNtlA4ZQ3NbO+lDn7+2dkn1RJq2rjstvJu3rqNo8BhihaVn3Of4pp9ROupGigaPBchtd3Lb9tYm3vyvF3jt1VVUVlYCMGrUqI5foHSSc7wlSVKXqqqqIhqNcu+997J69Wp27jl4TquZFI+4gdS+ejKNhwEIgizN29+guHrip+6TOrAdgH0vzGPPsvs49MoS2lsbgdzd8iASZeXKlYwbN47a2lqefPLJjl+gdJJ3vCVJUpfq0aMHL774IosXL2bOnDnsP3CAgsqxVNTOBqDtQAMfPHXPafsEba0ff46XVlAwcCTN29ZRNm4aqT2/JcimKRoy9lPHbG86QtO29fT9g+8SK+7J4V/8K0fXP03vL/057U1HyLY109DQwPr169mxYwe33XYbVVVVTJo0qXN+COoWDG9JktTlqqurWbRoEQAPP72axx+ey9ENz1BYOZpkv6ozPlx5qpLqiRzftJKycdNo2raO4qrriJycI354zVM0b1sPQNm4qZSNm0YklqSkppZE+YCPvz/w8g8BiMQTANx3330UFRVxxRVXcMstt7Bq1SrDW+fF8JYkSXll+IgRlI2q49hvXqGwcvRZ7VM07GqOrPsxrXt/S8vOt+k39W8+/ruK2llU1M46bftERSUQOeWbTz4nKwbnvol88t2pn6WOco63JEnqUtu2bePxxx9n7969AFREm2jatoFkv+FnfYxoooCiYeM5/Nq/ESvtTbLv5Z+5fUlNLU31a8gcP0A2k+L4uyspGnIVAPGyfgwYPppHHnmEVCrF1q1beemll5gyZUrHL1LCO96SJKmLlZSU8M477/DEE09w7NgxysrKiPUaRfm136Bl51tnf5zqiTRvXUv5NX/8uduWjqyjvfEQ+1csAKCwcgy9rv8mAAWJKHd9fyH//dQ/MXr0aPr06cOcOXOoq6vr2AVKJ0WCIAi6+iQkSZJONfPR19l9qLlLxi5MRPmX70xgRP8eXTK+Ll1ONZEkSXnn9glDKUx0Tab0Ly8yutUpDG9JkpR3br7qMrrid/JFyRh/WvvZ88OljjK8JUlS3ikpiDP1i4NIxsNNlXg0wpTRA0IdU92H4S1JkvLSPV+uoTgZC228wkSUB28dQ2EivDHVvRjekiQpL5UUxPn7mVdREMJd70QswvXVfakb2a/Tx1L3ZXhLkqS8dU1Vb2ZeN6RTH7SMRSP06VHIA1+7stPGkMDwliRJee6eL9dw89jLOiW+49EIvUuTLLnzWsqKEhf8+NKpXMdbkiTlvSAIWLJ6G0+v20kqk70gxyxMxKisKOLRWeOpKC24IMeUPovhLUmSLhq/3n2U7y9/l6NNbR0O8EgEkrEod940nDsmXk4sGrnAZymdmeEtSZIuKql0O0vX7uTZ9TtpzwY0t7Wf1X7JeJQggOure3P3lBqG9S3t3BOV/h/DW5IkXZQy7VnW1B/khY3v87u9x2lNZ3NxTS5tIkA2yG03uHcxU64cyIzxlU4rUZcxvCVJ0iXh0IkU2/afoDGVIZsNSMajDO5dwtA+JU4nUV4wvCVJkqQQuJygJEmSFALDW5IkSQqB4S1JkiSFwPCWJEmSQmB4S5IkSSEwvCVJkqQQGN6SJElSCAxvSZIkKQSGtyRJkhQCw1uSJEkKgeEtSZIkhcDwliRJkkJgeEuSJEkhMLwlSZKkEBjekiRJUggMb0mSJCkEhrckSZIUAsNbkiRJCoHhLUmSJIXA8JYkSZJCYHhLkiRJITC8JUmSpBAY3pIkSVIIDG9JkiQpBIa3JEmSFALDW5IkSQqB4S1JkiSFwPCWJEmSQmB4S5IkSSEwvCVJkqQQGN6SJElSCP4PPxvf9MuiYbQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10)) \n",
    "\n",
    "\n",
    "# nc = nx.draw_networkx_nodes(G, pos=pos, nodelist=G.nodes(), node_color=carac['predicted_ARG-class'].cat.codes, \n",
    "#                            with_labels=False, node_size=100, cmap=plt.cm.Set1)\n",
    "# Plot the network:\n",
    "nx.draw(G,pos=pos, with_labels=True,node_color=carac['predicted_ARG-class'].cat.codes,cmap=plt.cm.Set1,node_size=1050, edge_color=links_filtered['color'], \n",
    "        width=links_filtered[\"value\"]*3, edge_cmap=plt.cm.Blues,legend=carac['predicted_ARG-class'])\n",
    "\n",
    "\n",
    "#plt.legend(carac['predicted_ARG-class'],loc=\"best\",scatterpoints=1)\n",
    "\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/rcuadrat/.local/lib/python3.6/site-packages/ipykernel_launcher.py:19: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n"
     ]
    }
   ],
   "source": [
    "ARGS_CLASS_fornet=class_RPKG[class_RPKG.columns[0:-45]]\n",
    "corr = ARGS_CLASS_fornet.corr(method='spearman')\n",
    "corr\n",
    " \n",
    "# Transform it in a links data frame (3 columns only):\n",
    "links = corr.stack().reset_index()\n",
    "links.columns = ['var1', 'var2','value']\n",
    "links\n",
    " \n",
    "# Keep only correlation over a threshold and remove self correlation (cor(A,A)=1)\n",
    "links_filtered=links.loc[ ((links['value'] > 0.6 ) | (links['value'] < -0.6 )) & (links['var1'] != links['var2']) ]\n",
    "links_filtered\n",
    " \n",
    "# Build your graph\n",
    "G=nx.from_pandas_edgelist(links_filtered, 'var1', 'var2')\n",
    " \n",
    "# Plot the network:\n",
    "#nx.draw_random(G, with_labels=True, node_color='orange', node_size=400, edge_color='black', linewidths=5, font_size=10)\n",
    "links_filtered['color'] = np.where(links_filtered['value']>0, 'red', 'blue')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "G=nx.from_pandas_edgelist(links_filtered, 'var1', 'var2', create_using=nx.Graph(),edge_attr=True )\n",
    " \n",
    "# Custom the nodes:\n",
    "nx.draw_kamada_kawai(G, with_labels=True, node_color='skyblue', node_size=1500, edge_color=links_filtered['color'], width=links_filtered[\"value\"]*10, edge_cmap=plt.cm.Blues)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}