statistical_test.ipynb

{
 "cells": [
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "#Table of Contents\n* [1. Function Definitions](#1.-Function-Definitions)\n* [2. Statistical test](#2.-Statistical-test)\n* [3. Kolmogorov Smirnoff (KS)](#3.-Kolmogorov-Smirnoff-%28KS%29)\n\t* [3.1 Summary](#3.1-Summary)\n\t* [3.2 Apply to liver data](#3.2-Apply-to-liver-data)\n* [4. Analysis of Variance](#4.-Analysis-of-Variance)\n"
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "from pylab import *\nimport pandas as pd\nfrom scipy import stats\nimport matplotlib\n# matplotlib.style.use('ggplot')\n%matplotlib inline\nimport seaborn as sns\nsns.set_palette(sns.color_palette())\nsns.set_style('darkgrid')",
   "execution_count": 1,
   "outputs": []
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "# 1. Function Definitions"
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false
   },
   "cell_type": "code",
   "source": "def cumulative_fraction_step(d, **kwargs):\n    \"\"\"Calculate cumulative fraction function for given data.\n    \"\"\"\n    d = np.sort(d)\n    # take the first value where the lowest point is\n    x = d\n    cumfrac = 1.0 * np.arange(1, len(x) + 1) / len(x)\n    y = cumfrac\n    # add data points for plotting\n    dx = np.max(x) - np.min(x)\n    x0, x1 = np.min(x) - 0.05 * dx, np.max(x) + 0.05 * dx\n    x = np.array([x0] + list(x) + [x1])\n    y = np.array([0] + list(y) + [y[-1]])\n    plt.step(x, y, where='post', **kwargs)\n    return x, y\n\ndef ks_2sample(data1, data2):\n    \"\"\"Copy scipy version to have control over details.\n    \"\"\"\n    from scipy.stats import distributions\n    data1, data2 = map(asarray, (data1, data2))\n    n1 = data1.shape[0]\n    n2 = data2.shape[0]\n    n1 = len(data1)\n    n2 = len(data2)\n    data1 = np.sort(data1)\n    data2 = np.sort(data2)\n    data_all = np.concatenate([data1,data2])\n    cdf1 = np.searchsorted(data1,data_all,side='right')/(1.0*n1)\n    cdf2 = (np.searchsorted(data2,data_all,side='right'))/(1.0*n2)\n    d = np.max(np.absolute(cdf1-cdf2))\n    data_point = data_all[np.argmax(np.absolute(cdf1-cdf2))]\n    # Note: d absolute not signed distance\n    en = np.sqrt(n1*n2/float(n1+n2))\n    try:\n        d1_idx = np.where(data1>data_point)[0][0]\n    except IndexError:\n        d1_idx = -1\n    try:\n        d2_idx = np.where(data2>data_point)[0][0]\n    except IndexError:\n        d2_idx = -1\n    try:\n        prob = distributions.kstwobign.sf((en + 0.12 + 0.11 / en) * d)\n    except:\n        prob = 1.0\n    return d, prob, data_point, d1_idx, d2_idx\n\ndef ks_2s_statistics(data1, data2, xscale='linear'):\n    \"\"\"Summarize ks 2 sample statistics.\n    \"\"\"    \n    d1 = np.asarray(data1)\n    x1, cf1 = cumulative_fraction_step(d1);\n    d2 = np.asarray(data2)\n    x2, cf2 = cumulative_fraction_step(d2);\n    d, p, dp, d1_idx, d2_idx = ks_2sample(d1, d2)\n    plt.vlines(dp, cf1[d1_idx], cf2[d2_idx], zorder=10, color='red');\n    plt.xscale(xscale)\n    print 'KS D statistics: {} \\np-value: {}'.format(d, p)\n    return d, p",
   "execution_count": 5,
   "outputs": []
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "# 2. Statistical test"
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "**Purpose of this notebook:** evaluate available statistical tests"
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "Create some test data to have different scenarios ready."
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "data = dict()\ndata['controlA'] = [0.22, -0.87, -2.39, -1.79, 0.37, -1.54, 1.28, -0.31, -0.74, 1.72, 0.38, -0.17, -0.62, -1.10, 0.30, 0.15, 2.30, 0.19, -0.50, -0.09]\ndata['treatmentA'] = [-5.13, -2.19, -2.43, -3.83, 0.50, -3.25, 4.32, 1.63, 5.18, -0.43, 7.11, 4.87, -3.10, -5.81, 3.76, 6.31, 2.58, 0.07, 5.76, 3.50]\ndata['controlB'] = [1.26, 0.34, 0.70, 1.75, 50.57, 1.55, 0.08, 0.42, 0.50, 3.20, 0.15, 0.49, 0.95, 0.24, 1.37, 0.17, 6.98, 0.10, 0.94, 0.38]\ndata['treatmentB'] = [2.37, 2.16, 14.82, 1.73, 41.04, 0.23, 1.32, 2.91, 39.41, 0.11, 27.44, 4.51, 0.51, 4.50, 0.18, 14.68, 4.66, 1.30, 2.06, 1.19]",
   "execution_count": 6,
   "outputs": []
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "# 3. Kolmogorov Smirnoff (KS)"
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "Follow the explanation given here: http://www.physics.csbsju.edu/stats/KS-test.html"
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "## 3.1 Summary"
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "KS-test is non-parametric and distribution free and can be applied in a wider context than the Students t-test. It may, however, not be as sensitive. \n\nIt is used to assess whether a treatment has an observable effect in the collected data. \"How different must the outcomes be?\"\n\nApply the right test in the right situation. If applying t-test to non-normal data one increases the risk of errors. However, the central-limit theorem helps in a way that for very large data sets t-test is not outrageously wrong even for non-normal data. Hence, the test is *robust* as it continues to work to some extent outside its narrow boundaries."
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false,
    "scrolled": false
   },
   "cell_type": "code",
   "source": "df = pd.DataFrame(data)\ndf.describe()",
   "execution_count": 3,
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/html": "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>controlA</th>\n      <th>controlB</th>\n      <th>treatmentA</th>\n      <th>treatmentB</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>count</th>\n      <td> 20.000000</td>\n      <td> 20.000000</td>\n      <td> 20.000000</td>\n      <td> 20.000000</td>\n    </tr>\n    <tr>\n      <th>mean</th>\n      <td> -0.160500</td>\n      <td>  3.607000</td>\n      <td>  0.971000</td>\n      <td>  8.356500</td>\n    </tr>\n    <tr>\n      <th>std</th>\n      <td>  1.133687</td>\n      <td> 11.164638</td>\n      <td>  4.075968</td>\n      <td> 12.819383</td>\n    </tr>\n    <tr>\n      <th>min</th>\n      <td> -2.390000</td>\n      <td>  0.080000</td>\n      <td> -5.810000</td>\n      <td>  0.110000</td>\n    </tr>\n    <tr>\n      <th>25%</th>\n      <td> -0.772500</td>\n      <td>  0.315000</td>\n      <td> -2.597500</td>\n      <td>  1.272500</td>\n    </tr>\n    <tr>\n      <th>50%</th>\n      <td> -0.130000</td>\n      <td>  0.600000</td>\n      <td>  1.065000</td>\n      <td>  2.265000</td>\n    </tr>\n    <tr>\n      <th>75%</th>\n      <td>  0.317500</td>\n      <td>  1.415000</td>\n      <td>  4.457500</td>\n      <td>  7.165000</td>\n    </tr>\n    <tr>\n      <th>max</th>\n      <td>  2.300000</td>\n      <td> 50.570000</td>\n      <td>  7.110000</td>\n      <td> 41.040000</td>\n    </tr>\n  </tbody>\n</table>\n</div>",
      "text/plain": "        controlA   controlB  treatmentA  treatmentB\ncount  20.000000  20.000000   20.000000   20.000000\nmean   -0.160500   3.607000    0.971000    8.356500\nstd     1.133687  11.164638    4.075968   12.819383\nmin    -2.390000   0.080000   -5.810000    0.110000\n25%    -0.772500   0.315000   -2.597500    1.272500\n50%    -0.130000   0.600000    1.065000    2.265000\n75%     0.317500   1.415000    4.457500    7.165000\nmax     2.300000  50.570000    7.110000   41.040000"
     },
     "metadata": {},
     "execution_count": 3
    }
   ]
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "The basis for the KS test is the *cumulative fraction function* (cff) which is the faction of data reached until a specific data value.\nThe KS D statistics is then defined as the maximum difference between two cff. By comparison to a standard model one can obtain a p-value given the probability to observe a difference of at least that magnitude just by chance."
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false
   },
   "cell_type": "code",
   "source": "# create comparison cumulative fraction function plot\nd1 = np.asarray(data['controlA'])\n# x1, cf1 = cumulative_fraction_step(d1);\nd2 = np.asarray(data['treatmentA'])\n# x2, cf2 = cumulative_fraction_step(d2);\n# # calculate KS statistics\n# d, p, dp, d1_idx, d2_idx = ks_2sample(d1, d2)\n# plt.vlines(dp, cf1[d1_idx], cf2[d2_idx], zorder=10, color='red');\n# print 'KS D statistics: {} \\np-value: {}'.format(d, p)\nks_2s_statistics(d1, d2)",
   "execution_count": 7,
   "outputs": [
    {
     "output_type": "stream",
     "text": "KS D statistics: 0.45 \np-value: 0.0232132758544\n",
     "name": "stdout"
    },
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": "(0.44999999999999996, 0.023213275854449614)"
     },
     "metadata": {},
     "execution_count": 7
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeMAAAFVCAYAAADc5IdQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAIwRxgAAGCOMAQAw9n9pSrkr11FtbwAAAABJRU5ErkJggg==\n",
      "text/plain": "<matplotlib.figure.Figure at 0x18a18ba8>"
     },
     "metadata": {}
    }
   ]
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false
   },
   "cell_type": "code",
   "source": "d1 = np.asarray(data['controlB'])\n# x1, cf1 = cumulative_fraction_step(d1);\nd2 = np.asarray(data['treatmentB'])\nks_2s_statistics(d1, d2, xscale='log')",
   "execution_count": 8,
   "outputs": [
    {
     "output_type": "stream",
     "text": "KS D statistics: 0.45 \np-value: 0.0232132758544\n",
     "name": "stdout"
    },
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": "(0.45000000000000007, 0.023213275854449551)"
     },
     "metadata": {},
     "execution_count": 8
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAekAAAFaCAYAAAA6kKVMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": "<matplotlib.figure.Figure at 0x18ad9e80>"
     },
     "metadata": {}
    }
   ]
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false
   },
   "cell_type": "code",
   "source": "print stats.pearsonr(data['controlA'], data['treatmentA'])\nprint stats.ks_2samp(data['controlA'], data['treatmentA'])",
   "execution_count": 9,
   "outputs": [
    {
     "output_type": "stream",
     "text": "(0.42384570290462886, 0.062553208947154049)\n(0.44999999999999996, 0.023213275854449614)\n",
     "name": "stdout"
    }
   ]
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "## 3.2 Apply to liver data"
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "Now that the method is established it can be applied to the correlation data found in liver tissue."
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false
   },
   "cell_type": "code",
   "source": "import sys\nsys.path.insert(0, '../..')\nfrom mcs.data.tissue import LiverTissue\nfrom mcs.mcsio import liver_tmp_save, liver_tmp_load, MCSDOCPATH, create_pardirs",
   "execution_count": 10,
   "outputs": [
    {
     "output_type": "stream",
     "text": "WARNING:traits.has_traits:DEPRECATED: traits.has_traits.wrapped_class, 'the 'implements' class advisor has been deprecated. Use the 'provides' class decorator.\n",
     "name": "stderr"
    }
   ]
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false
   },
   "cell_type": "code",
   "source": "tmp_name = '/marino_paper/local_correlations/local_corrs'\nmany_corrs = dict()\nadult_lpns = ['adult_m1_s3_fv2', 'adult_m2_s6_fv3', 'adult_m3_s5_fv2']\nkd_lpns = ['kd_itgb1_ecm_fv2', 'kd_itgb1_flk1_fv1']\nctrl_lpns = ['kd_itgb1_ecm_luc_fv2', 'kd_itgb1_flk1_luc_fv1']\nall_lpns = adult_lpns + kd_lpns + ctrl_lpns\nfor lpn in all_lpns:\n    t = LiverTissue(lpn)\n    local_corrs = liver_tmp_load(t.lp, tmp_name)\n    many_corrs.update(local_corrs[0])",
   "execution_count": 11,
   "outputs": []
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false,
    "scrolled": true
   },
   "cell_type": "code",
   "source": "lpn1, lpn2, lpn3 = 'adult_m1_s3_fv2', 'adult_m2_s6_fv3', 'adult_m3_s5_fv2'\nd1 = many_corrs[lpn1]['a1sb1s']['corrs']\nd2 = many_corrs[lpn3]['a1sb1s']['corrs']\nks_2s_statistics(d1, d2)",
   "execution_count": 12,
   "outputs": [
    {
     "output_type": "stream",
     "text": "KS D statistics: 0.0763062956976 \np-value: 0.0091314340056\n",
     "name": "stdout"
    },
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": "(0.076306295697600945, 0.0091314340056042297)"
     },
     "metadata": {},
     "execution_count": 12
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAecAAAFVCAYAAADVDycqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XlgVNXd//H3zGQlCQmBsJNA2BFkV0AQN1wqbUUQgorW\nWvWx1bZWba1PpdjqA9raWgUff7VVK48VSl1qUbEiKLsiEDDIGjAJS4CQkH2Z5N7fHyMXxyQTlszc\nWT6vf3ruMuR7GpwP995zz3GYpmkiIiIiQcNpdwEiIiLiTeEsIiISZBTOIiIiQUbhLCIiEmQUziIi\nIkFG4SwiIhJkTiuct27dyqxZsxrtX7FiBdOmTSMrK4slS5a0enEiIiKRKKqlE1544QXefvttEhIS\nvPa73W7mzZvH66+/TlxcHDNnzuSyyy6jffv2fitWREQkErR45ZyRkcH8+fP55lwlubm5pKenk5SU\nRHR0NCNHjmTjxo1+K1RERCRStBjOV155JS6Xq9H+iooKkpKSrO2EhATKy8tbtzoREZEI1OJt7eYk\nJSVRWVlpbVdWVpKcnOzzM6Zp4nA4zvZHiojIOSo8XklVjZsPNu2mAfh4UwHR0S5KK2qb/Ywz8QSO\n6Fpieu44q58ZW9fhrD7nwMVtI6Zz6XmDzurzoeyswzkzM5O8vDxKS0uJj49n48aN3H777T4/43A4\nOHYscq+u09KS1H/13+4ybBHJfYdz679pmpyoqGv0aNFtuCmpPYFpmuTsL8YwDOtYdV0DG3cV4krf\nDg4DTM9Fkbvec44zsfTUHzQI6oD4M6ipZ2JPopyn4qOqvorh7UcA0CYqgYzEdABcDic92qfSsWPb\nc/r9h/rfnbS0pJZP+obTDueTV7xLly6lqqqK6dOn89BDD3H77bdjGAbTpk2jY8eOZ1yAiEgoc9cb\nXsH5ZWE5h4+fuqu4u+AExRV1mMaZrzHkjiol71ix1z5HXCXRPb/A4TSa+dRX+kHDybbpANOBMwZM\nANMJDoP02H60iY0iNiaKKGfzdzVNTBKiE+ibksmA1L4kRLc5477ImXEEelWqUP8X0LnQ1YP6H6n9\nD4e+V9fWszOvBMM0Wb3tMNEuJ1tzi6hvOL2vUEdsFVGd8qCFUHW1K/ScH+1u+Q81HbSv70uDAe2S\nYomPPTU+yOlwEBcHl/QYR6/kjNOq0V/C4fd/Lvx65SwiEi4M0ySvsJw6t+facnfBCYpKazh0vJLc\ng2VERzn55nVkXX3zodo9LZGUpBiqog9zIn4n0cTQLinO87Oop6A294xrbB+Xittwk57Uja6JXRod\nv6bnFcS4os/4z5XQoHAWkbBTVlXHexvycDk9L6TsPXCCA8cqiY3xXFmWlDc/+Ak8t6p7dWl8tXOi\noo4rR/fA6XDQu1sy7dpG0eCoo7juOEt2/4vDlZ6r3mqgrIkf0attBhd3H0t6UnefP79tTBJtos/k\nKbCEG4WziIQUd30DlTX1FJfVsiu/hI+zD1mhe1LB0YomP1tVW09aShxpKXGUVboZNSCN9m09V7gJ\ncdEM6d2e2GgXJUYhu0tycX7j7ZK1hz5lm9MzIdMnBbUcrjzS6GdEOVz8euzPcTlOfb127JBEbeTe\n1ZWzoHAWkZBQUe3mqcXZ5BU2nXLxsae+zuJiXNTUNfCjKUNITowBIDE+ms6pLQ9k+sfuf/HxgbXN\nHi+qPk6UM8rzaigOXE4XQzuch4HJ5F6T6JzQqdFn2sYlcUzzQMgZUDiLSFA6WFTJ0ZIqAI6VVLNo\nxV6v4xcM7EhtXQMTh3ejV5e2JCfEnNafW1R9nBUFq2kwDTYd2UqUwwUOcODAMA0q3KdGWv9w6Pcb\nfb5X23TaaLSy+JnCWURs12AY7Mo/wbqcQtblFBLlcjQ7CvrhWSPp063xhEduo77RvryyAkprS6lt\ncLPq4DoKyg82+Wd2jO/AyRFgcVFxJMe05Wcj7z77DomcI4WziNimrLKO3y3awsFjlV776xtMenRM\nJD42iuF9PbNLRbmcXDSkM3Ex3l9bpbVlPLz2sTP6ub3apnNJj/F0T+xCcmwy8VFx59YRkVamcBYR\nvzJNk9KKWtblHKaiyvPu7p4DpeTsL6bW3eB17uUju3PhoE706pJkjbRuTl2Dm08LN/HarjesfQNT\n+3md02A0kBSTSJ+UTMBkUPsBdIhPbZ2OifiRwllE/CL3UCmLlu8h91CZz/M6JMcx8/K+DO+X5vM8\nwzQoKD/IZ0eycTlcfJD/kdfxeeNnkxSTeK5liwQFhbOItIoGw2B3QSl17gZWbT3Elj1FXseH9+1A\n59Q21vPitokx9O7qe7EcgAp3Jf/a+y7rDje9JO3kXldxdc/LtKiOhBWFs4icsaoaN4ePe0ZS17gb\nWL31EJ/uONrkuf97/0S6dUmmqKjpd4+bY5omZXXljZ4nX9B5BMPSBtM2JonuSd2IduprTMKP/laL\nyGmrqatn7eeFvPrB7mbPmTSqBylJMXRObcPwvp5b1adzVWuYBpXuKvLKClh1cD3bj+/0Ov6jobeT\nmZxBnAZvSQRQOIuIF3d9A+tyCq3lBU1g+WcFJMRF8+U3JgC5ZoxnacCGBpNRAzrSu2tbn0Fc6a5i\n7cFPOFh5mMOVR4h1nXo3eV9pXpOf6ZrQmbuH3kZqXLtz7JlI6FA4iwjguSp+Y9U+ln92oMnjx6jB\n5XRgmjB5XAbXjs0gOsrV5LknFVUXs6ckly927mTf8XxO1JY2Osfp8IzKduDAxGR42hDaxibxncyr\niXXF6lmyRCSFs0iE27q3iE93HGH9du95oqdcnEmXr6a7dDigf3o7EuOjWXVgPYcrt/FG7jbr3Hqj\nnnWHPyU55tSVc1NBfNKsgdPp1KYjPdv2UPiKNEHhLBJh6hsM7nl6FdEuJw6Hg4pq73WD75g8iAsG\ndWz0nvG+0jz+e8Pz1Jve7yZ/XWldGR3i2wPQIb49dQ11dIhPZXCXfgxpO4SuiZ1bv0MiYUjhLBLm\nTNOkuKyWDzcdYMMXhZyoqAOgzm3QrUMCyQkxdGnfhqzL+5IQH01stPet6pKaEyzY+levFZi6J3bl\n1kFZXuc5HQ46tkmzblN/XVpaEseOaeEHkdOlcBYJQ69/nMsXXxbjdDrIPdj0JCDNzVF90pGqY6w7\n9CnL8z+29sVHxfHL0ffRPl6Ds0T8SeEsEma25RbxzvpTI5+dDgeGaTJ6QEc6pcZz/cW9fX5+Z/Ee\nXt35T4prSrz2/2T4XfRr5/uzItI6FM4iYaSotJqnl3gGaqUkxvCHe8af9mdN02T1wQ0s3v2m1/57\nh91B35RMXE7fI7NFpPUonEVCmLu+gUNFVdb2oy+fmuLyybvHtfj5/3y5kkOVRyiuKSa39EuvY09M\n+DWJ0QmtVquInD6Fs0iI+rKwjN+8/FmTx568eyxRLu+BWUeqjvFh/iqKa0qIcrr4vGhHk58d3Wk4\ntwya0eTALhEJDIWzSIhpMAw+/eIoLyz9wto3aVQPwPM+8uiBHemQHG8dq6ir5BdrHm32z5vYfRxX\npE8k2hmtVZ1EgoTCWSREbNheyKc7jpK913u1p/k/vZg2cY3/UzZNkz9t+X/sObHP2pee1J1re02i\nd0ovAGJdMbpCFglCCmeRIFVeVceiD/dYt6dXbzvsdfz6izO5aEgXK5gN02DNwU/4V+67xEfFU1J7\nwuv8h0b/lB5JXQNTvIicE4WzSJApLqvhgefWNXksyuXgj/eOJy7G5TWDV019LX/Y/BwHKzwBXtNQ\nS/u4dpTUlvKdzKuZlHFJIEoXkVaicBYJIuu3F/LCv089S06Mj+au75xHhxTPMompSbGNFpt47JOn\nvGbvGpjajx8MvllLK4qEMIWzSJD4aMtBXnl/l7Xd3LPkk/aX5vP7TfOt7Y7xHbim1xWM7jRci0mI\nhDiFs4hN6hsMduaVUFdvMP+Nz72O/f6H43wG8/L8j3lz7zvW9q2Dsrig8wi/1SoigaVwFrHJ//1n\nF6u2eg/y6tMtmQdnDvO5TvKK/FVewfybsb/UXNciYUbhLGKDkvJaK5gvGtKZHh2TyOza1udCFDX1\ntdy/6hFrOzM5g/tH/sjvtYpI4CmcRQJo/+Ey1uUU8uGmA9a+268d1OLnVhSs5vU9/7a2h6UN5geD\nZ/mlRhGxn8JZJEDeXp3LC2/leO37049bXpiiur7GK5hnj3mQTm3SWr0+EQkeCmcRPzMMk2de38a2\n3OPWvgdnDqd/jxSczuZHVZfXVfBB3kd8WLDK2rfgsif9WquIBAeFs4gflVbUct/8tdb24MxUfjpt\nqM9QBnh+20uNFqb41YX3+6VGEQk+CmcRP/rVXz6x2vfcMJQRvdu3+JldxXu9gvmeoT+gb7tMopz6\nz1UkUui/dhE/2bL7GJU19QD84sbhjB+ZzrFj5T4/U+Wu4pnsPwPQIS6VR8c95Pc6RST4aDkaET+o\nqnHz7FcTi/Tq0pb+6S2/h1zpruLB1XOs7f/WbWyRiKUrZxE/uOfp1Vb7lze3PHNX9tHPeSFnobX9\nm7G/JMYV7ZfaRCT4KZxFWpFpmtz37Bpr+/E7LrSWfGzOqzuWsO7wRmt77vhHaBuT5LcaRST4KZxF\nWom73mDBm59TVuUGYPqlfejSPqHZ849UHeP/dixhX+mX1r5nLpmLy9n81J0iEhkUziLnaHfBCZas\n3EvuoTJr3x3fHsTY8zo3+5lvrijVO7kXPxt5t1/rFJHQoXAWOUumafL5vmKeXrLVa//3rhngM5iL\na0q8gvmXo39Kt8QufqtTREKPwlnkLLjrDe76/Ude+174+SW4nM0/Xy6rrWBn8R6ezX7B2veHiY8R\n64rxV5kiEqIUziJn6MV3drDm81NLPV46vBtTLs70GcwF5QeZt+JPXvtmj3lQwSwiTVI4i5wBwzSt\nYHY6HMz+3ijSO/keWT0/+y/sKN5tbV/YeSRX9bxMi1eISLMUziKnqb7BYPGHewFIiIvi2Z9e3OJn\ndhXvtYI5MSaBX11wP0kxiX6tU0RCn2YIEzlN63IK+XCzZx3mCwZ1avH88roKayrO1Lh2vDjl9wpm\nETktunIWaYFpmryzPo83Vu0DYPyQLtx4RV+fn6k36nlozW+s7UfH/sKvNYpIeFE4izSjuKyG9z8t\n4IPPCrz233J1f5+Dv1YdWMfi3W9Z249c+ABOh25SicjpUziLNKG4rIYHnlvnte/K0T2YflkfnI6m\n12I2TZN7Vz6EiWntu/v82+ic0NGvtYpI+FE4izTh68H8oymDOa9XKnExvv9z+Xowj+40nOv7TtYc\n2SJyVhTOIl8xTZPsPUXWUo8As783ip6d2/r8nNuo56cfPWxtX9PzCiZnXum3OkUk/CmcRYAGw+C3\nf/uM/CMV1r4fTB7YYjCbpukVzOO7XqhgFpFz5jOcDcNgzpw57N69m+joaB5//HHS09Ot4x988AHP\nP/88DoeDqVOnMnPmTL8XLNLaTNPkjic/srZH9kvj1msGkBjvez1l0zS5Z+WpUdgPjLyHXsnpPj4h\nInJ6fIbz8uXLcbvdLFq0iK1btzJv3jyee+456/jcuXN56623iI+P59prr2Xy5MkkJekZm4SWF9/d\nYbV/NGUII/u3PHNXSc0JfrXuf6ztHw+7U8EsIq3GZzhv3ryZCRMmADB06FBycnK8jkdHR1NWVobT\n6cQ0TRzNjGIVCVZPvLqZXQUnALhpUr8Wg9kwDbYf38nz21629n0782r6p/bxZ5kiEmF8hnNFRQWJ\niadmNHK5XBiGgfOrdzxvu+02pk6dSnx8PFdeeaXXuc1JS4vsK2v1Pzj6X1pRy6w5yzC/eutp7JAu\nZF090Odn6urruPn1n3jte+ba39A58fTnyA6W/tshkvsO6n+k9/9M+QznxMREKisrre2vB/OhQ4d4\n9dVXWbFiBfHx8Tz44IMsW7aMq6++2ucPPHasvBXKDk1paUnqfxD0v7Silvvmr7W2LxnejVuu6u+z\ntm+OyB6WNoSbB96AqzqOY9Wn16dg6b8dIrnvoP6r/2f+DxOf4TxixAhWrlzJNddcQ3Z2Nv3797eO\n1dbW4nQ6iYmJwel0kpqaSnl55P6fL6Hhm5OL/OqWUWR2bX5EtmEa/O6z+eSXH7D2PTjqHnq21fNl\nEfEfn+E8adIk1q5dS1ZWFuAZALZ06VKqqqqYPn06U6ZMISsri9jYWDIyMpgyZUpAihY5Gys3H2Dh\nf04t3fjb2y+gW1rTj2JKa8tZfXAdy/NX4TbcAMS6Yviv87+nYBYRv3OYpmm2fFrrifRbG+q/Pf0v\nr6rjJ8+ssbbn//Ri2sQ1/W/Tt3OX8X7eCq99Nw2YxriuF5xTDZH8+4/kvoP6r/638m1tkXDx9WD+\n6y8ubfbNgjf2LOXDglXW9ox+UxjecYiWehSRgFI4S9ib8+KnVvuP91zUbDDXNdRZwTwwtR/3DPtB\nQOoTEfkmrWMnYW37/mLyj3qm5LxkWFeSE2ObPffBVb+22j8c+n2/1yYi0hyFs4S1JR/tBaB7WiK3\nXD2g2fP2leZRbzYAcP/IH2n9ZRGxlW5rS9iqqqm3FrKYeXnTM3g1GA18UriJV3f+E/Dczs5MzghY\njSIiTVE4S1ha/lkBf1++x9oekNGuyfOe+OwZDlYctrZ/MPhmv9cmItIShbOEneraeq9gfuTWUU0O\nAnM3uK1gvrjbOKb2nUyUU/9JiIj99E0kYeeZf26z2r5em3o2+wWrPaP/dX6vS0TkdCmcJawUl9VY\nq0zNvLxvk8G8sXALHxasoqD8IAAPjf5Jo3NEROykcJaw4a5v4KnF2db2pNE9Gp3z521/Y2vRdmu7\nQ1wqPZK6BaQ+EZHTpXCWsFBR7ebHf1ptbf946vmNzlmRv8oK5kGp/blxwFTaxaUErEYRkdOlcJaQ\nZximVzDffu1AhvXtYG3XG/U8vOYxKuurAOiW2IUfDbs94HWKiJwuhbOEtPU5hbyw9Atre85to0nv\n5D3J/JI9b1vB3K9dH83+JSJBT+EsIWvNtsO8+O4Oa/u+6UMbBfOhikLWHNwAwI39p3JRtwsDWqOI\nyNlQOEtI2rjzqBXMyQkxPDhzOF07JHidY5gGj3/6B2v7XJd8FBEJFIWzhJRadwNP/n0z+w+fWhv2\nybvHER3VeC7sRbvetNp/nPh4s+87i4gEG4WzhAzTNLn7qY+t7VH907j92kFNBnNdQx1rD30CwHW9\nv0WMKzpgdYqInCuFs4SMNdtOzYH946nne43I/rp/7nmblQVrrO1JGZf4uzQRkValdfEkJJRV1vHS\nezsBuHxk92aD+V+573kF8y9G/Tgg9YmItCZdOUtIeO7Nz632zMv7Nnvef/JWAjAsbTB3DLnF73WJ\niPiDrpwl6FVUu9l9oBSAh24agdPZ9MCuk69MAQpmEQlpCmcJel+f/atfj+an23xz77sApCd193tN\nIiL+pNvaEtRM07Tac+8a0+Q57+7/gI1HtlDTUAOgGcBEJOQpnCWo/fEfW612p3ZtGh2vqa/lnf0f\nWNsXdb2QpJjEgNQmIuIvCmcJajn7iwG48zuDGh0zTZP7Vz1ibS+47MmA1SUi4k965ixB69MdR6z2\nmEGdGx3ffnyn1f7J8LsCUpOISCDoylmC0ivLdvJR9iEA+nRPbvKcFz5/BYAxnUfRr13vgNUmIuJv\nunKWoPPRloNWMPfsnMQvbhze6JyjVceoNxsAmNp3ckDrExHxN105S1DZf7iMV97fBUBCXBSzvze6\n0TnLvlzBv/ct85wT3YY20Y0HiomIhDJdOUvQOHC0gt/+7TNr++kfj290TkH5QSuYAf77gp8FpDYR\nkUDSlbMEjfc+ybfaL/z8ElzOxv92nLfxT1Zbo7NFJFzpylmCxvrthQA8+v0LmgzmhV/8w2o/c8nc\ngNUlIhJoCmcJCtu/ep8ZoEv7xs+QDdNgQ6Hnlve3el6By+kKWG0iIoGmcJag8PQSz0xgw/p0IMrV\n+K/lpiOnZgr7Vq9JAatLRMQOeuYstso/Us7S9Xk0GJ45tO/67nlNnvdB/kcAXNh5JA5H06tSiYiE\nC4Wz2GrOSxut9tjzOhMb3fh29dqDn3Cw4jAAV/e8LGC1iYjYReEstvn6c+Z5/zWWDslxTZ634sAa\nwPNOc8c2aQGpTUTETgpnscXRkiqeWpwNwODMVDqmxDd5Xn75AQorPXNs/2bsLwNWn4iInTQgTGzx\n0P/bYLXvvX5Ik+ccrDjMExufsbbjomL9XpeISDBQOEvAGaZptZ++dzzRUY2fMxumwf98+kdr+w8T\nHwtIbSIiwUC3tSXg7n16tdVumxDT9DkrH7Laz1wyV+81i0hE0ZWzBFReYTnVtfUAfP9bA5s85z9f\nrrTaPx52p4JZRCKOwlkCpqS8hkdf9rw6NaJfGuPP79Lkef/a9x4Al3YfT//UPgGrT0QkWCicJWBu\nmfO+1b7j24OaPGdFwalb3lP7ftvvNYmIBCOFswTER9kHrfbT945vcrKRoupiXt/zbwBGdxqumcBE\nJGIpnMXvqmrcvLJsFwCjBnRsdhDYnz//m9W+dVBWQGoTEQlGGq0tfvfYK5v4y1/u8Gzs2dvkOeV1\nFdYUnf9z0SO6ahaRiKYrZ/G7wuIqAFKT4nA2E7r/3rfMaifHJgWkLhGRYKVwFr8qKa+12tFRzf91\nW3voUwDuGnKr32sSEQl2CmfxG8MwuX/BWgCft6mPV5dY7SEdmh7FLSISSfTMWfyitLKO+55dY20n\nNzMIDGDNIc88290Tu+pZs4gIunIWP/l4y6lXpx7MGkaUy8ct7YOfAHCV1moWEQF05Sx+si6nEICf\nzxzOgIx2zZ5XU19DZb1nwNiQ9k1P5ykiEml8hrNhGMyZM4fdu3cTHR3N448/Tnp6unV827ZtPPHE\nE5imSadOnXjiiSeIiWn+9qVEBtM0OXqiGoB+6Sk+z7t/1WwAuiV2IdoVHZD6RESCnc/b2suXL8ft\ndrNo0SIeeOAB5s2bZx0zTZPZs2czb948/v73vzN27FgOHDjg94Il+P1nYwEALqej2VenAB7d8KTV\nvmnANL/XJSISKnxeOW/evJkJEyYAMHToUHJycqxj+/fvJyUlhZdeeok9e/YwceJEMjMz/VuthIST\n4fytMRnNnlNUfZxj1ccBuH3wzWS07RGQ2kREQoHPK+eKigoSExOtbZfLhWEYAJSUlLBlyxZuvvlm\nXnrpJdavX8+GDRv8W60EvQPHKqx3m787vlez5/16/RMAtImKZ0TH8wNSm4hIqPB55ZyYmEhlZaW1\nbRgGTqcnz1NSUkhPT7eulidMmEBOTg5jxozx+QPT0iJ79qdw7//vF2cD0CYuik6d2p464PTc3k5L\nS2JXUa61e/7k35IYmxDQGu0U7r9/XyK576D+R3r/z5TPcB4xYgQrV67kmmuuITs7m/79+1vHevTo\nQVVVFfn5+aSnp7Np0yamTWv5ueGxY+XnXnWISktLCuv+l1bU8sX+YgDumTLEq6+phonL6eDYsXLW\n5m4GoH1cO6rLDKoJ3/9Pvi7cf/++RHLfQf1X/8/8HyY+w3nSpEmsXbuWrCzPCkFz585l6dKlVFVV\nMX36dB5//HHuv/9+TNNkxIgRTJw48ewql7Dw9rovrbav16eW5a0A4Po+k/1dkohISPIZzg6Hg0cf\nfdRrX69ep54jjhkzhiVLlvinMgk5H285BMBPpjX/DPlA+SGrPTRtsN9rEhEJRZohTFpFSXkthmkC\nMKR3+2bPyzm+E4BRnYZpqk4RkWYonKVVvPTeDqDld5tPLg3Zv13fgNQlIhKKFM7SKnL2eQaC3Td9\naLPnmF9rX9h5hJ8rEhEJXQpnOWcv/Hu71R7ocx5tz/vPyTFJuJwuv9clIhKqFM5yTj7fd5z1248A\n8J2Lejb7HNk0TcprKwA4TwtciIj4pHCWc/LHf2wFoGuHBK6b0Pz0rSW1J6z29X2v9XtdIiKhTOEs\nZ626tt5q//p7o32ea341kvuHQ79PfFS8X+sSEQl1Cmc5a8VfzaE9ID2F6Kim/yo1GA38O9czQtvl\ndHFe+wEBq09EJFT5nIRExJc3PvbMkZ0Q1/Q6zOV1FTy05jcALJt/M/079ObHAatORCR0KZzlrG3Z\nUwTAwJ5Nj9B+ZsufrXZW/+v5zvmXUXK8KiC1iYiEMoWznJXCYk/IupwOLhvRvdHx7KOfc6iyEIB7\nh93BgNS+ROn1KRGR06JnznJW3lq9D4DzeqU2efwfu98CIMoZxYBUzQYmInImFM5yVgqOet5ZnnFZ\nn0bHjlcXU1rnWR7ul6N/GtC6RETCgcJZzlhFtZvDXz077tjO+7WoCncls9fPs7Y7J3QMaG0iIuFA\n4Sxn7F+r9wOe580up/dfoTnrn7Daj417OKB1iYiEC4WznLEPNx8A4MYrvJ8l1xv1VNfXADD7wgdo\nF5cS8NpERMKBwlnOSEW122pPHNbN69jJV6diXTF00u1sEZGzpnCWM/KXpV8AnlnBnE7vRS5yS78E\n4Ia+3w10WSIiYUXhLKctZ/9xtuUeB2Bonw5ex1YUrAbAgYOxXX3Psy0iIr4pnOW0VFS7+cNizwpU\n7ZJiueqCdOuYYRq8vuffAFyefrEt9YmIhBOFs5yWx/72mdV+4r/Geh3bWLjFan8786qA1SQiEq4U\nztKiWncDR09UA3Dv9UOIcnn/tXllx2IArul5OVFOzQgrInKuFM7SouyvFrgAGN4vzetYSc0Jq31l\nxmUBq0lEJJwpnKVF23I94Xzd+F6Njr29z7NWc2ZyT2JcTS8dKSIiZ0bhLC1av/0I0PiqGeDTws0A\njO96YUBrEhEJZwpn8enzfcetdo+OiV7H1hzcYLVHdBoasJpERMKdwll82pXveaY8qn/jq+a1hz4B\noF9Kb6I1EExEpNUonMWn/KOepR+vGZPR+Fj5QQBuGnhDQGsSEQl3Cmdplmma5OwrBhovDZlfdsBq\nd4hPDWhdIiLhTuEszfriyxKrnRDnPRJ7Q+Fn3zxdRERaicJZmvXU4mwAJg7r6rXfNE0+PrAOgAdG\n3hPwukREwp3CWZq090Cp1Z5+aR+vYydfnwLo2bZHwGoSEYkUCmdp0v/+KweA83u3Jz7WeyT2hsJN\nAGQmZ+DgHzDcAAAWZ0lEQVRwOBp9VkREzo3CWRo5UlxFSXktAN/6xijt0tpydpfsBWB6vykBr01E\nJBIonKWROS9tBCCjcxL9eqR4HVuw9S9Wu0eS97NoERFpHQpn8WKaJrXuBgB+Nt171q/1hzZysOIw\nALPHPBjw2kREIoXCWbw88apnsFfbNtEktYmx9le6q/i/nUsA6JHUjU5tGs8YJiIirUPhLBbTNNn9\n1Sjta8f19Dq26sB6q/3Q6J8EsiwRkYijcBbL9v3FVnvSKO9XpLYc2wbAVVqzWUTE7xTOYtm8x7Nu\nc/e0BK/9NfW11rPmCd3GBLwuEZFIo3AWAEoravloy1cLWUzq53Vs3sanrXa7OO/R2yIi0voUzoJp\nmtw3f621/fXXp/aU7ONYtWdN57vPvy3gtYmIRCKFs/Bx9iGr/evvjfaa9evpLc8DMKTDQAZ3GBjw\n2kREIpHCWfhs11EAxp/fhYzOSdb+KneV1Z6h2cBERAJG4SzW0pBXX5DutX9Z3goAEqMT9KxZRCSA\nFM4RbvW2U7e0u7RvY7UN0+DD/FUAjO96YcDrEhGJZArnCPfehnwALh/R3etZ84qC1Vb7W70mBbwu\nEZFIpnCOYKZpUljsea58/cRMr2Nv7n0HgCvSJ+JyugJem4hIJFM4R7A9X03VGRPl9Fqzed2hjVb7\nO5lXB7wuEZFIp3COYGu2eWb9Gj2go7XPNE1e/WqBi4ndx+mqWUTEBgrnCHaiohaAy0Z2t/a9sXep\n1Z7W9zsBr0lERBTOEe3wcc/z5vZt4wDPCO2TA8Fu7D8Vp0N/PURE7KBv3whlmCbHy2oASGoTDcDh\nyiPW8Yu66fUpERG7+AxnwzCYPXs2WVlZzJo1i/z8/CbPe+SRR3jqqaf8UqD4x678E1b75CtUe0/s\nB/Res4iI3XyG8/Lly3G73SxatIgHHniAefPmNTpn0aJF7Nmzx+sdWQl+q7Z6Jh8ZM6iTte8fu98C\noHtSV1tqEhERD5/hvHnzZiZMmADA0KFDycnJaXR827ZtzJgxA9M0/VeltLrDxysBz3zaACdqS61j\nF+nKWUTEVj7DuaKigsTERGvb5XJhGAYAR48eZcGCBcyePVvBHIJOlHtGamd2bQvAqzv+CYADhwaC\niYjYLMrXwcTERCorK61twzBwOj1f3O+//z4lJSXccccdFBUVUVNTQ+/evbnuuut8/sC0tCSfx8Nd\nsPS/wfD8g6pblxScTgdfFO8C4KfjbvdrjcHSf7tEcv8jue+g/kd6/8+Uz3AeMWIEK1eu5JprriE7\nO5v+/ftbx2bNmsWsWbMAePPNN9m3b1+LwQxw7Fj5OZYcutLSkoKi/4ZpUllTT6d28Rw/XsHx6hLr\nWJ+4fn6rMVj6b5dI7n8k9x3Uf/X/zP9h4jOcJ02axNq1a8nKygJg7ty5LF26lKqqKqZPn+51rgaE\nhY531ucBp66eNx/dCkCH+Pa21SQiIqf4DGeHw8Gjjz7qta9Xr16NzpsyZUrrViV+9e5X4XzRkC40\nGA28lfsuoHm0RUSChUb+RJj6BoNadwMA3x3fi6c2PWcdOz/tPLvKEhGRr1E4R5gX393htZ1XXgDA\n9wbNJNrp80aKiIgEiMI5wmzY7pmic8qEXl7vNo/uPNyukkRE5Bt0qRRByqrqrPa3L+rFj1b8HIAu\nCZ2a+4iIiNhAV84Ror7B4LG/fQZA24QYr2N3DrnVjpJERKQZCucI8fbaLykq9axCddOkfvw7d5l1\nrGObDnaVJSIiTVA4R4iNOzzPmr9zUU8G90liWd4KAC7qeoGdZYmISBMUzhHi5PPmyeN6smjXm9b+\nGwdMs6skERFphsI5AjQYBtW1DURHOYlyOfnsSDYAPxz6fZsrExGRpiicI8CSlbkAtG0T7bV/UGr/\npk4XERGbKZwjwH82eiYaGdm/I0XVxQB0apOm+dBFRIKUwjnMlVbUWu2sy/tSVH0cgM56t1lEJGgp\nnMPcyi0HAeielgjAC58vBKBdbLJtNYmIiG8K5zC34QvPK1SXDO8KQE2D513nK9In2laTiIj4pnAO\nc0dLqgEYM6gzu0v2WvvbxaXYVZKIiLRA4RzGausarHZ0DPxpy58ByEjqYVdJIiJyGhTOYWzbPs/g\nr5H90/j4wFpr/4Oj7rGrJBEROQ0K5zB2cu3mmCgnb+59B4BbBs7QK1QiIkFO4Rym6twN1m3tqy5u\nZ+0f1WmYXSWJiMhpUjiHqRWbPa9QJcRF8fb+dwHITO6Jy+mysywRETkNCucw9frHnik7p07MZEfx\nbgCu7zPZzpJEROQ0KZzDVINhAtA1w23t65Wcblc5IiJyBhTOYWj11kNWe1vR5wBc2HmkXeWIiMgZ\nUjiHmfoGg5fe2wnANWPSWXlgDQBjuoyysywRETkDCucw8+6GPKs98YJTs4D1Tu5pQzUiInI2FM5h\n5p31nnD+weSBrDv0KeBZt1mjtEVEQofCOYzUuhtw1xsAZGTAB/kfATC262gbqxIRkTOlcA4jK796\ntzkuxsVfti+09o/oeL5dJYmIyFlQOIeRf6z0rDr1rTEZHK0qAuA3Y39pZ0kiInIWFM5h6KIRpwaC\ntY9v5+NMEREJRgrnMFFWVQdARqckntr0HAC9k3vZWZKIiJwlhXOYeP/TfAAaXDWU1J4AYGpfTdcp\nIhKKFM5h4r0NnnDumlkKQGJ0Ahlte9hZkoiInCWFc5jJYwsAV2VcanMlIiJythTOYaC4rMbTcDRQ\nVlcOwKU9JthYkYiInAuFcxhY+/lhABJGrrL2ORwOu8oREZFzpHAOA9l7iyCqFsNZC8DDF9xnc0Ui\nInIuFM4hrqyyjv2Hy4nusRuAhKg2dEvsYnNVIiJyLhTOIe7nz68DwJlUDMDUvt+2sxwREWkFCucQ\n9tGWg9S5PQtdOOOqARjdebidJYmISCtQOIewHXklALTtdsza53ToVyoiEur0TR7CNu48CoC72yYA\nxnbR0pAiIuFA4RziXGn5VvvGAVNtrERERFqLwjlEVVS7cSScIKbXFwB0S+yiW9oiImFC3+Yh6v/+\ns4uY3tus7QdG/sjGakREpDUpnEPUpzuO4IyrAuCxcQ8T44qxuSIREWktCucQVFVTjzO5yNpuF5di\nYzUiItLaFM4h6GBRBa52RwAYmNrP5mpERKS1KZxD0Hsb8onqeACAy9MvtrkaERFpbQrnEJSTd9Rq\n903JtLESERHxB4VziDFMEzrmAp7Xp6KcUTZXJCIirU3hHGI+zy0iupsnnC/sPNLmakRExB8UziGm\noKzQal/aY7yNlYiIiL8onEPMit2eiUd6xPTVjGAiImHK5wNLwzCYM2cOu3fvJjo6mscff5z09HTr\n+NKlS3nllVdwuVz069ePOXPm4HA4/F50JKtOzMUJDO7Ux+5SRETET3xeei1fvhy3282iRYt44IEH\nmDdvnnWspqaGP/3pTyxcuJDXXnuNiooKVq5c6feCI1lpTTnONhUAjE8fYXM1IiLiLz7DefPmzUyY\nMAGAoUOHkpOTYx2LjY1l8eLFxMbGAlBfX09cXJwfS5Xfb3weAKcRRUpsss3ViIiIv/i8rV1RUUFi\nYqK17XK5MAwDp9OJw+EgNTUVgIULF1JdXc24ceNa/IFpaUnnWHJoO9v+V9RWUuw+BkD/uutC9v/H\nUK27tURy/yO576D+R3r/z5TPcE5MTKSystLaPhnMX9/+3e9+R15eHs8+++xp/cBjx8rPstTQl5aW\ndNb9X1mwBgCjpg2zJg4Lyf8fz6X/4SCS+x/JfQf1X/0/83+Y+LytPWLECFatWgVAdnY2/fv39zo+\ne/Zs6urqWLBggXV7W/zjw3zP76G+sCdJbaJtrkZERPzJ55XzpEmTWLt2LVlZWQDMnTuXpUuXUlVV\nxeDBg3n99dcZNWoUt9xyCwC33norV1xxhf+rjkCltWUANBzvglMj4kVEwprPcHY4HDz66KNe+3r1\n6mW1d+zY4Z+qxEtB+UEMDEx3NOPPS2/5AyIiEtI0i0UI+OjAWgCM6kSmTNBCFyIi4U7hHAI2Fm4B\noP5gX9ol6dm+iEi4UzgHuX2lX9JgNgAwuKNmBRMRiQQK5yB3cpR2Q2l7hvdNs7kaEREJBIVzkMs+\n5pmVrf5gH3p2bmtzNSIiEggK5yBWUnPCahuVyaR3SvRxtoiIhAuFcxD7sqwAAKO6DZ3bJWrFLxGR\nCKFwDmJrDm4AwCjrQK8umpdWRCRSKJyD2M6SPYBnVrC6esPmakREJFAUzkHKbdRbbaOiHSP7a6S2\niEikUDgHqWVffuhpGJ5f0fA+CmcRkUihcA5SX5bmA+A+3JMol4PYGJfNFYmISKD4XPhC7GM9by7q\nhtlg2lyNiIgEkq6cg5Bhnhr8Zda24crRPWysRkREAk3hHIRO1JYCnvebwcGlI7rZW5CIiASUwjkI\nvZ+3EvDMCgbQqV0bO8sREZEAUzgHocMVhYDn/eaEOA0LEBGJNArnIJRb+iXgmRnskuG6pS0iEmkU\nzkGmoPzQqQ3TSa8uWolKRCTSKJyDzB82PweAUeF53nx+7/Z2liMiIjZQOAeRBqOBuoY6AGp3jwQg\nyqVfkYhIpNE3fxDZ99Wz5uj6JKiPIb2j1m8WEYlECucgsrNkLwDVxZ5b2ndfN9jOckRExCYK5yCS\ne2I/AEZFCgCdUvV+s4hIJFI4B5E9J/YB0FCaxsRhXW2uRkRE7KJwDhLV9dWnNtyx3DSpn33FiIiI\nrRTOQaKw8hgAZl0sbRNiNEpbRCSCKQGCxMqC1QA0nEizuRIREbGbwjlI5J7IA6ChpBP3zxhmczUi\nImInhXMQqKmv5UTdCQDMslR66P1mEZGIpnAOAm/sXWq1r7+4r42ViIhIMFA4B4F9RZ4lIuv2DmXS\nqB42VyMiInZTOAeBw7X5AAxI7UdMtMvmakRExG4KZ5udqC0FhwnAD7+tgWAiIqJwtt0H+9YAYFQl\nEh8bbXM1IiISDBTONluVUwBA2wY9axYREQ+Fs80aUj2LXdwy+jKbKxERkWChcLbRidpSHF89bx7U\nJd3makREJFgonG30/NaXAYiqTcHhcNhbjIiIBA2Fs40KKg4CEHd8qM2ViIhIMFE426S8rsJq927b\n075CREQk6CicbfLe/g8BME0HM6/QlJ0iInKKwtkmHx9cC0BC0QiS2sTYXI2IiAQThbNNTM8gbfol\nDLS3EBERCToKZxtU17pxOMCsi2Xm5QpnERHxpnC2wVvbNgDgwElivKbsFBERbwpnG3yUuwWAdtHt\nba5ERESCkcI5wI6X1uCIqQVgxpCrbK5GRESCkcI5wJZ9mocr5RgAfVIzbK5GRESCkcI5wNa5/mq1\n46PibKxERESClcI5gHYczrfaN/afamMlIiISzBTOAfTixtc9japkLup2ob3FiIhI0FI4B9Dh6gIA\nJnS43OZKREQkmPkMZ8MwmD17NllZWcyaNYv8/Hyv4ytWrGDatGlkZWWxZMkSvxYa6hoMgzqqARjU\nKdPmakREJJj5DOfly5fjdrtZtGgRDzzwAPPmzbOOud1u5s2bx0svvcTChQtZvHgxx48f93vBoWrr\nl4cBMOujGdhD7zeLiEjzonwd3Lx5MxMmTABg6NCh5OTkWMdyc3NJT08nKSkJgJEjR7Jx40auvvpq\nP5Zrr2MVZRiGeVafffWLN6ANxLiiiI5ytXJlIiISTnyGc0VFBYmJida2y+XCMAycTicVFRVWMAMk\nJCRQXl7uv0pt9uzqN9npXn/2f0Abz/9c1eXa1ilIRETCls9wTkxMpLKy0to+GcwASUlJXscqKytJ\nTk5u8QempSW1eE4w+s31twC32F1GyAvV339rieT+R3LfQf2P9P6fKZ/PnEeMGMGqVasAyM7Opn//\n/taxzMxM8vLyKC0tpa6ujo0bNzJs2DD/VisiIhIBHKZpNvsQ1TRN5syZw65duwCYO3cu27dvp6qq\niunTp7Ny5UoWLFiAYRhMmzaNG2+8MWCFi4iIhCuf4SwiIiKBp0lIREREgozCWUREJMgonEVERIKM\nwllERCTI+DWca2pquPfee7npppu48847KS4ubnTOyy+/zPTp05k+fTrz58/3ZzkBE8lzkrfU96VL\nlzJ9+nRmzpzJr3/9a8JtPGJL/T/pkUce4amnngpwdf7XUv+3bdvGTTfdxI033sh9991HXV2dTZX6\nR0v9/+CDD5g6dSrTpk3jtddes6lK/9q6dSuzZs1qtD+cv/e+rrn+n/F3n+lHL774ovnss8+apmma\n77zzjvnYY495Hc/Pzzevv/560zAM0zRNMysry9y5c6c/SwqI999/33zooYdM0zTN7Oxs8+6777aO\n1dXVmZMmTTLLysrMuro6c+rUqWZRUZFdpbY6X32vrq42r7jiCrOmpsY0TdP82c9+Zn744Ye21Okv\nvvp/0muvvWbOmDHDfOqppwJdnt/56r9hGOZ3v/tdMz8/3zRN01y8eLGZm5trS53+0tLv/9JLLzVL\nS0u9vgfCyZ///Gdz8uTJ5owZM7z2h/v33knN9f9svvv8euW8efNmLr74YgAmTJjA+vXe01926dKF\nv/71rzgcDgDq6+uJi4vzZ0kBcbpzkkdHR1tzkocLX32PjY1l8eLFxMbGAuHz+/46X/0/eXzbtm3M\nmDEj7O4agO/+79+/n5SUFF566SVmzZpFWVkZmZnhtUJbS7//6OhoysrKqK2txTRN67svXGRkZDB/\n/vxGf7fD/XvvpOb6fzbffT6n7zwTS5Ys4ZVXXvHa1759exISEoCm596OiooiJSUF0zR58sknGTRo\nEBkZGa1Vkm0ieU5yX313OBykpqYCsHDhQqqrqxk3bpxdpfqFr/4fPXqUBQsWsGDBAt59910bq/Qf\nX/0vKSlhy5YtzJ49m/T0dO666y4GDx7MmDFjbKy4dfnqP8Btt93G1KlTiY+P58orr/Q6NxxceeWV\nHDhwoNH+cP/eO6m5/p/Nd1+rhfMNN9zADTfc4LXv3nvvtebfrqyspG3bto0+V1tby8MPP0xiYiJz\n5sxprXJs5Y85yUOFr76f3P7d735HXl4ezz77rB0l+pWv/r///vuUlJRwxx13UFRURE1NDb179+a6\n666zq9xW56v/KSkppKenW1fLEyZMICcnJ6zC2Vf/Dx06xKuvvsqKFSuIj4/nwQcfZNmyZWG9kt9J\n4f69dzrO9LvPr7e1vz4396pVqxg1apTXcdM0+eEPf8iAAQN49NFHw+YWTyTPSe6r7wCzZ8+mrq6O\nBQsWWLd4womv/s+aNYs33niDhQsXcueddzJ58uSwCmbw3f8ePXpQVVVlDZLatGkTffv2taVOf/HV\n/9raWpxOJzExMTidTlJTU8Py6rEp4f69dzrO9Luv1a6cmzJz5kx+8YtfcOONNxITE2ONTn355ZdJ\nT0/HMAw2btyI2+22/kLff//9If9LmzRpEmvXriUrKwvwzEm+dOlSa07yhx56iNtvv92ak7xjx442\nV9x6fPV98ODBvP7664waNYpbbvGs8HXrrbdyxRVX2Flyq2rpd/914fKP0a9rqf+PP/44999/P6Zp\nMmLECCZOnGhzxa2rpf5PmTKFrKwsYmNjycjIYMqUKTZX7B8n/25HyvfeN32z/2fz3ae5tUVERIKM\nJiEREREJMgpnERGRIKNwFhERCTIKZxERkSCjcBYREQkyCmcREZEgo3AWEREJMv8f0Sxt8SnXYp4A\nAAAASUVORK5CYII=\n",
      "text/plain": "<matplotlib.figure.Figure at 0x21eb66d8>"
     },
     "metadata": {}
    }
   ]
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": true
   },
   "cell_type": "code",
   "source": "# calculate D statistics for all pairs of a given label\nlabel = 'a1sb1s'\n",
   "execution_count": 13,
   "outputs": []
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false,
    "scrolled": false
   },
   "cell_type": "code",
   "source": "lpn1, lpn2, lpn3 = 'adult_m1_s3_fv2', 'adult_m2_s6_fv3', 'adult_m3_s5_fv2'\nd1 = many_corrs[lpn1]['s1sgcs']['corrs']\nd2 = many_corrs[lpn2]['s1sgcs']['corrs']\nks_2s_statistics(d1, d2)\nplt.legend(labels=[lpn1, lpn2], loc='upper left');",
   "execution_count": 14,
   "outputs": [
    {
     "output_type": "stream",
     "text": "KS D statistics: 0.270614762475 \np-value: 1.02882553198e-28\n",
     "name": "stdout"
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAecAAAFVCAYAAADVDycqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niIhsDMOZiIjIxvw/Ne7lzl1Yz+QAAAAASUVORK5CYII=\n",
      "text/plain": "<matplotlib.figure.Figure at 0x19099b38>"
     },
     "metadata": {}
    }
   ]
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false
   },
   "cell_type": "code",
   "source": "# create nice color palettes\n# sns.palplot(sns.color_palette(\"Blues\"))\n# sns.palplot(sns.color_palette(\"Greens\"))\n# sns.palplot(sns.color_palette(\"Oranges\"))\ncolors = [sns.color_palette(\"Oranges\")[i] for i in [0, 2, 4]] \\\n         + [sns.color_palette(\"Blues\")[i] for i in [2, 4]] \\\n         + [sns.color_palette(\"Greens\")[i] for i in [2, 4]]\nprint len(colors)",
   "execution_count": null,
   "outputs": []
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false
   },
   "cell_type": "code",
   "source": "sns.set_style('whitegrid')\nlabel = 'a1sb1s'\n# sns.set_palette(colors)\nfor label in ['a1sb1s', 'a1sb3s', 'a3sb1s', 'a3sb3s', \n              'a1sc1s', 'a1sc3s', 'a3sc1s', 'a3sc3s', \n              'a1ss1s', 'a1ss3s', 'a3ss1s', 'a3ss3s', \n              'a1sshape1s', 'a1sshape3s', 'a3sshape1s', 'a3sshape3s', \n              'a1sgcs', 'a1sgcs', 'a3sgcs', 'a3sgcs', \n              'b1sc1s', 'b1sc3s', 'b3sc1s', 'b3sc3s', \n              'b1ss1s', 'b1ss3s', 'b3ss1s', 'b3ss3s', \n              'b1sshape1s', 'b1sshape3s', 'b3sshape1s', 'b3sshape3s', \n              'b1sgcs', 'b1sgcs', 'b3sgcs', 'b3sgcs', \n              'c1sshape1s', 'c1sshape3s', 'c3sshape1s', 'c3sshape3s', \n              'c1sgcs', 'c1sgcs', 'c3sgcs', 'c3sgcs', \n              's1sshape1s', 's1sshape3s', 's3sshape1s', 's3sshape3s', \n              's1sgcs', 's1sgcs', 's3sgcs', 's3sgcs', \n              's1sc1s', 's1sc3s', 's3sc1s', 's3sc3s', ]:\n    liver_data = [many_corrs[lpn][label]['corrs'] for lpn in all_lpns]\n    plt.clf()\n    sns.boxplot(data=liver_data, palette=colors)\n    sns.despine()\n    plt.ylim(-0.1, 1.1)\n    plt.axhline(0.5, color='black', zorder=0)\n    plt.xticks(np.arange(len(all_lpns)),all_lpns, rotation=10)\n    plt.title('{} {}'.format(label[:2], label[3:-1]))\n    plt.tight_layout()\n    fig_name = MCSDOCPATH + '/figures/notebooks/statistical_test/box_plots/adult_kd_luc/{}_{}.png'.format(label[:2], label[3:-1])\n    create_pardirs(fig_name)\n    plt.savefig(fig_name, transparent=False, dpi=300)\n#     plt.show()",
   "execution_count": null,
   "outputs": []
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "# 4. Analysis of Variance"
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "Compare variances of experimentally observed data to be able to make statements about significant differences."
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": false
   },
   "cell_type": "code",
   "source": "# plt.figure(figsize=(10, 7))\n# fig = plt.figure(figsize=(10, 10))\nsns.set_style(\"darkgrid\")\ndf.plot(kind='hist', stacked=False, bins=20, alpha=0.5, figure=fig)\nsns.despine()\nplt.show()",
   "execution_count": 59,
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEJCAYAAACDscAcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XlYVOX7x/H3MOy4gIYauCu55IKmlkpauJUYbim5byla\nWi65lGiWoeaGaEWIqCkmqbmlmaam5s8lTc1v5oYbuCFqKqIwMHN+f5ATI4szbHPI+3VdXMHzzDnn\nM4ecm7M9j0ZRFAUhhBDCTDbWDiCEEKJokcIhhBDCIlI4hBBCWEQKhxBCCItI4RBCCGERKRxCCCEs\nYrXCMXnyZIKCgkzaVq9ezWuvvYa3tzd+fn6sXbvWSumEEEJkp9ALh6IohIaGsmrVKjQajbF9586d\nTJkyhYEDB7Jp0yb69u3LpEmT2L17d2FHFEIIkYNCLRxxcXH07duX6OhoPDw8TPr2799PjRo16N69\nO+XLlycgIIDatWuzd+/ewowohBDiCQq1cBw9ehRPT082bdqEp6enSV+DBg2IiYnh4MGDKIrCoUOH\nOHv2LHXq1CnMiEIIIZ7AtjA35u/vj7+/f5Z97du35+zZs/Tr1w+tVoter2fQoEF07NixMCMKIYR4\ngkItHDn58ccfWbRoEePGjaNZs2YcPnyY2bNnU7VqVbp27WrteEIIIf6hmsIxf/58unbtysCBAwGo\nWbMmt2/fZtasWVI4hBBCRVTzHMf169czXc+oV68ed+7cITExMcdlZYBfIYQoPKo54qhUqRKnTp0y\naTtz5gxubm4UL148x2U1Gg0JCTkXFzVzdy9eZPMX5ewg+a1N8luPu3vOn6s5sWrhyHikMHjwYD78\n8EOqV69O8+bNOXbsGAsXLmT48OFWTCiEEOJxVi0cGR8A7NChAwCLFi3i888/x9PTk9GjR9OzZ09r\nxRNCCJEFqxWO5cuXZ2rr0KGDsYAIIYRQJ9VcHBdCCFE0SOEQQghhESkcQgghLCKFQwghhEWkcAgh\nhLCIFA4hhBAWUc2T40+n9AcgDQaD8XvzaZ78EiGEKABSOKxsw1/x2Drc4kGSzqzX29va0LF22QJO\nJYQQ2ZPCYWW6NAMGW4UUvblHHIYCzSOeFtYeGFSOmIsyKRxCPKXStn2PkmrekW5+0djZY9tWXdMk\nxMdf588/j9OqVdtcryMyMpyff/6J6Oh1Ju2nTv3F4MH9ePHFpsyePT+vUVVDCocQTyklVQe6wi0c\n1j7Oycr06Z9SpkzZPBWOdJmPon788QcqVKjIoUMHuX79OuXKlcvjNtRB7qoSQjzV8m86H9MVpaam\nsmPHNnr06EPx4iXYtGl9fm3I6qRwCCFU78GDJObM+Rx//3a0bduS0aNHEBt7CYC9e3czaFAfWrf2\noWvXDixevBC9Xg/AkSOH8fVtxq+/7qJnz674+jZj4MBeHD9+DIDg4CkcOXKILVs20aJFEwCGDx/C\nzJnBDBzYm9df92Xfvr3o9XpWrPiGt97qjK9vc/r1e4udO7fnmHnv3t3cu3ePF19syssvv8LmzRv/\nuYOy6LNa4Zg8eTJBQUEmbTExMQwcOBBvb29atGhBaGiozO4nhGDSpA85evQwU6YEs2jRMpydnRgz\nZgQ7dmxj4sRxtGrVlm++ieadd95nzZrvWLBgrnHZtLQ0lixZxIQJk1iy5FtcXIoxbdqnAIwc+QH1\n6zfA17cNGzb8ZFxm8+aN9Os3kC++WIi3d0MWLJjLypVRDBs2gmXLomnduh1TpnzEtm3bss28Zcsm\natasTZkyZfH1bcXNmwns2/drwe2kQlTohUNRFEJDQ1m1apXJfBy3b9+mT58+uLm5sX79ej7++GOi\noqJYvHhxYUcUQqhIbOxFfvttPx988CENGzaiYsVKjBs3kZYtfVm06GtatWpLz5598PQsT6tWbRgy\nZBjr13/PgwdJQPpnTmDgu9Sr502lSpUJCOjJlStx3L17BxeXYmi1tjg4OODmVsq4zdq1n6dlS1+q\nVauOohhYv/57AgPfpWVLX8qXr0CfPgN49dVWLFy4MMvMt27d5LffDuDr2waAF15ogptbKTZuXJfl\n64uaQi0ccXFx9O3bl+joaDw8PEz6oqKiKFGiBDNnzqRy5cq0atWK/v37c+TIkcKMKIRQmXPnYgCo\nVet5Y1uJEiUZPnwkCQk3qFOnnsnr69VrgF6v59Kli8a2ChUqGr93dnYBIDU1Ldttenh4Gr+/dOki\ner0+y+2cPXs2y+W3bt2CXq/n1VdbAWBjY0OLFq9y8OB+4uOv5/R2i4RCvavq6NGjeHp6Mm/ePEaO\nHGnSt3fvXlq3bo1WqzW2vfvuu4UZTwihQra22X9MOTg4ZGp7dB0h43L29vZZLJ1+Glzz2M1QGo0G\ne3vHDNtwJCsGgyHbbD/9tAmAgIBOmZbZtGkDgwYFZrlcUVGoRxz+/v7MmDGD0qVLZ+q7dOkSpUuX\nZurUqbRo0YLXX3+dhQsX/mcuJgkhcqdSpSpA+jMRjyQl3adDhzYkJiYaL3Q/cvz4Mezs7PD0LJ8v\n2y9fvjx2dnZZbsfLyyvT60+d+osLF84zZMi7LF260vi1ZMm3VK1a/T9xkVw1z3EkJiby9ddf06VL\nF8LDwzlz5gyfffYZycnJvPfee9aOJ8R/jsbOvtCfq9DYZfWXf84qVqyEj09L5syZwZgxH+Lq6srC\nhV9SvHhxJk+eytix7/PcczV4+eVXOHv2DIsXh/PGG52Mp6SexMXFhatXrxifs0i/IeffPePg4EhA\nQC8iIsIoWbIk1ap5sWvXTvbs+YWQkJBM6/vxxx8oVqw43bu/lelopXv3HsyYMZV9+/bi49PC4n2h\nFqopHLa2ttSoUYMJEyYAUKtWLW7dusVXX31lVuFwdy9e0BHzncFgwNnFHp1ewcXFvH9Q9loNpUsX\nw8ZGPXdSF8V9n9HTmF9RFJQe/QogzZNpNBqTG2PMyR8SMpvp06czceIH6PV6GjduzJIli6lQoQIG\nw+eEh4cTERFGuXLl6N+/P4GBgdjY2ODq6oxGo6FUKRfjdh5vGzCgH2PHjqVv3+78/PPP2Nvb4uRk\nb5JrwoQPKFbMkS++COHvv/+mevXqhISE0K5dOwCKFXPE1laLq6sjO3f+TJcunSlf3j3T++jZsxsR\nEV+xdesPdO7sl9ddaTUaxUr3u/bp04fKlSszdepUANq2bUuLFi1MbtHdvXs3gYGBHDx4kJIlS+a4\nvoSExALNWzAUVh+/hq2jHUlmDnLooNXQrd6zqGWsH3f34kV036eT/NYl+a0nL38wqebP1hdeeIHj\nx4+btJ05cwZXV9cnFg0hhBCFx6qFI+PBzqBBgzh9+jTTp0/n0qVLbNu2jYiICPr27WvFhEIIIR5n\n1WscGc9zVq9enSVLljBz5kxWrlxJ6dKlGTRoEEOGDLFiQiGEEI+zWuFYvnx5praGDRsSHR1thTRC\nCCHMpZprHEIIIYoGKRxCCCEsIoVDCCGERaRwCCGEsIhqnhwXQhQma89zo44HWEXuSOEQ4im14a94\ndGmFO9ieva0NHWuXLdRtPkl8/HX+/PN4nuYcj4wM5+effyI6On2+jZdfbmzS7+joSKVKVRg4cAjN\nmvnkKa8aSOEQ4imlSzOQoi/sIw/1jQo7ffqnlClTNk+FIyujR4/nlVd8URSFpKQktm/fykcffcCi\nRcupXj3zqLpFiVzjEEI81QpqtD4Xl2K4uZWiVKnSVKhQkQEDBuPh4cnPP//05IVVTgqHEEL1HjxI\nYs6cz/H3b0fbti0ZPXoEsbGXANi7dzeDBvWhdWsfunbtwOLFC9Hr9QAcOXIYX99m/PrrLnr27Iqv\nbzMGDuxlnFsjOHgKR44cYsuWTbRo0QSA4cOHMHNmMAMH9ub1133Zt28ver2eFSu+4a23OuPr25x+\n/d5i587tFr8PR0dHVY1snVtF/x0IIf7zJk36kKNHDzNlSjCLFi3D2dmJMWNGsGPHNiZOHEerVm35\n5pto3nnnfdas+Y4FC+Yal01LS2PJkkVMmDCJJUu+xcWlGNOmfQrAyJEfUL9+A3x927Bhw79HAps3\nb6Rfv4F88cVCvL0bsmDBXFaujGLYsBEsWxZN69btmDLlI7Zt25ZD6n8PZdLS0vj555+4dOki7dq1\nz/f9U9jkGocQQtViYy/y22/7WbAgHG/vhgCMGzeRZcuWsGjR17Rq1ZaePfsA4OlZnsTEu8ybN5sh\nQ94B0gdTDQx8l3r1vAEICOjJhx9+wN27dyhZ0hWt1hYHBwfc3EoZt1m79vO0bOkLpM82uH7994wZ\nM8HY1qfPAGJizrBw4ULCwppmmXvGjM+YNWs6ADpdCgaDga5dA6hUqXL+76RCJoVDCKFq587FAFCr\n1vPGthIlSjJ8+EjWr19Dt249TF5fr14D9Ho9ly5dNLZVqFDR+P2jmQFTU9Oy3aaHh6fx+0uXLqLX\n66lTp16m7YSFzc92HYGB7+Dj0xKAlJQUTp36iwULQtDr9YwZMz7b5YoCKRxCCFWztc3+Y8rBwSFT\n26P5vDMuZ2+f1Qyb6aeSNI89UqLRaLC3/3fK18enf824nZyyubmVNpn3vGrVaty8mcCiRV/zzjvv\n4eTklO2yame1axyTJ082me0vo9TUVDp16sSHH35YyKmEEGpTqVIVAE6d+svYlpR0nw4d2pCYmGi8\n0P3I8ePHsLOzM/nQzovy5ctjZ2eX5Xa8vCy7rfZRUVMU9d2WbIlCLxyKohAaGsqqVatM5uPIaP78\n+Zw6daqQkwkh1KhixUr4+LRkzpwZ/PHHMS5dusi0aZ9QvHhxZs+ezy+/bOfbb5cRFxfLzp3bWbw4\nnDfe6GQ8JfUkLi4uXL16hevXrwOPJpj798K2g4MjAQG9iIgIY9euHcTFxbJ8+VL27PmFAQMGZLve\n+/cTuXXrJrdu3SQh4QZ79uxi9eqV+Pi0NDubWhXqqaq4uDg++ugjYmJi8PDwyPI1v//+O2vXruW5\n554rzGhCPHXsbW0o7Afy0rdpuYkTp7BgwVw+/HAMBoMeb++GzJmzAA8PTyZO/ITlyxcTERGGu3sZ\nunXrQZ8+/36gZ/UHasa2Ll268+mnQfTp043vvlv/T5/pMoMGBWJjY8P8+XO5c+cOlStX4ZNPptGu\nXTsSEhKzXCYkZCYhITMB0Gq1PPOMO61btyMw8N1c7QM10ShKQT3+ktnGjRvZt28fY8eOZeTIkVSu\nXJmpU6ca+5OSkujUqRNjxozhu+++o1y5ckyfPt2sdRfNCeMVVh+/hq2jHUlJOrOWcNBq6FbvWdQy\n1o+7e/Eiuu/TPb351TFW1dO7/63P3b14rpct1FNV/v7+zJgxg9KlS2fZP23aNOrVq8drr71GIdYz\nIZ5CGit/iaJMNXdV7dixg19//ZVNmzYBWR9eCiGEsD5VFI7bt28zadIkpk+fTokSJQAsPuLIy2GX\ntRgMBpxd7NHpFVxcsrpdMDN7rYbSpYupatiCorjvM5L81iX5ix5VFI7du3dz+/ZtRo4caWzT6XRo\nNBq2bt3KkSNHnriOonmeUeFBks6iaxxpWg23bt1HLYf7RfkcL0h+a5P81pOXgqeKwtG2bVsaNWpk\n/FlRFMaPH4+7uztjx461YjIhhBCPs2rheHQ6ysXFBRcX0/uaHRwccHFxoUKFCtaIJoQQIhtWPVGe\n0wVwuTguhBDqZLUjjuXLl+fYv2TJkkJKIoQQwhLquTVHCCFEkaCKi+NCiMJm7Qds5VR0USaFQ4in\nVMypRPRphVtAtLYaqte0/DbQCxfOc/36VZo29SmAVOn27duLh4cnlStXKbBtPMm9e3f59dfd+Pn5\nZ+obN24k+/f/H+HhS6hdu44V0v1LTlUJ8ZTSpyno9RTuVy4L1YcfjuHUqZP5vAf+lZBwg/HjR3Hn\nzt8Ftg1zhIUt4KefNmdqv3XrJgcP7qdixUps3LjOCslMSeEQQhQJBTl+3aN1W3uMvOy2v3XrFtzd\ny9C585vs2PEzDx4kFXIyU1I4hBCqNnz4EK5cucySJRF06+ZPt27+fPHFPHr06EKHDm04e/Y0Op2O\nBQtC6NjxNdq1a8nw4UM4ceJP4zoMBgNLly7irbe68OqrTXnttVcJChrHnTt3AOjatQMA7703lGnT\nPuHatau8/HJjdu7cTr9+PfD1bc7gwf2Ii4slMjKcDh3a0L59Kz777DOTrHv27DK+vlevN1m5MspY\nDB6tc/funQwc2Atf32b07NmVX3/dBUBkZDibN2/k2LEjvPxyY+P8IAA//bSJRo2a0KLFq6SkJLNt\n25aC3OVPJIVDCKFq06bNplw5D3r06E1ExDIURWHDhu+ZMGESs2bNo3r15/jss485fvwYU6fOIDIy\nihdeaMyIEYHExcUCEB29gjVrvmP06HFER69jypRgjh//g2XLIgFYvDgKgODgWYwc+YFx2xERXzF6\n9DgiIr7h3r27BAYOID7+OmFhkQwZ8g5RUVEcOLAPgP379zJ16iQCAnoSFbWKYcPeY/XqlSxdusjk\n/Xz11XyGDh3B8uWr8PJ6juDgKaSkJNOzZ1/atHmNOnXqsXHjVsqUKQOkz3x44cJ5Wrb0pUyZsjz/\nfF02blxf4Ps9J1I4hBCqVqJECbRaG5ycnHF1dQXAx6cl9es3oFat57ly5TK//LKdjz76mHr1vClf\nvgIDBgymXr36REenF4RKlSoTFPQJTZq8RNmy5XjppWY0afIS586dA6BkSVfjtjLOzterV1/q129A\ntWrVjX/tjx37ERUqVKRTp66ULl2aCxfOA7Bs2RI6d+5G+/Zv4OHhiY9PC4YOHc6KFd+YvJ8ePfrQ\npMlLeHqWp0+fgSQlJXHhwgWcnJywt7fH1tYWN7dSxoFMf/zxB4oXL0Hjxi8C0Lp1W86ePW0ylW5h\nk7uqhBBFikajwcPD0/jzmTOnARgypL/J61JTdaSlpQHQvPnL/PnnccLDvyQuLpbY2ItcunSR+vUb\n5LgtT89/hzxydHSkdOlnsLOzM7Y5ODiQmpo+QOnZs6c5ffok69atMfYrigGdTse1a1eNbRUrVjJ+\n/2iopbS01Cy3n5qayvbt2/DxaYGtbfrH9SuvtGL+/Lls2LCWmjVr55i/oEjhEEIUOQ4ODsbv7ezS\nP8bCw5eYtKf3pX/IL126iBUrluHn50/Tps3p128gq1at5Pr1azlu59GH9SM5TWdgZ2dPQEBP2rR5\nLVOfu3sZEhJumGTKKLuL4nv37iYx8R5bt/7I1q0/GtsNBgPbt29jxIjRODs75/geCoIUDiFEEZD9\nA4NVqlQD4PbtWzRq1MTYPmfO51SuXIWuXbuzYsUyBg8eSvfuPY39cXGxxg/x/Bgbr0qVqsTFXcLT\ns7yxbc+eXezYsZVJk6bmsOS/Hs+xZcsm3N3LMGfOApP2P/44ypw5M9i2bQudOnXNc3ZLyTUOIZ5S\nWlsNWi2F+2Wbuw9oZ2dnYmMvcfNmQqa/zsuXr4CvbxtmzgzmwIF9XLlymfDwL9m4cS1VqlQFoGzZ\nchw4sJ/Y2IucP3+OuXM/58yZU+h0OuP6AWJiznDv3t1cZezXbxDbt28jKmopcXGx7N+/l1mzpuHg\n4JjpyCX79+lCQkIC165d5caNeA4e3E+nTl2pUqWqyZe/f2fKlfPghx+sc5FcjjiEeErl5glua3nr\nrV6EhMzi0KEDODo6ZeqfMGESX3+9gOnTP+H+/ftUrlyV4OBZNGyYPs9PUNAnzJ37OQMG9KJkSVf8\n/PwZPz6IkJCZpKSk4OJSjK5dAwgLW8DRo78zYsToTNtIPxrIvvC9+GJTgoI+ISrqGxYvXoirqxuv\nv+7HkCHvPraOrNabzs/vDX79dRe9e3dj0KChaLW2vPFG50zL2NjY0K1bAF98MY/Tp09Ro0bNHPdf\nftMoVnriZfLkyRgMBpP7oKOiooiKiiI+Ph4PDw/69+9Pt27dzFpf0ZyFS2H18WsWzQDooNXQrd6z\nqGWsn6I8AxpIfmuT/NaTlxkAC/1UlaIohIaGsmrVKpNK++233zJnzhzeffddNm7cSP/+/fnkk0/Y\nsGFDYUcUQgiRg0I9VRUXF8dHH31ETEwMHh4eJn3fffcdvXv35o033gCgQoUKHDt2jLVr19KxY8fC\njCmEECIHhXrEcfToUTw9Pdm0aROenp4mfUFBQQQEBJi0aTQaEhOL5mGgEEL8VxXqEYe/vz/+/pmH\nCwZo3Lixyc9Xr15l8+bN9O3btzCiCSGEMJMqb8e9ffs2gYGBlClThsGDB1s7jhBCiAxUdztuXFwc\nb7/9NjqdjuXLl1OsWDGzlsvLHQLWYjAYcHaxR6dXcHGxN2sZe62G0qWL5fgEa2Erivs+I8lvXZK/\n6FFV4Thx4gSDBw/Gzc2NZcuWUbZsWbOXLZq3xCk8SNJZdDtumlbDrVv3kdtx84fkty7Jbz1F6nbc\n7Jw7d46BAwdSoUIFvv32W4uKhhBCiMJj1SOOjM8ejh8/HgcHBz7//HN0Oh0JCQkAaLVaSpUqZa2I\nQvxHWXemO7UcMYvcsWrhePQA4IULF/jzzz/RaDS89prpyJKVKlVi69at1ognxH/akSOHjMOOFxZb\nW1saNmz85Bc+5sKF81y/fpWmTX0KIFW6ffv24uHhSeXKVQpsG09y795dfv11N35+6XefRkaGm0wE\nZWNjg6urKz4+LXnnnfdwcTHvGnB+s1rhWL58ufH7KlWqcOrUKWtFEeKplJaWVuiFI7c+/HAM7dq1\nL7DCkZBwg/HjR7FgQThgvcIRFraAy5fjjIUD4NlnPQgPXwKk/86uXLnM7NnTmTHjM6ZOnWGVnKq5\nxiGEEDkpyGH1Hq3bSkP3ZcqRkY2NDW5upXBzK4W7exm8vRvSv//b7NnzCykpyVZIqbK7qoQQ4nHD\nhw/hypXLLFkSwZYtmwBo2dKX//u/PSQmJhIS8gWVKlUhPPxLtm/fSnLyQ7y8ajBs2Hs8/3wdIP3W\n92XLFvPTTz8SH38NBwdHGjVqzAcffISrqytdu3YA4L33hvL66x0YMGAw3bt35JNPpvPNN5HExcVS\nrVp1Jk+eyrZtW1i3bg0Gg4GOHf0JDHzfmHXPnl1ERoYTFxfLs88+S4cOnXjrrV5oNBquXbtK9+4d\n+eyzz/nmm0guXrxAuXLPMmzYCF5++RUiI8PZvHkjAC1aNGHVquzH6XN0dCyo3W0WOeIQQqjatGmz\nKVfOgx49ehMRsQxFUdiw4XsmTJjErFnzqF79OT777GOOHz/G1KkziIyM4oUXGjNiRCBxcbEAREev\nYM2a7xg9ehzR0euYMiWY48f/YNmySAAWL06fmzw4eBYjR35g3HZExFeMHj2OiIhvuHfvLoGBA4iP\nv05YWCRDhrxDVFQUBw7sA2D//r1MnTqJgICeREWtYtiw91i9eqXJNQqAr76az9ChI1i+fBVeXs8R\nHDyFlJRkevbsS5s2r1GnTj02bPiJMmWyvrP05s0EoqNX0Lp1OxwcrFNA5IhDCKFqJUqUQKu1wcnJ\nGVdXVwB8fFoa5wu/fDmOX37ZzrJl3xknbhowYDB//HGU6Ogoxo79iEqVKhMU9AlNmrwEpE/s1KTJ\nS5w7dw6AkiVdjdtydnbh7t30yZx69epr3E6LFq+ydu0qxo79CDs7OypUqMjSpRFcuHCel15qxrJl\nS+jcuRvt26cP1Orh4cmDB0nMnBnMgAH/joDRo0cfY44+fQayc+d2Lly4QM2atbC3t8fW1hY3t3/v\nJL169Qpt2rQAwGDQo9PpKFnSlbFjPyqAvW0eKRxCiCJFo9Hg4fHvIKlnzpwGYMiQ/iavS03VGS/+\nN2/+Mn/+eZzw8C+Ji4slNvYily5dNBaF7Hh6VjB+7+joSOnSz5jMGe7g4EBqavrDu2fPnub06ZOs\nW7fG2K8oBnQ6HdeuXTW2VaxYyfi9i4sLAGlpqdlmKFOmLKGhYf+sT+HWrVusXr2SwMABRER8Q4UK\nFXN8DwVBCodVGbDXgtYGHLTmLWGvTV8OzFxAiP8gBwcH4/d2dukfY+HhS0za0/vSP+SXLl3EihXL\n8PPzp2nT5vTrN5BVq1Zy/fq1HLfz+JSvOQ31Y2dnT0BAT9q0eS1Tn7t7GRISbphkyiini/K2trYm\n85iXL1+B2rWfx8+vNT/8sI533nk/22ULihQOK3t4aj/JD5NRDOa93snFEZ7vVLChhFCd7B8YrFKl\nGgC3b9+iUaMmxvY5cz6ncuUqdO3anRUrljF48FC6d+9p7I+LizV+iGc1paulqlSpSlzcJZMP+T17\ndrFjx1YmTZpq1josyaHXp2Gtm8AsKhxHjhxh37593Lhxg8DAQM6fP0/t2rUpXbp0QeX7z0tJTiEl\nOQWD3rzKYaOVJ27F08fZ2ZnY2EvcvJmQ6a/z8uUr4Ovbhpkzgxk9ejwVKlRk06YNbNy4lpCQL4H0\naxoHDuznpZeakZamZ/36NZw5cwovrxrG9QPExJyhatVqucrYr98gxo0bSZUq1WjZ0pfLl2OZNWsa\nTZs2z3Tkkv37dCEhIYFr164aL47r9QZu375lfN/37t0jKmopaWlptGnTLldZ88qsd6PT6fjggw/Y\ntm0bdnZ2pKWl0b17d5YuXcrp06f59ttvqVix8M+zCSFyz9wPMzVs8623ehESMotDhw7g6OiUqX/C\nhEl8/fUCpk//hPv371O5clWCg2fRsGEjAIKCPmHu3M8ZMKAXJUu64ufnz/jxQYSEzCQlJQUXl2J0\n7RpAWNgCjh79nREjRmfaRvrRQPZ/uL34YlOCgj4hKuobFi9eiKurG6+/7seQIe8+to6s1pvOz+8N\nfv11F717d+PLLyPQaDRcv36Vjh3/Pf3l7OyMl1cNZsyYw3PP1TRr/+U3jWLGEy8zZ87ku+++Y+bM\nmfj4+FC/fn2+//573NzcGDRoEDVq1GDevHmFkTdbRXOESj3frPiO5IfJZh9xOBdzpl+vANRyjaMo\njw4KT3N+dYxV9fTuf+sr8NFxf/jhB0aPHk2rVq3Qav/9wPLw8GD48OH89ttvuQ4ghLAGjZW/RFFm\nVuG4e/culSpVyrLP1dWV+/fv52soIYQQ6mVW4ahevTqbN2/Osm/37t1Ur17d4g1PnjyZoKAgk7a9\ne/fSsWP0S6ViAAAboElEQVRH6tevj7+/P3v27LF4vUIIIQqWWYVj2LBhrFu3zvhfgN9//53g4GCi\noqJ4++23zd6goiiEhoayatUqk4tCMTExDBs2jPbt27N+/XpatWrFu+++S0xMjIVvSQghREEyq3C0\nadOGWbNm8ddffzFp0iQApk2bxubNm5kyZQrt27c3a2NxcXH07duX6OhoPDw8TPqWLVtGgwYNCAwM\npEqVKrz//vs0aNCAZcuWWfiWhBBCFCSz741744036NChA+fPn+fOnTsUL16catWqmVwsf5KjR4/i\n6enJvHnzGDlypEnf4cOHMxWgJk2a8OOPP5q9fiGEEAXPopuqdTodt2/f5saNG1SrVo2EhATKlStn\n9vL+/v74+/tn2RcfH59pnvEyZcpw7VrOQwIIIYQoXGYXjqioKEJDQ0lMTESj0bB69Wq++OILHj58\nSFhYmPHJy9xKTk7ONM6Mvb09Op0uT+sVQgiRv8y6xrFmzRqCg4Pp0qULS5cuRVEUNBoN3bp143//\n+x/z58/PcxAHB4dMRUKn0+HklPkpUSGEENZj1hFHZGQk/fv3Z/z48SZzFLdq1YpRo0axZMkSJkyY\nkKcgzz77LDdu3DBpu3HjhtmnwvLyFKS1pKamYmOTfmeZjda8ObVsbDS4urpkOcKmtRTFfZ+R5Lcu\nyV/0mFU4Ll++zMsvv5xln5eXV6YP/Nx44YUXOHTokEnbwYMHadSokVnLF83H/vUYDOlDP5g75IjB\noHDnThIy5Ej+kPzWJfmtp8CHHClbtix//PFHln0nT5606AJ5RhmHyerduzeHDh1iwYIFnDt3jtDQ\nUP73v//Rt2/fXK1bCCFEwTCrcLz55pt89dVXLF26lMuXLwPw8OFDtm/fTlhYGJ065W5+iIwPAD73\n3HN8+eWXbN26lc6dO7Nr1y7CwsKoWrVqrtYthBCiYJg1Oq7BYGDKlCmsXr0601j4fn5+fP7551YZ\nojmjonm4KKPjWpvkty7Jbz15OVVl1qe9jY0Nn376KQMGDODAgQPGBwAbN25MjRo1cr1xIYQQRY9Z\nhaNTp06MGjWKli1bUqVKlYLOJIQQQsXMusYRGxuLo6NjQWcRQghRBJhVOPz8/Fi6dCk3b94s6DxC\nCCFUzqxTVVevXuXgwYP4+PjwzDPP4OLiYnx6/NF/t27dWtBZhRBCqIBZhcPd3Z0OHTpk25/VBOxC\nCCH+m8wqHDNmzCjoHEXcE+9ozuflhBDCesw+VZUdGxsbnJ2dKVGiRL6FKorStn2PkmrZSL6a0u4F\nlEYIIQqOWYXD19fXeD3jcY/aXV1d6d27N8OHD8/3kEWBkqoDC4eAV9JSCyiNEEIUHLNPVU2aNImm\nTZvy+uuv88wzz3Dr1i127NjBzp07CQwMJDk5mYiICGMBEUII8d9kVuH44Ycf8Pf3Jzg42KS9U6dO\nTJ06lRMnThAeHo6rqyvR0dFSOIQQ4j/MrOc4spoP/BFfX1/2798PgLe3N7GxsfmXTgghhOqYVThK\nlSrFkSNHsuw7evQorq6uANy7d49ixYrlXzohhBCqY9apqrfeeov58+eTnJxMu3btcHNz4/bt22zf\nvp0lS5YwdOhQbty4QUREhNkTL2Xn7t27fP755+zZs4eUlBS8vb2ZMGEC1apVy9N6hRBC5A+zCkdg\nYCApKSlERkYSGRlpbC9WrBjDhg3jnXfeYcOGDdy7d49Zs2blKVBQUBDnz59nwYIFlChRgpCQEN5+\n+222bt2Kvb19ntYthBAi78yeROO9997j7bff5tixY9y+fZuyZctSq1Yt46mpDh065HpCp4wOHDjA\nyJEjadCgAQAjR46kQ4cOnDt3jlq1auV5/UIIIfLGrGscj2i1WmxtbdFoNHh5eXH//n1jX35N5OTt\n7c3mzZu5ffs2Op2ONWvWULJkSSpUqJAv6xdCCJE3Zn/aR0VFERoaSmJiIhqNhtWrV/PFF1/w8OFD\nwsLCcHZ2zpdAs2fPpl+/fjRr1gytVoujoyNLliyRi+5CCKESZh1xrFmzhuDgYLp06cLSpUuNI+J2\n69aN//3vf8yfPz/fAo0dO5bk5GQWLlzIypUr8fHxYcSIEcTHx+fbNoQQQuSeWYUjMjKS/v378+GH\nH5rcNdWqVStGjRrFtm3b8iXMsWPH2LNnDzNnzqRFixbUq1ePOXPm4ODgwNKlS/NlG0IIIfLGrFNV\nly9f5uWXX86yz8vLixs3buRLmEeDKdapU8fYZmtrS61atZ74YGFeJl7PK4PBwD1nezR2Fi7naIeN\nTfqQ9DZa8y432dhocHV1wc7Owo0VIGvu+/wg+a1L8hc9ZhWOsmXL8scff9CsWbNMfSdPnqRcuXL5\nEqZy5coAnDp1itq1awOgKAoxMTG0bNkyx2UTEhLzJUPuKKQ+sHyQQ5xTMRjSB4406A1mLWIwKNy5\nkwRoLcxYMNzdi1t53+eN5LcuyW89eSl4ZhWON998ky+//BInJydeeeUVAB4+fMj27dsJCwujb9++\nuQ6QUe3atWnevDkTJkzg448/xtXVlW+++Ybr16/Tp0+ffNmGEEKIvDGrcAwZMoSrV68yY8YM46RO\njwYy9PPzY+jQofkWKDQ0lJCQED744APu379P3bp1WbFiBc8++2y+bUMIIUTuaZSsJtnIxsWLFzlw\n4AB37tyhePHiNGrUiBo1ahRkPrNZ/VTV5pWWn6pyL8O3l+6Q/DDZ7FNVzsWc6dcrADlVlT8kv3VJ\nfusp0FNVhw8fZtWqVRw5coSbN2+iKArlypXjhRdeoG7durnesBBCiKIpx8IRHBzM8uXLcXR0pG7d\nusa7neLj49m0aRNr165lyJAhjB49ulDCCiGEsL5sC8f333/P8uXLGTx4MEOHDsXFxcWkPzExkYUL\nF7Jw4UJq1arF66+/XuBhhRBCWF+2Dw+sXr2aTp06MWbMmExFA6B48eKMGTOGTp06ER0dXaAhhRBC\nqEe2hSMmJoa2bds+cQWtW7fm5MmT+RpKCCGEemVbOB48eGCc2S8nbm5uJCYWzbsKhBBCWC7bwmEw\nGMwaKl2r1WLBHb1CCCGKOIvm48iKRqPJjxxCCCGKiBwPKT777LMnzoMhp6mEEOLpkm3haNy4MQCp\nqak5rsDR0dH4WiGEEP992RaO5cuXF2YOIYQQRUSer3EIIYR4ukjhEEIIYRFVFo7Vq1fTrl076tev\nT5cuXThw4IC1IwkhhPiH6grHunXr+PTTTwkMDGTTpk00adKEYcOGceXKFWtHE0IIgcoKh6IoLFiw\ngCFDhtClSxcqVKjA+PHjqVSpEr///ru14wkhhMDMGQALy/nz57l69Srt27c3tmk0GtavX2/FVEII\nITJS1RHHxYsXAbh79y59+/alWbNm9O7dm6NHj1o3mBBCCCNVFY779+8DMGHCBAICAoiMjMTLy4t+\n/fpx7tw5K6cTQggBKjtVZWdnB8CwYcPw8/MD4OOPP+bw4cOsXLmSoKCgbJfNy/y5eWUwGLjnbI/G\nzsLlHO2wsUkf68tGa14Nt7HR4OrqYtxXamDNfZ8fJL91Sf6iR1WFo0yZMgA899xzJu1Vq1Z94l1V\n1p0wXiH1gQ50OssWc07FYEgfWdigN5i1iMGgcOdOEqC1MGPBcHcvbuV9nzeS37okv/XkpeCp6lTV\n888/j5OTE8ePHze2KYrCuXPnqFixohWTCSGEeERVRxxOTk7079+fefPm8cwzz+Dl5cW3337L5cuX\n6dGjh7XjCSGEQGWFA+D999/HycmJadOmcevWLWrXrk1kZCSVK1e2djQhhBCosHAADBkyhCFDhlg7\nhhBCiCyo6hqHEEII9ZPCIYQQwiJSOIQQQlhECocQQgiLSOEQQghhESkcQgghLKLK23GtSa83kJpq\n3vAfGWls7bGxdMiRXEgf2Ur55yv3axBCiNySwvGYlGQ9F2MeWLiUQkV7Fxwe3C+QTBnZaG04fzaJ\n1BTLltPaaqhe8+kbjE0Ikf+kcBRBhjQFvd7SI4fcHqEIIYQpucYhhBDCIlI4hBBCWEQKhxBCCItI\n4RBCCGER1RaOY8eOUbt2bQ4dOmTtKEIIITJQZeF48OAB48aNQ1HkTiAhhFAbVRaOGTNmUK5cOSkc\nQgihQqorHLt372bPnj0EBQVZO4oQQogsqOoBwNu3bzNx4kRmzJhBiRIlrB1HCCFEFlRVOD7++GNa\ntWqFj48P169ft3acQqHV2qDV2qAxcwwpW60NNloFrdayJ8e1Wkh/elzGqhJC5I1qCse6des4efIk\nGzduNGk39zqHu3v+jMN0xzYZF5c0i5ZRFAXHJDucXewtWs5QzAk70tDpbcw+Z+hsX4K4q3+hS7Es\no72DLY2btcbWNv9/5fm1761F8luX5C96VFU4rl+/TvPmzU3aBw8eTOfOnZkyZUqOyyckJOZLjgdJ\nqSQlWTrKrUJySiqKpcslp5GWpqDXGzD3PoC0NIXkh8kkP9RbtCm9Xs/ffyeR35e13N2L59u+twbJ\nb12S33ryUvBUUzhmzZqFLsOw5Ddu3KBXr14EBwfTrFkzKyYTQgiRkWoKR9myZU1+trOzM7aXKlXK\nGpGEEEJkQXW342ak0ciFXCGEUBvVHHE8rly5cpw8edLaMYQQQjxG1UccQggh1EcKhxBCCItI4RBC\nCGERKRxCCCEsIoVDCCGERaRwCCGEsIhqb8ctajR2dmBv2VhV2NoW8pCDyj9flpLnaYQQ/5LCkQ+0\nWohNLgs6J4uWc0xxQ2Nzr4BSmdJo4PzZ+6SmmL+M1lZD9ZpP3wBuQoicSeHIJ2mpejQ6y0asNaRZ\nNlBhXhnSFPR6S44eZAZGIURmco1DCCGERaRwCCGEsIgUDiGEEBaRwiGEEMIiqiscN2/eZPz48fj4\n+NC4cWMGDRrE2bNnrR1LCCHEP1RVOAwGA8OHD+fSpUuEhYURHR1N8eLF6d+/P3fu3LF2PCGEEKis\ncJw6dYpjx44xbdo06tatS7Vq1Zg5cyYPHjxg165d1o4nhBAClRUODw8PwsPDqVKlirHt0SyAiYlF\nc0J4IYT4r1HVA4Curq60bNnSpG358uUkJyfTvHlzK6USQgiRkaoKx+N27NjB3LlzGTBgAFWrVrV2\nHFGg8vqUuoynpX6Zf8cGgyHL9qzJ71gtVFs41q5dy+TJk/Hz82PcuHFPfL27e/6MqXTHNhkXF8uG\nDtHYKBhsbLCx01q0nFZrYzwVpzHz34TGBrRaLXYWj6eoxdHJHjs78//x2dhA6dLFsLHJ+Yxmfux7\ng8HA3r17SUuzbN/b2tri4+PzxIw5ya//d6ylqOS35u+4IBWV/Z+fVFk4wsLCCA0NpXfv3gQFBZm1\nTEJC/lwDeZCUSlKSzqJltFqFNIMBTaplY0/Z6g0oSvpfW4qZf3QpBtDr9aTqLNuWVqsh+aEOnc78\nwqHVwq1b98npLz139+L5tO8VEhMf5upD5UkZc5J/+a2jaOXP/Dt2drbnwYOc/73l9XdckIrW/jeV\nl4KnusIRERFBaGgoI0eOZOjQodaOI4QQ4jGqKhynTp0iJCSEN998kzfffJOEhARjX7FixXBysmzY\nciGEEPlPVYVjy5YtGAwG1qxZw5o1a0z65AhECCHUQVWFY9SoUYwaNcraMYQQQuRAnbcpCCGEUC0p\nHEIIISwihUMIIYRFpHAIIYSwiBQOIYQQFlHVXVXiyfL27GxuxoNS8rzVp5uMwVW0WPb7Mh1r6+n5\nXUnhKGps4F6ynrsWDotSUgOX7iTzd6L5Q5U4OdhQgxKWJhSPOXLkUK6GUmnYsHEBJRI52fBXPLo0\ng1mvdXaxJy0ljY61yxZwKnWRwlEEKQoYzB3cKsMyeoOC3oLl9Ia8/rUsANLS0iwuHMJ6dGkGUvTm\n/b9vq1dIM7PI/JfINQ4hhBAWkcIhhBDCIlI4hBBCWEQKhxBCCIuornDo9XrmzJmDj48PDRo04L33\n3uPWrVvWjiWEEOIfqiscCxYsYP369cyaNYsVK1YQHx/PiBEjrB1LCCHEP1RVOHQ6HcuXL2fMmDE0\nbdqU2rVrM3fuXI4cOcLRo0etHU8IIQQqKxynTp0iKSmJJk2aGNs8PT3x9PTk8OHDVkwmhBDiEVUV\njuvXrwNQtqzpU5hlypQhPj7eGpGEEEI8RlWF4+HDh9jY2KDVak3a7e3tSUlJsVIqIYQQGalqyBFH\nR0cMBgMGgwEbm39rmk6nw8nJqVAy2NhAqWcs2y0aG4X795zR25g/DhSAbTFHnJ0dUQDFzFELnJ0d\nSDVoSLNwOBBnJzuKOdtaNFSJg70WMPzzlTWdTgfkz3AatrY2WPq3TPoylu33jPIzf3YK8n0VRv78\n9Pi+sLGx+aftScvk/ndsKTdnLalmDiPiYG+DwUb75Bf+x2gUxcJBjwrQ8ePH6d69O7t37zY5XeXr\n60uvXr0YNGiQFdMJIYQAlZ2qqlmzJi4uLhw8eNDYdvnyZa5evUrjxjJSqBBCqIGqTlXZ29vTs2dP\nZs6ciZubG6VKleKTTz6hSZMm1KtXz9rxhBBCoLJTVZD+5Pjs2bNZt24daWlptGjRgsmTJ+Pq6mrt\naEIIIVBh4RBCCKFuqrrGIYQQQv2kcAghhLBIkS8cOp0Of39/Nm7cmKlv6dKlvPrqq3h7ezNw4EAu\nXbpkhYRZK6qjAE+ePJmgoCCTtr1799KxY0fq16+Pv78/e/bssVK6rN28eZPx48fj4+ND48aNGTRo\nEGfPnjX2qz3/9evXee+993jxxRdp3Lgxo0eP5saNG8Z+tefP6NixY9SuXZtDhw4Z29SePyYmhpo1\na2b6OnLkCKD+/ACrV6+mXbt21K9fny5dunDgwAFjX67yK0VYYmKi8vbbbys1atRQNm7caNK3atUq\npWHDhsrWrVuV06dPK0OHDlVat26tpKSkWCmtqZCQEMXHx0fZt2+fcuLECaV79+5Kjx49rB0rWwaD\nQZk3b55So0YNJSgoyNh+9uxZpU6dOsrXX3+tnD9/Xpk3b55Sp04d5ezZs1ZM+y+9Xq8EBAQoAQEB\nyvHjx5WYmBjl/fffV5o1a6b8/fffqs9vMBiUN954QxkwYIBy6tQp5eTJk0rv3r2Vzp07K4qi/v2f\nUVJSktKmTRulZs2aym+//aYoStHIv3nzZuWll15Sbt68afKVmppaJPKvXbtWqVOnjvL9998rsbGx\nyvTp0xVvb2/l8uXLuc5fZAvH//3f/ymtWrVSOnfunGXhaNu2rbJgwQLjz0lJSUqDBg2UH374obCj\nZpKSkqI0bNhQWbdunbHt8uXLSo0aNZQjR45YMVnWYmNjld69eysvvfSS8uqrr5oUjkmTJil9+vQx\neX2fPn2USZMmFXbMLJ04cUKpUaOGcu7cOWNbSkqK4u3traxbt071+RMSEpTRo0crV65cMbb9/PPP\nSo0aNZS7d++qPn9Gj7LWqFHDWDiKQv6QkBCld+/eWfapPb/BYFBeffVVZf78+SZtnTp1UtavX5/r\n/EX2VNUvv/xC586diY6OztR369YtLl26xIsvvmhsc3Z2pk6dOqoYZbeojQJ89OhRPD092bRpE56e\nniZ9hw8fNnkfAE2aNFHN+/Dw8CA8PJwqVaoY2zQaDQD37t3j999/V3X+Z555hjlz5uDh4QGkn7b6\n7rvvqFevHiVKlFD9/n9k9+7d7NmzJ9NpzqKQ/+zZs1SrVi3LPrXnP3/+PFevXqV9+/bGNo1Gw7p1\n6+jYsWOu86vqAUBLTJw4Mdu+nEbZfdRnTUVtFGB/f3/8/f2z7IuPj8/yfVy7dq0woj2Rq6srLVu2\nNGlbvnw5KSkpNG/enNDQUFXnz+idd95h586dlCxZkmXLlgHq3/8At2/fZuLEicyYMYMSJUqY9BWF\n/GfPnkWn0xEQEMCVK1fw8vJi1KhR1KtXT/X5L168CMDdu3fp27cvMTExVK1alTFjxtCgQYNc51fl\nEcfly5ezvBhVs2ZN6tev/8TlHz58CICDg4NJu52dnSpG2f0vjQKcnJycaT/b29v/M/ie+uzYsYO5\nc+cyYMAAqlWrVqTyjxw5klWrVtGwYUMGDBhAfHx8kcj/8ccf06pVK3x8fDL1qT1/cnIyly9f5sGD\nB4wbN46vvvqKMmXK0KdPH86dO6f6/Pfv3wdgwoQJBAQEEBkZiZeXF/369ctTflUecZQrV44tW7Zk\n2Zdx1NzsODo6AmR68zqdDmdn57wHzCM1jAKcXxwcHLLcz2p8H2vXrmXy5Mn4+fkxduxYoGjlf+65\n5wAICQmhZcuWrF+/XvX5161bx8mTJzPd9aj889yx2vM7Ojry+++/Y2dnh61t+sdl3bp1OXHiBCtX\nrlR9fjs7OwCGDRuGn58fkF7IDx8+nKf8qiwctra2JuekLfXss88CcOPGDSpUqGBsv3HjBl5eXnnO\nl1eP8iUkJJgcJsbHx9O6dWtrxcqVZ5991uTWUEjfz+XKlbNSoqyFhYURGhpK7969Tc6zqz3/rVu3\nOHDggPEfPaR/mFWsWJH4+HjV51+3bh3Xr1+nefPmJu2DBw+mU6dOqs8PZPoQ1Wg0VK9enWvXrqk+\nf5kyZYB//+h4pFq1aly+fDnX+VV5qiqvSpcuTaVKlfjtt9+MbUlJSZw4cYJGjRpZMVm6/9IowC+8\n8ILJPfkABw8eVMV+fiQiIoLQ0FBGjhyZ6eKs2vNfuXKFMWPG8OeffxrbEhMTuXDhAtWrV1d9/lmz\nZrFlyxY2btzIxo0bWbRoEQDBwcG8//77qs//559/0qBBA06cOGFs0+v1nDx5Ei8vL9Xnf/7553Fy\ncuL48ePGNkVRiImJoVKlSrnPXwB3gBW6rG7HXblypeLt7a1s3rxZOX36tBIYGKi0a9dOSU1NtVJK\nU7Nnz1aaN2+u7NmzR/nzzz+Vbt26ZbotTo169+6tTJw40fjz6dOnleeff16ZP3++EhMTo8ybN0+p\nX7++ye2v1nTy5EmlVq1aysSJE5WEhATlxo0bxq8HDx6oPr/BYFB69eqldOzYUfnjjz+UEydOKAMH\nDlTatm1bJPI/7tq1aya346o9f1pamtKpUyelS5cuyh9//KGcOXNGGTt2rNKkSRPl1q1bqs+vKIoy\nb948pUmTJsq2bduUCxcuKMHBwUr9+vWVCxcu5Dr/f7ZwKIqihIeHKz4+Poq3t7fy9ttvK3FxcVZI\nl7W0tDRlxowZyosvvqi88MILyqhRo5S///7b2rGeqHfv3ibPcSiKouzatUvx8/NT6tatq3Tq1EnZ\nt2+fldJlNnfuXKVGjRpZfoWFhSmKou78iqIot2/fViZMmKA0bdpUadiwofL+++8r8fHxxn6158/o\n2rVrJg8AKor688fHxysffPCB0rRpU8Xb21sZNGiQyQNyas+vKOmfha+88opSt25dJSAgQDl8+LCx\nLzf5ZXRcIYQQFvlPXuMQQghRcKRwCCGEsIgUDiGEEBaRwiGEEMIiUjiEEEJYRAqHEEIIi0jhEEII\nYREpHEIIISwihUMIIYRF/h/pD/3beEGDOwAAAABJRU5ErkJggg==\n",
      "text/plain": "<matplotlib.figure.Figure at 0x1f0425c0>"
     },
     "metadata": {}
    }
   ]
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true,
    "format": "column"
   },
   "cell_type": "code",
   "source": "sns.set_style('whitegrid')\nsns.boxplot(data=[data[l] for l in ['controlA', 'treatmentA']])\nsns.despine()\nplt.show()",
   "execution_count": 66,
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAd4AAAFVCAYAAABB6Y7YAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": "<matplotlib.figure.Figure at 0x1c1fc550>"
     },
     "metadata": {}
    }
   ]
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "axes = df.hist(cumulative=True, normed=1, figsize=(10, 10))\naxes[0, 1].set_xscale('log')\nplt.show()",
   "execution_count": 31,
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmMAAAJkCAYAAABdzSbFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n6P8BycW3vCC7g5YAAAAASUVORK5CYII=\n",
      "text/plain": "<matplotlib.figure.Figure at 0x1c511dd8>"
     },
     "metadata": {}
    }
   ]
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "# sns.set_style('darkgrid')\nsns.set_context('notebook')\nsns.kdeplot(df['controlB'], shade=True, cumulative=True)",
   "execution_count": 64,
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": "<matplotlib.axes._subplots.AxesSubplot at 0x1aa19d68>"
     },
     "metadata": {},
     "execution_count": 64
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeYAAAFVCAYAAAA6zUwUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nW7B66gAAAABJRU5ErkJggg==\n",
      "text/plain": "<matplotlib.figure.Figure at 0x1925fcc0>"
     },
     "metadata": {}
    }
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "name": "python2",
   "display_name": "Python 2",
   "language": "python"
  },
  "language_info": {
   "mimetype": "text/x-python",
   "nbconvert_exporter": "python",
   "name": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.9",
   "file_extension": ".py",
   "codemirror_mode": {
    "version": 2,
    "name": "ipython"
   }
  },
  "gist_id": "3810f2e0cd32ef798aef"
 },
 "nbformat": 4,
 "nbformat_minor": 0
}