bioinfo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 背景简介\n",
    "eMaize问题要求我们以SNP作为特征，通过训练一个模型，对玉米的三个性状进行预测 <br>\n",
    "接下来的教程会展示从原始数据开始，如何对数据进行转换，存取，特征选择以及回归和后续分析的整个过程。<br>\n",
    "本问题最基本的目标是使用6210个样本中的前4754个样本作为训练集，预测其他样本的性状<br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "本教程使用jupyter notebook，可以让使用者获得更好的体验，方便对代码进行修改，以及对结果进行查看和分析\n",
    "#### jupyter notebook基本使用指南：\n",
    "一段相关的代码在同一个代码框中书写，同时按住shift与enter即可运行选中的代码框的代码<br>\n",
    "仅仅按enter键具有回车的效果\n",
    "##### 使用上方的编辑栏：\n",
    "点击加号在两个代码框中间插入新的代码框，删除代码框点击剪刀，中止程序点击方框"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#导入必需的库\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "import numpy as np\n",
    "from sklearn.random_projection import SparseRandomProjection\n",
    "from scipy.sparse import load_npz, save_npz\n",
    "import scipy.stats\n",
    "import pandas as pd\n",
    "import sklearn\n",
    "import h5py\n",
    "from sklearn.metrics import r2_score\n",
    "from scipy.stats.stats import pearsonr\n",
    "import pickle\n",
    "from sklearn.linear_model import LinearRegression\n",
    "import xgboost\n",
    "from xgboost.sklearn import XGBRegressor\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.kernel_ridge import KernelRidge\n",
    "from sklearn import neighbors\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.gaussian_process import GaussianProcessRegressor\n",
    "from sklearn.gaussian_process.kernels import DotProduct\n",
    "from tqdm import tqdm_notebook as tqdm\n",
    "from IPython.display import display, Image\n",
    "%pylab inline\n",
    "plt.style.use('ggplot')\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Share/home/chenxupeng/projects/emaize/bioinfo_test\n"
     ]
    }
   ],
   "source": [
    "cd .."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Errno 2] No such file or directory: 'emaize/bioinfo_test/'\n",
      "/Share/home/chenxupeng/projects/emaize/bioinfo_test/jupyter_notebook\n"
     ]
    }
   ],
   "source": [
    "cd emaize/bioinfo_test/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1查看原始数据\n",
    "genotype：SNP数据，每个位点可能有三种情况，如AA，AT，TT  <br>\n",
    "trait：共三种，trait1开花期，trait2株高，trait3产量，为连续值 <br>\n",
    "原始数据中有6210个样本，每个样本SNP位点约为190万个 <br>\n",
    "因为计算资源的原因，这里仅仅选取其中的5000个SNP作为示例 <br>\n",
    "因为数据量的原因，结果肯定不够理想"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 数据格式\n",
    "txt存储格式不适合大数据读取的问题，对内存的占用过多 <br>\n",
    "对于结构化的、能够存储为矩阵的数据，可以使用HDF5格式存取，内存占用小，读取速度快"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### 读取SNP数据，数据格式为HDF5\n",
    "##### 在命令行查看数据shape的方法为：\n",
    "cd至文件路径下，输入：h5ls snp_5000  <br>\n",
    "若使用了新版h5py，可能出现无法打开的情况，此时输入HDF5_USE_FILE_LOCKING=FALSE h5ls snp_5000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5001, 6214)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#使用h5py读取5000个SNP：\n",
    "with h5py.File('data/snp_5000','r') as f:\n",
    "    snps = f['snp'][:]\n",
    "#查看数据shape,h5py读取出的snps是一个矩阵，可以用.shape查看其shape\n",
    "snps.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "#查看数据内容\n",
    "#np.array(snps)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### 读取性状数据\n",
    "使用numpy/pandas均可读取性状数据并显示，这里用pandas展示，真正计算时一般用numpy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       type     id pedigree    trait1    trait2    trait3\n",
      "0  training  L0001  f1_X_m1 -1.746103 -0.785525 -0.331637\n",
      "1  training  L0002  f2_X_m1 -1.679248 -1.569490 -2.572614\n",
      "2  training  L0003  f3_X_m1 -2.748920 -0.608644 -1.108812\n",
      "3  training  L0004  f4_X_m1 -2.414647 -0.672045 -1.315051\n",
      "4  training  L0005  f5_X_m1 -1.879812 -0.740913 -1.879181\n",
      "          type     id    pedigree    trait1    trait2    trait3\n",
      "0     training  L0001     f1_X_m1 -1.746103 -0.785525 -0.331637\n",
      "1     training  L0002     f2_X_m1 -1.679248 -1.569490 -2.572614\n",
      "2     training  L0003     f3_X_m1 -2.748920 -0.608644 -1.108812\n",
      "3     training  L0004     f4_X_m1 -2.414647 -0.672045 -1.315051\n",
      "4     training  L0005     f5_X_m1 -1.879812 -0.740913 -1.879181\n",
      "5     training  L0006     f6_X_m1 -2.347793 -0.401772 -1.442920\n",
      "6     training  L0007     f7_X_m1 -2.013521 -0.526802 -0.688812\n",
      "7     training  L0008     f8_X_m1 -2.214084 -0.691969 -1.589630\n",
      "8     training  L0009     f9_X_m1 -1.879812 -0.759537 -1.872018\n",
      "9     training  L0010    f10_X_m1 -1.946666 -0.627000 -1.918008\n",
      "10    training  L0011    f11_X_m1 -2.682065 -0.264470 -1.626809\n",
      "11    training  L0012    f12_X_m1 -2.347793 -0.854826 -1.567208\n",
      "12    training  L0013    f13_X_m1 -1.812957 -0.973503 -0.791896\n",
      "13    training  L0014    f14_X_m1 -2.147230 -0.776863 -1.861917\n",
      "14    training  L0015    f15_X_m1 -1.679248 -0.979567 -1.267843\n",
      "15    training  L0016    f16_X_m1 -2.615211 -1.953243 -1.973740\n",
      "16    training  L0017    f17_X_m1 -2.748920 -0.393110 -0.579711\n",
      "17    training  L0018    f18_X_m1 -1.946666 -1.038473 -0.536085\n",
      "18    training  L0019    f19_X_m1 -2.347793 -0.166150 -1.017190\n",
      "19    training  L0020    f20_X_m1 -2.347793 -0.989294 -2.165867\n",
      "20    training  L0021    f21_X_m1 -2.615211 -0.570504 -2.406563\n",
      "21    training  L0022    f22_X_m1 -1.612394 -0.785092 -1.630749\n",
      "22    training  L0023    f23_X_m1 -2.548356 -0.342000 -1.633328\n",
      "23    training  L0024    f24_X_m1 -3.216901 -1.185737 -1.781112\n",
      "24    training  L0025    f25_X_m1 -2.214084 -1.106041 -1.751670\n",
      "25    training  L0026    f26_X_m1 -1.545539  0.059944 -1.143770\n",
      "26    training  L0027    f27_X_m1 -2.949483 -0.529545 -2.126826\n",
      "27    training  L0028    f28_X_m1 -2.682065  0.241859 -1.417848\n",
      "28    training  L0029    f29_X_m1 -2.481502 -0.520450 -1.334177\n",
      "29    training  L0030    f30_X_m1 -2.080375 -1.150617 -2.152901\n",
      "...        ...    ...         ...       ...       ...       ...\n",
      "6180      test  L6181  f179_X_m30       NaN       NaN       NaN\n",
      "6181      test  L6182  f180_X_m30       NaN       NaN       NaN\n",
      "6182      test  L6183  f181_X_m30       NaN       NaN       NaN\n",
      "6183      test  L6184  f182_X_m30       NaN       NaN       NaN\n",
      "6184      test  L6185  f183_X_m30       NaN       NaN       NaN\n",
      "6185      test  L6186  f184_X_m30       NaN       NaN       NaN\n",
      "6186      test  L6187  f185_X_m30       NaN       NaN       NaN\n",
      "6187      test  L6188  f186_X_m30       NaN       NaN       NaN\n",
      "6188      test  L6189  f187_X_m30       NaN       NaN       NaN\n",
      "6189      test  L6190  f188_X_m30       NaN       NaN       NaN\n",
      "6190      test  L6191  f189_X_m30       NaN       NaN       NaN\n",
      "6191      test  L6192  f190_X_m30       NaN       NaN       NaN\n",
      "6192      test  L6193  f202_X_m30       NaN       NaN       NaN\n",
      "6193      test  L6194  f191_X_m30       NaN       NaN       NaN\n",
      "6194      test  L6195  f199_X_m30       NaN       NaN       NaN\n",
      "6195      test  L6196  f203_X_m23       NaN       NaN       NaN\n",
      "6196      test  L6197  f204_X_m23       NaN       NaN       NaN\n",
      "6197      test  L6198  f205_X_m23       NaN       NaN       NaN\n",
      "6198      test  L6199  f206_X_m23       NaN       NaN       NaN\n",
      "6199      test  L6200  f207_X_m23       NaN       NaN       NaN\n",
      "6200      test  L6201  f203_X_m24       NaN       NaN       NaN\n",
      "6201      test  L6202  f204_X_m24       NaN       NaN       NaN\n",
      "6202      test  L6203  f205_X_m24       NaN       NaN       NaN\n",
      "6203      test  L6204  f206_X_m24       NaN       NaN       NaN\n",
      "6204      test  L6205  f207_X_m24       NaN       NaN       NaN\n",
      "6205      test  L6206  f203_X_m25       NaN       NaN       NaN\n",
      "6206      test  L6207  f204_X_m25       NaN       NaN       NaN\n",
      "6207      test  L6208  f205_X_m25       NaN       NaN       NaN\n",
      "6208      test  L6209  f206_X_m25       NaN       NaN       NaN\n",
      "6209      test  L6210  f207_X_m25       NaN       NaN       NaN\n",
      "\n",
      "[6210 rows x 6 columns]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([['training', 'L0001', 'f1_X_m1', -1.74610282478836,\n",
       "        -0.7855251215738208, -0.33163696544539506],\n",
       "       ['training', 'L0002', 'f2_X_m1', -1.67924837319023,\n",
       "        -1.5694898694514998, -2.57261413504835],\n",
       "       ['training', 'L0003', 'f3_X_m1', -2.7489195987604504,\n",
       "        -0.608643883224438, -1.10881183080429],\n",
       "       ..., \n",
       "       ['test', 'L6208', 'f205_X_m25', nan, nan, nan],\n",
       "       ['test', 'L6209', 'f206_X_m25', nan, nan, nan],\n",
       "       ['test', 'L6210', 'f207_X_m25', nan, nan, nan]], dtype=object)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traits = pd.read_csv('data/pheno_emaize.txt','\\t')\n",
    "#仅显示前五个,4754之后的样本的性状是未知的\n",
    "print(traits.head(n=5))\n",
    "#numpy.array\n",
    "#traits['trait1'][:4000]\n",
    "print(traits)\n",
    "numpy.array(traits)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(6210, 6)\n"
     ]
    }
   ],
   "source": [
    "#pandas dataframe也可查看shape\n",
    "print traits.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,u'Normalized Trait3: Yield')"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2adb63fea090>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#查看性状的分布情况\n",
    "trait1 = np.array(traits['trait1'])[:4754]\n",
    "trait2 = np.array(traits['trait2'])[:4754]\n",
    "trait3 = np.array(traits['trait3'])[:4754]\n",
    "fig, ax = plt.subplots(1,3, figsize=(15,4))\n",
    "ax[0].hist(trait1,bins = 50,color='b',alpha=0.6)\n",
    "ax[0].set_title('Normalized Trait1: Flowering Time',fontsize=14)\n",
    "ax[1].hist(trait2,bins = 50,color='g',alpha=0.6)\n",
    "ax[1].set_title('Normalized Trait2: Height',fontsize=14)               \n",
    "ax[2].hist(trait3,bins = 50,color='r',alpha=0.6)\n",
    "ax[2].set_title('Normalized Trait3: Yield',fontsize=14)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### 查看训练集与测试集的划分\n",
    "下图中彩色部分为训练集性状，白色部分为待预测性状 <br>\n",
    "可以发现其划分方式并不随机，这会导致常规的机器学习方法出现一些问题，由于是基础介绍，这里不讨论如何解决这个问题。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_parent_table(phenotype_file):\n",
    "    phenotypes = pd.read_table(phenotype_file)\n",
    "    pedigree = phenotypes['pedigree'].str.split('_', expand=True)\n",
    "    pedigree.columns = ['f', 'X', 'm']\n",
    "    phenotypes = pd.concat([phenotypes, pedigree], axis=1)\n",
    "    phenotypes['number'] = np.arange(phenotypes.shape[0])\n",
    "    parent_table = phenotypes.pivot_table(values='number'\n",
    "                                          , index=['m'], columns=['f'], dropna=False)\n",
    "    male_ids = ['m%d' % i for i in range(1, parent_table.shape[0] + 1)]\n",
    "    female_ids = ['f%d' % i for i in range(1, parent_table.shape[1] + 1)]\n",
    "    parent_table = parent_table.loc[male_ids, female_ids]\n",
    "    return parent_table\n",
    "phenotype_file = 'data/pheno_emaize.txt'\n",
    "parent_table = generate_parent_table(phenotype_file)\n",
    "phenotypes = pd.read_table('data/pheno_emaize.txt')\n",
    "fig, ax = subplots(3,1, figsize=(20, 10))\n",
    "for i in range(3):\n",
    "    trait = ['trait1','trait2','trait3'][i]\n",
    "    ax[i].matshow(np.take(np.ravel(phenotypes[trait].values), parent_table), cmap=cm.RdBu)\n",
    "    ax[i].set_title('Phenotypes of training data (%s)'%trait)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  2将SNP数据编码为向量\n",
    "每个位点的碱基只有三种情况，不会出现更多碱基组合的可能，比如某位点仅有AA，AT，TT三种可能的情况<br>\n",
    "我们可以采取三种方式对其编码：\n",
    "- 转化为0、1、2。找到minor allele frequency（MAF），即两种碱基（如A、T）中出现频率低的那个，以A作为MAF为例，则TT为0，AT为1，AA为2，这样可以突出MAF\n",
    "- 转化为3-bit one hot vector,$[1,0,0]^T,[0,1,0]^T,[0,0,1]^T$这样可以保持三种向量在空间距离的一致\n",
    "- 转化为2-bit vector,则AA，AT，TT分别编为$[1,0]^T,[1,1]^T,[0,1]^T$,不需要考虑MAF\n",
    "我们采取第三种方式"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def convert_2bit(seq):\n",
    "    genotypes = np.zeros([6210,2])\n",
    "    a = seq[1].split('/')\n",
    "    for i in range(6210):\n",
    "        if seq[4:][i] == a[0] + a[0]:\n",
    "            genotypes[i] = np.array([0,1])\n",
    "        if seq[4:][i] == a[0] + a[1]:\n",
    "            genotypes[i] = np.array([1,0])\n",
    "        if seq[4:][i] == a[1] + a[1]:\n",
    "            genotypes[i] = np.array([1,1])\n",
    "    genotypes = genotypes.astype('int').T\n",
    "    return genotypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 注意，接下来的步骤耗时14:11\n",
    "真实计算时此步骤使用C加速计算，这里为了连贯性仅仅展示python的方法 <br>\n",
    "可以跳过接下来的代码框步骤，直接使用处理好的结果，data/2bit_geno"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#该代码框可跳过以节约时间，直接运行下一个代码框\n",
    "geno_conv = convert_2bit(snps[1])\n",
    "for i in tqdm(range(4999)):\n",
    "    geno_conv = np.concatenate((geno_conv,convert_2bit(snps[i+2])),axis =0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "#读取处理成2bit格式的SNP\n",
    "with h5py.File('data/2bit_geno') as f:\n",
    "    geno_conv = f['data'][:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#查看SNP的大致情况\n",
    "fig, ax = plt.subplots(figsize=(5,10))\n",
    "ax.matshow(geno_conv[:300,:200],cmap = cm.binary_r)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3特征提取与降维\n",
    "原数据每个样本有190万个SNP，转化为2bit coding后有大约380万个feature，大多数的feature可能是冗余的 <br>\n",
    "而且过多的feature使得机器学习模型无法承受，一个考虑时间开销及效果的feature数量应该在几千至几万量级 <br>\n",
    "#### 特征选择：\n",
    "特征选择的方法包括filter，wrapper和embedding三大类 <br>\n",
    "我们使用过如下方法： <br>\n",
    "Mutual information:劣势在于需要将连续的性状值离散化，损失信息<br>\n",
    "ANOVA:通过p-value筛选feature，速度较慢，我们设计了加速ANOVA计算的算法。<br>\n",
    "基于模型的方法:基于广义线性模型或其他带有feature权重的机器学习模型，根据权重挑选feature<br>\n",
    "\n",
    "#### 降维：\n",
    "PCA、SVD：劣势在于降维后的feature数量不能超过样本数量，一次性损失的feature过多<br>\n",
    "Random projection:基于LSH的降维方式，速度较快<br>\n",
    "\n",
    "通过对问题的后续分析，我们发现对于预测绝大多数样本，基本的降维方法就已经够用<br>\n",
    "但是对于部分很难预测的样本，简单的特征选择方法也无法取得好的效果<br>\n",
    "我们根据后续开发的针对性的模型，设计了基于模型的特征选择方法，因为内容限制，不在这里使用。\n",
    "\n",
    "#### 接下来分别使用ANOVA和  Random projection演示特征选择和降维，对于后续的计算来说，选择其中一种就可以，也可以把不同的方法拼起来使用"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 我们会提供ANOVA、Random projection处理上面5000个snps后的数据，以及在完整数据集上用Random projection降维至10000个feature的数据。用于送入下一部分的回归模型。下面的三种方法的处理后的数据可以在feature_selection文件夹下找到，存储格式为HDF5，用h5py打开\n",
    "##### ANOVA数据： \n",
    "feature_selection/anova    包含三个性状各自的feature，大小为4000*6210\n",
    "##### Random projection(5000)数据：      \n",
    "feature_selection/randomproj_5000    从5000个SNPs降维得到，三个性状使用同一组feature,大小为1000*6210\n",
    "##### Random projection(whole SNPs)数据： \n",
    "feature_selection/randomproj_whole 从所有SNPs降维得到，三个性状使用同一组feature，大小为10000*6210"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### ANOVA\n",
    "方差分析方法可以利用p值挑选feature <br>\n",
    "调用scipy.stats.f_oneway,利用SNPs和性状可以很容易地计算出p-value，但是对于大量数据来说速度较慢 <br>\n",
    "这里我们使用一种加速ANOVA计算的方法完成计算，相比于scipy.stats的方法可以提升计算速度数百倍。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "#加速ANOVA算法\n",
    "def fast_anova_2bit(X, y):\n",
    "    y = y - y.mean()\n",
    "    y2 = y*y\n",
    "    N = X.shape[0]\n",
    "    SS_tot = np.sum(y2)\n",
    "    # 10, 01, 11\n",
    "    masks = [np.logical_and(X[:, 0::2], np.logical_not(X[:, 1::2])),\n",
    "             np.logical_and(np.logical_not(X[:, 0::2]), X[:, 1::2]),\n",
    "             np.logical_and(X[:, 0::2], X[:, 1::2])]\n",
    "    Ni = np.concatenate([np.sum(mask, axis=0) for mask in masks]).reshape((3, -1))\n",
    "    at_least_one = Ni > 0\n",
    "    SS_bn = [np.sum(y.reshape((-1, 1))*mask, axis=0) for mask in masks]\n",
    "    SS_bn = np.concatenate(SS_bn).reshape((3, -1))\n",
    "    SS_bn **= 2\n",
    "    SS_bn = np.where(at_least_one, SS_bn/Ni, 0)\n",
    "    SS_bn = np.sum(SS_bn, axis=0)\n",
    "    SS_wn = SS_tot - SS_bn\n",
    "    M = np.sum(at_least_one, axis=0)\n",
    "    DF_bn = M - 1\n",
    "    DF_wn = N - M\n",
    "    SS_bn /= DF_bn\n",
    "    SS_wn /= DF_wn\n",
    "    F = SS_bn/SS_wn\n",
    "    \n",
    "    p_vals = np.ones(F.shape[0])\n",
    "    ind = np.nonzero(M == 2)[0]\n",
    "    if ind.shape[0] > 0:\n",
    "        p_vals[ind] = scipy.stats.f.sf(F[ind], 1, N - 2)\n",
    "    ind = np.nonzero(M == 3)[0]\n",
    "    if ind.shape[0] > 0:\n",
    "        p_vals[ind] = scipy.stats.f.sf(F[ind], 2, N - 3)\n",
    "    return F, p_vals"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### 注意X和y分别是什么\n",
    "我们需要输入进 fast_anova_2bit(X, y)的X是处理过的SNPs中前4754个样本的，y是trait1、trait2、trait3\n",
    "因为需要分别预测三个性状，我们需要针对三个性状分别挑选特征"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10000, 4754)\n",
      "(10000, 1456)\n"
     ]
    }
   ],
   "source": [
    "geno_conv_train = geno_conv[:,:4754]\n",
    "geno_conv_test = geno_conv[:,4754:]\n",
    "print geno_conv_train.shape\n",
    "print geno_conv_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5000,)"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#分别计算三种性状下5000个SNPs做完ANOVA的p-value\n",
    "F,pval_1 = fast_anova_2bit(geno_conv_train.T,trait1)\n",
    "F,pval_2 = fast_anova_2bit(geno_conv_train.T,trait2)\n",
    "F,pval_3 = fast_anova_2bit(geno_conv_train.T,trait3)\n",
    "pval_1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,u'5000 SNPs p-value for trait3 distribution')"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3cAAADSCAYAAADpCjnWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvAOZPmwAAIABJREFUeJzt3Xu8ZWV95/nPN6AYRQWkUOQSUMskYI+INUjG0cbWIKIR7Ve0IV7QECsY6I49To+onYFG6RDHy0i3wca2BjAKkqixOpaSktEQHVEKJVwllFhKSQUKEBBRIvCbP9ZzrF2nzmXXOfvsc/Y+n/frtV9nr2c9a63f2pff2c9az3pWqgpJkiRJ0mj7lcUOQJIkSZI0fzbuJEmSJGkM2LiTJEmSpDFg406SJEmSxoCNO0mSJEkaAzbuJEmSJGkM2LjTjJKcn+S9C7Tu9ya5M8k/LcT65yLJF5OcuIDr35TkJe35u5L89wGu+/4kT2vPB/q+Jflokj8Z1PqkQTA/DXz95idpAMxNA1+/uWknjEzjLslXk/y8vQn3J7lp0vzfS/KDJD9N8tdJ9uqZt1eSz7V5P0jye/0uO0UcxyW5Osl97ct1WZKD2rwzklSS1/TU37WVTdQ5P8k/t324O8n6JL8xiNdolCQ5AHg7cEhVPWVA66wkz5jPOqrqZVV1QVvfm5J8bdI2XpTkK0nuTbJpntv6z1X1B7PVa5/9WetV1e5Vdct8Ymrb22G/q+rkqnrPfNc9rsxP42WE89N/SHJdkp8k+X6S/zCPbZmfxoC5abyMcG56W5Jb2vt/W5IPJdl1jtsyN81iZBp3zantTdi9qn59ojDJocB/A94APBl4APjznuU+Avxzm/c64Ny2TD/L/lL78F9I98V6InBwq/tIT7W7gTOT7DLDfryvqnYH9gfuAM7vZ+fHzK8Bd1XVHTu74FwTwlyXm+SnwBpgzj+aBm1A+6X5Mz+Nj1HNTwHeCOwJHAOcmuT4Aax37gGZn5YCc9P4GNXc9D+Aw6vqCcCzgGcD/24A652zsc5NVTUSD+CrwB9MM+8/A5/qmX46XUJ6PPC49vyZPfM/AZw927JTbOd3gatniPEM4JPAPwAntrJdgQIOatPnA+/tWeblwP3t+RHABuA+4Hbgg9Ns5yhgM/Au4E5gE/C6GeK6EXhFz/SubbnD2/RfAv8E3AtcDhzaU/eX8QJvAr42ad0FPKM93w14P/DDFv9HgV+dIp6XAD+jS+z3A+e38lcC1wP3tPf7N3uW2QS8A7gGeBDYddI6L2+x/LSt89/0vE7vaPv3CbofPX8DbAV+3J7vP/lzBvwm8HPg4ba+e6bYh019fG7fAPwAuAt4d9uPl/R8Xv6iPX8M8Bet3j3AlXT/MM9qMfy8xfFfe173U4Cbge9P8V6c317/9cBPgL8Dfq3NO6jV3bXf/WbHz+1bgI10/5DXAk+d9Jk4ucX2Y7ofCFnsHLKQD8xP5qcllJ966p8D/JcZXnvz05jnJ8xN5qYllpuAJwFfBv58htfe3DSP3DRqZ+7+tJ3O/3qSo3rKD6VLCgBU1fdoSak9Hq6qf+yp/w9tmdmWnezbwG+008kvSrL7FHUK+BPg9CSPmmln2vKvA77Tij4MfLi6IxtPBy6ZYfGnAHsD+wEnAucl+fVp6l4EnNAz/VLgzqr6dpv+IrAS2Kft4ydninsGf0b3uh0GPKPF9n9OrlRVXwZeBtxW3ZHENyV5ZovzbcAKYB3wP5I8umfRE+gS+h5V9dCkdb6wPX12W+en2/RTgL3ojnatpjtb/f+06QPpEuV/nSLGG+m+ZN9o69tjZ1+MJIcA59IlqafSJbT9p6l+It0RzQNavZOBn1XVu4G/Z9uR11N7lnkV8DzgkGnW+TrgPXSfk6vp433tZ7+T/CvgT4HXAvvSJeCLJ1V7BfA/0x2dey3dZ27cmZ+2MT9tv86h56ckAV5A96NvB+anZZWfzE3bmJu2X+fQclPrxnsfXQP52XRnfndgbpp/bhqlxt07gKfRfejPo/vwPr3N253uyEmve+mOPs00b7Zlt1Ndn9yjWgyXAHe2fuC7T6q3lu7oxnR9ff/3JPfQtd53pzuqA/AL4BlJ9q6q+6vqimmWn/AnVfVgVf0d8AW6D8JUPgW8Mslj2/TvtbKJeNdU1U+q6kG6IyLPTvLEWba9nfZD4i3Av6+qu6vqJ3RH9vrtEvRvgC9U1fqq+gXdUaxfBf6XnjrnVNWtVfWznQjtEeD09jr9rKruqqrPVNUDLcazgH+5E+vbGb8L/E1VXd5e2z9h+24ovX5Bl5ieUVUPV9VVVXXfLOv/0/ZaT/d6fKFn2+8Gfqv115+v1wFrqurbbd3vbOs+qKfO2VV1T1X9EPgK3T+tcWZ+2pH5aXYLmZ/OYNsPsqmYn5ZHfjI37cjcNLuB56aq+lRrgD+T7uzY7dNUNTfNMzeNTOOuqr458SWq7qLNrwPHttn3A0+YtMgT6E6pzjRvtmWniuOKqnptVa2gOyr6Qro3f7L/2MofM8W891fVHlX1lKp6ZXVHvABOovvQfzfJlUleMVUMzY+r6qc90z8AnprkwGy7cPr+FvNGuu4Fv9OS1CtpCSrJLknOTvK9dkRlU1vf3jNseyorgMcCVyW5pyXgL7Xyfjy17QMt5keAW+n+GUy4dSdjAthaVT+fmEjy2CT/Ld1F4PfRdUnYY5Z+/nP1VHpibu/XXdPU/QRwKXBxuouN3zfb0Utmfz16t30/XTeAp84a9ewmv1f30+1X73vVO4rXA3T/iMeW+WkH5qf+LEh+SnIq3bV3L28/IqZifuqMdX4yN+3A3NSfBfvtVFU30/UomPIaTcxNE+acm0amcTeFort4HLoPybMnZqQb0nQ34B/bY9ckK3uWfTbbuqrMtOzMAVRdCXyW7uLQyfPW0x1d+qO+d6jq5qo6ge4U/58Bf5XkcdNU33PSvAPpTtX/sLZdON37oZjoXnAccENLWtAdiTqOri/3E+n6FMO217bXT+mSUFch6R2p6U660/SHtuS7R1U9cVIMM7mN7nT/xLpDd5r9Rz11qs919Zq8zNuBXwee144gTXRJmGp/57K9Xlvo9qHbQPfP4UlTBln1i6r6T1V1CN0Rt1fQ/TibKY7Z4uvd9u50XSxuo3sfoee9pOuC0e96J79Xj6Pbrx9Nu8TyY34yP/Vj4Pkpye8DpwEvrqrNM2zb/LQ8mZvMTf1Y6N9Ou9J1oZ2KuWmeRqJxl2SPJC9N8ph0w+O+ju6DdWmr8km6IysvaC/WmcBn29Gqn9IlkTOTPC7J8+m+kJ+Ybdkp4vhfk7wlyT5t+jfojuRM1wXg3cD/sRP7+fokK9qRl3ta8cMzLPKfkjw6yQvoPtB/OUPdi4GjgbfS062ArgvFg3RHDx5L1x1gOv8AHJrksCSPoeuGAPzyaNHHgA/1vD77Jem3v/AlwMuTvLgddXl7i+v/63N56E7xP22WOo+nS6T3pBu2+fRZ1rd/evquJ/mVtu+P6ibzmGzft73XXwGvaJ+bR9N9tqb8zqW7DuFftKNg99F1NZh47/vZr6kc27Pt9wDfrK5rxla6ZPL6dvTx99k+ye6w35N8Cnhz+xzsRveZ+WZVbZpDjCPP/DQt89P2hpGfXkf3Gv12zT60t/lpzJmbpmVu2t4wctMf9OzfIXRdEi+bZnlz0zyNROOO7of0e+n6Yt8J/FvgVVV1E0BVXU93IeMn6YbHfTzbH/X5I7o+yHfQHYV5a1umn2V73UOXkK5Nd9r+S8DngPdNVbmqvg58ayf28xjg+rbuDwPH954Wn+Sf6EbTua3FfnJVfXe6FVfVFuAbdEc2Pt0z60K608Q/Am5g+mRLdRdWn0k3ytHNwNcmVXkH3RG3K9Kdtv8y3ZGeWbX38vXAf6F7j38H+J2q+ud+lm/OAC5I17Vhuj70/zfdZ+FOun390gzr+3/pjk7+U5I7W9kL6RLcOrZdVPy30+zT9XSjMn2K7kjUj+lGoJrKU+gS2n103UD+jm4EKOg+C7+b5MdJzpkh3sk+RZeA7waeS9ffe8Jb6G7ncBfdhfG9/wim2u/e/bqMrg/8Z9p+PZ3+rw8YR+anHZmfdnQGC5+f3kt3JPjKbOti9tFp9sn8NP7MTTsyN+3oDBY+Nz2f7v3/Kd3vp3V0o5ZOtU/mpnlK1Xx7nmnY0o129RdVNd3oQZK0KMxPkpYic5OWi1E5cydJkiRJmoGNO0mSJEkaA7M27pKsSXJHkut6yj6d5Or22JTk6lZ+UJKf9cz7aM8yz01ybZKNSc5JMtUIO+pDVX3VbgVa7pIckOQrSW5Mcn2SP27lZyT5UU8eOrZnmXe2HHRT7wXrSY5pZRuTnLYY+zMuzE+S+WkpMjdpuZj1mrskL6S7n8mFVbXDsLVJPgDcW1VnprsR399MU+9bwB/TXYi5ju6mil+c9x5IWpaS7AvsW1XfTvJ44CrgVXQ3pL2/qt4/qf4hdIMCHEF3v5kv090bCbrhu3+b7qLtK4ETquqGoeyIpLFjfpK0WHadrUJVXZ7t757+S+3s22uBfzXTOlqSe0JVfaNNX0iX5GzcSZqTNorZlvb8J0luZPubgU52HHBxdTd1/n6SjXQ/pAA2Tgwdn+TiVtcfT5LmxPwkabHM2ribxQuA26u72/yEg5N8h25Y0v9YVX9Pl9B6hzHdzMxJ7pf23nvvOuigg+YZpqSl5KqrrrqzqlYMan3tANRzgG/SDbl8apI3AhuAt1fVj+lyTu9w1b156NZJ5c+bbZvmJmn8DDo3wfDzk7lJGk/95qf5Nu5OoOtGMGELcGBV3ZXkucBfJzmUnbyDfZLVwGqAAw88kA0bNswzTElLSZIfDHBdu9PdN+ZtVXVfknPpbjxa7e8HgN9n+jw01bXHU+Ync5M03gaZm9r6hpKfzE3S+Os3P815tMwkuwL/mp6bOlbVg1V1V3t+FfA9uj7jm4Hei1j3p7uB5JSq6ryqWlVVq1asGOgBNEljJMmj6H44fbKqPgtQVbdX1cNV9QjwMbZ1bdoMHNCz+EQemq58B+YmSf0aZn4yN0maMJ9bIbwE+G5V/bK7ZZIVSXZpz58GrARuaX3Pf5LkyHad3huBz89j25KWuZZLPg7cWFUf7Cnft6faq4GJkX7XAscn2S3JwXT56Vt0AxSsTHJwkkcDx7e6kjQn5idJi2XWbplJLgKOAvZOshk4vao+TpdgLppU/YXAmUkeAh4GTq6qu9u8twLnA79KN5CKg6lImo/nA28Arp24HQvwLuCEJIfRdV3aBPwhQFVdn+QSuoEIHgJOqaqHAZKcClwK7AKsqarrh7kjksaO+UnSopj1VgiLbdWqVWXfcWm8JLmqqlYtdhzzYW6Sxo+5SdJS1W9+mk+3TEmSJEnSEjHf0TKXlINO+8J205vOfvkiRSJJ20zOTWB+krQ0+NtJGi+euZMkSZKkMWDjTpIkSZLGgI07SZIkSRoDNu4kSZIkaQzYuJMkSZKkMWDjTpIkSZLGgI07SZIkSRoDNu4kSZIkaQzYuJMkSZKkMWDjTpIkSZLGgI07SZIkSRoDNu4kSZIkaQzM2rhLsibJHUmu6yk7I8mPklzdHsf2zHtnko1Jbkry0p7yY1rZxiSnDX5XJEmSJGn56ufM3fnAMVOUf6iqDmuPdQBJDgGOBw5ty/x5kl2S7AJ8BHgZcAhwQqsrSZIkSRqAXWerUFWXJzmoz/UdB1xcVQ8C30+yETiizdtYVbcAJLm41b1hpyOWJEmSJO1gPtfcnZrkmtZtc89Wth9wa0+dza1sunJJkiRJ0gDMtXF3LvB04DBgC/CBVp4p6tYM5VNKsjrJhiQbtm7dOscQJUmSJGn5mFPjrqpur6qHq+oR4GNs63q5GTigp+r+wG0zlE+3/vOqalVVrVqxYsVcQpQkSZKkZWVOjbsk+/ZMvhqYGElzLXB8kt2SHAysBL4FXAmsTHJwkkfTDbqydu5hS5IkSZJ6zTqgSpKLgKOAvZNsBk4HjkpyGF3Xyk3AHwJU1fVJLqEbKOUh4JSqerit51TgUmAXYE1VXT/wvZEkSZKkZaqf0TJPmKL44zPUPws4a4rydcC6nYpOkiRJktSX+YyWKUmSJElaImzcSRpZSQ5I8pUkNya5Pskft/K9kqxPcnP7u2crT5Jzkmxst3I5vGddJ7b6Nyc5cbH2SdLoMzdJWiw27iSNsoeAt1fVbwJHAqckOQQ4DbisqlYCl7VpgJfRDfS0ElhNd1sXkuxFdz3x8+hG/z295/6dkrSzzE2SFoWNO0kjq6q2VNW32/OfADcC+wHHARe0ahcAr2rPjwMurM4VwB5t9N+XAuur6u6q+jGwHjhmiLsiaYyYmyQtFht3ksZCkoOA5wDfBJ5cVVug+5EF7NOq7Qfc2rPY5lY2XfnkbaxOsiHJhq1btw56FySNIXOTpGGycSdp5CXZHfgM8Laqum+mqlOU1Qzl2xdUnVdVq6pq1YoVK+YWrKRlw9wkadhs3EkaaUkeRffj6ZNV9dlWfHvr0kT7e0cr3wwc0LP4/sBtM5RL0pyYmyQtBht3kkZWktDdd/PGqvpgz6y1wMSocicCn+8pf2Mbme5I4N7WNepS4Ogke7bBCo5uZZK008xNkhbLrDcxl6Ql7PnAG4Brk1zdyt4FnA1ckuQk4IfAa9q8dcCxwEbgAeDNAFV1d5L3AFe2emdW1d3D2QVJY8jcJGlR2LiTNLKq6mtMfU0KwIunqF/AKdOsaw2wZnDRSVquzE2SFovdMiVJkiRpDNi4kyRJkqQxYONOkiRJksaAjTtJkiRJGgOzNu6SrElyR5Lresr+ryTfTXJNks8l2aOVH5TkZ0mubo+P9izz3CTXJtmY5Jw2TLAkSZIkaQD6OXN3PnDMpLL1wLOq6n8C/hF4Z8+871XVYe1xck/5ucBqYGV7TF6nJEmSJGmOZm3cVdXlwN2Tyv62qh5qk1cA+8+0jiT7Ak+oqm+04X4vBF41t5AlSZIkSZMN4pq73we+2DN9cJLvJPm7JC9oZfsBm3vqbG5lkiRJkqQBmNdNzJO8G3gI+GQr2gIcWFV3JXku8NdJDmXqG3nWDOtdTdeFkwMPPHA+IUqSJEnSsjDnM3dJTgReAbyudbWkqh6sqrva86uA7wHPpDtT19t1c3/gtunWXVXnVdWqqlq1YsWKuYYoSZIkScvGnBp3SY4B3gG8sqoe6ClfkWSX9vxpdAOn3FJVW4CfJDmyjZL5RuDz845ekiRJkgT00S0zyUXAUcDeSTYDp9ONjrkbsL7d0eCKNjLmC4EzkzwEPAycXFUTg7G8lW7kzV+lu0av9zo9SZIkSdI8zNq4q6oTpij++DR1PwN8Zpp5G4Bn7VR0kiRJkqS+DGK0TEmSJEnSIrNxJ0mSJEljwMadJEmSJI0BG3eSJEmSNAZs3EmSJEnSGLBxJ0mSJEljwMadJEmSJI0BG3eSJEmSNAZs3EmSJEnSGLBxJ0mSJEljwMadpJGVZE2SO5Jc11N2RpIfJbm6PY7tmffOJBuT3JTkpT3lx7SyjUlOG/Z+SBov5iZJi8XGnaRRdj5wzBTlH6qqw9pjHUCSQ4DjgUPbMn+eZJckuwAfAV4GHAKc0OpK0lydj7lJ0iLYdbEDkKS5qqrLkxzUZ/XjgIur6kHg+0k2Ake0eRur6haAJBe3ujcMOFxJy4S5SdJi8cydpHF0apJrWteoPVvZfsCtPXU2t7LpyiVp0MxNkhaUjTtJ4+Zc4OnAYcAW4AOtPFPUrRnKd5BkdZINSTZs3bp1ELFKWj7MTZIWXF+Nu2kuDN4ryfokN7e/e7byJDmnXfx7TZLDe5Y5sdW/OcmJg98dSctdVd1eVQ9X1SPAx9jWvWkzcEBP1f2B22Yon2rd51XVqqpatWLFisEHL2lsmZskDUO/Z+7OZ8cLg08DLquqlcBlbRq6C39XtsdquiNVJNkLOB14Hl1CO72nS4IkDUSSfXsmXw1MHJRaCxyfZLckB9PlqG8BVwIrkxyc5NF0AxusHWbMksafuUnSMPQ1oMo0FwYfBxzVnl8AfBV4Ryu/sKoKuCLJHi2hHQWsr6q7AZKsp2swXjSvPZC0bCW5iC637J1kM90BpKOSHEbXfWkT8IcAVXV9kkvoBiN4CDilqh5u6zkVuBTYBVhTVdcPeVckjRFzk6TFMp/RMp9cVVsAqmpLkn1a+bwvDE6ymu6sHwceeOA8QpQ0zqrqhCmKPz5D/bOAs6YoXwesG2BokpYxc5OkxbIQA6rM+8Jg+45LkiRJ0s6ZT+Pu9on+4+3vHa183hcGS5IkSZJ2znwad2uBiREvTwQ+31P+xjZq5pHAva375qXA0Un2bAOpHN3KJEmSJEnz1Nc1d9NcGHw2cEmSk4AfAq9p1dcBxwIbgQeANwNU1d1J3kM3+hPAmRODq0iSJEmS5qff0TKnujAY4MVT1C3glGnWswZY03d0kiRJkqS+LMSAKpIkSZKkIbNxJ0mSJEljwMadJEmSJI0BG3eSJEmSNAZs3EmSJEnSGLBxJ0mSJEljwMadJEmSJI0BG3eSJEmSNAZs3EmSJEnSGLBxJ0mSJEljwMadJEmSJI0BG3eSJEmSNAZs3EmSJEnSGLBxJ0mSJEljYM6NuyS/nuTqnsd9Sd6W5IwkP+opP7ZnmXcm2ZjkpiQvHcwuSJIkSZJ2neuCVXUTcBhAkl2AHwGfA94MfKiq3t9bP8khwPHAocBTgS8neWZVPTzXGCRJkiRJnUF1y3wx8L2q+sEMdY4DLq6qB6vq+8BG4IgBbV+SJEmSlrVBNe6OBy7qmT41yTVJ1iTZs5XtB9zaU2dzK9tBktVJNiTZsHXr1gGFKEmSJEnja96NuySPBl4J/GUrOhd4Ol2XzS3AByaqTrF4TbXOqjqvqlZV1aoVK1bMN0RJkiRJGnuDOHP3MuDbVXU7QFXdXlUPV9UjwMfY1vVyM3BAz3L7A7cNYPuSlqnWO+COJNf1lO2VZH2Sm9vfPVt5kpzTBnW6JsnhPcuc2OrfnOTExdgXSePD3CRpsQyicXcCPV0yk+zbM+/VwERiWwscn2S3JAcDK4FvDWD7kpav84FjJpWdBlxWVSuBy9o0dAeiVrbHarpeBiTZCzgdeB7dwajTe7qTS9JcnI+5SdIimFfjLsljgd8GPttT/L4k1ya5BngR8O8Bqup64BLgBuBLwCmOlClpPqrqcuDuScXHARe05xcAr+opv7A6VwB7tINRLwXWV9XdVfVjYD07/iiTpL6ZmyQtljnfCgGgqh4AnjSp7A0z1D8LOGs+25SkWTy5qrYAVNWWJPu08ukGddqpwZ7ojqxz4IEHDjhsSWPO3CRpwQ1qtExJWuqmG9TJwZ4kLSZzk6SBsXEnadzcPnHtb/t7RyufblAnB3uSNAzmJkkLzsadpHGzFpgYVe5E4PM95W9sI9MdCdzbukhdChydZM82WMHRrUySBsncJGnBzeuaO0laTEkuAo4C9k6ymW5kubOBS5KcBPwQeE2rvg44FtgIPAC8GaCq7k7yHuDKVu/Mqpo8EIIk9c3cJGmx2LiTNLKq6oRpZr14iroFnDLNetYAawYYmqRlzNwkabHYLVOSJEmSxoCNO0mSJEkaAzbuJEmSJGkM2LiTJEmSpDFg406SJEmSxoCNO0mSJEkaAzbuJEmSJGkM2LiTJEmSpDFg406SJEmSxsC8G3dJNiW5NsnVSTa0sr2SrE9yc/u7ZytPknOSbExyTZLD57t9SZIkSdLgzty9qKoOq6pVbfo04LKqWglc1qYBXgasbI/VwLkD2r4kSZIkLWsL1S3zOOCC9vwC4FU95RdW5wpgjyT7LlAMkiRJkrRsDKJxV8DfJrkqyepW9uSq2gLQ/u7TyvcDbu1ZdnMr206S1Uk2JNmwdevWAYQoSZIkSeNt1wGs4/lVdVuSfYD1Sb47Q91MUVY7FFSdB5wHsGrVqh3mS5IkSZK2N+8zd1V1W/t7B/A54Ajg9onulu3vHa36ZuCAnsX3B26bbwySJEmStNzNq3GX5HFJHj/xHDgauA5YC5zYqp0IfL49Xwu8sY2aeSRw70T3TUmSJEnS3M23W+aTgc8lmVjXp6rqS0muBC5JchLwQ+A1rf464FhgI/AA8OZ5bl+SJEmSxDwbd1V1C/DsKcrvAl48RXkBp8xnm5IkSZKkHS3UrRAkSZIkSUNk406SJEmSxoCNO0mSJEkaAzbuJEmSJGkM2LiTJEmSpDFg407SWEqyKcm1Sa5OsqGV7ZVkfZKb2989W3mSnJNkY5Jrkhy+uNFLGlfmJkkLycadpHH2oqo6rKpWtenTgMuqaiVwWZsGeBmwsj1WA+cOPVJJy4m5SdKCsHEnaTk5DrigPb8AeFVP+YXVuQLYI8m+ixGgpGXJ3CRpIGzcSRpXBfxtkquSrG5lT66qLQDt7z6tfD/g1p5lN7ey7SRZnWRDkg1bt25dwNAljTFzk6QFs+tiByBJC+T5VXVbkn2A9Um+O0PdTFFWOxRUnQecB7Bq1aod5ktSH8xNkhaMZ+4kjaWquq39vQP4HHAEcPtEl6b2945WfTNwQM/i+wO3DS9aScuFuUnSQrJxJ2nsJHlcksdPPAeOBq4D1gIntmonAp9vz9cCb2wj0x0J3DvRRUqSBsXcJGmh2S1T0jh6MvC5JNDluU9V1ZeSXAlckuQk4IfAa1r9dcCxwEbgAeDNww9Z0jJgbpK0oGzcSRo7VXUL8Owpyu8CXjxFeQGnDCE0ScuYuUnSQptz4y7JAcCFwFOAR4DzqurDSc4A3gJMDNf0rqpa15Z5J3AS8DDw76rq0nnEPquDTvvCDmWbzn75Qm5SkiRJkhbFfM7cPQS8vaq+3fqPX5VkfZv3oap6f2/lJIcAxwOHAk8FvpzkmVX18DxikCRJkiQxjwFVqmpLVX27Pf8JcCNT3Hulx3HAxVX1YFV9n67/+BFz3b4kSZIkaZuBjJaZ5CDgOcA3W9GpSa5JsibJnq2srxtxSpIkSZJ23rwbd0l2Bz4DvK2q7gPOBZ4OHAZsAT4wUXWKxae80WaS1Uk2JNmwdevWqapIkiRJknrMq3GX5FF0DbtPVtVnAarq9qp6uKoeAT7Gtq6Xfd+Is6rOq6pVVbVqxYoV8wlRkiRJkpaFOTe9rYIkAAAIJUlEQVTu0t2k5ePAjVX1wZ7yfXuqvZru5pzQ3Yjz+CS7JTkYWAl8a67blyRJkiRtM5/RMp8PvAG4NsnVrexdwAlJDqPrcrkJ+EOAqro+ySXADXQjbZ7iSJmSJEmSNBhzbtxV1deY+jq6dTMscxZw1ly3KUmSJEma2kBGy5QkSZIkLa75dMscSQed9oXtpjed/fJFikSSJEmSBmfZNe4kSZI0tckHwcED4dIosXEnSYvAXgSSJGnQvOZOkiRJksbAsj9zZ/cDSZIkSePAM3eSJEmSNAaW/Zm7qUx1Nm8uPAMoSZIkaVhs3EnSEmAXcUlLlQNASaPDxt0CMhlKmg9ziKSlyNwkLV027pY4j+ZLkiRJ6oeNuyEa1LV8czliNqhGoo1NafH4/ZO0FJmbpKXDxt2YWqiG5LgaVBcTu6po2Pr5jvo5lDRs5iZpcdi4GwNzbYAttTOJk021noXa1kIZ5hnTue5nP+uZSx0tHR50kLQUzeW3gP9/pJmlqoa7weQY4MPALsB/r6qzZ6q/atWq2rBhQ1/rXi5nmbQ09NP4XE525p9rkquqatUChrPTzE07by4HGOb6vfEHnoZhKeYm2Ln8tDO5CcYzPy3k/2fzjBZLv/lpqGfukuwCfAT4bWAzcGWStVV1wzDjkAZhHP8hLlfmprkZ5negn20t1I+3QZ05n0tPg6XeOwEWbr/6Wc9cYhk15qedt5DfiWF+32b7PC92r55BWWrf27ns6zDz4GyGeuYuyW8BZ1TVS9v0OwGq6k+nW8aj49JoGOUzd+YmaXyNcm6Cnc9PnrmTRsPONu76zU+/MueI5mY/4Nae6c2tTJIWk7lJ0lJlfpLUt2EPqJIpynY4dZhkNbC6Td6f5KY+1783cOccY1ssoxbzqMULxjwU+bOdivnXFjKWOTA37ciYF96oxQsjGPOI5yboIz/NIzfB6L2noxYvGPOwjFTMO5mboM/8NOzG3WbggJ7p/YHbJleqqvOA83Z25Uk2LLXuFLMZtZhHLV4w5mEZxZh7mJsmMeaFN2rxgjEvklnz01xzE4ze6zNq8YIxD8uoxbxQ8Q67W+aVwMokByd5NHA8sHbIMUjSZOYmSUuV+UlS34Z65q6qHkpyKnAp3XC+a6rq+mHGIEmTmZskLVXmJ0k7Y+g3Ma+qdcC6BVr9nLokLLJRi3nU4gVjHpZRjPmXzE07MOaFN2rxgjEvCvPTdkYtXjDmYRm1mBck3qHfxFySJEmSNHjDvuZOkiRJkrQARrJxl+SYJDcl2ZjktCnm75bk023+N5McNPwot4tntnj/tyQ3JLkmyWVJFn0o5tli7qn3u0kqyaKPTtRPzEle217r65N8atgxThHPbJ+NA5N8Jcl32ufj2MWIsyeeNUnuSHLdNPOT5Jy2P9ckOXzYMS6mUctNLaaRyk/mpuEwN40Xc9NwmJ8W3qjlphbTcPNTVY3Ug+5i4u8BTwMeDfwDcMikOn8EfLQ9Px749BKP90XAY9vzty5mvP3G3Oo9HrgcuAJYtdRjBlYC3wH2bNP7jEDM5wFvbc8PATYtcswvBA4Hrptm/rHAF+nuy3Qk8M3FjHcJvp9LJjftRMxLJj+Zm5ZUzOamEXmYm5ZOzK2e+Wlh411SuanFMdT8NIpn7o4ANlbVLVX1z8DFwHGT6hwHXNCe/xXw4iRT3QR0GGaNt6q+UlUPtMkr6O5hs5j6eY0B3gO8D/j5MIObRj8xvwX4SFX9GKCq7hhyjJP1E3MBT2jPn8gU914bpqq6HLh7hirHARdW5wpgjyT7Die6RTdquQlGLz+Zm4bD3DRezE3DYX5aeCOXm2D4+WkUG3f7Abf2TG9uZVPWqaqHgHuBJw0luh31E2+vk+ha74tp1piTPAc4oKr+ZpiBzaCf1/mZwDOTfD3JFUmOGVp0U+sn5jOA1yfZTDdS2r8dTmhztrOf93Eyarlpu3iapZ6fzE3DYW4aL+am4TA/LbxxzE0w4Pw09FshDMBUR5ImD/nZT51h6TuWJK8HVgH/ckEjmt2MMSf5FeBDwJuGFVAf+nmdd6XrXnAU3RG+v0/yrKq6Z4Fjm04/MZ8AnF9VH0jyW8AnWsyPLHx4c7KUvnvDNmq5CUYvP5mbhsPcNF7MTcNhflp445ibYMDfv1E8c7cZOKBnen92POX6yzpJdqU7LTvT6dCF1E+8JHkJ8G7glVX14JBim85sMT8eeBbw1SSb6PoHr13kC4P7/Vx8vqp+UVXfB26iS1iLpZ+YTwIuAaiqbwCPAfYeSnRz09fnfUyNWm7aLp5mqecnc9NwmJvGi7lpOMxPC28ccxMMOj8N6mLBYT3ojiDcAhzMtospD51U5xS2vzD4kiUe73PoLhBdudivb78xT6r/VRb/ouB+XudjgAva873pToE/aYnH/EXgTe35b7Yvexb5tT6I6S8KfjnbXxT8rcWMdQm+n0smN+1EzEsmP5mbllTM5qYReZiblk7Mk+qbnxYm3iWXm1osQ8tPi7qj83iBjgX+sX2p393KzqQ7cgNdK/0vgY3At4CnLfF4vwzcDlzdHmuX+ms8qe6iJ6g+X+cAHwRuAK4Fjh+BmA8Bvt4S2NXA0Ysc70XAFuAXdEeaTgJOBk7ueY0/0vbn2qXwuVhi7+eSyk19xryk8pO5acnEbG4aoYe5aWnEPKmu+Wlh4l1SuanFNNT8lLZSSZIkSdIIG8Vr7iRJkiRJk9i4kyRJkqQxYONOkiRJksaAjTtJkiRJGgM27iRJkiRpDNi4kyRJkqQxYONOkiRJksaAjTtJkiRJGgP/P6BjVbNC1RetAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2b48cdcd9c50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,3, figsize=(15,3))\n",
    "ax[0].hist(pval_1,bins = 50)\n",
    "ax[0].set_title('5000 SNPs p-value for trait1 distribution')\n",
    "ax[1].hist(pval_2,bins = 50)\n",
    "ax[1].set_title('5000 SNPs p-value for trait2 distribution')               \n",
    "ax[2].hist(pval_3,bins = 50)\n",
    "ax[2].set_title('5000 SNPs p-value for trait3 distribution')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "可以设定一个阈值，比如留下p-value前百分之四十的SNPs <br>\n",
    "根据index选取留下的SNPs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "threshold1: 0.044328\n",
      "threshold2: 0.004042\n",
      "threshold3: 0.004392\n"
     ]
    }
   ],
   "source": [
    "threshold1 = np.percentile(pval_1,40)\n",
    "threshold2 = np.percentile(pval_2,40)\n",
    "threshold3 = np.percentile(pval_3,40)\n",
    "print 'threshold1: %f' %threshold1\n",
    "print 'threshold2: %f' %threshold2\n",
    "print 'threshold3: %f' %threshold3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "#返回符合条件的p-value的坐标，即可找到需要留下的SNPs的位置，注意每个SNPs占据两行\n",
    "anova_index_1 = np.where(pval_1<threshold1)[0]\n",
    "anova_index_1 = np.sort(np.concatenate((anova_index_1,anova_index_1 +1)))\n",
    "anova_index_2 = np.where(pval_2<threshold2)[0]\n",
    "anova_index_2 = np.sort(np.concatenate((anova_index_2,anova_index_2 +1)))\n",
    "anova_index_3 = np.where(pval_3<threshold3)[0]\n",
    "anova_index_3 = np.sort(np.concatenate((anova_index_3,anova_index_3 +1)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "#根据p-value选取保留的SNPs，注意这一步要在所有的6210个样本上做\n",
    "feature1_anova = np.take(geno_conv,anova_index_1,axis=0)\n",
    "feature2_anova = np.take(geno_conv,anova_index_2,axis=0)\n",
    "feature3_anova = np.take(geno_conv,anova_index_3,axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "feature*_anova 即为回归模型可以使用的feature"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Random projection\n",
    "Random projection 不依赖于性状，仅仅在原SNPs数据进行降维 <br>\n",
    "Random projection可以使用scikit-learn下的sklearn.random_projection模块计算 <br>\n",
    "包括generate，transform和normalize等步骤 <br>\n",
    "这里演示从5000个feature（10000行）降维"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### generate\n",
    "产生一个稀疏矩阵"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 319,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1000, 10000)\n"
     ]
    }
   ],
   "source": [
    "#10000为操作前的feature个数：2*5000\n",
    "X = np.zeros((2, 10000))\n",
    "#确定降维后的个数，这里定为1000，使用sklearn random_projection 模块下的 SparseRandomProjection 函数\n",
    "proj = SparseRandomProjection(1000)\n",
    "proj.fit(X)\n",
    "print proj.components_.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### transform"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 325,
   "metadata": {},
   "outputs": [],
   "source": [
    "X= geno_conv.T\n",
    "X_ = proj.transform(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### normalize\n",
    "对每个feature进行normalize，避免出现过大的值"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 334,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "normalized_feeature = StandardScaler().fit_transform(X_).T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "最终大小为1000*6210，结果在feature_selection/randomproj_5000"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4回归模型\n",
    "这部分通过几个常用的机器学习模型对上一部处理过的feature进行拟合和预测\n",
    "这里使用sklearn和xgboost提供的模块，这些模块具有很好的封装，使用风格统一，使用时可以查看其官方文档\n",
    "这里不介绍具体的机器学习模型的算法原理，可以参考周志华老师的《机器学习》等书进行学习。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 机器学习模型\n",
    "接下来会使用一些常用的可以用于回归的机器学习模型，可以选择其中的一种或几种对feature_selection/文件夹下的三种数据进行回归和预测。下面我们将几个模型列出，并且选择其中的一种作为示例，其他的模型可同理调用。 <br>\n",
    "我们使用$r^2,pcc$作为衡量预测结果的指标<br>\n",
    "$r^2 = 1-\\frac{SS_{res}}{SS_{tot}}$<br>\n",
    "$pcc = \\frac{cov(X,Y)}{\\sigma_X \\sigma_Y} = \\frac{E[(X-\\mu_X)(Y-\\mu_Y)]}{\\sigma_X \\sigma_Y} $<br>\n",
    "我们可以绘制结果的heatmap图，散点图等进行可视化。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 交叉验证\n",
    "交叉验证(Cross validation)可以帮助调参，寻找机器学习模型中的超参数 <br>\n",
    "一般可以使用10折或者5折交叉验证，注意在最终预测时，使用调参后的模型在整个训练集上训练，这时不再交叉验证 <br>\n",
    "因为交叉验证需要额外增加计算时间，因此这里只在整个训练集上训练一次，不再展示交叉验证的过程。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### 如果深究的话，本问题还有其特殊性，可以设计特殊的交叉验证方式\n",
    "不同的样本具有关联性CV，有的样本可能来自同一亲本，而且训练集和测试集的划分并不是随机的\n",
    "因此在真正解决这个问题的时候，需要考虑不同的抽样方式下的调参与训练，我们可以使用下图所示的几种抽样方式"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 375,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2aaea0998e90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2aae0ee0f550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2aae0f295250>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2aae0eecc150>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for method in ('random', 'by_female', 'by_male', 'cross'):\n",
    "    with h5py.File('data/cv_index.%s'%method, 'r') as f:\n",
    "        index_train = f['0/train'][:]\n",
    "        index_test = f['0/test'][:]\n",
    "    fig, ax = subplots(2, 1, figsize=(16, 6))\n",
    "    sampling_table = np.zeros(np.prod(parent_table.shape))\n",
    "    sampling_table[index_train] = 1\n",
    "    sampling_table = np.take(sampling_table, parent_table)\n",
    "    ax[0].matshow(sampling_table, cmap=cm.Greys)\n",
    "    ax[0].set_title('Training samples (%s)'%method)\n",
    "\n",
    "    sampling_table = np.zeros(np.prod(parent_table.shape))\n",
    "    sampling_table[index_test] = 1\n",
    "    sampling_table = np.take(sampling_table, parent_table)\n",
    "    ax[1].matshow(sampling_table, cmap=cm.Greys)\n",
    "    ax[1].set_title('Test samples (%s)'%method)\n",
    "    plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 准备数据\n",
    "sklearn的机器学习模型一般需要提供X和y以供模型训练，然后提供新的X，模型就可以预测新的y <br>\n",
    "通过前面的工作，我们获得了三种不同的X(ANOVA、random projection(5000/whole))，我们还需要将X和y划分为训练集和测试集 <br>\n",
    "注意我们需要分别对三个性状进行预测，因此ANOVA的X是三种  <br>\n",
    "##### 评价模型的时候要注意，y_true的部分值缺失"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "先处理y，y 的train和test是统一的，不受方法影响"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 386,
   "metadata": {},
   "outputs": [],
   "source": [
    "#y 的train和test的统一的，不受方法影响\n",
    "pheno_whole = pd.read_csv('data/emaize_pheno_whole',delimiter=',')\n",
    "wholepheno = {}\n",
    "for trait in ['trait1','trait2','trait3']:\n",
    "    wholepheno[trait] = np.array(pheno_whole[trait])\n",
    "y_train = {}\n",
    "y_test = {}\n",
    "for trait in ['trait1','trait2','trait3']:\n",
    "    y_train[trait] = wholepheno[trait][:4754]\n",
    "    y_test[trait] = wholepheno[trait][4754:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "再处理X，使用ANOVA时X要区分不同性状，Random projection由于是与性状无关的降维方法，三种性状下的feature都一样 <br>\n",
    "因为机器学习模型要求一般数据形式为sample*feature，因此需要对结果转置"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 572,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_data(method):\n",
    "    if method == 'randomproj_5000':\n",
    "        with h5py.File('feature_selection/randomproj_5000') as f:\n",
    "            X_train = f['data'][:][:,:4754].T\n",
    "            X_test = f['data'][:][:,4754:].T\n",
    "    if method == 'randomproj_whole':\n",
    "        with h5py.File('feature_selection/randomproj_whole') as f:\n",
    "            X_train = f['X'][:][:4754,:]\n",
    "            X_test = f['X'][:][4754:,:]\n",
    "    if method == 'anova':\n",
    "        X_train = {}\n",
    "        X_test = {}\n",
    "        with h5py.File('feature_selection/anova') as f:\n",
    "            X_train['trait1'] = f['feature1'][:][:,:4754].T\n",
    "            X_test['trait1'] = f['feature1'][:][:,4754:].T\n",
    "            X_train['trait2'] = f['feature2'][:][:,:4754].T\n",
    "            X_test['trait2'] = f['feature2'][:][:,4754:].T\n",
    "            X_train['trait3'] = f['feature3'][:][:,:4754].T\n",
    "            X_test['trait3'] = f['feature3'][:][:,4754:].T\n",
    "    return X_train,X_test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "选择三种方法之一作为X,注意ANOVA方法返回的X需要指明性状"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 432,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "anova method X_train shape: (4000, 4754)\n",
      "anova method X_test shape: (4000, 1456)\n"
     ]
    }
   ],
   "source": [
    "#查看anova方法的X_train X_test\n",
    "X_train, X_test = prepare_data('anova')\n",
    "print 'anova method X_train shape: %s' %(X_train['trait1'].shape,)\n",
    "print 'anova method X_test shape: %s' %(X_test['trait1'].shape,)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 433,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "random projection method X_train shape: (1000, 4754)\n",
      "random projection method X_test shape: (1000, 1456)\n"
     ]
    }
   ],
   "source": [
    "#查看random projection方法的X_train X_test\n",
    "X_train, X_test = prepare_data('randomproj_5000')\n",
    "print 'random projection method X_train shape: %s' %(X_train.shape,)\n",
    "print 'random projection method X_test shape: %s' %(X_test.shape,)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 选择需要的机器学习模型\n",
    "接下来会提供多种机器学习模型，并且讲解其使用方法，可以选择自己喜欢的模型进行回归，也可以用sklearn或其他package提供的模型\n",
    "模型包括：\n",
    "lr: Linear regression\n",
    "ridge: Ridge regression\n",
    "kr: Kernel Ridge regression\n",
    "rfr: Random Forest regression\n",
    "xgbr: XGBoost regression\n",
    "knr: K-nearest neigbour regression\n",
    "gpr: Gaussian Process regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 436,
   "metadata": {},
   "outputs": [],
   "source": [
    "def Model(model):\n",
    "    if model=='lr':\n",
    "        reg = LinearRegression()\n",
    "    elif model=='xgbr':\n",
    "        reg = XGBRegressor()\n",
    "    elif model=='ridge':\n",
    "        reg = Ridge()\n",
    "    elif model=='kr':\n",
    "        reg = KernelRidge(alpha = 10000, kernel = 'polynomial',degree = 3)\n",
    "    elif model=='knr':\n",
    "        reg = neighbors.KNeighborsRegressor(n_neighbors=4, algorithm='brute')\n",
    "    elif model=='rfr':\n",
    "        reg = RandomForestRegressor(n_estimators=10, criterion='mse', max_depth=12, n_jobs=5)\n",
    "    elif model=='gpr':\n",
    "        kernel = 1.0 * DotProduct(sigma_0=1.0)**4\n",
    "        reg = GaussianProcessRegressor(kernel = kernel, optimizer=None)\n",
    "    return reg"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "接下来可以使用三种特征（X）中的一种以及七种方法中的一种进行训练和预测 <br>\n",
    "这里我们以randomproj_5000作为特征，用Ridge作为回归模型演示 <br>\n",
    "如果用anova得到的feature，需要注意不同性状的feature不一样 <br>\n",
    "某个机器学习模型的使用方法如下：reg.fit(X,y)用于拟合，reg.predict(X)用于预测。更多用法可以参考sklearn官方文档"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 594,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6fe63da81a78441186e6881a589c2ed8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/html": [
       "<p>Failed to display Jupyter Widget of type <code>HBox</code>.</p>\n",
       "<p>\n",
       "  If you're reading this message in the Jupyter Notebook or JupyterLab Notebook, it may mean\n",
       "  that the widgets JavaScript is still loading. If this message persists, it\n",
       "  likely means that the widgets JavaScript library is either not installed or\n",
       "  not enabled. See the <a href=\"https://ipywidgets.readthedocs.io/en/stable/user_install.html\">Jupyter\n",
       "  Widgets Documentation</a> for setup instructions.\n",
       "</p>\n",
       "<p>\n",
       "  If you're reading this message in another frontend (for example, a static\n",
       "  rendering on GitHub or <a href=\"https://nbviewer.jupyter.org/\">NBViewer</a>),\n",
       "  it may mean that your frontend doesn't currently support widgets.\n",
       "</p>\n"
      ],
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=3), HTML(value=u'')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test = prepare_data('randomproj_whole')\n",
    "reg = Model('gpr')\n",
    "y_predict = {}\n",
    "y_predict_train = {}\n",
    "for trait in tqdm(['trait1','trait2','trait3']):\n",
    "    reg.fit(X_train,y_train[trait])\n",
    "    y_predict[trait] = reg.predict(X_test)\n",
    "    y_predict_train[trait] = reg.predict(X_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "计算预测结果与真实值的$r^2,PCC$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 596,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_nonan = np.where(np.isnan(np.array(pheno_whole['trait1'])[4754:]) ==0) # 返回非零元素的索引\n",
    "pcc_train = {}\n",
    "pcc_test = {}\n",
    "r2_train = {}\n",
    "r2_test = {}\n",
    "for trait in ['trait1','trait2','trait3']:\n",
    "    pcc_test[trait] = pearsonr(y_predict[trait][test_nonan],np.array(pheno_whole[trait])[4754:][test_nonan])\n",
    "    pcc_train[trait] = pearsonr(y_predict_train[trait],y_train[trait])\n",
    "    r2_test[trait] = r2_score(y_predict[trait][test_nonan],np.array(pheno_whole[trait])[4754:][test_nonan])\n",
    "    r2_train[trait] = r2_score(y_predict_train[trait],y_train[trait])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 598,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'trait1': (0.43154634151808663, 4.3991562935411105e-50),\n",
       " 'trait2': (0.4616095517270678, 6.0543813523966813e-58),\n",
       " 'trait3': (0.56324735668802994, 3.5568566668991143e-91)}"
      ]
     },
     "execution_count": 598,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pcc_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 599,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'trait1': (1.0, 0.0), 'trait2': (1.0, 0.0), 'trait3': (1.0, 0.0)}"
      ]
     },
     "execution_count": 599,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pcc_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 602,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'trait1': -8.1960562982322926,\n",
       " 'trait2': -7.4624092986264596,\n",
       " 'trait3': -1.9911988726354415}"
      ]
     },
     "execution_count": 602,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r2_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 603,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'trait1': 1.0, 'trait2': 1.0, 'trait3': 1.0}"
      ]
     },
     "execution_count": 603,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r2_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "可以看到预测结果并不是很好，在测试集上的pcc只有0.5左右。后续的分析可以发现，这是因为样本之间具有相关性导致的<br>\n",
    "具体的原因分析比较复杂，简单来说，因为这组测试集与训练集的样本的亲本之间亲缘关系较远，模型难以从SNPs得到的feature推断出亲本信息，导致预测结果较差。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "绘制heatmap图观察预测结果 <br>\n",
    "GPR具有很强的拟合能力，总可以在训练集上得到接近1的PCC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 601,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2aafeb261390>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(2,3, figsize=(15,10))\n",
    "for i in range(3):\n",
    "    traits = ['trait1','trait2','trait3']\n",
    "    ax[0,i].scatter(y_predict[traits[i]][test_nonan],np.array(pheno_whole[traits[i]])[4754:][test_nonan])\n",
    "    ax[0,i].set_title('%s test set predict & true value plot' %traits[i])\n",
    "    line1 = [(-4, -4), (4, 4)]\n",
    "    (line1_xs, line1_ys) = zip(*line1)\n",
    "    ax[0,i].add_line(Line2D(line1_xs, line1_ys, linewidth=1, color='red'))\n",
    "    ax[0,i].set_xlim(left=-4, right=4)\n",
    "    ax[0,i].set_ylim(bottom=-4, top=4)\n",
    "    ax[1,i].scatter(y_predict_train[traits[i]],y_train[traits[i]])\n",
    "    ax[1,i].set_title('%s train set predict & true value plot' %traits[i])\n",
    "    ax[1,i].add_line(Line2D(line1_xs, line1_ys, linewidth=1, color='red'))\n",
    "    ax[1,i].set_xlim(left=-4, right=4)\n",
    "    ax[1,i].set_ylim(bottom=-4, top=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'pheno_whole' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m\u001b[0m",
      "\u001b[0;31mNameError\u001b[0mTraceback (most recent call last)",
      "\u001b[0;32m<ipython-input-72-5e4b2ba4453a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mpheno_whole\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'pheno_whole' is not defined"
     ]
    }
   ],
   "source": [
    "pheno_whole"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "绘制完整真实值与预测值的heatmap图 <br>\n",
    "从图中我们可以清晰地看出一个基本的模型的问题： <br>\n",
    "模型强烈地依赖已有信息进行预测，当未知样本的父本与已知训练集的亲缘关系较远时，模型只能依赖母本（横坐标）进行预测 <br>\n",
    "导致预测的heatmap图有明显的与母本相关的特征，而实际上子代的性状更容易被父本主导 <br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 640,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2aafeb4db610>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2aaed1771950>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2aafaf822fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "wholepre = np.concatenate((y_predict['trait1'],y_predict['trait2'],y_predict['trait3'])).reshape(3,-1)\n",
    "predictions =  pd.DataFrame(wholepre.T)\n",
    "predictions.columns = ['trait1', 'trait2', 'trait3']\n",
    "predictions = predictions.set_index(np.arange(4754,6210))\n",
    "def normalize_phenotype(x, range_pheno=4.0):\n",
    "    return (np.clip(x, -range_pheno, range_pheno) + range_pheno)/2.0/range_pheno\n",
    "for trait in traits:\n",
    "    fig, ax = subplots(2, 1, figsize=(16, 6))\n",
    "    ax[0].matshow(np.take(np.ravel(normalize_phenotype(pheno_whole[trait].values)), parent_table), cmap=cm.RdBu_r)\n",
    "    ax[0].set_title('Phenotypes of whole true data (%s)'%trait)\n",
    "\n",
    "    trait_pred = np.full(phenotypes.shape[0], np.nan)\n",
    "    trait_pred[predictions.index.tolist()] = normalize_phenotype(predictions[trait].values)\n",
    "    ax[1].matshow(np.take(trait_pred, parent_table), cmap=cm.RdBu)\n",
    "    ax[1].set_title('Prediction on test data (%s 1)'%traits)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 后续分析\n",
    "#### 不同样本具有不同的预测难度\n",
    "普通的机器学习模型在测试集上表现结果不好，但是通过多次的十字交叉抽样模拟，可以发现不同样本的预测难度不同，在大多数样本上，不需要专门设计的机器学习模型就足够表现很好\n",
    "\n",
    "#### 样本之间具有关联性\n",
    "不服从一些基本的假设，比如线性模型下，残差并不是独立的，需要考虑问题的特殊性进行额外的设计。<br>\n",
    "由于存储空间和计算时间的限制，无法展示其他有效的方法，有兴趣的同学可以查找育种领域的其他模型进行尝试。\n",
    "\n",
    "\n",
    "#### 复杂的机器学习模型并不一定有效\n",
    "育种领域目前最好的模型依然是线性模型，通过特殊的设计，可以考虑到亲缘关系、显著相关的SNP(causal)以及随机效应部分\n",
    "而寻找合适的feature是预测结果好坏的决定性因素，至今没有非常好的方法。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们通过模拟特殊的十字交叉抽样方式发现，虽然测试集的样本不好预测，但是大多数的样本使用简单的机器学习方法就可以在大多数样本上取得较好的结果 <br>\n",
    "由于计算资源限制，下面直接展示模拟结果"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们使用一种特殊的十字交叉抽样，在训练集上抽样1000次，用来测试基本的机器学习模型结果 <br>\n",
    "我们使用了2bit coding编码的SNPs,通过random projection降维至80000 <br>\n",
    "然后使用Gaussian Process Regression作为回归模型 <br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 502,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 502,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename='plots/0.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 503,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABhAAAARiCAYAAABF4737AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvAOZPmwAAIABJREFUeJzs3V+opHd9P/D355fVXqjQiJsQYqxBwg9toat7CEJAIqKN8oPVCyFe2FDkt14koOBN6o1e5uKn0kIrxBoSQQ0BDeYiqCEIodDanJWgialNsKmuWbIrQrUUKomf38WZ4DHfPX/2zJl5Zs6+XnCYeb7zzHk+M2fmM9+d9z7PU90dAAAAAACA7f7X1AUAAAAAAACrR4AAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADCYJEKrqlqr6SVU9W1V3TlEDAPOrqueq6kdV9URVbc7GXl9Vj1TVM7PLK6euE4CdVdU9VXW+qp7cNnbRXl5b/nY2j/9hVb1jusoB2MkOvf2zVfWL2dz9iar6wLbb/nrW239SVX8xTdUA7Kaqrquq71XV01X1VFV9Yja+0Ln70gOEqroiyd8leX+StyX5SFW9bdl1AHBo3t3dJ7p7Y7Z8Z5JHu/uGJI/OlgFYXfcmueUVYzv18vcnuWH2czrJF5dUIwCX5t6MvT1JvjCbu5/o7oeTZPadzK1J/nR2n7+ffXcDwGp5McmnuvutSd6Z5PZZD1/o3H2KPRBuTPJsd/+0u3+b5P4kpyaoA4DFOJXkvtn1+5J8cMJaANhDdz+W5FevGN6pl59K8pXe8s9J/riqrllOpQDs1w69fSenktzf3f/T3f+e5NlsfXcDwArp7nPd/YPZ9d8keTrJtVnw3H2KAOHaJD/ftnx2NgbA+ukk362qM1V1ejZ2dXefS7Y+3JJcNVl1ABzUTr3cXB5gvd0xO4zFPdsONaq3A6yZqnpzkrcn+X4WPHefIkCoi4z10qsA4DDc1N3vyNZucbdX1bumLgiAhTKXB1hfX0zyliQnkpxL8rnZuN4OsEaq6rVJvpHkk939691WvcjYJff3KQKEs0mu27b8xiTPT1AHAHPq7udnl+eTPJitXZ1feHmXuNnl+ekqBOCAdurl5vIAa6q7X+jul7r7d0m+lN8fpkhvB1gTVfWqbIUHX+3ub86GFzp3nyJAeDzJDVV1fVW9Olsn6nlogjoAmENVvaaqXvfy9STvS/Jktnr6bbPVbkvyrWkqBGAOO/Xyh5L8ZW15Z5L/fHl3aQBW2yuOe/2hbM3dk63efmtV/VFVXZ+tk23+y7LrA2B3VVVJvpzk6e7+/LabFjp3PzZHzQfS3S9W1R1JvpPkiiT3dPdTy64DgLldneTBrc+vHEvyte7+dlU9nuSBqvpYkp8l+fCENQKwh6r6epKbk7yhqs4m+UySu3LxXv5wkg9k6wSb/53kr5ZeMAB72qG331xVJ7J1+Irnknw8Sbr7qap6IMmPk7yY5PbufmmKugHY1U1JPprkR1X1xGzs01nw3L26HdYOAAAAAAD4Q1McwggAAAAAAFhxAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABpMGCFV1esrtA3D49HaAo0VfBzh69HaAo2dRvX3qPRB8YAEcPXo7wNGirwMcPXo7wNFzJAMEAAAAAABgBVV3H/zOVbck+ZskVyT5h+6+a4/1D74xAAAAAADYwcmTJ6cuYTIXLlzI8ePH973+mTNnftnde97hwAFCVV2R5N+SvDfJ2SSPJ/lId/94l/sIEAAAAAAAOHTz/Gf5y01Vnenujb3Wm+cQRjcmeba7f9rdv01yf5JTc/w+AAAAAABgRcwTIFyb5Ofbls/OxgAAAAAAgDV3bI771kXGhn1Equp0FnQGaAAAAAAAYDHmCRDOJrlu2/Ibkzz/ypW6++4kdyfOgQAAAAAAAOtinkMYPZ7khqq6vqpeneTWJA8dTlkAAAAAAMCUDrwHQne/WFV3JPlOkiuS3NPdTx1aZQAAAAAAwGSqe3lHFdrY2OjNzc2lbQ9gJ1UXO43L7y2zNwKwHLv1fn0fAGC5ppqbmRPClqo6090be603zyGMAAAAAACAI0qAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAyqu5e3sarlbQzgiNmrX1fVkir5Q7vVNU9Nq/p44XLhPcgUFvWZAgDA5WGZ33Xv1yLnsfM83qo6090be61nDwQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGBwbJkbO3nyZDY3N5e5SYCLqqpdb+/uJVVyeFax5kXWtIqPFy4n6/ge3K33r+Pjudz4GwHA0TLV3MyckGVb99eVPRAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYHJu6AAAAAADg4Kpqx9u6e4mVAEeNPRAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYHJvnzlX1XJLfJHkpyYvdvXEYRQEAAAAAANOaK0CYeXd3//IQfg8AAAAAALAiHMIIAAAAAAAYzBsgdJLvVtWZqjp9GAUBAAAAAADTm/cQRjd19/NVdVWSR6rqX7v7se0rzIKF00nypje9ac7NAQAAAAAAyzDXHgjd/fzs8nySB5PceJF17u7uje7eOH78+DybAwAAAAAAluTAAUJVvaaqXvfy9STvS/LkYRUGAAAAAABMZ55DGF2d5MGqevn3fK27v30oVQEAAAAAAJM6cIDQ3T9N8ueHWAsAAAAAALAi5joHAgAAAAAAcDQJEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgEF19/I2VrW8jQEcMXv166paUiXLcbk93suNvy8AAMBq2+3fbav6b7Zlfte9X/M8V4t8PFV1prs39lrPHggAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMDg2DI3dvLkyWxubi5zkwAXVVW73t7dS6rk8KxjzfO43B7v5cbfl0XYrfd7zQEA62wd5zlT1byOz9Uq8lzt37o/V/ZAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYLBngFBV91TV+ap6ctvY66vqkap6ZnZ55WLLBAAAAAAAlmk/eyDcm+SWV4zdmeTR7r4hyaOzZQAAAAAA4IjYM0Do7seS/OoVw6eS3De7fl+SDx5yXQAAAAAAwIQOeg6Eq7v7XJLMLq86vJIAAAAAAICpLfwkylV1uqo2q2rzwoULi94cAAAAAABwCA4aILxQVdckyezy/E4rdvfd3b3R3RvHjx8/4OYAAAAAAIBlOmiA8FCS22bXb0vyrcMpBwAAAAAAWAV7BghV9fUk/5Tkf1fV2ar6WJK7kry3qp5J8t7ZMgAAAAAAcEQc22uF7v7IDje955BrAQAAAAAAVsTCT6IMAAAAAACsHwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAACDY1MX8LKqmroEODTdfeD7zvNe2Gu7q/g+W9WaV/G5goPa7X3mtc5OVrU/L8qqPh7v38Mx1dwMAFbBOn6W+S4Atszz77J55sDb2QMBAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAwZ4BQlXdU1Xnq+rJbWOfrapfVNUTs58PLLZMAAAAAABgmaq7d1+h6l1J/ivJV7r7z2Zjn03yX939/y5pY1W7bwyAHe2jXy+pkuW43B7v5cbfF2B3+iQAwKXbaw61inab1y3y8VTVme7e2Gu9PfdA6O7HkvzqUKoCAAAAAADWwjznQLijqn44O8TRlYdWEQAAAAAAMLmDBghfTPKWJCeSnEvyuZ1WrKrTVbVZVZsH3BYAAAAAALBkBwoQuvuF7n6pu3+X5EtJbtxl3bu7e2M/x1MCAAAAAABWw4EChKq6Ztvih5I8eTjlAAAAAAAAq+DYXitU1deT3JzkDVV1NslnktxcVSeSdJLnknx8gTUCAAAAAABLVt29tI1tbGz05qZTIQDTq6pdb19mbwRgOXbr/fo+ALDO1nGeM1XN6/hcwSJU1Zn9nHbgoCdRBgAAAAAAjjABAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAACDPQOEqrquqr5XVU9X1VNV9YnZ+Our6pGqemZ2eeXiywUAAAAAAJZhP3sgvJjkU9391iTvTHJ7Vb0tyZ1JHu3uG5I8OlsGAAAAAACOgD0DhO4+190/mF3/TZKnk1yb5FSS+2ar3Zfkg4sqEgAAAAAAWK5LOgdCVb05yduTfD/J1d19LtkKGZJcddjFAQAAAAAA09h3gFBVr03yjSSf7O5fX8L9TlfVZlVtXrhw4SA1AgAAAAAAS7avAKGqXpWt8OCr3f3N2fALVXXN7PZrkpy/2H27++7u3ujujePHjx9GzQAAAAAAwILtGSBUVSX5cpKnu/vz2256KMlts+u3JfnW4ZcHAAAAAABM4dg+1rkpyUeT/KiqnpiNfTrJXUkeqKqPJflZkg8vpkQAAAAAAGDZ9gwQuvsfk9QON7/ncMsBAAAAAABWwb5PogwAAAAAAFw+BAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADKq7l7exqgtJ/mPb0BuS/HJpBQCwDHo7wNGirwMcPXo7wNFzqb39T7r7+F4rLTVAGDZetdndG5MVAMCh09sBjhZ9HeDo0dsBjp5F9XaHMAIAAAAAAAYCBAAAAAAAYDB1gHD3xNsH4PDp7QBHi74OcPTo7QBHz0J6+6TnQAAAAAAAAFbT1HsgAAAAAAAAK0iAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADCYJEKrqlqr6SVU9W1V3TlEDAPOrqueq6kdV9URVbc7GXl9Vj1TVM7PLK6euE4CdVdU9VXW+qp7cNnbRXl5b/nY2j/9hVb1jusoB2MkOvf2zVfWL2dz9iar6wLbb/nrW239SVX8xTdUA7Kaqrquq71XV01X1VFV9Yja+0Ln70gOEqroiyd8leX+StyX5SFW9bdl1AHBo3t3dJ7p7Y7Z8Z5JHu/uGJI/OlgFYXfcmueUVYzv18vcnuWH2czrJF5dUIwCX5t6MvT1JvjCbu5/o7oeTZPadzK1J/nR2n7+ffXcDwGp5McmnuvutSd6Z5PZZD1/o3H2KPRBuTPJsd/+0u3+b5P4kpyaoA4DFOJXkvtn1+5J8cMJaANhDdz+W5FevGN6pl59K8pXe8s9J/riqrllOpQDs1w69fSenktzf3f/T3f+e5NlsfXcDwArp7nPd/YPZ9d8keTrJtVnw3H2KAOHaJD/ftnx2NgbA+ukk362qM1V1ejZ2dXefS7Y+3JJcNVl1ABzUTr3cXB5gvd0xO4zFPdsONaq3A6yZqnpzkrcn+X4WPHefIkCoi4z10qsA4DDc1N3vyNZucbdX1bumLgiAhTKXB1hfX0zyliQnkpxL8rnZuN4OsEaq6rVJvpHkk939691WvcjYJff3KQKEs0mu27b8xiTPT1AHAHPq7udnl+eTPJitXZ1feHmXuNnl+ekqBOCAdurl5vIAa6q7X+jul7r7d0m+lN8fpkhvB1gTVfWqbIUHX+3ub86GFzp3nyJAeDzJDVV1fVW9Olsn6nlogjoAmENVvaaqXvfy9STvS/Jktnr6bbPVbkvyrWkqBGAOO/Xyh5L8ZW15Z5L/fHl3aQBW2yuOe/2hbM3dk63efmtV/VFVXZ+tk23+y7LrA2B3VVVJvpzk6e7+/LabFjp3PzZHzQfS3S9W1R1JvpPkiiT3dPdTy64DgLldneTBrc+vHEvyte7+dlU9nuSBqvpYkp8l+fCENQKwh6r6epKbk7yhqs4m+UySu3LxXv5wkg9k6wSb/53kr5ZeMAB72qG331xVJ7J1+Irnknw8Sbr7qap6IMmPk7yY5PbufmmKugHY1U1JPprkR1X1xGzs01nw3L26HdYOAAAAAAD4Q1McwggAAAAAAFhxAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABpMGCFV1esrtA3D49HaAo0VfBzh69HaAo2dRvX3qPRB8YAEcPXo7wNGirwMcPXo7wNFzJAMEAAAAAABgBVV3H/zOVbck+ZskVyT5h+6+a4/1D74xgEN08uTJXW8/c+bMkioBYFl26/36PgCwztZxnjNVzev4XLF/e33fc5RduHAhx48f3/f6Z86c+WV373mHAwcIVXVFkn9L8t4kZ5M8nuQj3f3jXe4jQABWwl69r6qWVAkAy7Jb79f3AYB1to7znKlqXsfniv2b5z/LX26q6kx3b+y13jyHMLoxybPd/dPu/m2S+5OcmuP3AQAAAAAAK2KeAOHaJD/ftnx2NgYAAAAAAKy5Y3Pc92L79Az7iFTV6SzoDNAAAAAAAMBizBMgnE1y3bblNyZ5/pUrdffdSe5OnAMBAAAAAADWxTyHMHo8yQ1VdX1VvTrJrUkeOpyyAAAAAACAKR14D4TufrGq7kjynSRXJLmnu5/a7T4nT57M5ubmQTcJsDTddpgCuJzo+wDAUbWO85ypal7H5woWbZ5DGKW7H07y8CHVAgAAAAAArIh5DmEEAAAAAAAcUQIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAbV3cvbWNXyNgawi716X1UtqZL9W9Wad6trnppW9fHC5eIovgcX1a84PP5GAHAw6/gZOlXN6/hcsX/L/K57v6Z6Pe+lqs5098Ze69kDAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGBxb5sZOnjyZzc3NZW4S4EC6e+oSLtkq1rzImlbx8cLl5Ki9B4/a4zmK/I0A4GDW8TN0qprX8bli9a3768oeCAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADI7Nc+eqei7Jb5K8lOTF7t44jKIAAAAAAIBpzRUgzLy7u395CL8HAAAAAABYEQ5hBAAAAAAADOYNEDrJd6vqTFWdPoyCAAAAAACA6c17CKObuvv5qroqySNV9a/d/dj2FWbBwukkedOb3jTn5gAAAAAAgGWYaw+E7n5+dnk+yYNJbrzIOnd390Z3bxw/fnyezQEAAAAAAEty4AChql5TVa97+XqS9yV58rAKAwAAAAAApjPPIYyuTvJgVb38e77W3d8+lKoAAAAAAIBJHThA6O6fJvnzQ6wFAAAAAABYEXOdAwEAAAAAADiaBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMqruXt7Gq5W0MYBd79b6qWlIl+7eONc/jcnu8lxt/X6aw2+vOaw4AWGfrOM+ZquZ1fK6mso7P1TK/696veZ6rRT6eqjrT3Rt7rWcPBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYHBsmRs7efJkNjc3l7lJgAPp7qlLuGTrWPM8LrfHe7nx92XZvOYAgKNqHec5U9W8js/VVDxX+7fuz5U9EAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgMGeAUJV3VNV56vqyW1jr6+qR6rqmdnllYstEwAAAAAAWKb97IFwb5JbXjF2Z5JHu/uGJI/OlgEAAAAAgCNizwChux9L8qtXDJ9Kct/s+n1JPnjIdQEAAAAAABM66DkQru7uc0kyu7zq8EoCAAAAAACmtvCTKFfV6ararKrNCxcuLHpzAAAAAADAIThogPBCVV2TJLPL8zut2N13d/dGd28cP378gJsDAAAAAACW6aABwkNJbptdvy3Jtw6nHAAAAAAAYBXsGSBU1deT/FOS/11VZ6vqY0nuSvLeqnomyXtnywAAAAAAwBFxbK8VuvsjO9z0nkOuBQAAAAAAWBELP4kyAAAAAACwfgQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADA4NjUBbysqqYuAQ5Ndx/4vvO8F/ba7iq+z6aqeR2fKzio3V7vXuvs5Cj2yXV8L6xjzatoqrkZACzLOs4Zpqp5HZ8rLl/z/LtsnjnwdvZAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABtXdu69QdU+S/5PkfHf/2Wzss0n+b5ILs9U+3d0P77mxqt03BrAk++h9S6pk/9ax5nlcbo/3cuPvyxR2e915zbFq9EkALsU6znOmqnkdnyv2b6851Cra7XW3yMdTVWe6e2Ov9fazB8K9SW65yPgXuvvE7GfP8AAAAAAAAFgfewYI3f1Ykl8toRYAAAAAAGBFzHMOhDuq6odVdU9VXXloFQEAAAAAAJM7aIDwxSRvSXIiybkkn9tpxao6XVWbVbV5wG0BAAAAAABLdqAAobtf6O6Xuvt3Sb6U5MZd1r27uzf2c0IGAAAAAABgNRwoQKiqa7YtfijJk4dTDgAAAAAAsAqO7bVCVX09yc1J3lBVZ5N8JsnNVXUiSSd5LsnH97OxkydPZnPTkYyA1dfdU5dwydax5nlcbo/3cuPvy7J5zbFuvGYB2K91/MyYquZ1fK5Yf6v+utszQOjuj1xk+MsLqAUAAAAAAFgRBz2JMgAAAAAAcIQJEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAY7BkgVNV1VfW9qnq6qp6qqk/Mxl9fVY9U1TOzyysXXy4AAAAAALAM+9kD4cUkn+rutyZ5Z5Lbq+ptSe5M8mh335Dk0dkyAAAAAABwBOwZIHT3ue7+wez6b5I8neTaJKeS3Ddb7b4kH1xUkQAAAAAAwHJd0jkQqurNSd6e5PtJru7uc8lWyJDkqsMuDgAAAAAAmMa+A4Sqem2SbyT5ZHf/+hLud7qqNqtq88KFCwepEQAAAAAAWLJ9BQhV9apshQdf7e5vzoZfqKprZrdfk+T8xe7b3Xd390Z3bxw/fvwwagYAAAAAABZszwChqirJl5M83d2f33bTQ0lum12/Lcm3Dr88AAAAAABgCsf2sc5NST6a5EdV9cRs7NNJ7kryQFV9LMnPknx4MSUCAAAAAADLtmeA0N3/mKR2uPk9h1sOAAAAAACwCvZ9EmUAAAAAAODyIUAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCDBvKBiAAAgAElEQVRAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAbV3cvbWNWFJP+xbegNSX65tAIAWAa9HeBo0dcBjh69HeDoudTe/ifdfXyvlZYaIAwbr9rs7o3JCgDg0OntAEeLvg5w9OjtAEfPonq7QxgBAAAAAAADAQIAAAAAADCYOkC4e+LtA3D49HaAo0VfBzh69HaAo2chvX3ScyAAAAAAAACraeo9EAAAAAAAgBUkQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYDBJgFBVt1TVT6rq2aq6c4oaAJhfVT1XVT+qqieqanM29vqqeqSqnpldXjl1nQDsrKruqarzVfXktrGL9vLa8rezefwPq+od01UOwE526O2frapfzObuT1TVB7bd9tez3v6TqvqLaaoGYDdVdV1Vfa+qnq6qp6rqE7Pxhc7dlx4gVNUVSf4uyfuTvC3JR6rqbcuuA4BD8+7uPtHdG7PlO5M82t03JHl0tgzA6ro3yS2vGNupl78/yQ2zn9NJvrikGgG4NPdm7O1J8oXZ3P1Edz+cJLPvZG5N8qez+/z97LsbAFbLi0k+1d1vTfLOJLfPevhC5+5T7IFwY5Jnu/un3f3bJPcnOTVBHQAsxqkk982u35fkgxPWAsAeuvuxJL96xfBOvfxUkq/0ln9O8sdVdc1yKgVgv3bo7Ts5leT+7v6f7v73JM9m67sbAFZId5/r7h/Mrv8mydNJrs2C5+5TBAjXJvn5tuWzszEA1k8n+W5Vnamq07Oxq7v7XLL14ZbkqsmqA+Cgdurl5vIA6+2O2WEs7tl2qFG9HWDNVNWbk7w9yfez4Ln7FAFCXWSsl14FAIfhpu5+R7Z2i7u9qt41dUEALJS5PMD6+mKStyQ5keRcks/NxvV2gDVSVa9N8o0kn+zuX++26kXGLrm/TxEgnE1y3bblNyZ5foI6AJhTdz8/uzyf5MFs7er8wsu7xM0uz09XIQAHtFMvN5cHWFPd/UJ3v9Tdv0vypfz+MEV6O8CaqKpXZSs8+Gp3f3M2vNC5+xQBwuNJbqiq66vq1dk6Uc9DE9QBwByq6jVV9bqXryd5X5Ins9XTb5utdluSb01TIQBz2KmXP5TkL2vLO5P858u7SwOw2l5x3OsPZWvunmz19lur6o+q6vpsnWzzX5ZdHwC7q6pK8uUkT3f357fdtNC5+7E5aj6Q7n6xqu5I8p0kVyS5p7ufWnYdAMzt6iQPbn1+5ViSr3X3t6vq8SQPVNXHkvwsyYcnrBGAPVTV15PcnOQNVXU2yWeS3JWL9/KHk3wgWyfY/O8kf7X0ggHY0w69/eaqOpGtw1c8l+TjSdLdT1XVA0l+nOTFJLd390tT1A3Arm5K8tEkP6qqJ2Zjn86C5+7V7bB2AAAAAADAH5riEEYA8P/Zu78Q287yfuDfh4z2QoVGziSEGHtEQtFeNOoQhECJiDZ6c/RC0AsbinC8SEDBm9QbvcyNigUrxBqSgn8QNJiLoIYghEJrM0eCJqb+EtKox4ScEaEKhUri87uYnXaa98zMObP37DV75/OBYe/1rrX2es46c555z3xZawEAAABwwgkQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABhMGiBU1dkpjw/A4untAOtFXwdYP3o7wPo5rt4+9RUIfmABrB+9HWC96OsA60dvB1g/axkgAAAAAAAAJ1B199F3rrolyReTXJHkH7v7zoO2P3XqVJ8+ffp/lnd2drK5uXnk4wNw8ujtAOtFXwdYP3o7wPq53N5+7ty533T3oTtsHLWgqroiyZeSvCfJ+SSPVNX93f2z/fY5ffp0tre3j3pIAAAAAABgTlX1i0vZbp5bGN2Y5Knufrq7/5Dkm0nOzPF5AAAAAADACTFPgHBtkl/tWT4/GwMAAAAAAFbcPAFCXWRseKBCVZ2tqu2q2t7Z2ZnjcAAAAAAAwLLMEyCcT3LdnuU3JHn25Rt1913dvdXdWx7QAwAAAAAAq2GeAOGRJNdX1Zuq6tVJPpzk/sWUBQAAAAAATGnjqDt29wtVdXuS7ye5Isnd3f34wioDOEZVF7sL2//qHu7IBsCKO6j36/sAAMs11dzMnBAuz5EDhCTp7geSPLCgWgAAAAAAgBNinlsYAQAAAAAAa0qAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBgY+oCAKbQ3VOXAMCS6f0AACfHVHMzc0K4PK5AAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABhtTFwAwhao6cH13L6kSAJbloN6v7wMALNdUczNzQrg8rkAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgsDF1AQAAAADA0VXVvuu6e4mVAOvGFQgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBgY56dq+qZJL9P8mKSF7p7axFFAQAAAAAA05orQJh5V3f/ZgGfAwAAAAAAnBBuYQQAAAAAAAzmDRA6yQ+q6lxVnV1EQQAAAAAAwPTmvYXRTd39bFVdleTBqvr37n547wazYOFskrzxjW+c83AAAAAAAMAyzHUFQnc/O3u9kOS+JDdeZJu7unuru7c2NzfnORwAAAAAALAkRw4Qquo1VfW6l94neW+SxxZVGAAAAAAAMJ15bmF0dZL7quqlz/l6d39vIVUBAAAAAACTOnKA0N1PJ/nLBdYCAAAAAACcEHM9AwEAAAAAAFhPAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGG1MXADCF7p66BACWTO8HANbVKs5zpqp5Fc8VTMkVCAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwGBj6gIAplBVB67v7iVVAsCyHNT79X0AYJWt4jxnqppX8VzBlFyBAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADA4NEKrq7qq6UFWP7Rl7fVU9WFVPzl6vPN4yAQAAAACAZbqUKxDuSXLLy8buSPJQd1+f5KHZMgAAAAAAsCYODRC6++Ekv33Z8Jkk987e35vkAwuuCwAAAAAAmNBRn4FwdXc/lySz16sWVxIAAAAAADC1Y3+IclWdrartqtre2dk57sMBAAAAAAALcNQA4fmquiZJZq8X9tuwu+/q7q3u3trc3Dzi4QAAAAAAgGU6aoBwf5JbZ+9vTfLdxZQDAAAAAACcBIcGCFX1jST/kuTPq+p8VX0syZ1J3lNVTyZ5z2wZAAAAAABYExuHbdDdH9ln1bsXXAsAAAAAAHBCHPtDlAEAAAAAgNUjQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYbUxcAMIXunroEAJZM7wcA1tUqznOmqnkVzxVMyRUIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADA4NAAoarurqoLVfXYnrHPVtWvq+rR2df7j7dMAAAAAABgmS7lCoR7ktxykfEvdPcNs68HFlsWAAAAAAAwpUMDhO5+OMlvl1ALAAAAAABwQszzDITbq+ons1scXbmwigAAAAAAgMkdNUD4cpI3J7khyXNJPrffhlV1tqq2q2p7Z2fniIcDAAAAAACW6UgBQnc/390vdvcfk3wlyY0HbHtXd29199bm5uZR6wQAAAAAAJboSAFCVV2zZ/GDSR5bTDkAAAAAAMBJsHHYBlX1jSQ3JzlVVeeTfCbJzVV1Q5JO8kySjx9jjQALV1UHru/uJVUCwLIc1Pv1fQBgla3iPGeqmlfxXMGUDg0QuvsjFxn+6jHUAgAAAAAAnBBHfYgyAAAAAACwxgQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwODRCq6rqq+mFVPVFVj1fVJ2bjr6+qB6vqydnrlcdfLgAAAAAAsAyXcgXCC0k+1d1vSfLOJLdV1VuT3JHkoe6+PslDs2UAAAAAAGANHBogdPdz3f3j2fvfJ3kiybVJziS5d7bZvUk+cFxFAgAAAAAAy3VZz0CoqtNJ3pbkR0mu7u7nkt2QIclViy4OAAAAAACYxiUHCFX12iTfTvLJ7v7dZex3tqq2q2p7Z2fnKDUCAAAAAABLdkkBQlW9Krvhwde6+zuz4eer6prZ+muSXLjYvt19V3dvdffW5ubmImoGAAAAAACO2aEBQlVVkq8meaK7P79n1f1Jbp29vzXJdxdfHgAAAAAAMIWNS9jmpiQfTfLTqnp0NvbpJHcm+VZVfSzJL5N86HhKBAAAAAAAlu3QAKG7/zlJ7bP63YstBwAAAAAAOAku+SHKAAAAAADAK4cAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYVHcv72BVO0l+sWfoVJLfLK0AAJZBbwdYL/o6wPrR2wHWz+X29j/r7s3DNlpqgDAcvGq7u7cmKwCAhdPbAdaLvg6wfvR2gPVzXL3dLYwAAAAAAICBAAEAAAAAABhMHSDcNfHxAVg8vR1gvejrAOtHbwdYP8fS2yd9BgIAAAAAAHAyTX0FAgAAAAAAcAIJEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGEwSIFTVLVX186p6qqrumKIGAOZXVc9U1U+r6tGq2p6Nvb6qHqyqJ2evV05dJwD7q6q7q+pCVT22Z+yivbx2/f1sHv+Tqnr7dJUDsJ99evtnq+rXs7n7o1X1/j3r/m7W239eVX89TdUAHKSqrquqH1bVE1X1eFV9YjZ+rHP3pQcIVXVFki8leV+Styb5SFW9ddl1ALAw7+ruG7p7a7Z8R5KHuvv6JA/NlgE4ue5JcsvLxvbr5e9Lcv3s62ySLy+pRgAuzz0Ze3uSfGE2d7+hux9IktnvZD6c5C9m+/zD7Hc3AJwsLyT5VHe/Jck7k9w26+HHOnef4gqEG5M81d1Pd/cfknwzyZkJ6gDgeJxJcu/s/b1JPjBhLQAcorsfTvLblw3v18vPJPmn3vWvSf60qq5ZTqUAXKp9evt+ziT5Znf/d3f/R5Knsvu7GwBOkO5+rrt/PHv/+yRPJLk2xzx3nyJAuDbJr/Ysn5+NAbB6OskPqupcVZ2djV3d3c8luz/cklw1WXUAHNV+vdxcHmC13T67jcXde241qrcDrJiqOp3kbUl+lGOeu08RINRFxnrpVQCwCDd199uze1ncbVX1V1MXBMCxMpcHWF1fTvLmJDckeS7J52bjejvACqmq1yb5dpJPdvfvDtr0ImOX3d+nCBDOJ7luz/Ibkjw7QR0AzKm7n529XkhyX3YvdX7+pUviZq8XpqsQgCPar5ebywOsqO5+vrtf7O4/JvlK/vc2RXo7wIqoqldlNzz4Wnd/ZzZ8rHP3KQKER5JcX1VvqqpXZ/dBPfdPUAcAc6iq11TV6156n+S9SR7Lbk+/dbbZrUm+O02FAMxhv15+f5K/qV3vTPKfL10uDcDJ9rL7Xn8wu3P3ZLe3f7iq/qSq3pTdh23+27LrA+BgVVVJvprkie7+/J5Vxzp335ij5iPp7heq6vYk309yRZK7u/vxZdcBwNyuTnLf7s+vbCT5end/r6oeSfKtqvpYkl8m+dCENQJwiKr6RpKbk5yqqvNJPpPkzly8lz+Q5P3ZfcDmfyX526UXDMCh9untN1fVDdm9fcUzST6eJN39eFV9K8nPkryQ5LbufnGKugE40E1JPprkp1X16Gzs0znmuXt1u60dAAAAAADwf01xCyMAAAAAAOCEEyAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMJg0QKiqs1MeH4DF09sB1ou+DrB+9HaA9XNcvX3qKxD8wAJYP3o7wHrR1wHWj94OsH7WMkAAAAAAAABOoOruo+9cdUuSLya5Isk/dvedB21/6tSpPn369P8s7+zsZHNz88jHBziqc+fOHbj+He94x5IqWT96O3BSHdT79f396esA60dvXz+rOM+ZquZVPFdwKS63t587d+433X3oDkcOEKrqiiT/L8l7kpxP8kiSj3T3z/bbZ2trq7e3t490PIBFqqoD188TrgJwMh3U+/V9AGCVreI8Z6qaV/FcwXGoqnPdvXXYdvPcwujGJE9199Pd/Yck30xyZo7PAwAAAAAAToh5AoRrk/xqz/L52RgAAAAAALDi5gkQLna9z3CdT1Wdrartqtre2dmZ43AAAAAAAMCyzBMgnE9y3Z7lNyR59uUbdfdd3b3V3Vse0AMAAAAAAKthngDhkSTXV9WbqurVST6c5P7FlAUAAAAAAExp46g7dvcLVXV7ku8nuSLJ3d39+MIqAzhG3cMd1wBYc3o/ALCuVnGeM1XNq3iuYEpHDhCSpLsfSPLAgmoBAAAAAABOiHluYQQAAAAAAKwpAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAg42pCwCYQlUduL67l1QJAMtyUO/X9wGAVbaK85ypal7FcwVTcgUCAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAw2Ji6AIApdPfUJQCwZHo/ALCuVnGeM1XNq3iuYEquQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYLAxz85V9UyS3yd5MckL3b21iKIAAAAAAIBpzRUgzLyru3+zgM8BAAAAAABOCLcwAgAAAAAABvMGCJ3kB1V1rqrOLqIgAAAAAABgevPewuim7n62qq5K8mBV/Xt3P7x3g1mwcDZJ3vjGN855OAAAAAAAYBnmugKhu5+dvV5Icl+SGy+yzV3dvdXdW5ubm/McDgAAAAAAWJIjBwhV9Zqqet1L75O8N8ljiyoMAAAAAACYzjy3MLo6yX1V9dLnfL27v7eQqgAAAAAAgEkdOUDo7qeT/OUCawEAAAAAAE6IuZ6BAAAAAAAArCcBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAIONqQsAmEJVHbi+u5dUCQDLclDv1/cBgFW2ivOcqWpexXMFU3IFAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMNiYugCAKXT31CUAsGR6PwCwrlZxnjNVzat4rmBKrkAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGhwYIVXV3VV2oqsf2jL2+qh6sqidnr1ceb5kAAAAAAMAyXcoVCPckueVlY3ckeai7r0/y0GwZAAAAAABYE4cGCN39cJLfvmz4TJJ7Z+/vTfKBBdcFAAAAAABM6KjPQLi6u59LktnrVYsrCQAAAAAAmNqxP0S5qs5W1XZVbe/s7Bz34QAAAAAAgAU4aoDwfFVdkySz1wv7bdjdd3X3VndvbW5uHvFwAAAAAADAMh01QLg/ya2z97cm+e5iygEAAAAAAE6CQwOEqvpGkn9J8udVdb6qPpbkziTvqaonk7xntgwAAAAAAKyJjcM26O6P7LPq3QuuBQAAAAAAOCGO/SHKAAAAAADA6hEgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAg42pCwCYQlUduL67l1QJAMtyUO/X9wGAVbaK85ypal7FcwVTcgUCAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwODRAqKq7q+pCVT22Z+yzVfXrqnp09vX+4y0TAAAAAABYpku5AuGeJLdcZPwL3X3D7OuBxZYFAAAAAABM6dAAobsfTvLbJdQCAAAAAACcEPM8A+H2qvrJ7BZHVy6sIgAAAAAAYHJHDRC+nOTNSW5I8lySz+23YVWdrartqtre2dk54uEAAAAAAIBlOlKA0N3Pd/eL3f3HJF9JcuMB297V3VvdvbW5uXnUOgEAAAAAgCU6UoBQVdfsWfxgkscWUw4AAAAAAHASbBy2QVV9I8nNSU5V1fkkn0lyc1XdkKSTPJPk48dYI8DCdffUJQCwZHo/ALCuVnGeM1XNq3iuYEqHBgjd/ZGLDH/1GGoBAAAAAABOiKM+RBkAAAAAAFhjAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgIEAAAAAAAAAGAgQAAAAAAGAgQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABocGCFV1XVX9sKqeqKrHq+oTs/HXV9WDVfXk7PXK4y8XAAAAAABYhku5AuGFJJ/q7rckeWeS26rqrUnuSPJQd1+f5KHZMgAAAAAAsAYODRC6+7nu/vHs/e+TPJHk2iRnktw72+zeJB84riIBAAAAAIDluqxnIFTV6SRvS/KjJFd393PJbsiQ5KpFFwcAAAAAAEzjkgOEqnptkm8n+WR3/+4y9jtbVdtVtb2zs3OUGgEAAAAAgCW7pAChql6V3fDga939ndnw81V1zWz9NUkuXGzf7r6ru7e6e2tzc3MRNQMAAAAAAMfs0AChqirJV5M80d2f37Pq/iS3zt7fmuS7iy8PAAAAAACYwsYlbHNTko8m+WlVPTob+3SSO5N8q6o+luSXST50PCUCAAAAAADLdmiA0N3/nKT2Wf3uxZYDAAAAAACcBJf8EGUAAAAAAOCVQ4AAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQDTVvRUAABRmSURBVIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAyqu5d3sKqdJL/YM3QqyW+WVgAAy6C3A6wXfR1g/ejtAOvncnv7n3X35mEbLTVAGA5etd3dW5MVAMDC6e0A60VfB1g/ejvA+jmu3u4WRgAAAAAAwECAAAAAAAAADKYOEO6a+PgALJ7eDrBe9HWA9aO3A6yfY+ntkz4DAQAAAAAAOJmmvgIBAAAAAAA4gQQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMJgkQquqWqvp5VT1VVXdMUQMA86uqZ6rqp1X1aFVtz8ZeX1UPVtWTs9crp64TgP1V1d1VdaGqHtszdtFeXrv+fjaP/0lVvX26ygHYzz69/bNV9evZ3P3Rqnr/nnV/N+vtP6+qv56magAOUlXXVdUPq+qJqnq8qj4xGz/WufvSA4SquiLJl5K8L8lbk3ykqt667DoAWJh3dfcN3b01W74jyUPdfX2Sh2bLAJxc9yS55WVj+/Xy9yW5fvZ1NsmXl1QjAJfnnoy9PUm+MJu739DdDyTJ7HcyH07yF7N9/mH2uxsATpYXknyqu9+S5J1Jbpv18GOdu09xBcKNSZ7q7qe7+w9JvpnkzAR1AHA8ziS5d/b+3iQfmLAWAA7R3Q8n+e3Lhvfr5WeS/FPv+tckf1pV1yynUgAu1T69fT9nknyzu/+7u/8jyVPZ/d0NACdIdz/X3T+evf99kieSXJtjnrtPESBcm+RXe5bPz8YAWD2d5AdVda6qzs7Gru7u55LdH25JrpqsOgCOar9ebi4PsNpun93G4u49txrV2wFWTFWdTvK2JD/KMc/dpwgQ6iJjvfQqAFiEm7r77dm9LO62qvqrqQsC4FiZywOsri8neXOSG5I8l+Rzs3G9HWCFVNVrk3w7ySe7+3cHbXqRscvu71MECOeTXLdn+Q1Jnp2gDgDm1N3Pzl4vJLkvu5c6P//SJXGz1wvTVQjAEe3Xy83lAVZUdz/f3S929x+TfCX/e5sivR1gRVTVq7IbHnytu78zGz7WufsUAcIjSa6vqjdV1auz+6Ce+yeoA4A5VNVrqup1L71P8t4kj2W3p9862+zWJN+dpkIA5rBfL78/yd/Urncm+c+XLpcG4GR72X2vP5jduXuy29s/XFV/UlVvyu7DNv9t2fUBcLCqqiRfTfJEd39+z6pjnbtvzFHzkXT3C1V1e5LvJ7kiyd3d/fiy6wBgblcnuW/351c2kny9u79XVY8k+VZVfSzJL5N8aMIaAThEVX0jyc1JTlXV+SSfSXJnLt7LH0jy/uw+YPO/kvzt0gsG4FD79Pabq+qG7N6+4pkkH0+S7n68qr6V5GdJXkhyW3e/OEXdABzopiQfTfLTqnp0NvbpHPPcvbrd1g4AAAAAAPi/priFEQAAAAAAcMIJEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYTBogVNXZKY8PwOLp7QDrRV8HWD96O8D6Oa7ePvUVCH5gAawfvR1gvejrAOtHbwdYP2sZIAAAAAAAACdQdffRd666JckXk1yR5B+7+86Dtj916lSfPn36f5Z3dnayubl55OMDcPLo7QDrRV8HWD96O8D6udzefu7cud9096E7bBy1oKq6IsmXkrwnyfkkj1TV/d39s/32OX36dLa3t496SAAAAAAAYE5V9YtL2W6eWxjdmOSp7n66u/+Q5JtJzszxeQAAAAAAwAkxT4BwbZJf7Vk+PxsDAAAAAABW3DwBQl1kbHigQlWdrartqtre2dmZ43AAAAAAAMCyzBMgnE9y3Z7lNyR59uUbdfdd3b3V3Vse0AMAAAAAAKthngDhkSTXV9WbqurVST6c5P7FlAUAAAAAAExp46g7dvcLVXV7ku8nuSLJ3d39+EH7nDt3LlUXu/MRwHJ1D3dc+z9WsVcd9Gc67M8zz77AK9Mq9pzjOu46/kxhtfmevHTOFezyb2H1reL/6dZtTriOnKvpHdafl+HIAUKSdPcDSR5YUC0AAAAAAMAJMc8tjAAAAAAAgDUlQAAAAAAAAAYCBAAAAAAAYCBAAAAAAAAABgIEAAAAAABgsLHMg73jHe/I9vb2Mg8JcCTdPXUJCzXPn2fdzgVw/Fax5xzncfVRThrfk5fOuYJd/i2stlX8+1vHOeG6ca5eOVyBAAAAAAAADAQIAAAAAADAQIAAAAAAAAAMBAgAAAAAAMBAgAAAAAAAAAwECAAAAAAAwECAAAAAAAAADKq7l3ewquUdDOAAh/W+qlpSJZdunn59Ev88vLKt4r9BVt9B33e+5wBg+cwJF2cV5zlT1byK5+qVZp6/o+P6Xfc83xvH+fucOT/7XHdvHbadKxAAAAAAAICBAAEAAAAAABgIEAAAAAAAgIEAAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYbExdAACLUVVTlwCXbKrv1+4+cP08dR322cd1XABgtRw0Z3ilzQleaX9e4NKsYm+Yp7cftO88/4ed5/+oe7kCAQAAAAAAGAgQAAAAAACAgQABAAAAAAAYCBAAgP/f3v2ESHLVcQD//WA8qYeEbEIQ/4DkoBcjOwRBD5GABC+Jhxw87UFYDwoKXoKXePSiHoUVQ3LQQEBDchAxBCF6EWZFdCVCQGLQLLsbcjDXJD8P24GNb6eqp7urqqv684FhZrq6u37v1avX1f2lugAAAAAaAgQAAAAAAKAhQAAAAAAAABoCBAAAAAAAoHG0zYMz87WIeDsi3o2Id6rqeBdFAdDKzI0fW1WDPfdQ+mruso/tYT8MOTaMO4D5meMxEvN3aOOqaz87tL5YItuXfdM17rb5nGGbY4YhH7tNm9a1VYCw8uWqenMHzwMAAAAAAOwJX2EEAAAAAAA0tg0QKiJ+l5mXM/PiLgoCAAAAAACmt+1XGH2xqt7IzLsj4sXM/EdVvXzrHVbBgnABAAAAAABmZKszEKrqjdXv6xHxXEQ8cJv7XKqqYxdYBgAAAACA+dg4QMjMD2fmR9//OyK+EhFXdlUYAAAAAAAwnW2+wuieiHguM99/nl9W1W93UhUAjarqXL6aj8+8bK6W2KZDss14BoBd8XoDw7OfLdtQ29f7BTbVN3aWtt4xbBwgVNU/I+JzO6wFAAAAAADYE1tdAwEAAAAAAFgmAQIAAAAAANAQIAAAAAAAAA0BAgAAAAAA0BAgAAAAAAAADQECAAAAAADQOJq6AJajqk5dlpkjVgLLdGj70aG1F4DpbHMc6xgYYHNdc2iEefSQ2fbzN9UxUtdz9805mz5vn23W22eo9t7KGQgAAAAAAEBDgAAAAAAAADQECAAAAAAAQEOAAAAAAAAANAQIAAAAAABAQ4AAAAAAAAA0jsZc2fnz5+Pk5GTMVbInqmrqEuBMjFmYln2QsRlznGabsWFcAWzHPLobc+zHqWqeY1/N0Rz7ecia9/140xkIAAAAAABAQ4AAAAAAAAA0BAgAAAAAAEBDgAAAAAAAADQECAAAAAAAQEOAAAAAAAAANAQIAAAAAABAI6tqvJVljrcygA59c19mjlTJ+uZY8zYOrb2HxvZlCl3jzpgDAOZsjsc5U9U8x75ifWN+1r0rXeNuyPZk5uWqOu67nzMQAAAAAACAhgABAAAAAABoCBAAAAAAAICGAAEAAAAAAGgIEAAAAAAAgIYAAQAAAAAAaBxNXQAAcHgyc+oS4KBVVedy+ygAAEyv77i877h+F3rPQMjMJzPzemZeueW2OzPzxcx8dfX7jmHLBAAAAAAAxrTOVxg9FREP/99tj0fES1V1X0S8tPofAAAAAABYiN4Aoapejoi3/u/mRyLi6dXfT0fEozuuCwAAAAAAmNCmF1G+p6quRkSsft+9u5IAAAAAAICpDX4R5cy8GBEXh14PAAAAAACwO5uegXAtM++NiFj9vn7aHavqUlUdV9XxhusCAAAAAABGtmmA8EJEXFj9fSEint9NOQAAAAAAwD7o/QqjzHwmIh6MiLsy898R8URE/DAins3Mb0TE6xHx2JBFAstVVZ3LM3OkStg3tj3AcMyxTKHruM+YZApTjUn7AsBwuubRvs+gNn3epesNEKrq66csemjHtQAAAAAAAHti068wAgAAAAAAFkyAAAAAAAAANAQIAAAAAABAQ4AAAAAAAAA0BAgAAAAAAEDjaOoCgOWrqqlLYIG6xlVmjlgJm+ibF2zDZbP/wmGyf7NvphqT9gWA4SztM6h9aI8zEAAAAAAAgIYAAQAAAAAAaAgQAAAAAACAhgABAAAAAABoCBAAAAAAAICGAAEAAAAAAGgIEAAAAAAAgMbRmCs7f/58nJyc3HZZZo5ZCmytqk5dZjx/kP7Yjb5+PLQxucQ2HRLb77DZ/uvbZm6f6rHATfYjmJZ9ELa3j/tRV00R89y/h+rnvr5alzMQAAAAAACAhgABAAAAAABoCBAAAAAAAICGAAEAAAAAAGgIEAAAAAAAgIYAAQAAAAAAaBxNXcD7qmrqEmBnjOf5W9o2XFp7AHZtjvPkNjVP9VjgJvsRTOvQ9sE5tneqmufYV1PZ177a17o2tQ/tcQYCAAAAAADQECAAAAAAAAANAQIAAAAAANAQIAAAAAAAAA0BAgAAAAAA0BAgAAAAAAAADQECAAAAAADQ6A0QMvPJzLyemVduue0HmfmfzPzL6uerw5YJAAAAAACMaZ0zEJ6KiIdvc/tPqur+1c9vdlsWAAAAAAAwpd4Aoapejoi3RqgFAAAAAADYE9tcA+HbmfnX1Vcc3bGzigAAAAAAgMltGiD8NCI+HRH3R8TViPjRaXfMzIuZeZKZJzdu3NhwdQAAAAAAwJg2ChCq6lpVvVtV70XEzyLigY77Xqqq46o6Pnfu3KZ1AgAAAAAAI9ooQMjMe2/592sRcWU35QAAAAAAAPsgq6r7DpnPRMSDEXFXRFyLiCdW/98fERURr0XEN6vqau/KMrtXBjCSNea+kSpZ3xxr3sahtRcYXte8Yk4BAOZsjsc5U9U8x75ifX2fJeyjrnE3ZHsy83JVHffeb8xOFSAA+2KOH07PseZtHFp7geF5swgALNUcj3MECAxBgHCm9a4VIGx6EWUAAAAAAGDBBAgAAAAAAEBDgAAAAAAAADQECAAAAAAAQEOAAAAAAAAANAQIAAAAAABA42jqAoB5qKpTl2XmiJUcrkPr50Nrb9c+FrG8/ji09gIAADBffe9Rl/y5mTMQAAAAAACAhgABAAAAAABoCBAAAAAAAICGAAEAAAAAAGgIEAAAAAAAgIYAAQAAAAAAaBxNXQAwD5k5dQkcmKrqXL60Mbm09nDYDm3/BYCz6Hqd9BoJsJ/63uNs89h9n/udgQAAAAAAADQECAAAAAAAQEOAAAAAAAAANAQIAAAAAABAQ4AAAAAAAAA0BAgAAAAAAEBDgAAAAAAAADSOpi4AgPVUVefyzBypEoBu5iMAOJ3XSQDmxBkIAAAAAABAQ4AAAAAAAAA0BAgAAAAAAEBDgAAAAAAAADQECAAAAAAAQEOAAAAAAAAANI6mLgAAODyZOXUJAAAADKSqTl3m/eC89J6BkJkfz8zfZ+Yrmfn3zPzO6vY7M/PFzHx19fuO4csFAAAAAADGsM5XGL0TEd+rqs9ExBci4luZ+dmIeDwiXqqq+yLipdX/AAAAAADAAvQGCFV1tar+vPr77Yh4JSI+FhGPRMTTq7s9HRGPDlUkAAAAAAAwrjNdRDkzPxURn4+IP0XEPVV1NeJmyBARd++6OAAAAAAAYBprX0Q5Mz8SEb+KiO9W1X/XvdhFZl6MiIublQcAAAAAAExhrTMQMvNDcTM8+EVV/Xp187XMvHe1/N6IuH67x1bVpao6rqrjXRQMAAAAAAAMrzdAyJunGvw8Il6pqh/fsuiFiLiw+vtCRDy/+/IAAAAAAIApZFV13yHzSxHxh4j4W0S8t7r5+3HzOgjPRsQnIuL1iHisqt7qea7ulQGMZI25b6RK5q+rL5fYj4fW3qHMcR+cY818kP2XIRhXwFmZNxjCHMfVVDXPsa9YX9/7tk1tMzaGfC+5TXsz8/I63xrUew2EqvpjRJzWiofOWhgAAAAAALD/1roGAgAAAAAAcFgECAAAAAAAQEOAAAAAAAAANAQIAAAAAABAQ4AAAAAAAAA0jqYuAID1VFXn8swcqZL9WO9UDq29Q5ljP86xZnanbw7elHE1f7bhbgx5nNP13LYfUzDugLPyWra+oY4Z9lVXe3fVHmcgAAAAAAAADQECAAAAAADQECAAAAAAAAANAQIAAAAAANAQIAAAAAAAAA0BAgAAAAAA0BAgAAAAAAAAjaOpCwCA26mqzuWZOVIlDMH2/aCu/ji0vthXtsOy2QenN2Q/24aHy/EGsBTmq/X1zf1DPW/XNtpm+w3VnrNwBgIAAAAAANAQIAAAAAAAAA0BAgAAAAAA0BAgAAAAAAAADQECAAAAAADQECAAAAAAAACNo6kLAGA9mTl1CbdVVacu26bmIds7VM1Drneqmocyx5qHpD9gvoaan7ued9vnhkNhP1mfOQe4nTm+D93XuubMGQgAAAAAAEBDgAAAAAAAADQECAAAAAAAQEOAAAAAAAAANAQIAAAAAABAQ4AAAAAAAAA0BAgAAAAAAEAjq2q8lWXeiIh/3XLTXRHx5mgFADAGczvAspjXAZbH3A6wPGed2z9ZVef67jRqgNCsPPOkqo4nKwCAnTO3AyyLeR1gecztAMsz1NzuK4wAAAAAAICGAAEAAAAAAGhMHSBcmnj9AOyeuR1gWczrAMtjbgdYnkHm9kmvgQAAAAAAAOynqc9AAAAAAAAA9pAAAQAAAAAAaAgQAAAAAACAhgABAAAAAABoCBAAAAAAAIDG/wAaDW4cJLdL0AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 503,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename='plots/1.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "可以看到一千次抽样的测试结果，大多数测试的PCC都比较高"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 504,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 504,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename='plots/2.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "按照样本查看每个样本多次抽样的平均PCC，注意这里是有bias没有消除的"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 505,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUYAAAJOCAYAAADLbK4AAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvAOZPmwAAIABJREFUeJzs3Xu4XFV9//H3h4SbckkgBwxJ4KAGBLSixoDFC4JyrRIrVKjVhAajFgvUG8H62NSnVvRnpVAQf1Gs4AVEEEgRf4hAalUuBrkIRCRCJMdEciAJELlY8Pv7Y60Dm5U558ycmTlzLp/X88wze6+99t7fPXvPd9a+jiICMzN7zmadDsDMbKRxYjQzKzgxmpkVnBjNzApOjGZmBSdGM7OCE2MbSFok6ZudjqMRkpZKOqHTcVj/JL1b0g87OP83SLqnU/MfTk6M1jBJvvgVkNQtKSRtzK+VkhZWhkvSSZLulPQHST2SvivpFZU6syVdJWmDpHWSbpZ0fK35RcS3IuKQIcba9I91RPxPROxZmeZKSW8p5rNY0j2S/iRpXjPz6yQnRhtTJE3swGwnRcQ2wHHApyQdlsvPBE4GTgJ2APYALgeOzLG+DrgO+G/gpcCOwAeBw4c1ep5N4q3IB7cDfwf8ogXT6pyIGNcvYBfgUqAXuB84qTJsNnADsAFYA5wNbFEZvg9wDbAOeBD4RC5fBFwMXAA8BtwFzOpn/gLOANYCjwB3AC/Pw44EbgUeBVYBiyrjdQMBHJ+HrQc+ALw2T2MDcHal/jzgp8B/5Pn8Cji4MnwpcEKl/2+B5Xm6VwO7VYZFMd378nLeD7y7n+Xs97MEvgx8oah/BfDhOtbRIuAS4Jv5czqhjvV2CHBP/hy+REpMdS17EWPfOphYKfs58FFgJvAMMHuAbe8nwDkNbKvzgJ9U10Ne5/fmWM8BVGO8w4A/Av8LbARur6zzz+Tt4glScj4+L/tjeb2+vzKdA4Ge3P0N4E95vI3Ax2ss27xOf7+HnBc6HUBHFz61mG8BPgVsAbw4bwyH5uGvAfYHJuYvwXLglDxs2/yl+wiwVe7fLw9bBDwJHAFMAD4L3NhPDIfmGCaRkuRewNTKhviKHOefkZLvnDys70v55Tz/Q/I8Lwd2AqaRku2bcv15wNPAPwCbA+8iJYYdKl+SE3L3HGBFjmUi8EngZzVifyEpGe2Z+6cC+/SznAN9lm8kJXfl/sn5C7dLHetoEekLPyfX3XqQeU3JMf9lHn5yHr+hZS/WwcS87g4AHgcOJiWs3w6w7b2AlDjf3MD2Oo9NE+OVedvZlfTDcVg/4y4CvlmULQUeIP3AT8zbxZHAS/LyvCkvz6sr22NPZfyVwFv6mZ8T42h9AfsBDxRlpwH/2U/9U4DLcvdxwK391FsE/KjSvzfwRD91DwJ+nb/Imw0S778DZ+Tuvi/ltMrwh4F3VfovrSSEecBqKi0K4GbgPbl7aSU5/ACYX6m3Wf6C7FbE80JSq+ydwNYNfvbVz1L5C/rG3P8+4Lp61lH+rH/cwLzeC9xQGSZSUm5o2Yt1sIHUYltObs0C/0g/P4Z5+LQ87ssa+MzmsWlifH2l/2Jg4QDbZK3E+OlB5nk5cHLuPpBxkhjH+zHG3YBd8oHvDZI2AJ8AdgaQtIekKyX9XtKjwL+SWhwAM4DfDDDt31e6Hwe2qnX8KyKuI+3qnQM8mA9eb5fnv5+k6yX1SnqE1AqZUkziwUr3EzX6t6n0/y7yVpv9ltQqK+0GnFn5TNaREsi0IvY/kFqeHwDWSPq+pJfVmN6An2WO6SLSjw3AXwPfqsTS7zrKVtU7r7y8z9bP8+5pdNkLUyJickTsFRFn5bKHSS3o/qwn7YoOVKce5Xa2TX8V+1F+dodLujGfCNpA2uspt7kxb7wnxlXA/RExqfLaNiKOyMPPJR2LmxkR25G+kKqM+5JWBBERZ0XEa0i7NHsAH8uDvg0sAWZExPak3WbVnkpdpkmqjr8rqRVZWkU6tlT9XLaOiJ/ViP3qiHgr6Qv+K+Ar/cx7oM8S4ELgaEm7kVqJl1ZiGWgdQWo51TuvNcD0vor585heGbfuZR/EtcB0SbNqDYyIx0nHQd/Z4HSHqr8rCZ4tl7Ql6XP/ArBzREwCrqL/bW7MXp0w3hPjzcCjkk6VtLWkCZJeLum1efi2pONRG3NL6IOVca8EXiTpFElbStpW0n6NBiDptblluDnwB9Jxwmcq818XEU9Kmk1qSTVjJ+AkSZtLOoZ0HO2qGvW+DJwmaZ8c4/a5fhn7zpLeLumFwFOkg/DPlPUqy9LfZ0lE3Eo6RvZV4OqI2JAHDbaOGp3X94FXSJqTW/AnAi9qdNkHExH3kk7sXCjpQElbSNpK0rGVS3o+DsyT9DFJO+b5vVLSRY3Orw4PAt2DnHneAtiStB6elnQ46dj1QNN8cbWgbzlJyXTzvMyjLs+MuoBbKSKeAd4G7Es62/kQ6Yu5fa7yUVIyeozUEvpOZdzHgLfm8X9POjP45iGEsV2e9nrSru3DpF9sSJc9fFrSY6STDxcPYfpVN5HOlj5EOht5dEQ8XFaKiMuAzwEX5V3RO6l9CclmpJNPq0m7nG/KMdfS72dZcSHwFlJLuS+WwdZRQ/OKiIeAY4DPkz7rvYFlpMTeyLLX4ySeO0yygXTo5R3Af+V5/Yx0jPkg4D5J64DF1P6xatZ38/vDkmpeSpO36ZNI29l60me4ZIBpfhb4ZD7s8NFc9kPSIZw/Jy3LE6STa6NK31lAG+PyxbYnRMTrOx3LSJJbMz2ky4yu73Q8NjKM6xajjU+SDpU0KR9T6zv+eGOHw7IRxInRxqPXkXZrHyLtps+JiCc6G5KNJN6VNjMruMVoZlboxA33m5gyZUp0d3d3OgwzG2NuueWWhyKiq9HxRkRi7O7uZtmyZZ0Ow8zGGEm/Hcp43pU2Mys4MZqZFZwYzcwKToxmZgUnRjOzghOjmVlhRFyuY2bP6V74/U3KVp5+ZAciGb/cYjQzKzgxmpkVnBjNzApOjGZmBSdGM7OCE6OZWcGJ0cys4MRoZlZwYjQzKzgxmpkVnBjNzAp1J0ZJEyTdKunK3L+7pJsk3SvpO5K2yOVb5v4VeXh3e0I3M2uPRlqMJwPLK/2fA86IiJnAemB+Lp8PrI+IlwJn5HpmZqNGXYlR0nTgSOCruV/AQcAlucr5wJzcfVTuJw8/ONc3MxsV6n3s2L8DHwe2zf07Ahsi4unc3wNMy93TgFUAEfG0pEdy/YeqE5S0AFgAsOuuuw41frNRy48XG7kGbTFK+gtgbUTcUi2uUTXqGPZcQcTiiJgVEbO6uhr+21czs7app8V4APB2SUcAWwHbkVqQkyRNzK3G6cDqXL8HmAH0SJoIbA+sa3nkZmZtMmiLMSJOi4jpEdENHAtcFxHvBq4Hjs7V5gJX5O4luZ88/LqI2KTFaGY2UjVzHeOpwIclrSAdQzwvl58H7JjLPwwsbC5EM7Ph1dB/vkTEUmBp7r4PmF2jzpPAMS2IzcysI3zni5lZwYnRzKzgxGhmVnBiNDMrODGamRWcGM3MCk6MZmYFJ0Yzs4ITo5lZwYnRzKzgxGhmVnBiNDMrODGamRUaerqOmXWG/wZheLnFaGZWcGI0Mys4MZqZFZwYzcwKToxmZgUnRjOzgi/XMRulfAlP+7jFaGZWcGI0Mys4MZqZFZwYzcwKgyZGSVtJulnS7ZLukvTPuXx3STdJulfSdyRtkcu3zP0r8vDu9i6CmVlr1dNifAo4KCJeCewLHCZpf+BzwBkRMRNYD8zP9ecD6yPipcAZuZ6Z2agx6OU6ERHAxty7eX4FcBDw17n8fGARcC5wVO4GuAQ4W5LydMxsFBmvlwTVdYxR0gRJtwFrgWuA3wAbIuLpXKUHmJa7pwGrAPLwR4Ada0xzgaRlkpb19vY2txRmZi1UV2KMiGciYl9gOjAb2KtWtfyuAYZVp7k4ImZFxKyurq564zUza7uG7nyJiA2SlgL7A5MkTcytwunA6lytB5gB9EiaCGwPrGtdyGajT61dUhu56jkr3SVpUu7eGngLsBy4Hjg6V5sLXJG7l+R+8vDrfHzRzEaTelqMU4HzJU0gJdKLI+JKSXcDF0n6F+BW4Lxc/zzgG5JWkFqKx7YhbjOztqnnrPQdwKtqlN9HOt5Ylj8JHNOS6MzMOsB3vpiZFZwYzcwKToxmZgUnRjOzghOjmVnBidHMrODEaGZWcGI0Mys4MZqZFZwYzcwKToxmZgUnRjOzghOjmVmhoQfVmo124/U/TKwxbjGamRWcGM3MCk6MZmYFJ0Yzs4ITo5lZwYnRzKzgxGhmVnBiNDMrODGamRWcGM3MCk6MZmaFQROjpBmSrpe0XNJdkk7O5TtIukbSvfl9ci6XpLMkrZB0h6RXt3shzMxaqZ4W49PARyJiL2B/4ERJewMLgWsjYiZwbe4HOByYmV8LgHNbHrWZWRsN+nSdiFgDrMndj0laDkwDjgIOzNXOB5YCp+byCyIigBslTZI0NU/HzNrITw9qjYaOMUrqBl4F3ATs3Jfs8vtOudo0YFVltJ5cVk5rgaRlkpb19vY2HrmZWZvUnRglbQNcCpwSEY8OVLVGWWxSELE4ImZFxKyurq56wzAza7u6EqOkzUlJ8VsR8b1c/KCkqXn4VGBtLu8BZlRGnw6sbk24ZmbtV89ZaQHnAcsj4ouVQUuAubl7LnBFpfy9+ez0/sAjPr5oZqNJPX9tcADwHuCXkm7LZZ8ATgculjQfeAA4Jg+7CjgCWAE8Dhzf0ojNRrhaJ0BsdKnnrPRPqH3cEODgGvUDOLHJuMxshBoPZ75954uZWcGJ0cys4MRoZlbw/0rbmOWTIDZUbjGamRXcYjQzwC3sKrcYzcwKToxmZgUnRjOzghOjmVnBidHMrOCz0jbujYd7f60xToxmTfAlLmOTd6XNzApuMdqY4JabtZJbjGZmBSdGM7OCE6OZWcGJ0cys4MRoZlZwYjQzKzgxmpkVnBjNzApOjGZmBSdGM7OCbwk0q8FP3BnfBm0xSvqapLWS7qyU7SDpGkn35vfJuVySzpK0QtIdkl7dzuDNzNqhnhbj14GzgQsqZQuBayPidEkLc/+pwOHAzPzaDzg3v5uNen5QxfgxaIsxIn4MrCuKjwLOz93nA3Mq5RdEciMwSdLUVgVrZjYchnryZeeIWAOQ33fK5dOAVZV6PblsE5IWSFomaVlvb+8QwzAza71Wn5VWjbKoVTEiFkfErIiY1dXV1eIwzMyGbqiJ8cG+XeT8vjaX9wAzKvWmA6uHHp6Z2fAb6uU6S4C5wOn5/YpK+YckXUQ66fJI3y63mVktI/HSqEETo6QLgQOBKZJ6gH8iJcSLJc0HHgCOydWvAo4AVgCPA8e3IWYzs7YaNDFGxHH9DDq4Rt0ATmw2KDOzTvItgWZmBSdGM7OCE6OZWcGJ0cys4KfrmI1xI/FymJHOidHMhs1oeRCHE6PZODRaElSn+BijmVnBLcYW8DEcs7HFiXEMcqI2a453pc3MCk6MZmYFJ0Yzs4KPMY4TY+m4oy81sXZzYmyT/r68rU5GThJmreddaTOzghOjmVnBu9IN8q6r2abG0jFscIvRzGwTbjHaiOYWunWCW4xmZgW3GEcRt57MhocT4zAbSQepR1IsZlXDdR1wf7wrbWZWcItxAN51NRu60fz9aVtilHQYcCYwAfhqRJzernmNdqN5A2olfw42UrQlMUqaAJwDvBXoAX4uaUlE3N2O+TXKX0AzG0i7WoyzgRURcR+ApIuAo4CWJcZ6Txw4CTZmOE7IeJ3YSKeIaP1EpaOBwyLihNz/HmC/iPhQpc4CYEHu3RO4p8akpgAPtTzA5o3EuBxT/UZiXI6pfo3EtVtEdDU6g3a1GFWj7HkZOCIWA4sHnIi0LCJmtTKwVhiJcTmm+o3EuBxT/YYjrnZdrtMDzKj0TwdWt2leZmYt1a7E+HNgpqTdJW0BHAssadO8zMxaqi270hHxtKQPAVeTLtf5WkTcNYRJDbir3UEjMS7HVL+RGJdjql/b42rLyRczs9HMtwSamRWcGM3MCh1LjJIOk3SPpBWSFtYYfoak2/Lr15I25PJ9Jd0g6S5Jd0h6V6djqgzfTtLvJJ09EmKStKukH0paLuluSd0jJK7P5/W3XNJZkmpd3tWOmHaVdL2kW/O2c0Rl2Gl5vHskHdqKeJqNS9JbJd0i6Zf5/aBOx1QM3yjpoyMhJkl/VskLv5S0VVPBRMSwv0gnZH4DvBjYArgd2HuA+n9POoEDsAcwM3fvAqwBJnUypkrZmcC3gbM7/Tnl/qXAW3P3NsALOh0X8OfAT/M0JgA3AAcOR0ykg/YfzN17Aysr3bcDWwK75+lMGK7PaoC4XgXskrtfDvyu0zFVhl8KfBf4aKdjIp1EvgN4Ze7fsdn116kW47O3DEbEH4G+Wwb7cxxwIUBE/Doi7s3dq4G1QMNXtrcyJgBJrwF2Bn7YgliajknS3sDEiLgGICI2RsTjnY6LdKH/VqSNf0tgc+DBYYopgO1y9/Y8d23tUcBFEfFURNwPrMjTa4UhxxURt+ZtHOAuYCtJW3YyJgBJc4D7ckyt0kxMhwB3RMTtABHxcEQ800wwnUqM04BVlf6eXLYJSbuRfsWvqzFsNukL9ptOxiRpM+DfgI+1II6WxERqWW+Q9L286/F/lB7u0dG4IuIG4HpSS38NcHVELB+mmBYBfyOpB7iK1JKtd9xOxFX1TuDWiHiqkzFJeiFwKvDPLYijJTGRtvWQdLWkX0j6eLPBdCoxDnrLYMWxwCXlL4CkqcA3gOMj4k8djunvgKsiYlU/9TsR00TgDcBHgdeSdlHmdTouSS8F9iLdDTUNOEjSG4cppuOAr0fEdOAI4Bv5R62R5RnOuNIEpH2AzwHvHwEx/TNwRkRsbFEsrYhpIvB64N35/R2SDm4mmE49qLaRWwaPBU6sFkjaDvg+8MmIuHEExPQ64A2S/o50LG8LSRsjYpMDyMMYUw+phdH3hKPLgf2B85qMqdm43gHc2PfFkvSDHNePhyGm+cBhkFqu+QD9lDrH7URcayVNBy4D3hsRrdgzajam/YCjJX0emAT8SdKTEdHsCcdm199/R8RDAJKuAl4NXDvkaFpx4HQIB1onko5R7M5zB1r3qVFvT2Al+UL0XLZFXuBTRkpMxfB5tO7kSzOf04Rcvyv3/ydw4giI613Aj/I0Ns/r8m3DERPwA2Be7t6L9MUTsA/PP/lyH607+dJMXJNy/XcO97beX0xFnUW07uRLM5/TZOAXwAvydH4EHNlUPK38wBv8II4Afk06PviPuezTwNuLD/70Yry/Af4XuK3y2reTMRXTmEeLEmOzMZEeFHwH8Evg68AWnY6LlLD/L7Cc9HzOLw5XTKQzmT/NX7rbgEMq4/5jHu8e4PDh3Nb7iwv4JPCHYlvfqdOfVbF+W5IYW7D+/oZ0MuhO4PPNxuJbAs3MCr7zpQ0kLZL0zU7H0QhJSyWd0Ok4rH+S3i2plZeDNTr/N0iq9UDpMceJ0RomybsZgKRuSZHvANkoaWX1jg0lJ0m6U9IfJPVI+q6kV1TqzJZ0laQNktZJulnS8bXmFxHfiohDhhhr0z/WEfE/EbFnZZorJb2l0r+HpCsk9eZluVrSnrWnNrI5MdqYIqkTV1pMiohtSJeTfErpHzIh3Ql1MnASsAPpervLgSNzrK8jXd/538BLSXdsfBA4fFij59kk3mw+mER67uqepJsdbgauaDa2jmjlQebR+CLdVngp0AvcD5xUGTabdMvaBtLFyGdTOYFBOpt5DbCOdPfGJyoHpS8GLgAeIx0UntXP/AWcQbqD5xHSyZKX52FHArcCj5Iufl1UGa+bdJ3X8XnYeuADpGsW78gxn12pP4904Po/8nx+BRxcGb4UOKHS/7ekEyTrSc/V3K0yLIrp3peX837g3f0sZ7+fJfBl4AtF/SuAD9exjhYBlwDfzJ/TCXWst0NIJ1keAb5ESkx1LXsRY986mFgp+znp2tGZwDPA7AG2vZ8A5zSwrc4DflJdD3md35tjPYcaV0uQLnH5I+mk5Ubg9so6/0zeLp4gJefj87I/ltfr+yvTORDoyd3fAP6Ux9sIfLzGfHfIMe7Y6e95w3mh0wF0dOFTi/kW4FOkSwRenDeGQ/Pw15CusZuYvwTLyZcJAdvmL91HSLe4bUv6w6++L+uTpLNsE4DPkq7dqxXDoTmGSaQkuRcwtbIhviLH+Wek5DsnD+v7Un45z/+QPM/LgZ1IF0+vBd6U688Dngb+gXSZzLtIiWGHypfkhNw9h3Rb3F552T8J/KxG7C8kJaM9c/9Ualy2U8dn+UZScu87GTg5f+F2qWMdLSJ94efkulsPMq8pOea/zMNPzuM3tOzFOpiY190BwOPAwaSE9dsBtr0XkBLnmxvYXuexaWK8Mm87u5J+OA7rZ9xFwDeLsqXAA6Qf+L7Lp44EXpKX5015eV5d2R57KuOvBN4yQLxzgDWd/p4PKTd0OoCOLny6WPWBouw04D/7qX8KcFnuPo50AXV/G+GPKv17A0/0U/cg0iUK+wObDRLvv5PuOqh+KadVhj8MvKvSf2klIcyjuBaNtKvznty9tJIcfgDMr9TbLH9BdivieSGpVfZOYOsGP/vqZ6n8BX1j7n8fcF096yh/1j9uYF7vBW6oDBMpKTe07MU62EBqsS0nt2ZJl//U/DHMw6flcV/WwGc2j00T4+sr/RcDCwfYJmslxk8PMs/LgZNz94HUmRhJF2j/Djiuke1ipLzG+zHG3YBd8oHvDUqPxvoE6fhI38HkKyX9XtKjwL+SWhyQrtIf6E6E31e6Hyc9AGCT418RcR1pV+8c4EFJi/OdPUjaLz9mqVfSI6RWyJRiEtUHMDxRo3+bSv/vIm+12W9JrbLSbsCZlc9kHSmBPO/e1Yj4A6nl+QFgjaTvS3pZjekN+FnmmC4i/dgA/DXwrUos/a6j7Hm3Yg6y3nap1s/z7ml02QtTImJyROwVEWflsodJLej+rCftig5Upx7ldrZNfxX7UX52h0u6MZ882UDa6ym3uQFJ6iI9TOVLEXHhYPVHovGeGFcB90fEpMpr24joe87buaRjcTMjYjvSF1KVcV/SiiAi4qyIeA1pl2YPnnsYxbdJB7NnRMT2pN3mZp5dOE163rMPd6X2rW+rSMeWqp/L1hHxsxqxXx0RbyV9wX8FfKWfeQ/0WUJ6+s7R+aET+5Fau32xDLSOYNN7agea1xpSawZIJx2q/Y0s+yCuBaZLqvk3n5GedHQDqbU9HPq7kuDZcqUn91wKfAHYOSImkR7W0N82t8k0JU0mJcUlEfGZpiLuoPGeGG8GHpV0qqStJU2Q9HJJr83DtyUdj9qYW0IfrIx7JfAiSadI2lLStpL2azQASa/NLcPNSXc5PEk69tQ3/3UR8aTSk4T+eojL2Wcn4CRJm0s6hnQc7aoa9b4MnJYfXoCk7XP9MvadJb1d6YkrT5EOwvf3uKeBPksi4lbSMbKvkp640/dg28HWUaPz+j7wCklzcgv+ROBFjS77YCI9Gu9LwIWSDpS0haStJB1buaTn48A8SR+TtGOe3yslXdTo/OrwINA9yJnnvkfB9QJPSzqcdOx6oGm+uK8n7+lcDfw0mn9OQEeN68QY6YkvbwP2JZ3tfIj0xdw+V/koKRk9RmoJfacy7mOkW+7eRtqduRd48xDC2C5Pez1p1/Zh0i82pKf2fFrSY6STDxcPYfpVN5HOlj5EOht5dEQ8XFaKiMtIT3O5KO+K3kntS0g2I518Wk3a5XxTjrmWfj/LiguBt5Bayn2xDLaOGppXpAcNHAN8nvRZ7w0sIyX2Rpa9Hifx3GGSDaRDL+8A/ivP62ekY8wHAfdJWkd6GGutH6tmfTe/PyzpF7Uq5G36JNJ2tp70GQ70t8efBT6ZDzt8lLRsrwWO13PXdm6UtGvLlmKY+JbAcULSPNIJhtd3OpaRJLegekiXGV3f6XhsZBjXLUYbnyQdKmlSPqbWd/yxVY+vszHAidHGo9eRdmsfIu2mz4mIJzobko0k3pU2Myu4xWhmVujUXxs8z5QpU6K7u7vTYZjZGHPLLbc8FBEN/4voiEiM3d3dLFu2rNNhmNkYI+m3QxnPu9JmZgUnRjOzghOjmVnBidHMrODEaGZWcGI0MyuMiMt1zGxg3Qu/v0nZytOP7EAk44NbjGZmBSdGM7OCE6OZWcGJ0cys4MRoZlZwYjQzKzgxmpkVnBjNzApOjGZmBSdGM7OCE6OZWaGuxChppaRfSrpN0rJctoOkayTdm98n53JJOkvSCkl3SHp1OxfAzKzVGmkxvjki9o2IWbl/IXBtRMwErs39AIcDM/NrAXBuq4I1MxsOzTxd5yjgwNx9PrAUODWXXxDpD6tvlDRJ0tSIWNNMoGb2fH7iTvvUmxgD+KGkAP5vRCwGdu5LdhGxRtJOue40YFVl3J5c9rzEKGkBqUXJrrvuOvQlMLMBOYE2rt7EeEBErM7J7xpJvxqgrmqUxSYFKbkuBpg1a9Ymw83MOqWuY4wRsTq/rwUuA2YDD0qaCpDf1+bqPcCMyujTgdWtCtjMrN0GTYySXihp275u4BDgTmAJMDdXmwtckbuXAO/NZ6f3Bx7x8UUzG03q2ZXeGbhMUl/9b0fE/5P0c+BiSfOBB4Bjcv2rgCOAFcDjwPEtj9rMrI0GTYwRcR/wyhrlDwMH1ygP4MSWRGdm1gG+88XMrODEaGZWcGI0Mys4MZqZFZwYzcwKToxmZoVmHiJhZiNMrfuirXFuMZqZFZwYzcwKToxmZgUnRjOzghOjmVnBidHMrODEaGZWcGI0Mys4MZqZFZwYzcwKToxmZoW6E6OkCZJulXRl7t9d0k2S7pX0HUlb5PItc/+KPLy7PaGbmbVHIy3Gk4Hllf7PAWdExExgPTA/l88H1kfES4Ezcj0zs1GjrsQoaTpwJPDV3C/gIOCSXOV8YE7uPir3k4cfnOubmY0K9bYY/x34OPCn3L8jsCEins79PcC03D0NWAWQhz+S6z+PpAWSlkla1tvbO8Twzcxab9DEKOkvgLURcUu1uEbVqGPYcwURiyNiVkTM6urqqitYM7PhUM+Dag8A3i56TQ/sAAAgAElEQVTpCGArYDtSC3KSpIm5VTgdWJ3r9wAzgB5JE4HtgXUtj9zMrE0GbTFGxGkRMT0iuoFjgesi4t3A9cDRudpc4IrcvST3k4dfFxGbtBjNzEaqZq5jPBX4sKQVpGOI5+Xy84Adc/mHgYXNhWhmNrwa+s+XiFgKLM3d9wGza9R5EjimBbGZmXWE73wxMys4MZqZFZwYzcwKToxmZgUnRjOzghOjmVnBidHMrODEaGZWaOgCbzMbG7oXfn+TspWnH9mBSEYmtxjNzApOjGZmBSdGM7OCE6OZWcGJ0cys4MRoZlZwYjQzKzgxmpkVfIG3mQG+6LvKLUYzs0I9/yu9laSbJd0u6S5J/5zLd5d0k6R7JX1H0ha5fMvcvyIP727vIpiZtVY9LcangIMi4pXAvsBhkvYHPgecEREzgfXA/Fx/PrA+Il4KnJHrmZmNGvX8r3RExMbcu3l+BXAQcEkuPx+Yk7uPyv3k4QdLUssiNjNrs7qOMUqaIOk2YC1wDfAbYENEPJ2r9ADTcvc0YBVAHv4I6X+ny2kukLRM0rLe3t7mlsLMrIXqSowR8UxE7AtMJ/2X9F61quX3Wq3D2KQgYnFEzIqIWV1dXfXGa2bWdg2dlY6IDcBSYH9gkqS+y32mA6tzdw8wAyAP3x5Y14pgzcyGQz1npbskTcrdWwNvAZYD1wNH52pzgSty95LcTx5+XURs0mI0Mxup6rnAeypwvqQJpER6cURcKelu4CJJ/wLcCpyX658HfEPSClJL8dg2xG1m1jaDJsaIuAN4VY3y+0jHG8vyJ4FjWhKdmVkH+M4XM7OCE6OZWcGJ0cys4MRoZlbwY8fMrF/j9VFkbjGamRWcGM3MCt6VNhthau2+2vByi9HMrODEaGZWcGI0Mys4MZqZFZwYzcwKToxmZgUnRjOzghOjmVnBidHMrODEaGZWqOfPsGZIul7Sckl3STo5l+8g6RpJ9+b3yblcks6StELSHZJe3e6FMDNrpXpajE8DH4mIvUh/m3qipL2BhcC1ETETuDb3AxwOzMyvBcC5LY/azKyN6vkzrDXAmtz9mKTlwDTgKODAXO180v9Nn5rLL8h/mXqjpEmSpubpmFnmh0WMXA0dY5TUTfrHwJuAnfuSXX7fKVebBqyqjNaTy8ppLZC0TNKy3t7exiM3M2uTuhOjpG2AS4FTIuLRgarWKItNCiIWR8SsiJjV1dVVbxhmZm1XV2KUtDkpKX4rIr6Xix+UNDUPnwqszeU9wIzK6NOB1a0J18ys/eo5Ky3gPGB5RHyxMmgJMDd3zwWuqJS/N5+d3h94xMcXzWw0qecJ3gcA7wF+Kem2XPYJ4HTgYknzgQeAY/Kwq4AjgBXA48DxLY3YzKzN6jkr/RNqHzcEOLhG/QBObDIuM7OO8Z0vZmYFJ0Yzs4ITo5lZwYnRzKzgxGhmVnBiNDMr1HMdo9mYUevBDStPP7IDkdhI5hajmVnBLUazJrgFOja5xWhmVnCL0WwY+KG0o4tbjGZmBSdGM7OCE6OZWcHHGM3q5OOE44dbjGZmBSdGM7OCE6OZWcGJ0cysUM+/BH5N0lpJd1bKdpB0jaR78/vkXC5JZ0laIekOSa9uZ/BmZu1QT4vx68BhRdlC4NqImAlcm/sBDgdm5tcC4NzWhGlmNnwGTYwR8WNgXVF8FHB+7j4fmFMpvyCSG4FJkqa2Klgzs+Ew1OsYd46INQARsUbSTrl8GrCqUq8nl60pJyBpAalVya677jrEMMya5yfkWKnVF3jX+v/pqFUxIhYDiwFmzZpVs46ZjX0j8YdpqInxQUlTc2txKrA2l/cAMyr1pgOrmwnQbKh8p4oN1VAT4xJgLnB6fr+iUv4hSRcB+wGP9O1ym40XTsij36CJUdKFwIHAFEk9wD+REuLFkuYDDwDH5OpXAUcAK4DHgePbELOZWVsNmhgj4rh+Bh1co24AJzYblJlZJ/nOFzOzghOjmVnBz2M0s2EzWk5MucVoZlZwi9GshtHSsrH2cIvRzKzgFuMARuKtSmbWfk6MDXKyNBv7vCttZlZwYjQzKzgxmpkVfIwx8+UZZvUZD8fZ3WI0Mys4MZqZFZwYzcwKPsbYAuPhmIvZeOLEaGOCT55ZKzkxmllbjOYfq7YlRkmHAWcCE4CvRsTp7ZpXo4ZjhfU3D+9imw2u09+ftiRGSROAc4C3kv5S9eeSlkTE3e2Yn40vo7klYqNDu1qMs4EVEXEfQP471aOAYU+M/hKZtd9Y+561KzFOA1ZV+ntI/zP9LEkLgAW5d6Oke9oUy3CZAjw0WCV9bhgiaY+6lm+UG+vLOOqXr47vT7mMuw1lPu1KjKpRFs/riVgMLG7T/IedpGURMavTcbTLWF8+GPvLONaXD1q3jO26wLsHmFHpnw6sbtO8zMxaql2J8efATEm7S9oCOBZY0qZ5mZm1VFt2pSPiaUkfAq4mXa7ztYi4qx3zGkHGzGGBfoz15YOxv4xjffmgRcuoiBi8lpnZOOKHSJiZFZwYzcwKToyDkHSYpHskrZC0sJ86fyXpbkl3Sfp2pfwZSbfl14g9+TTYMko6o7Icv5a0oTJsrqR782vu8EZenyaXb6ysw10lXS/pVkl3SDqiMuy0PN49kg4d3sjrM9Tlk9Qt6YnKOvxyXTOMCL/6eZFOHP0GeDGwBXA7sHdRZyZwKzA59+9UGbax08vQimUs6v896WQawA7Affl9cu6e3OllatXyjaV1SDop8cHcvTewstJ9O7AlsHuezoROL1MLl68buLPRebrFOLBnb22MiD8Cfbc2Vr0POCci1gNExNphjrFZ9Sxj1XHAhbn7UOCaiFiXl/8a4LC2Rtu4ZpZvtKhnGQPYLndvz3PXFR8FXBQRT0XE/cCKPL2RpJnlGxInxoHVurVxWlFnD2APST+VdGN+qlCfrSQty+Vz2h3sENWzjABI2o3Uqriu0XE7qJnlg7GzDhcBfyOpB7iK1DKud9xOa2b5AHbPu9j/LekN9czQz2Mc2KC3NpI+w5nAgaQ7fP5H0ssjYgOwa0SslvRi4DpJv4yI37Q14sbVs4x9jgUuiYhnhjBupzSzfDB21uFxwNcj4t8kvQ74hqSX1zlupzWzfGtI6/BhSa8BLpe0T0Q8OtAM3WIcWD23NvYAV0TE/+ZdkXtIiZKIWJ3f7wOWAq9qd8BD0Mjtm8fy/N3M0XDrZzPLN5bW4XzgYoCIuAHYivTAhbGyDmsuXz5E8HAuv4V0rHKPQefY6QOrI/lFag3eR9q96jvou09R5zDg/Nw9hdTk35F0MmLLSvm9DHDQfyQvY663J7CSfFNALtsBuD8v6+TcvUOnl6mFyzdm1iHwA2Be7t4rJxYB+/D8ky/3MfJOvjSzfF19y0M6efO7erbRji/0SH8BRwC/Jv3S/GMu+zTw9twt4IukZ03+Ejg2l/957r89v8/v9LIMdRlz/yLg9Brj/i3pgP0K4PhOL0srl28srUPSmdqf5mW5DTikMu4/5vHuAQ7v9LK0cvmAdwJ35fJfAG+rZ36+JdDMrOBjjG0gaZGkb3Y6jkZIWirphE7HYf2T9G5JP+zg/N8wBh4oXRcnRmuYJO9m8OxdFSFpY36trN6VoeQkSXdK+oOkHknflfSKSp3Zkq6StEHSOkk3Szq+1vwi4lsRccgQY236xzoi/ici9qxMc6Wkt1T6p+TL1h7Oy3ODpAOamWenODHamCKpE5egTYqIbUiXjHyqci3rmcDJwEmkE1V7AJcDR+ZYX0e6ZvK/gZeSTtp9EDh8WKPn2STebD7YSDrm3EU6cfU54L86tE6a0+mDqp1+AbsAlwK9pLOqJ1WGzQZuADaQroc6G9iiMnwf0t0e64AHgU/k8kWkSwcuAB4jHfyd1c/8BZwBrAUeAe4AXp6HHUm63fBR0tnuRZXxuknXch2fh60HPgC8Nk9jA3B2pf480sHp/8jz+RVwcGX4UuCESv/fAsvzdK8GdqsMi2K69+XlvB94dz/L2e9nCXwZ+EJR/wrgw3Wso0XAJcA38+d0Qh3r7RDSiYZHgC+RElNdy17E2LcOJlbKfg58lHTJ1jPA7AG2vZ+Q7pqqd1udB/ykuh7yOr83x3oOlbPqlXqHAX8E/peUvG6vrPPP5O3iCVJyPj4v+2N5vb6/Mp0DgZ7c/Q3gT3m8jcDHi3luBrwtx7hTvcs4Ul4dD6CjC59W3i3Ap0iXAbw4bwyH5uGvAfYnXS7QnTeYU/KwbfOX7iOka6a2BfbLwxYBT5LOpE0APgvc2E8Mh+YYJpGS5F7A1MqG+Ioc55+Rku+cPKzvS/nlPP9D8jwvB3Yi3RmwFnhTrj8PeBr4B2Bz4F2kxLBD5UtyQu6eQzrLvFde9k8CP6sR+wtJyWjP3D+VGpfC1PFZvpGU3PtOBk7OX7hd6lhHi0hf+Dm57taDzGtKjvkv8/CT8/gNLXuxDibmdXcA8DhwMClh/XaAbe8FpMT55ga213lsmhivzNvOrqQfjsP6GXcR8M2ibCnwAOkHfmLeLo4EXpKX5015eV5d2R57KuOvBN5SY153kBJxAF/p9Pd8SLmh0wF0dOHTPxc+UJSdBvxnP/VPAS7L3ccBt/ZTbxHwo0r/3sAT/dQ9iHQZwv7AZoPE++/AGbm770s5rTL8YeBdlf5LKwlhHvnarsrwm4H35O6lleTwAyqXppASzuMULSdSYtxAuiRi6wY/++pnqfwFfWPufx9wXT3rKH/WP25gXu8FbqgMEykpN7TsxTrYQGqxLSe3ZkmXwNT8MczDp+VxX9bAZzaPTRPj6yv9FwMLB9gmayXGTw8yz8uBk3P3gdSRGPOwrUjfkbmNbBcj5TXejzHuBuySDxRvyI+b+gSwM4CkPSRdKen3kh4F/pXU4oB0Jf5At4b9vtL9OOme202OtUTEdaRdvXOAByUtlrRdnv9++VFKvZIeIbVCphSTeLDS/USN/m0q/b+LvNVmvyW1ykq7AWdWPpN1pATyvPtTI+IPpJbnB4A1kr4v6WU1pjfgZ5ljuoj0RQL4a+BblVj6XUdZ9T7awdbbLtX6ed49jS57YUpETI6IvSLirFz2MKkF3Z/1pF3RgerUo9zOtumvYj/Kz+7wfF/4urz8R7DpNjeoiHgyIi4EFkp6ZaPjd9p4T4yrgPsjYlLltW1E9D2r7lzSsbiZEbEd6QupyrgvaUUQEXFWRLyGtEuzB/CxPOjbpD8RmxER25N2m2vdN1qvaZKq4+9K7du/VpGOLVU/l60j4mc1Yr86It5K+oL/CvhKP/Me6LOEdCve0flBDvuRWrt9sQy0jmDT+2YHmtca0i1lQDrpUO1vZNkHcS0wXVLNv/KMiMdJx0Hf2eB0h6q/KwmeLZe0Jelz/wKwc0RMIj2Qob9trp6rEzYnHf4YVcZ7YrwZeFTSqZK2ljRB0sslvTYP35Z0PGpjbgl9sDLulcCLJJ0iaUtJ20rar9EAJL02tww3B/5AOk7Y9xCDbYF1EfGkpNmkllQzdgJOkrS5pGNIx9GuqlHvy8BpkvbJMW6f65ex7yzp7ZJeCDxFOgj/TFmvsiz9fZZExK2kY2RfBa6O9BAOGHwdNTqv7wOvkDQnt+BPBF7U6LIPJiLuJZ3YuVDSgZK2kLSVpGMrl/R8HJgn6WOSdszze6WkixqdXx0eBLoHOfO8BenWwF7gaUmHk45dDzTNZ5OepP0lvT4v69aSTiW17G9qPvzhNa4TY6SnqLwN2Jd0tvMh0hdz+1zlo6Rk9BipJfSdyriPAW/N4/+edGbwzUMIY7s87fWkXduHSb/YAH8HfFrSY6STDxcPYfpVN5HOlj5EOht5dOQb7Ksi4jLSpRYX5V3RO6l9CclmpJNPq0m7nG/KMdfS72dZcSHwFlJLuS+WwdZRQ/OKiIeAY4DPkz7rvYFlpMTeyLLX4ySeO0yygXTo5R3Af+V5/Yx0jPkg4D5J60gPXK31Y9Ws7+b3hyX9olaFvE2fRNrO1pM+w4GeWv5Z4JP5sMNHSUn1HNLn+jvSbviRkR/EMZr4lsBxQtI80gmG13c6lpEkt6B6SJcZXd/peGxkGNctRhufJB0qaVI+ptZ3/PHGDodlI4gTo41HryPt1j5E2k2fExFPdDYkG0m8K21mVnCL0cysMOjN3ZK2An5MOuM0kfSfGP8kaXfSRbk7kB4A+Z6I+GM+bnMB6basvjsxVg40jylTpkR3d3czy2FmtolbbrnloYjoanS8ep568RRwUERszNfa/UTSD4APk25Pu0jpT6znky6snQ+sj4iXSjqWdOnDuwaaQXd3N8uWLWs0djOzAUn67VDGG3RXOpKNuXfz/ArStVeX5PLzSTffQ/q/1/Nz9yXAwcXdFmZmI1pdxxjz3Qa3kZ7Wcg3pjN6GiHg6V6n+z+uz/wGbhz9Ces5cOc0FSv/Xu6y3t7e5pTAza6G6EmNEPBMR+5LuKZ1NupVsk2r5va7/qY2IxRExKyJmdXU1fAjAzKxtGjorne9fXUp6RNakytNiqv/z+ux/wObh25NuFzMzGxUGTYySuiRNyt1bk+5lXQ5cDxydq80lPXEZ0r2Vc3P30aTn6vliSTMbNeo5Kz0VOF/SBFIivTgirpR0N+lG+38hPX7/vFz/POAbklaQWorHtiFuM7O2GTQxRsQdwKtqlN9HOt5Ylj9JenqJWUO6F35/k7KVpx/ZgUhsvPOdL2ZmBSdGM7OCE6OZWcGJ0cysUM9ZabMxzSd9rOQWo5lZwYnRzKzgxGhmVnBiNDMrODGamRWcGM3MCk6MZmYFJ0Yzs4Iv8LZxpdbF3GYltxjNzApuMdqwa6TV5tv1rBPcYjQzK9Tzny8zJF0vabmkuySdnMsXSfqdpNvy64jKOKdJWiHpHkmHtnMBzMxarZ5d6aeBj0TELyRtC9wi6Zo87IyI+EK1sqS9Sf/zsg+wC/AjSXtExDOtDNysnbwLP74N2mKMiDUR8Yvc/RjpHwKnDTDKUcBFEfFURNwPrKDGf8OYmY1UDR1jlNRN+mOsm3LRhyTdIelrkibnsmnAqspoPdRIpJIWSFomaVlvb2/DgZuZtUvdiVHSNsClwCkR8ShwLvASYF9gDfBvfVVrjL7J/0pHxOKImBURs7q6uhoO3MysXepKjJI2JyXFb0XE9wAi4sGIeCYi/gR8hed2l3uAGZXRpwOrWxeymVl71XNWWsB5wPKI+GKlfGql2juAO3P3EuBYSVtK2h2YCdzcupDNzNqrnrPSBwDvAX4p6bZc9gngOEn7knaTVwLvB4iIuyRdDNxNOqN9os9Im9loMmhijIifUPu44VUDjPMZ4DNNxGXWEF9eY63kWwJt1Kn3lkI/MMKGyrcEmpkVnBjNzApOjGZmBSdGM7OCE6OZWcGJ0cys4MRoZlZwYjQzKzgxmpkVnBjNzApOjGZmBSdGM7OCE6OZWcGJ0cys4MRoZlZwYjQzKwz6oFpJM4ALgBcBfwIWR8SZknYAvgN0k/7a4K8iYn3+j5gzgSOAx4F5ff9LbeOPHxZro1E9LcangY9ExF7A/sCJkvYGFgLXRsRM4NrcD3A46Q+wZgILSH+zamY2agyaGCNiTV+LLyIeA5YD04CjgPNztfOBObn7KOCCSG4EJhX/KGhmNqI1dIxRUjfwKuAmYOeIWAMpeQI75WrTgFWV0XpyWTmtBZKWSVrW29vbeORmZm1Sd2KUtA1wKXBKRDw6UNUaZbFJQcTiiJgVEbO6urrqDcPMrO3qSoySNiclxW9FxPdy8YN9u8j5fW0u7wFmVEafDqxuTbhmZu03aGLMZ5nPA5ZHxBcrg5YAc3P3XOCKSvl7lewPPNK3y21mNhrU87/SBwDvAX4p6bZc9gngdOBiSfOBB4Bj8rCrSJfqrCBdrnN8SyM2M2uzQRNjRPyE2scNAQ6uUT+AE5uMy8ysY3zni5lZoZ5daTOj9l08K08/sgORWLu5xWhmVnBiNDMrODGamRWcGM3MCk6MZmYFJ0Yzs4ITo5lZwYnRzKzgxGhmVnBiNDMrODGamRWcGM3MCk6MZmYFJ0Yzs4ITo5lZYdDnMUr6GvAXwNqIeHkuWwS8D+j739NPRMRVedhpwHzgGeCkiLi6DXHbAPzcQLPm1NNi/DpwWI3yMyJi3/zqS4p7A8cC++RxviRpQquCNTMbDoMmxoj4MbCuzukdBVwUEU9FxP2kP8Sa3UR8ZmbDrpljjB+SdIekr0manMumAasqdXpy2SYkLZC0TNKy3t7eWlXMzDpiqInxXOAlwL7AGuDfcnmtfxOMWhOIiMURMSsiZnV1dQ0xDDOz1htSYoyIByPimYj4E/AVnttd7gFmVKpOB1Y3F6KZ2fAaUmKUNLXS+w7gzty9BDhW0paSdgdmAjc3F6KZ2fCq53KdC4EDgSmSeoB/Ag6UtC9pN3kl8H6AiLhL0sXA3cDTwIkR8Ux7QrdG+BIes/oNmhgj4rgaxecNUP8zwGeaCcrMrJMGTYw2drkVaVabbwk0Mys4MZqZFZwYzcwKToxmZgUnRjOzghOjmVnBl+uMcrUuuTGz5rjFaGZWcGI0Mys4MZqZFZwYzcwKPvli1gTfbz42ucVoZlZwi9FaxpcO2VjhFqOZWcGJ0cysMGhizH+PulbSnZWyHSRdI+ne/D45l0vSWZJW5L9WfXU7gzcza4d6WoxfBw4ryhYC10bETODa3A9wOOkPsGYCC0h/s2pmNqoMmhgj4sfAuqL4KOD83H0+MKdSfkEkNwKTin8UNDMb8YZ6jHHniFgDkN93yuXTgFWVej25bBOSFkhaJmlZb2/vEMMwM2u9Vp98UY2yqFUxIhZHxKyImNXV1dXiMMzMhm6oifHBvl3k/L42l/cAMyr1pgOrhx6emdnwG2piXALMzd1zgSsq5e/NZ6f3Bx7p2+U2MxstBr3zRdKFwIHAFEk9wD8BpwMXS5oPPAAck6tfBRwBrAAeB45vQ8xmZm01aGKMiOP6GXRwjboBnNhsUGZmneR7pdukv/uG/eQVs5HPtwSamRWcGM3MCk6MZmYFJ0Yzs4ITo5lZwWelW8BPrjYbW5wYh5n/PMls5HNiNBsF/IM6vJwYzUaYsXJoZjQncydGG9Ro3sA7wZ/X6OfEaM8zVlorZs3w5TpmZgUnRjOzghOjmVnBidHMrOCTL2Yd4hNdI5cTow2Jv9Q2ljWVGCWtBB4DngGejohZknYAvgN0AyuBv4qI9c2FaWY2fFrRYnxzRDxU6V8IXBsRp0tamPtPbcF8xj230syGRzt2pY8i/asgwPnAUpwYbZzzj9ro0uxZ6QB+KOkWSQty2c59/yWd33eqNaKkBZKWSVrW29vbZBhmZq3TbIvxgIhYLWkn4BpJv6p3xIhYDCwGmDVrVjQZx7DxL7/Z0I2W+8ibajFGxOr8vha4DJgNPChpKkB+X9tskGZmw2nILUZJLwQ2i4jHcvchwKeBJcBc4PT8fkUrAjWz5xstra/RqJld6Z2ByyT1TefbEfH/JP0cuFjSfOAB4JjmwxzbvIGbjSxDTowRcR/wyhrlDwMHNxOUmVkn+V5pM7OCbwkcoXz220aTsba9usVoZlZwYjQzKzgxmpkVnBjNzApOjGZmBSdGM7OCL9cZwFi7BMHM6uMWo5lZwYnRzKzgXWmzMWQ0PpBkJMbsFqOZWcGJ0cys4F3pzGegzayPE6OZNWQ8NCK8K21mVmhbi1HSYcCZwATgqxFxervmZWb9G4lnfUe6tiRGSROAc4C3Aj3AzyUtiYi72zG/Ro2HXQGzRvl78Zx2tRhnAyvy/8Ig6SLgKKBlibHeX0GvbLNNjfTvRX/xDVdLt12JcRqwqtLfA+xXrSBpAbAg926UdE+N6UwBHqp3pvpcg1EOTUMxDZORFtNIiwccU71GWkzPi2cI3/HdhjLTdiVG1SiL5/VELAYWDzgRaVlEzGplYM1yTIMbafGAY6rXSIupU/G066x0DzCj0j8dWN2meZmZtVS7EuPPgZmSdpe0BXAssKRN8zIza6m27EpHxNOSPgRcTbpc52sRcdcQJjXgrnaHOKbBjbR4wDHVa6TF1JF4FBGD1zIzG0d854uZWcGJ0cys0LHEKOkwSfdIWiFpYY3h8yT1Srotv06oDPu8pLskLZd0lqRalwe1PKZc568k3Z3n/+1K+VxJ9+bX3E7GI2lfSTfksjskvasV8TQTU2XYdpJ+J+nskRCTpF0l/TBvS3dL6h4BMbV8+67j+3ZG5bv2a0kbKsNavm03E1M7t+9nRcSwv0gnZH4DvBjYArgd2LuoMw84u8a4fw78NE9jAnADcOAwxTQTuBWYnPt3yu87APfl98m5e3IH49kDmJm7dwHWAJM6+RlVhp8JfLvWuu1ETMBS4K25exvgBR3ellq+fdcTT1H/70knTNuybbcgprZs39VXp1qMz94yGBF/BPpuGaxHAFuRPswtgc2BB4cppvcB50TEeoCIWJvLDwWuiYh1edg1wGGdiicifh0R9+bu1cBaoKvJeJqKCUDSa4CdgR+2IJamY5K0NzAxIq7J5Rsj4vFOxkR7tu9Gv2/HARfm7nZs203F1Mbt+1mdSoy1bhmcVqPeO3NT+RJJMwAi4gbgetKvxBrg6ohYPkwx7QHsIemnkm5UeoJQveMOZzzPkjSb9CX7TZPxNBWTpM2AfwM+1oI4WhJTLt8g6XuSbpX0f5QegNKxmNq0fde9fUraDdgduK7RcYcxpuqwVm7fz+rUg2oHvWUQ+C/gwoh4Sv+/nfOHsSmIwvjvZNeWEqxiewlC4U9DQUQiEQmJyiuEpZJQ6ClEQyHbiwgblURBgQKR0GxHiAgRu6HWaWRxFHOefSab7Hvu3DvN90smmTtvZu6XmS9z75yXuWZngFlgv5ltADaTTtMAPDGzve7+ogNN46Qt0L64/0sz2zpk2870uHs/FjMF3AFOuvvvhnoaaQKOA4/c/WuhkHAJTePAHmA78AW4Swrh3KyoaZLy/h7Fnz3gnrv/+o+2XQkGQ/sAAAF9SURBVGlKHZT3919qvTGueGTQ3b+5+4+4vAHsjPxRYC62Pd+Bx8CuLjRFnQfuvuju88AHkrnbOALZRA9mthp4CFx097mGWkpo2g2cM7MF4BpwwsxKfKOz6by9iu3cT+A+sKOypjb8PYo/eyxto0dt25Wmtvy9RMmA5bCJ9LT8THo97gdet2R1pgbyfbMAHAOeRh+rgGfA4Y40HQRmIz9J2gqsIwWm50nB6TWRX1tRz0SMy/kK87aspqzONOX+fGkyTmNRf338dgs4W1lTcX8PoyfqbQQWiIMfUVbc2wU0teLvf+7bVsdDDMwh4CMpNnAhyi4DRyJ/BXgXA/Yc2BTlY8B14D3p+44zHWoyYCbu+xboDbQ9DXyKdKqmHtK2dRF4PZC21R6jgT6mKbQwFpi3A8CbKL8NTFSeu1b8vZKeuL4EXF2mbXFvN9HUpr/7SUcChRAiQydfhBAiQwujEEJkaGEUQogMLYxCCJGhhVEIITK0MAohRIYWRiGEyPgDO5D5zNEDJmwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 505,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename='plots/3.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "这里绘制了每个样本的平均PCC heatmap图像 <br>\n",
    "可以发现大多数的样本是很好预测的，样本性状基本由父本性状主导（纵坐标为父本） <br>\n",
    "但是少数亲缘关系较远的父本（图中蓝色线）就很难预测。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 506,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 506,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename='plots/4.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 507,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 507,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename='plots/5.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 508,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 508,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename='plots/6.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "不同父本的性状有显著差别，而子代的性状由于设计原因，主要由父本控制。<br>\n",
    "我们可以通过绘图查看不同父本的性状的变化"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 495,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,u'different males have varied values')"
      ]
     },
     "execution_count": 495,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2aaf601147d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "male_index = np.ndarray([6210,]).astype('int')\n",
    "for i in range(6210):\n",
    "    male_index[i] = int(np.array(phenotypes['pedigree'])[i].split('_')[2][1:])\n",
    "male_trait1 = np.concatenate((male_index.reshape(1,-1),np.array(pheno_whole['trait1']).reshape(1,-1))).T\n",
    "male_trait1_bysort = male_trait1[male_trait1[:,0].argsort()]\n",
    "fig, ax = plt.subplots(figsize=(16,8))\n",
    "ax.plot(male_trait1_bysort[:,1])\n",
    "ax.set_title('different males have varied values')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "以上内容简要地介绍了eMaize问题使用的一些基本的常用的机器学习方法，由于实际工作的步骤、数据量、变量等问题，还需要慎重考虑计算时间、任务管理等工作 <br>\n",
    "想要预测较难预测的样本仅仅靠常规的机器学习方法并不够用，将机器学习应用于生物学问题时，不能简单套用模型，还需要根据问题进行针对性的设计，才有可能取得更好的结果。"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.14"
  },
  "nbpresent": {
   "slides": {
    "05f815a7-3844-479b-aec7-dfa667a64b1f": {
     "id": "05f815a7-3844-479b-aec7-dfa667a64b1f",
     "prev": "689a5d0c-5693-4013-b61b-d45a2e2f71b7",
     "regions": {
      "c500a5d6-3a04-4065-b946-3fdaad82545d": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "58f3ed60-5c2f-451f-bb4d-56a5e7fbc22f",
        "part": "whole"
       },
       "id": "c500a5d6-3a04-4065-b946-3fdaad82545d"
      }
     }
    },
    "08e24f22-16dd-4ac6-ba88-ae953cc0a1d7": {
     "id": "08e24f22-16dd-4ac6-ba88-ae953cc0a1d7",
     "prev": "42eea50e-aea9-41b2-8258-5ad236ee3a9b",
     "regions": {
      "0279fe89-f2dc-4000-989e-2f5d7659e877": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "9bc7e0e0-6a5d-4653-814e-37d611583e39",
        "part": "whole"
       },
       "id": "0279fe89-f2dc-4000-989e-2f5d7659e877"
      }
     }
    },
    "0ffb0625-9562-4680-876c-c85b0b9afe10": {
     "id": "0ffb0625-9562-4680-876c-c85b0b9afe10",
     "prev": "b8e8d50b-67f3-498e-b8f1-1a907707bd46",
     "regions": {
      "67f059a3-46ae-4397-9336-2babe32a1a37": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "fefb2ce9-2b13-4011-ba6a-7fe7df7ab617",
        "part": "whole"
       },
       "id": "67f059a3-46ae-4397-9336-2babe32a1a37"
      }
     }
    },
    "238c9334-cf28-4ed4-ade3-dca41af0bd35": {
     "id": "238c9334-cf28-4ed4-ade3-dca41af0bd35",
     "prev": "f8b02102-7ba4-4d1b-a453-01b1b94ae0de",
     "regions": {
      "dc74a3b6-b9dc-4a28-9567-73e0c0e5dccd": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "d9be0ffd-4713-4a03-97f5-4bd7f1673c5b",
        "part": "whole"
       },
       "id": "dc74a3b6-b9dc-4a28-9567-73e0c0e5dccd"
      }
     }
    },
    "365b70d8-939a-4db1-b5e0-ad27e1afb18f": {
     "id": "365b70d8-939a-4db1-b5e0-ad27e1afb18f",
     "prev": null,
     "regions": {
      "be69cdb2-ad32-4a45-94f0-375907054c03": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "7f867b5b-aea2-48e5-9b63-725d88035641",
        "part": "whole"
       },
       "id": "be69cdb2-ad32-4a45-94f0-375907054c03"
      }
     }
    },
    "37ec19b9-5c5c-4be1-a8f3-ca88a518f045": {
     "id": "37ec19b9-5c5c-4be1-a8f3-ca88a518f045",
     "prev": "71d97322-58c3-46e8-9ac0-86ad7bf7b3e4",
     "regions": {
      "64255ac6-6b59-459e-aa6c-90447c2d9359": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "3ef87107-6478-4f90-a378-17a50e1227fe",
        "part": "whole"
       },
       "id": "64255ac6-6b59-459e-aa6c-90447c2d9359"
      }
     }
    },
    "42eea50e-aea9-41b2-8258-5ad236ee3a9b": {
     "id": "42eea50e-aea9-41b2-8258-5ad236ee3a9b",
     "prev": "ce7d5c0a-5177-4afe-8fe5-2af3f2cfaa85",
     "regions": {
      "56f964f6-15b7-40ba-82d4-61882f7d88f6": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "eedb5b88-78ab-4a1a-ae6c-eda78b68021b",
        "part": "whole"
       },
       "id": "56f964f6-15b7-40ba-82d4-61882f7d88f6"
      }
     }
    },
    "53fa24d6-3139-4b5b-9c1e-87d11eafe95d": {
     "id": "53fa24d6-3139-4b5b-9c1e-87d11eafe95d",
     "prev": "8d5eb191-49d9-4147-8d9b-757de59ef8ba",
     "regions": {
      "f23ded98-a527-4d10-8550-a84ed2c4a8a0": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "207b70ad-5e68-46c7-9a44-36223f6c9024",
        "part": "whole"
       },
       "id": "f23ded98-a527-4d10-8550-a84ed2c4a8a0"
      }
     }
    },
    "5614b3ef-1ded-49a6-9b8e-85eb3117b8ef": {
     "id": "5614b3ef-1ded-49a6-9b8e-85eb3117b8ef",
     "prev": "671e01b3-07d5-4b96-9460-adec4a64909c",
     "regions": {
      "fe72b7b4-678d-4cf3-8049-e7a7a166e719": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "9b300800-9a95-41ed-8264-dbb9b1e23e68",
        "part": "whole"
       },
       "id": "fe72b7b4-678d-4cf3-8049-e7a7a166e719"
      }
     }
    },
    "5b943f44-aff3-4e2c-91df-0ef380841628": {
     "id": "5b943f44-aff3-4e2c-91df-0ef380841628",
     "prev": "37ec19b9-5c5c-4be1-a8f3-ca88a518f045",
     "regions": {
      "971e1a49-62e6-4c9d-9aa7-957d7ec08ffa": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "2a052dab-019a-497a-a44e-353817ae86e3",
        "part": "whole"
       },
       "id": "971e1a49-62e6-4c9d-9aa7-957d7ec08ffa"
      }
     }
    },
    "5eb52529-1008-4d47-8090-30f9a40f28ba": {
     "id": "5eb52529-1008-4d47-8090-30f9a40f28ba",
     "prev": "b7cfecdc-1c9f-4b9b-9b45-411271c3055b",
     "regions": {
      "ad70f308-8695-4ef4-91b6-9ae9459505e2": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "b7dbf4d1-d250-425f-bc50-efb5119d0493",
        "part": "whole"
       },
       "id": "ad70f308-8695-4ef4-91b6-9ae9459505e2"
      }
     }
    },
    "5ecb1979-048b-4bac-98e6-9e70f2ed526d": {
     "id": "5ecb1979-048b-4bac-98e6-9e70f2ed526d",
     "prev": "df4ce82e-7463-4996-8f83-d6cee9e27302",
     "regions": {
      "e9ef853c-04e8-4311-8870-150c10a01770": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "f5112f97-f3a2-4436-808c-b35bba73af63",
        "part": "whole"
       },
       "id": "e9ef853c-04e8-4311-8870-150c10a01770"
      }
     }
    },
    "671e01b3-07d5-4b96-9460-adec4a64909c": {
     "id": "671e01b3-07d5-4b96-9460-adec4a64909c",
     "prev": "08e24f22-16dd-4ac6-ba88-ae953cc0a1d7",
     "regions": {
      "b265844c-44da-409b-b33c-a01e2511dd82": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "75d0eefd-d9fd-4edf-a631-6be9d95c463f",
        "part": "whole"
       },
       "id": "b265844c-44da-409b-b33c-a01e2511dd82"
      }
     }
    },
    "689a5d0c-5693-4013-b61b-d45a2e2f71b7": {
     "id": "689a5d0c-5693-4013-b61b-d45a2e2f71b7",
     "prev": "5b943f44-aff3-4e2c-91df-0ef380841628",
     "regions": {
      "61275914-fc53-4855-873d-fde90eaba4dd": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "e5a36fc8-4fe7-414e-bc84-a41b000b4c4c",
        "part": "whole"
       },
       "id": "61275914-fc53-4855-873d-fde90eaba4dd"
      }
     }
    },
    "71d97322-58c3-46e8-9ac0-86ad7bf7b3e4": {
     "id": "71d97322-58c3-46e8-9ac0-86ad7bf7b3e4",
     "prev": "93ebbb30-7f19-452c-a00f-37a27eb6d0e7",
     "regions": {
      "40b744ec-8e3c-4197-a4b1-b8d9e573a133": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "e153977f-f46b-4e46-8911-ba559bdf2a9c",
        "part": "whole"
       },
       "id": "40b744ec-8e3c-4197-a4b1-b8d9e573a133"
      }
     }
    },
    "85b12cab-f206-4d25-b05b-04db8b717af4": {
     "id": "85b12cab-f206-4d25-b05b-04db8b717af4",
     "prev": "0ffb0625-9562-4680-876c-c85b0b9afe10",
     "regions": {
      "028c2a39-d44a-4dea-be10-2045520de99f": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "854b016d-67cf-4579-8b03-9cac1c935f33",
        "part": "whole"
       },
       "id": "028c2a39-d44a-4dea-be10-2045520de99f"
      }
     }
    },
    "88f46e6d-006a-44f6-b6a7-28b8c169ed03": {
     "id": "88f46e6d-006a-44f6-b6a7-28b8c169ed03",
     "prev": "365b70d8-939a-4db1-b5e0-ad27e1afb18f",
     "regions": {
      "4d6bd003-0d3a-4131-9742-125dad092bd8": {
       "attrs": {
        "height": 0.8000000000000002,
        "width": 0.8,
        "x": 0.1,
        "y": 0.10000000000000002
       },
       "content": {
        "cell": "2ec4d70d-82ba-4479-841c-d9ee406d94cb",
        "part": "whole"
       },
       "id": "4d6bd003-0d3a-4131-9742-125dad092bd8"
      }
     }
    },
    "8d5eb191-49d9-4147-8d9b-757de59ef8ba": {
     "id": "8d5eb191-49d9-4147-8d9b-757de59ef8ba",
     "prev": "05f815a7-3844-479b-aec7-dfa667a64b1f",
     "regions": {
      "da986ed3-cbff-460e-86e1-dbd06caafe7f": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "32c8b061-1eb5-4a29-90a7-aa84b8de3634",
        "part": "whole"
       },
       "id": "da986ed3-cbff-460e-86e1-dbd06caafe7f"
      }
     }
    },
    "93ebbb30-7f19-452c-a00f-37a27eb6d0e7": {
     "id": "93ebbb30-7f19-452c-a00f-37a27eb6d0e7",
     "prev": "d787c4d9-78d4-471a-ba1b-b0c01752fb27",
     "regions": {
      "0b3d1fa2-1aa2-467f-a537-1280d5278274": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "69e1b48b-b302-459c-b5fe-67d9c2dfb288",
        "part": "whole"
       },
       "id": "0b3d1fa2-1aa2-467f-a537-1280d5278274"
      }
     }
    },
    "94a7fd34-303f-4f1a-8dc9-27542868e3ec": {
     "id": "94a7fd34-303f-4f1a-8dc9-27542868e3ec",
     "prev": "238c9334-cf28-4ed4-ade3-dca41af0bd35",
     "regions": {
      "f90a2d40-2c1b-4dc3-9e77-4b1f41ce695d": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "92b51c33-92d6-4beb-9c4a-10c24d7f4ffa",
        "part": "whole"
       },
       "id": "f90a2d40-2c1b-4dc3-9e77-4b1f41ce695d"
      }
     }
    },
    "a4c23815-a792-4702-8520-3f0b68b452c7": {
     "id": "a4c23815-a792-4702-8520-3f0b68b452c7",
     "prev": "5eb52529-1008-4d47-8090-30f9a40f28ba",
     "regions": {
      "08294c15-92ea-4643-a89d-94a57d2c4c7c": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "116bf7c5-2e65-4200-b026-8d65e92b5bb3",
        "part": "whole"
       },
       "id": "08294c15-92ea-4643-a89d-94a57d2c4c7c"
      }
     }
    },
    "b7cfecdc-1c9f-4b9b-9b45-411271c3055b": {
     "id": "b7cfecdc-1c9f-4b9b-9b45-411271c3055b",
     "prev": "53fa24d6-3139-4b5b-9c1e-87d11eafe95d",
     "regions": {
      "17c10bc8-18a0-46ba-a5c0-b44c25c48d88": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "f98823cc-e142-4c09-9536-0e8f88007b26",
        "part": "whole"
       },
       "id": "17c10bc8-18a0-46ba-a5c0-b44c25c48d88"
      }
     }
    },
    "b8e8d50b-67f3-498e-b8f1-1a907707bd46": {
     "id": "b8e8d50b-67f3-498e-b8f1-1a907707bd46",
     "prev": "ece5d686-b296-4bde-9081-3ff962150a02",
     "regions": {
      "adc195d3-f510-4e9b-9508-47dd219e4a2a": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "2ed31be9-b75c-4991-a1fa-c77d6056112e",
        "part": "whole"
       },
       "id": "adc195d3-f510-4e9b-9508-47dd219e4a2a"
      }
     }
    },
    "ce7d5c0a-5177-4afe-8fe5-2af3f2cfaa85": {
     "id": "ce7d5c0a-5177-4afe-8fe5-2af3f2cfaa85",
     "prev": "85b12cab-f206-4d25-b05b-04db8b717af4",
     "regions": {
      "7101574c-c164-4f95-ba8c-5c527d8750f9": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "d93d55a5-9275-4177-b23c-a8e9280258f0",
        "part": "whole"
       },
       "id": "7101574c-c164-4f95-ba8c-5c527d8750f9"
      }
     }
    },
    "d787c4d9-78d4-471a-ba1b-b0c01752fb27": {
     "id": "d787c4d9-78d4-471a-ba1b-b0c01752fb27",
     "prev": "5ecb1979-048b-4bac-98e6-9e70f2ed526d",
     "regions": {
      "3eecddc1-9554-4984-ba94-d5433dc8179c": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "e87b3c56-8afa-4e6e-af8f-c11e1ea248a9",
        "part": "whole"
       },
       "id": "3eecddc1-9554-4984-ba94-d5433dc8179c"
      }
     }
    },
    "df4ce82e-7463-4996-8f83-d6cee9e27302": {
     "id": "df4ce82e-7463-4996-8f83-d6cee9e27302",
     "prev": "94a7fd34-303f-4f1a-8dc9-27542868e3ec",
     "regions": {
      "d5d485e7-f3e5-41fa-bff5-2e9925530d44": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "396b61a6-0d05-4b74-be99-4303082218dc",
        "part": "whole"
       },
       "id": "d5d485e7-f3e5-41fa-bff5-2e9925530d44"
      }
     }
    },
    "ece5d686-b296-4bde-9081-3ff962150a02": {
     "id": "ece5d686-b296-4bde-9081-3ff962150a02",
     "prev": "88f46e6d-006a-44f6-b6a7-28b8c169ed03",
     "regions": {
      "e2a259dc-dfb6-46dc-b4b5-40fafd34cc85": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "b6cf2364-5980-471e-bf7f-5ce1c8843529",
        "part": "whole"
       },
       "id": "e2a259dc-dfb6-46dc-b4b5-40fafd34cc85"
      }
     }
    },
    "f8b02102-7ba4-4d1b-a453-01b1b94ae0de": {
     "id": "f8b02102-7ba4-4d1b-a453-01b1b94ae0de",
     "prev": "fc03fb44-5733-43a3-a5f2-9a8b30501d5d",
     "regions": {
      "d2d3f1bf-1dc0-4f57-9c94-2624c3097d22": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "df962b76-bd82-49c6-a6d3-08dc553b5bdc",
        "part": "whole"
       },
       "id": "d2d3f1bf-1dc0-4f57-9c94-2624c3097d22"
      }
     }
    },
    "fc03fb44-5733-43a3-a5f2-9a8b30501d5d": {
     "id": "fc03fb44-5733-43a3-a5f2-9a8b30501d5d",
     "prev": "5614b3ef-1ded-49a6-9b8e-85eb3117b8ef",
     "regions": {
      "bad7c836-af18-40f4-a04b-b9b95ddf14a7": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "7eac45b9-8585-4b6f-8ec9-b44b526e56e2",
        "part": "whole"
       },
       "id": "bad7c836-af18-40f4-a04b-b9b95ddf14a7"
      }
     }
    }
   },
   "themes": {}
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}