Results_reproduce.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluation and Comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## sr_mats_pre is the predicted outcome"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/liuqiao/anaconda2/lib/python2.7/site-packages/ipykernel_launcher.py:5: RuntimeWarning: divide by zero encountered in log10\n",
      "  \"\"\"\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.019148212020000423"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "14.560460842143709"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "0.24753643106770698"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mse_hicGAN_norm=map(compare_mse,hr_mats_test[:,:,:,0],sr_mats_pre[:,:,:,0])\n",
    "psnr_hicGAN_norm=map(calculate_psnr,hr_mats_test[:,:,:,0],sr_mats_pre[:,:,:,0])\n",
    "ssim_hicGAN_norm=map(calculate_ssim,hr_mats_test[:,:,:,0],sr_mats_pre[:,:,:,0])\n",
    "np.median(mse_hicGAN_norm)\n",
    "np.median(psnr_hicGAN_norm)\n",
    "np.median(ssim_hicGAN_norm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-18T08:35:24.284769Z",
     "start_time": "2018-12-18T08:35:22.406252Z"
    },
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17.342128202329857"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "12.999959273493769"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "0.27598158962572117"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Best for hicPlus\n",
    "mse_hicPlus=map(compare_mse,Y_test_crop,Y_pre)\n",
    "psnr_hicPlus=map(calculate_psnr,Y_test_crop,Y_pre)\n",
    "ssim_hicPlus=map(calculate_ssim,Y_test_crop,Y_pre)\n",
    "np.median(mse_hicPlus)\n",
    "np.median(psnr_hicPlus)\n",
    "np.median(ssim_hicPlus)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Record the sample subscripts with the same genomic distance (offset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-18T08:38:42.827667Z",
     "start_time": "2018-12-18T08:38:42.794478Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\n",
    "ind_0,ind_40,ind_80,ind_120,ind_160=[],[],[],[],[]\n",
    "for i in range(len(distance_all)):\n",
    "    if not (Y_test_crop[i]==0).all():\n",
    "        if abs(distance_all[i][0])==0:\n",
    "            ind_0.append(i)\n",
    "        elif abs(distance_all[i][0])==40:\n",
    "            ind_40.append(i)\n",
    "        elif abs(distance_all[i][0])==80:\n",
    "            ind_80.append(i)\n",
    "        elif abs(distance_all[i][0])==120:\n",
    "            ind_120.append(i)\n",
    "        elif abs(distance_all[i][0])==160:\n",
    "            ind_160.append(i)\n",
    "        else:\n",
    "            print 'error'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Group the metrices according to the offset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-18T13:41:25.181905Z",
     "start_time": "2018-12-18T13:41:25.167662Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data_mse_hicGAN_norm = [np.array(mse_hicGAN_norm)[ind_0],np.array(mse_hicGAN_norm)[ind_40],np.array(mse_hicGAN_norm)[ind_80],np.array(mse_hicGAN_norm)[ind_120],np.array(mse_hicGAN_norm)[ind_160]]\n",
    "data_psnr_hicGAN_norm = [np.array(psnr_hicGAN_norm)[ind_0],np.array(psnr_hicGAN_norm)[ind_40],np.array(psnr_hicGAN_norm)[ind_80],np.array(psnr_hicGAN_norm)[ind_120],np.array(psnr_hicGAN_norm)[ind_160]]\n",
    "data_ssim_hicGAN_norm = [np.array(ssim_hicGAN_norm)[ind_0],np.array(ssim_hicGAN_norm)[ind_40],np.array(ssim_hicGAN_norm)[ind_80],np.array(ssim_hicGAN_norm)[ind_120],np.array(ssim_hicGAN_norm)[ind_160]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-18T08:46:50.529282Z",
     "start_time": "2018-12-18T08:46:50.510070Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data_mse_hicPlus = [np.array(mse_hicPlus)[ind_0],np.array(mse_hicPlus)[ind_40],np.array(mse_hicPlus)[ind_80],np.array(mse_hicPlus)[ind_120],np.array(mse_hicPlus)[ind_160]]\n",
    "data_psnr_hicPlus = [np.array(psnr_hicPlus)[ind_0],np.array(psnr_hicPlus)[ind_40],np.array(psnr_hicPlus)[ind_80],np.array(psnr_hicPlus)[ind_120],np.array(psnr_hicPlus)[ind_160]]\n",
    "data_ssim_hicPlus = [np.array(ssim_hicPlus)[ind_0],np.array(ssim_hicPlus)[ind_40],np.array(ssim_hicPlus)[ind_80],np.array(ssim_hicPlus)[ind_120],np.array(ssim_hicPlus)[ind_160]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 355,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAMYAAAFiCAYAAACtYacRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nLHBhcDhZ4MLgcLLAhcHhZOH/A53ueNn3JUAeAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 216x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data={}\n",
    "data['dis']=['0']*len(ind_0)+['40']*len(ind_40)+['80']*len(ind_80)+['120']*len(ind_120)\n",
    "data['mse'] = np.concatenate([np.array(mse_hicGAN_norm)[ind_0],np.array(mse_hicGAN_norm)[ind_40],np.array(mse_hicGAN_norm)[ind_80],np.array(mse_hicGAN_norm)[ind_120]],axis=0)\n",
    "data['psnr'] = np.concatenate([np.array(psnr_hicGAN_norm)[ind_0],np.array(psnr_hicGAN_norm)[ind_40],np.array(psnr_hicGAN_norm)[ind_80],np.array(psnr_hicGAN_norm)[ind_120]],axis=0)\n",
    "data['ssim'] = np.concatenate([np.array(ssim_hicGAN_norm)[ind_0],np.array(ssim_hicGAN_norm)[ind_40],np.array(ssim_hicGAN_norm)[ind_80],np.array(ssim_hicGAN_norm)[ind_120]],axis=0)\n",
    "fig, axes = plt.subplots(nrows=1, ncols=1, figsize=(3,6), sharey=True)\n",
    "axes = sns.violinplot(x=\"dis\", y=\"psnr\",palette='muted',width=1,saturation=0.85,linewidth=0.5,data=data,order=['0','40','80','120'])\n",
    "axes.grid(False)\n",
    "#plt.savefig('fig/boxplot_psnr2.pdf')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use a dic to record pairs of true value and predicted value with the same genomic distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-18T13:41:39.181769Z",
     "start_time": "2018-12-18T13:41:36.544815Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#distance_all = [[offset,chrom],[offset,chrom],...,[offset,chrom]]\n",
    "#hr_mats_test: true test samples\n",
    "#sr_mats_pre: predicted test samples\n",
    "dic_norm={}\n",
    "for i in range(200):\n",
    "    dic_norm[i]=[[],[]]#true and pre\n",
    "for i in range(len(distance_all)):\n",
    "    for j in range(-39,40,1):\n",
    "        dis = distance_all[i][0]-j\n",
    "        dic_norm[abs(dis)][0]+=hr_mats_test[i,:,:,0].diagonal(offset=j).tolist()\n",
    "        dic_norm[abs(dis)][1]+=sr_mats_pre[i,:,:,0].diagonal(offset=j).tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dic_hicPlus_norm={}\n",
    "for i in range(200):\n",
    "    dic_hicPlus_norm[i]=[[],[]]#true and pre\n",
    "for i in range(len(distance_all)):\n",
    "    for j in range(-27,28,1):\n",
    "        dis = distance_all[i][0]-j\n",
    "        dic_hicPlus_norm[abs(dis)][0]+=Y_test_crop_norm[i].diagonal(offset=j).tolist()\n",
    "        dic_hicPlus_norm[abs(dis)][1]+=Y_pre_norm[i].diagonal(offset=j).tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dic_baseline={}\n",
    "for i in range(200):\n",
    "    dic_baseline[i]=[[],[]]#true and pre\n",
    "for i in range(len(distance_all)):\n",
    "    for j in range(-27,28,1):\n",
    "        dis = distance_all[i][0]-j\n",
    "        dic_baseline[abs(dis)][0]+=hr_mats_test[i,:,:,0].diagonal(offset=j).tolist()\n",
    "        dic_baseline[abs(dis)][1]+=lr_mats_test[i,:,:,0].diagonal(offset=j).tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(6623, 40, 40, 1)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(6623, 40, 40, 1)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hr_mats_test.shape\n",
    "lr_mats_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dic_gassian={}\n",
    "for i in range(200):\n",
    "    dic_gassian[i]=[[],[]]#true and pre\n",
    "for i in range(len(distance_all)):\n",
    "    for j in range(-27,28,1):\n",
    "        dis = distance_all[i][0]-j\n",
    "        dic_gassian[abs(dis)][0]+=hr_mats_test[i,:,:,0].diagonal(offset=j).tolist()\n",
    "        dic_gassian[abs(dis)][1]+=lr_mats_test_gaussian_blur[i].diagonal(offset=j).tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dic_rep={}\n",
    "for i in range(200):\n",
    "    dic_rep[i]=[[],[]]#true and pre\n",
    "for i in range(len(distance_all)):\n",
    "    for j in range(-27,28,1):\n",
    "        dis = distance_all[i][0]-j\n",
    "        dic_rep[abs(dis)][0]+=hr_mats_test[i,:,:,0].diagonal(offset=j).tolist()\n",
    "        dic_rep[abs(dis)][1]+=hr_mats_test_rep[i,:,:,0].diagonal(offset=j).tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-18T09:51:48.436139Z",
     "start_time": "2018-12-18T09:51:46.508601Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dic_hicPlus={}\n",
    "for i in range(200):\n",
    "    dic_hicPlus[i]=[[],[]]#true and pre\n",
    "for i in range(len(distance_all)):\n",
    "    if i in filter_idx:\n",
    "        for j in range(-27,28,1):#the size of sample of hicPlus is 28x28\n",
    "            dis = distance_all[i][0]-j\n",
    "            dic_hicPlus[abs(dis)][0]+=Y_test_crop[i].diagonal(offset=j).tolist()\n",
    "            dic_hicPlus[abs(dis)][1]+=Y_pre[i].diagonal(offset=j).tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-18T13:41:44.567185Z",
     "start_time": "2018-12-18T13:41:42.428256Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/liuqiao/anaconda2/lib/python2.7/site-packages/scipy/stats/stats.py:3033: RuntimeWarning: Mean of empty slice.\n",
      "  mx = x.mean()\n",
      "/home/liuqiao/anaconda2/lib/python2.7/site-packages/numpy/core/_methods.py:80: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  ret = ret.dtype.type(ret / rcount)\n",
      "/home/liuqiao/anaconda2/lib/python2.7/site-packages/scipy/stats/stats.py:3034: RuntimeWarning: Mean of empty slice.\n",
      "  my = y.mean()\n",
      "/home/liuqiao/anaconda2/lib/python2.7/site-packages/scipy/stats/stats.py:3038: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  r = r_num / r_den\n"
     ]
    }
   ],
   "source": [
    "y=[]\n",
    "y_baseline=[]\n",
    "y_gaussian=[]\n",
    "y_hicPlus=[]\n",
    "#y_rep=[]\n",
    "for i in range(200):\n",
    "    y.append(scipy.stats.pearsonr(dic_norm[i][0],dic_norm[i][1])[0])\n",
    "    y_hicPlus.append(scipy.stats.pearsonr(dic_hicPlus_norm[i][0],dic_hicPlus_norm[i][1])[0])\n",
    "    y_baseline.append(scipy.stats.pearsonr(dic_baseline[i][0],dic_baseline[i][1])[0])\n",
    "    y_gaussian.append(scipy.stats.pearsonr(dic_gassian[i][0],dic_gassian[i][1])[0])\n",
    "    #y_rep.append(scipy.stats.pearsonr(dic_rep[i][0],dic_rep[i][1])[0])\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MannwhitneyuResult(statistic=6965.0, pvalue=7.931797403600585e-07)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mannwhitneyu(y[:100],y_hicPlus[:100],alternative='greater')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1174,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fbf1c3eb590>]"
      ]
     },
     "execution_count": 1174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fb358992cd0>]"
      ]
     },
     "execution_count": 1174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fbefeb8a190>]"
      ]
     },
     "execution_count": 1174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fb35a3d0490>]"
      ]
     },
     "execution_count": 1174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(0.3, 1)"
      ]
     },
     "execution_count": 1174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5,0,'Genomic distance(10kb)')"
      ]
     },
     "execution_count": 1174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Pearson r')"
      ]
     },
     "execution_count": 1174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(0, 100)"
      ]
     },
     "execution_count": 1174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7fb35a3fa290>"
      ]
     },
     "execution_count": 1174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEHCAYAAAC0pdErAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.set_style(\"whitegrid\") \n",
    "fig, axes = plt.subplots(nrows=1, ncols=1, figsize=(6, 4), sharey=True)\n",
    "axes.grid(False)\n",
    "axes.plot(list(range(200)),y,'r',label='hicGAN')\n",
    "axes.plot(list(range(200)),y_hicPlus,'g',label='HiCPlus')\n",
    "axes.plot(list(range(200)),y_gaussian,'k',linestyle='dashed',label='2D Gaussian')\n",
    "#axes.plot(list(range(200)),y_rep,'k',linestyle='dashed',label='Biological Rep2')\n",
    "axes.plot(list(range(200)),y_baseline,'k',label='Baseline')\n",
    "\n",
    "plt.ylim(0.3,1)\n",
    "plt.xlabel('Genomic distance(10kb)')\n",
    "plt.ylabel('Pearson r')\n",
    "plt.xlim(0,100)\n",
    "plt.legend()\n",
    "#plt.savefig('fig/comparison_pearson.pdf')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "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"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}