Figures/Figure 2B.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "8bf7138c",
   "metadata": {},
   "source": [
    "### Differences between RBP binding plots\n",
    "\n",
    "Figure 2B. \n",
    "\n",
    "The RBP coordinates and PSI values can be downloaded from Zenodo. Here, we show how to calculate the MSE as explained in Figure 2A. \n",
    "\n",
    "The MSE can also be downloaded from Zenodo directly to plot Figure 2B. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "848b5ee5",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "import copy as cp\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from tqdm.notebook import tqdm\n",
    "from scipy.stats import pearsonr\n",
    "\n",
    "import seaborn as sns\n",
    "from matplotlib import pyplot as plt\n",
    "from sklearn.metrics import mean_squared_error as mse\n",
    "\n",
    "import matplotlib\n",
    "matplotlib.rcParams['pdf.fonttype'] = 42\n",
    "\n",
    "from mse_peaks import mse_peaks"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e5a4560",
   "metadata": {},
   "source": [
    "### Calculate MSE between RBP profiles"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f1a9624",
   "metadata": {},
   "source": [
    "#### Hippocampus"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "abe4544c",
   "metadata": {},
   "source": [
    "First, we load the peaks from the eCLIP data. In the dataframe, every row is a gene and every column is an RBP. The values indicate begin and end of peaks measured."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "67dcabe9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AATF</th>\n",
       "      <th>ABCF1</th>\n",
       "      <th>AKAP1</th>\n",
       "      <th>APOBEC3C</th>\n",
       "      <th>AQR</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ENSG00000107554</th>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[[99875689, 99875747], [99875702, 99875735], [...</td>\n",
       "      <td>[]</td>\n",
       "      <td>[[99877075, 99877107], [99877291, 99877325], [...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000107566</th>\n",
       "      <td>[[100151393, 100151426], [100179512, 100179588...</td>\n",
       "      <td>[[100151980, 100152011]]</td>\n",
       "      <td>[[100150218, 100150240], [100150281, 100150333...</td>\n",
       "      <td>[[100150776, 100150813]]</td>\n",
       "      <td>[[100151395, 100151432], [100151536, 100151567...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000213341</th>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[[100188774, 100188812]]</td>\n",
       "      <td>[[100188586, 100188621], [100188761, 100188789...</td>\n",
       "      <td>[[100193190, 100193249], [100193223, 100193264...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000095485</th>\n",
       "      <td>[[100237058, 100237180], [100237063, 100237133...</td>\n",
       "      <td>[[100237084, 100237172], [100237091, 100237148...</td>\n",
       "      <td>[[100232429, 100232464], [100232929, 100233112...</td>\n",
       "      <td>[[100232368, 100232393], [100233104, 100233140...</td>\n",
       "      <td>[[100233336, 100233418], [100233370, 100233429...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000196072</th>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[[100274793, 100274830], [100274812, 100274866...</td>\n",
       "      <td>[[100274718, 100274745], [100274719, 100274760...</td>\n",
       "      <td>[[100274979, 100275024], [100275386, 100275428...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                              AATF  \\\n",
       "ENSG                                                                 \n",
       "ENSG00000107554                                                 []   \n",
       "ENSG00000107566  [[100151393, 100151426], [100179512, 100179588...   \n",
       "ENSG00000213341                                                 []   \n",
       "ENSG00000095485  [[100237058, 100237180], [100237063, 100237133...   \n",
       "ENSG00000196072                                                 []   \n",
       "\n",
       "                                                             ABCF1  \\\n",
       "ENSG                                                                 \n",
       "ENSG00000107554                                                 []   \n",
       "ENSG00000107566                           [[100151980, 100152011]]   \n",
       "ENSG00000213341                                                 []   \n",
       "ENSG00000095485  [[100237084, 100237172], [100237091, 100237148...   \n",
       "ENSG00000196072                                                 []   \n",
       "\n",
       "                                                             AKAP1  \\\n",
       "ENSG                                                                 \n",
       "ENSG00000107554  [[99875689, 99875747], [99875702, 99875735], [...   \n",
       "ENSG00000107566  [[100150218, 100150240], [100150281, 100150333...   \n",
       "ENSG00000213341                           [[100188774, 100188812]]   \n",
       "ENSG00000095485  [[100232429, 100232464], [100232929, 100233112...   \n",
       "ENSG00000196072  [[100274793, 100274830], [100274812, 100274866...   \n",
       "\n",
       "                                                          APOBEC3C  \\\n",
       "ENSG                                                                 \n",
       "ENSG00000107554                                                 []   \n",
       "ENSG00000107566                           [[100150776, 100150813]]   \n",
       "ENSG00000213341  [[100188586, 100188621], [100188761, 100188789...   \n",
       "ENSG00000095485  [[100232368, 100232393], [100233104, 100233140...   \n",
       "ENSG00000196072  [[100274718, 100274745], [100274719, 100274760...   \n",
       "\n",
       "                                                               AQR  \n",
       "ENSG                                                                \n",
       "ENSG00000107554  [[99877075, 99877107], [99877291, 99877325], [...  \n",
       "ENSG00000107566  [[100151395, 100151432], [100151536, 100151567...  \n",
       "ENSG00000213341  [[100193190, 100193249], [100193223, 100193264...  \n",
       "ENSG00000095485  [[100233336, 100233418], [100233370, 100233429...  \n",
       "ENSG00000196072  [[100274979, 100275024], [100275386, 100275428...  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "peaks_path = '../../PSI_prediction/Zenodo/Human/HPC/RBP/RBP_coor_repl.pickle'\n",
    "RBP_coor_repl, rows, columns = pd.read_pickle(peaks_path)\n",
    "peaks = pd.DataFrame(RBP_coor_repl, index=rows, columns=columns)\n",
    "peaks.iloc[:5,:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f8d2e90",
   "metadata": {},
   "source": [
    "Read the PSI values, so we can split later on for exons < 0.5 and exons > 0.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d1996af0",
   "metadata": {},
   "outputs": [],
   "source": [
    "PSI_glia = pd.read_csv('../../PSI_prediction/Zenodo/Human/HPC/PSI/PSI_glia_norm.csv', \n",
    "                       index_col=0)\n",
    "PSI_neur = pd.read_csv('../../PSI_prediction/Zenodo/Human/HPC/PSI/PSI_neur_norm.csv', \n",
    "                       index_col=0)\n",
    "PSI = pd.concat((PSI_glia, PSI_neur), axis=1)\n",
    "PSI.columns = ['Glia', 'Neurons']\n",
    "\n",
    "# Only keep the exons with PSI value defined in neurons and glia\n",
    "PSI = PSI.loc[PSI.notna().sum(axis=1) > 1]\n",
    "PSI_glia = PSI_glia.loc[PSI.index]\n",
    "PSI_neur = PSI_neur.loc[PSI.index]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13d06537",
   "metadata": {},
   "source": [
    "Next, we read the RBP counts. This is slightly different from the peaks. Here, there are no coordinates per gene/RBP, but there is counted per exon already how many RBPs bind at a specific location. We use this to filter out the exons with no RBP binding sites since these genes were probably not expressed/measured in the eCLIP data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "0ca827b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "RBP_counts = pd.read_csv('../../PSI_prediction/Zenodo/Human/HPC/RBP/RBP_peaks.csv', \n",
    "                         index_col=0)\n",
    "RBP_counts = RBP_counts.loc[PSI.index]\n",
    "tokeep = np.sum(RBP_counts, axis=1) > 0\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e29ef0a4",
   "metadata": {},
   "source": [
    "Split the names of the exons based so that we get the chromosome, begin, end, gene name and strand"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3110eb04",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>chr10</td>\n",
       "      <td>100002942</td>\n",
       "      <td>100003023</td>\n",
       "      <td>ENSG00000107554</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>chr10</td>\n",
       "      <td>1000677</td>\n",
       "      <td>1000868</td>\n",
       "      <td>ENSG00000107937</td>\n",
       "      <td>+</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>chr10</td>\n",
       "      <td>100158984</td>\n",
       "      <td>100159055</td>\n",
       "      <td>ENSG00000107566</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>chr10</td>\n",
       "      <td>100167348</td>\n",
       "      <td>100167406</td>\n",
       "      <td>ENSG00000107566</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>chr10</td>\n",
       "      <td>100174208</td>\n",
       "      <td>100174281</td>\n",
       "      <td>ENSG00000107566</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       0          1          2                3  4\n",
       "0  chr10  100002942  100003023  ENSG00000107554  -\n",
       "1  chr10    1000677    1000868  ENSG00000107937  +\n",
       "2  chr10  100158984  100159055  ENSG00000107566  -\n",
       "3  chr10  100167348  100167406  ENSG00000107566  -\n",
       "4  chr10  100174208  100174281  ENSG00000107566  -"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exons_info = pd.DataFrame(RBP_counts.index)[0].str.split('_',expand=True)\n",
    "exons_info.iloc[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17648743",
   "metadata": {},
   "source": [
    "Calculate the following:\n",
    "- MSE between variable and not variable for exons with low PSI \n",
    "- MSE between variable and not variable for exons with high PSI \n",
    "- Number of variable exons with low PSI and a binding site for that RBP\n",
    "- Number of variable exons with high PSI and a binding site for that RBP\n",
    "- Number of not variable exons with low PSI and a binding site for that RBP\n",
    "- Number of not variable exons with high PSI and a binding site for that RBP\n",
    "- Number of peaks in variable exons with a low PSI and a binding site for that RBP\n",
    "- Number of peaks in variable exons with a high PSI and a binding site for that RBP\n",
    "- Number of peaks in not variable exons with a low PSI and a binding site for that RBP\n",
    "- Number of peaks in not variable exons with a high PSI and a binding site for that RBP\n",
    "\n",
    "! This takes ~15 minutes to run"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "510d0014",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "132c555cbc9a4692b93123a502d83834",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/122 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "349e277ec9af453d865c2095cbab213f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ce061d41ca444eda905a86464ebf6f02",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "02625742b56349f49fcf1f65027d195d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5c90eafaa20a49669516e584eff95712",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ecfeb4edfa3d4f238a0c41d51cc5662d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "da4fd0819d9e48e08231f9e687405499",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1145411a0d764740828029db025de0f9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e460101f1a9544cf8379cc3b79293ed1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "406b0aa861b24dac89c916c73bf2d01f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8dd99894a25c4d608aa37196e8cb69d0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dd23add0478d4ace8d664e79dfc8571f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bf35c3e5be6349db92e7ea07f75f7e6d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c9e1f50e3a8d4ab085b0c9f5011dd734",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "97df9e28837046af924cc7f20e0d2461",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b6f4a8fe434c47709967187e2580f9c0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b1f1ddd74d9040c5b3ca85d31fb09ab8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5986b6f66e2842558dcbf7f1d0f1e340",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "96aaa88a8e8b46669143f2d8eb3e3635",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6fa0b023064b42c0a3dbd3a40f60c18b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e53ce05ae63843dea0390287282995ea",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f0064a4019634eec9c705d00fe3c35aa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b8772387161a4e20b1139c31363fb4a1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7e547a7676744a7baf31501c4bdaf4c1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2b221aee4c494b86810267742aa938c7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ee2d3677cec44935886b8c9ae1c74fdd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cbaa9adae43f4c468f1bff381e5d6759",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f1115f05a965405ba59ca8f18dcf013d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a798950583044e019e4545b692fb7895",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5cb2c8b9aaa046c89e095c0d00e5f8d6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2cdcc49264f740efbd72a3fb5ca6fbf2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c6d58799021640649cf2b20c4e9a22c6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7711a79f45af46d09c5a9f9d20cbecbd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "36d437f3033248c99e7dda493b6fdc6d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "342c5694e02e41b381601e6484a6c281",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c3858f1a59ad4f0590c3d4d72825896d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "39569bd9b7a344daaa1ca37b7d464e01",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ed35642be6874baab0f57039eab353b4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c52011c468394c008ef7ecaf65362964",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "03e6406b5a214f978d4cc7a2d68ad9e3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0d8097a25ee04660b4bc13eb10eb70f1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f33f40ea399749bea105591a52aaa0c0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0a71db0c11f646038ef8fe8323dba70b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8466e2aa680f4250a802962832a1a274",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0946fb32f0fe4d0288c2df2eba5213d2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7d3ea7e91fe54dd7bcb46c6cdfbd3fab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3cdae2177e054c39be01ea922aa86ab7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5280edfd4c824f47bf12067e16934041",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f7dd9c0bf08c484587b30513a68385c5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fcfb53370be343f380ab8ef523e22fa7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0a6eb592be214e68bdddac5b6e9d967e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "33d78ff9405744f0b946259180a75f7e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "97e8161356764f868e948b82094cfb6e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cbe342128bf54fe8ba9a70e98e29dc44",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f2232c2e2e184e55a111b8a0b35c3a9b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "59604461c4e94d34ad15a5913e917e5d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e4128707a50742baaf81ded764040310",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4f5bb515ab3949548e749ddf335b7bfb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bf12031ed315415385ccd7a18d18748e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b07f541ab54542fd8ce9900be2defd83",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "76dc803dd3ce4e1082a52fee36632aab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a6299c297c86425ba3a68ef451c38063",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3d65a19b529b47809594b3ba6c7558a0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d9e1b69e2bff44b6b28007ea5333c163",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "946cc2c52a4748d1bbb658f9c1528d01",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "18124cde24054120a69fdff6e8bc5dde",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9ab18414f1304c0bb0fdda773f546a06",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "07f2996e67304a5786f43e905e6800e4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1bc90f509f8b472ca430067dff15e7cc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2ab7d53283ea4d38a9a56fd5ff80b592",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "639671ad72fc4775adbf9bfa88977fd1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fa99add97b0d4a788b62479110dd382f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bdcd1bf36d684c7da7682a0ecbd1539e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8e1dd18694d34912971c28704c749874",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "57b86092b2164348b544e3f4fbb07a17",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4d613b67f7c545889a9e13bad452fac4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eaccd9117ba242d992582c7386f15063",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eb7d05e1b3974f9ca2e439590e306800",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0b9ce88e07414d29aae0d19b99e2180d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ca3dd348c35144c8a04d4ed615291b45",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6b15eee3b3b74ca4929cbe57623b9632",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a74a08a0339d4d65a9f6fb0aa28d8785",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cc7949f6a79c48089350fd1005062359",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "146778576c35461f85db383e14309f68",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e2a2d22a79684b8c8e65efc4926226e9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "16fb2d34fc6c488ab907cc5b50f9a4ed",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b868b5e1f7b2429c81e289dfff8d23cf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "235f640477884a49b0c557d5de646180",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cb55e65f0dfc45d997c79152af602a99",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "255f2de022db4dbc965af03ef26a6da0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6ca7ae786c4546f0b66701bdf64b3945",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7819af0f3f884ae89772a64617a547cf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "acfe51fa49d049ff9cf6fdba8c16efb9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "795ccaf9be874ce09a3db8485a427fd7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "30e8ea64eaa74715bd457d74170b79cd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dae24e1772264106bc017c939698f624",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3e03cf34106d4e77b94a09a740f4c5f2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "db5c2f08d1f44125822e44adf661f555",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8435541fc1494136a686f153ebc0d098",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8d701142836f485da4f589490a5d48b3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9f9320fead73431a8ada41d36a2e56ae",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "973b9be811804d1fb08fff2769a72b54",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4570affb0fc14909b4009b9c572ae519",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "37766050415d408d8fd042b386dda4ba",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cc3978f412314dfda0ab3ec4ce6f6e38",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "42723f83f11c4ae981488537d52b18c3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a6cf2f90d5984477941e262affd2820e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6849869a2f7f476d8147e0db5f92242f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d8299563c1de42928789b3dbbf13ba50",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b6ebd0350de84dc8a75e6f92c4670969",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "26ba70a8c92d4ff1ac088462baf2d272",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "42e02e3271d24af383813d3ee1dbacf5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c44d1f68b7184e59a04fa53187d45c5b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f11041400d6445ba06f19226d3b43c4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dadafcb5f66a41c48f7ef611f3df4bf8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2a5e10d5973e4a5da18d35bb3ffbc71b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "50a760c4f9ae4ee59dc42ef0bc669513",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1897dfcf35df409ba44266be7a1b70f3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "194a8954187a45589fbe2b15143afc62",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "43c4bd0c72b74612a25e2ebfc56cec8d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8bc4eaac6db04d4aad7c33e00e90b798",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c1dfb77d650547679d35950c5098079e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4de41b2fff134845bdab5efffaf8dfbe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/37382 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "HPC_mse_neur = pd.DataFrame(data=np.zeros((len(peaks.columns),10)),\n",
    "                       index=peaks.columns, columns=['Low', 'High', 'Exons low var',\n",
    "                                                    'Exons high var', 'Exons low cons',\n",
    "                                                    'Exons high cons', 'Peaks low var',\n",
    "                                                    'Peaks high var', 'Peaks low cons',\n",
    "                                                    'Peaks high cons'])\n",
    "HPC_mse_glia = pd.DataFrame(data=np.zeros((len(peaks.columns),10)),\n",
    "                       index=peaks.columns, columns=['Low', 'High', 'Exons low var',\n",
    "                                                    'Exons high var', 'Exons low cons',\n",
    "                                                    'Exons high cons', 'Peaks low var',\n",
    "                                                    'Peaks high var', 'Peaks low cons',\n",
    "                                                    'Peaks high cons'])\n",
    "\n",
    "for RBP in tqdm(peaks.columns):\n",
    "    res_glia, res_neur = mse_peaks(RBP, exons_info.iloc[tokeep.values], \n",
    "                                   peaks, PSI_glia[tokeep], PSI_neur[tokeep])    \n",
    "    HPC_mse_neur.loc[RBP] = res_neur\n",
    "    HPC_mse_glia.loc[RBP] = res_glia"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "63441e35",
   "metadata": {},
   "source": [
    "We follow the same steps for the Frontal Cortex"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "30e0093c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AATF</th>\n",
       "      <th>ABCF1</th>\n",
       "      <th>AKAP1</th>\n",
       "      <th>APOBEC3C</th>\n",
       "      <th>AQR</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ENSG00000107566</th>\n",
       "      <td>[[100151393, 100151426], [100179512, 100179588...</td>\n",
       "      <td>[[100151980, 100152011]]</td>\n",
       "      <td>[[100150218, 100150240], [100150281, 100150333...</td>\n",
       "      <td>[[100150776, 100150813]]</td>\n",
       "      <td>[[100151395, 100151432], [100151536, 100151567...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000213341</th>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[[100188774, 100188812]]</td>\n",
       "      <td>[[100188586, 100188621], [100188761, 100188789...</td>\n",
       "      <td>[[100193190, 100193249], [100193223, 100193264...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000095485</th>\n",
       "      <td>[[100237058, 100237180], [100237063, 100237133...</td>\n",
       "      <td>[[100237084, 100237172], [100237091, 100237148...</td>\n",
       "      <td>[[100232429, 100232464], [100232929, 100233112...</td>\n",
       "      <td>[[100232368, 100232393], [100233104, 100233140...</td>\n",
       "      <td>[[100233336, 100233418], [100233370, 100233429...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000196072</th>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[[100274793, 100274830], [100274812, 100274866...</td>\n",
       "      <td>[[100274718, 100274745], [100274719, 100274760...</td>\n",
       "      <td>[[100274979, 100275024], [100275386, 100275428...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000119906</th>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[[100913049, 100913088], [100913318, 100913336...</td>\n",
       "      <td>[[100962847, 100962901]]</td>\n",
       "      <td>[[100913258, 100913289], [100916016, 100916056...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                              AATF  \\\n",
       "ENSG                                                                 \n",
       "ENSG00000107566  [[100151393, 100151426], [100179512, 100179588...   \n",
       "ENSG00000213341                                                 []   \n",
       "ENSG00000095485  [[100237058, 100237180], [100237063, 100237133...   \n",
       "ENSG00000196072                                                 []   \n",
       "ENSG00000119906                                                 []   \n",
       "\n",
       "                                                             ABCF1  \\\n",
       "ENSG                                                                 \n",
       "ENSG00000107566                           [[100151980, 100152011]]   \n",
       "ENSG00000213341                                                 []   \n",
       "ENSG00000095485  [[100237084, 100237172], [100237091, 100237148...   \n",
       "ENSG00000196072                                                 []   \n",
       "ENSG00000119906                                                 []   \n",
       "\n",
       "                                                             AKAP1  \\\n",
       "ENSG                                                                 \n",
       "ENSG00000107566  [[100150218, 100150240], [100150281, 100150333...   \n",
       "ENSG00000213341                           [[100188774, 100188812]]   \n",
       "ENSG00000095485  [[100232429, 100232464], [100232929, 100233112...   \n",
       "ENSG00000196072  [[100274793, 100274830], [100274812, 100274866...   \n",
       "ENSG00000119906  [[100913049, 100913088], [100913318, 100913336...   \n",
       "\n",
       "                                                          APOBEC3C  \\\n",
       "ENSG                                                                 \n",
       "ENSG00000107566                           [[100150776, 100150813]]   \n",
       "ENSG00000213341  [[100188586, 100188621], [100188761, 100188789...   \n",
       "ENSG00000095485  [[100232368, 100232393], [100233104, 100233140...   \n",
       "ENSG00000196072  [[100274718, 100274745], [100274719, 100274760...   \n",
       "ENSG00000119906                           [[100962847, 100962901]]   \n",
       "\n",
       "                                                               AQR  \n",
       "ENSG                                                                \n",
       "ENSG00000107566  [[100151395, 100151432], [100151536, 100151567...  \n",
       "ENSG00000213341  [[100193190, 100193249], [100193223, 100193264...  \n",
       "ENSG00000095485  [[100233336, 100233418], [100233370, 100233429...  \n",
       "ENSG00000196072  [[100274979, 100275024], [100275386, 100275428...  \n",
       "ENSG00000119906  [[100913258, 100913289], [100916016, 100916056...  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "peaks_path = '../../PSI_prediction/Zenodo/Human/FC/RBP/RBP_coor_repl.pickle'\n",
    "RBP_coor_repl, rows, columns = pd.read_pickle(peaks_path)\n",
    "peaks = pd.DataFrame(RBP_coor_repl, index=rows, columns=columns)\n",
    "peaks.iloc[:5,:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "81e2e716",
   "metadata": {},
   "outputs": [],
   "source": [
    "PSI_glia = pd.read_csv('../../PSI_prediction/Zenodo/Human/FC/PSI/PSI_glia_norm.csv', \n",
    "                       index_col=0)\n",
    "PSI_neur = pd.read_csv('../../PSI_prediction/Zenodo/Human/FC/PSI/PSI_neur_norm.csv', \n",
    "                       index_col=0)\n",
    "PSI = pd.concat((PSI_glia, PSI_neur), axis=1)\n",
    "PSI.columns = ['Glia', 'Neurons']\n",
    "PSI = PSI.loc[PSI.notna().sum(axis=1) > 1]\n",
    "PSI_glia = PSI_glia.loc[PSI.index]\n",
    "PSI_neur = PSI_neur.loc[PSI.index]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "d7d8ded3",
   "metadata": {},
   "outputs": [],
   "source": [
    "RBP_counts = pd.read_csv('../../PSI_prediction/Zenodo/Human/FC/RBP/RBP_peaks.csv', \n",
    "                         index_col=0)\n",
    "\n",
    "# We have to get rid of the .version in the gene name\n",
    "xx = pd.DataFrame(RBP_counts.index)['GE'].str.split('_', expand=True)\n",
    "RBP_counts.index = xx[0]+'_'+xx[1]+'_'+xx[2]+'_'+xx[3].str.split('.', expand=True)[0]+'_'+xx[4]\n",
    "\n",
    "RBP_counts = RBP_counts.loc[PSI.index]\n",
    "tokeep = np.sum(RBP_counts, axis=1) > 0\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "7b0caca0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>chr10</td>\n",
       "      <td>100158984</td>\n",
       "      <td>100159055</td>\n",
       "      <td>ENSG00000107566</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>chr10</td>\n",
       "      <td>100167348</td>\n",
       "      <td>100167406</td>\n",
       "      <td>ENSG00000107566</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>chr10</td>\n",
       "      <td>100202088</td>\n",
       "      <td>100202149</td>\n",
       "      <td>ENSG00000213341</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>chr10</td>\n",
       "      <td>100204506</td>\n",
       "      <td>100204657</td>\n",
       "      <td>ENSG00000213341</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>chr10</td>\n",
       "      <td>100245799</td>\n",
       "      <td>100245913</td>\n",
       "      <td>ENSG00000095485</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       0          1          2                3  4\n",
       "0  chr10  100158984  100159055  ENSG00000107566  -\n",
       "1  chr10  100167348  100167406  ENSG00000107566  -\n",
       "2  chr10  100202088  100202149  ENSG00000213341  -\n",
       "3  chr10  100204506  100204657  ENSG00000213341  -\n",
       "4  chr10  100245799  100245913  ENSG00000095485  -"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exons_info = pd.DataFrame(RBP_counts.index)[0].str.split('_',expand=True)\n",
    "exons_info.iloc[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "177a791c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "aa0024a09d95436a891573059ebfca6b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/122 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "235ac5385f114b41ab63be00a70e3bd6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f404f327ea98437da6686bf91ce56a72",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "866e09c8610e487290f9d0e2b1c24136",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cb627729ff5c4016a66c33dd28dc6378",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ee81df3a3a5d455ab7cbc5d9a0fe4ae0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7cd5ba24f4fc4e009342823780c36e90",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ad638fb62a6445569e3c51299aab6896",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7673b0bfc08a453a8a717113e15a075f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4df718c6bba0464792a8bb13a4669586",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e906020b9f004aaf87b58a7d3aabd232",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4108913647bf41418b39369dd315d850",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9f4c0ffc5aab4d589d0bb368c0feeb00",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d905715ce28148728c88065dac2ff078",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4f015e91456b4f8c8df962830d7ffd02",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c7579b9660c6411da2e8abfe023cb302",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "635b2bb2d8944973b4558262ea5e2cc6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d5c893c5afd9429ba72fa42c31cacea4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "61b77de781044cad9712b4fd62dc3d01",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "602b7563140b425b9af7350b3e87f64d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8e1301933dd2416c92da2a6201859c67",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8ba963bd29a640909dca2b5e466f61e5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "12da3ec254f24b9a860e5b33dc2ee2fa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6e66313c5d4b4214ba2da3756efb3dad",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f3386900303b4376af967f19b8afd903",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6ec3dc05d690433d8ca3e48fd5a58720",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a7568cd73744456faad3a42f26dd39cc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2d45faf633ce4bdf9d433f47175c4f33",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dc1d49cb91b3442fb5600b185e7f9ff5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "de1a970d49784f7fa68998d12305a860",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eaa1a85e912b4f52ae1fd09f74c9bf15",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f819b1d8d3384a9eb2153498277348e3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3bc3b0c378714dbc830d6eb634a7c2c3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "725566495c814e18b424f9330c573ff8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "12c8aee981a341a280057e7052f4689a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dc81964bee344d46889f778cf749b29a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "288b00d213924af0a8ef1a9931e01aba",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c796ef5566b54422b8fdb1bb6e447c6a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "802f1e134efe477a913c1cfdb8403746",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b8a3150d2e6744069fe6b600290e6fcc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d0f1523bc9fc42ed991aed1d3e48d086",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6b273f6089b04046b5f4905e15968cf0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b40cab0ce3a046d4a6643824464a753a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3e4cc18a994f40a7bba318f6c8629294",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "29807877843e4cb5b0ff6c6fcb9451eb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "971e636d1512454082610492a849119f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "88af237994b4471e94368ad9b173a482",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ad7dd64045e745269299c5a9f957d6e2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "678257c040804e719d9977490b359717",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fd34e5643eb5446d8346c09382d73c82",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1e1d608596d648a896304ce2bc4336cd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dff81e1989ac4b828593ce47f57d058e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f3eae989494b435a8888593636673500",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8c29ea6c3ffe49cd8ee75f8694ba84d5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c7eea1bbfdb6454db0b817ab15979df7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "41128252c7fe4cfe83bc4b6577df580b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a4f3fc982f1a4dcba883ff6fef9500fd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c3307c385ba94a5c9142b1732fd27f83",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6d8cfee401f441c1a5fad3af0abe6745",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c0ec6ddf80724725a8c19f9b2eefab9a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "884cde2a1ef34f3681cf5ea76003af01",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a14793d94c7344768181be3602a6aeec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "25870a54b69f417ab85fb5232a5548d5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "48c1c3ed282849c996b6713e122590ab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2f7c96c39f9c4e3ca0be4925a1dc5eff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f1bae0e3be8c405c8d67e7b50dbbf060",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8fa2831ed66246f48a5acdbcbf0a6289",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3e3e21d173514f0dab728c3cc9e925f5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2a46a64b658141f6b9cb00170e7f0b65",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "548cf4f82349471085fb72018de5b991",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6b64099604a44874bc44a64fb22d45ff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "67ea6aa15b6a4edb8a2a963645e3e554",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "91bf6ae7bd2d466c8c85407eba1bf667",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "203aab4cf06b461cad0d8887f077f562",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "55c7e2e6efdf4afe969eed05cd7cc5b8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "815aa16c21d04863b141b64745c7a5c0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "578787d101344080a87fd31cc4db2eea",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5d3946159fcb42158522c617cab8d35c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2017c0cd5a5941999557ed8ae2eed30f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a180a1e723c14417ac538ac526e58f31",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4357734513294a30ad11c6f17468a2a3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4d6dd0c7f8f2438ba7a220eba63ff6e0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "64678ec8a701499190903970d24f524b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "695e121a201d416086336eea8d18ea51",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4ec70e4272724ffd9c8abbf542fa951c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8fca8a142fa84e139122159d2680f93f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f222c662b6584e8ea0294a2610166d4d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "09986b29e40f4e40955faea099e02724",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eaf7ef4d7203430fb40e0af409a280e5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "83cf0bc9c7a846cd9c6dd61589f56326",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f540bfa84cd94a80ae3c0b1c399123a3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bd91a9e78c3b42e199da4b0e9bba01ec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "20ab3c463ae74d76b9671b521ce32d89",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7bddf4db7da04e6cb60ac83d388507a6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "32b539cc84cd49958c17897a05cb3553",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d8ee2d55b6f245f68f93670940aaf4ea",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6f195d66180c46daba70fda2da3f477c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4bc82a62eee84ba38876e84db552fafc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bd984f096bb541298b63637b86800f93",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "20e4af664cf34ae0bd15ebb43e4813a1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6f915aa97fb0439482b35919feb11433",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "88e70b94795748cda2a623b5fce0176f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fe4b3ecd032b46e78ea06051ac37e8df",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9b20765706e446f88f73fe456493f3c3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3cd7c04739ea420cbf861243652eb3e5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b6b42af9b9af4a38b4b97f7fe00a776f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fdb2ca53ffc34e76a731349cf467513c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7c06d3ddc02a44b9a5ea2b658e34f6b4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "88aec6623d32403787f7454f50f18595",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "662bc0525cf34b36ad347f6a656a4560",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "269c39369f4c4e60b62de976ce0c3390",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "76d68777bfdd42f6a860daa4f06b40bf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "75d8b7f68f5e4b1e8d89693063e843a2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "af0f1d2d20bd420586a4ba355d5e524c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bea7d932fbae4ad0b8e66390313ccb86",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "33f1af9ce2e94682b76a6e3148b57276",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "65e19c15c10c406885f9f2560f77fc84",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "283f1cc61ce944ad971a5adcfc0d79f5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "863e4c781cd842999c7e48a50927146b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "876c726d1a7247ee8d0eba3511dd5025",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2bf362068c7f4e5d94605353beb6e9c9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f1df41280d8f4b6884fa8cf968aac068",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "55b7894e37c04248b69e978a32e3c791",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/14077 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "FC_mse_neur = pd.DataFrame(data=np.zeros((len(peaks.columns),10)),\n",
    "                       index=peaks.columns, columns=['Low', 'High', 'Exons low var',\n",
    "                                                    'Exons high var', 'Exons low cons',\n",
    "                                                    'Exons high cons', 'Peaks low var',\n",
    "                                                    'Peaks high var', 'Peaks low cons',\n",
    "                                                    'Peaks high cons'])\n",
    "FC_mse_glia = pd.DataFrame(data=np.zeros((len(peaks.columns),10)),\n",
    "                       index=peaks.columns, columns=['Low', 'High', 'Exons low var',\n",
    "                                                    'Exons high var', 'Exons low cons',\n",
    "                                                    'Exons high cons', 'Peaks low var',\n",
    "                                                    'Peaks high var', 'Peaks low cons',\n",
    "                                                    'Peaks high cons'])\n",
    "\n",
    "for RBP in tqdm(peaks.columns):\n",
    "    res_glia, res_neur = mse_peaks(RBP, exons_info.iloc[tokeep.values], \n",
    "                                      peaks, PSI_glia[tokeep],\n",
    "                                      PSI_neur[tokeep])    \n",
    "    FC_mse_neur.loc[RBP] = res_neur\n",
    "    FC_mse_glia.loc[RBP] = res_glia"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c9d1d5c",
   "metadata": {},
   "source": [
    "### Plotting the MSE (Figure 2B)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "23df241e",
   "metadata": {},
   "outputs": [],
   "source": [
    "HPC_mse_neur = pd.read_csv('../../PSI_prediction/Zenodo/Human/HPC/RBP/mse_neur.csv', index_col=0)\n",
    "HPC_mse_glia = pd.read_csv('../../PSI_prediction/Zenodo/Human/HPC/RBP/mse_glia.csv', index_col=0)\n",
    "\n",
    "# Only keep exons with at least 50 exons in every category\n",
    "tk = (HPC_mse_neur['Exons low var'] > 50) & (HPC_mse_glia['Exons low var'] > 50) & (HPC_mse_neur['Exons high var'] > 50) & (HPC_mse_glia['Exons high var'] > 50)\n",
    "HPC_mse_neur = HPC_mse_neur[tk]\n",
    "HPC_mse_glia = HPC_mse_glia[tk]\n",
    "\n",
    "HPC_mse_glia2 = HPC_mse_glia[['Low', 'High']].melt()\n",
    "HPC_mse_glia2['cell type'] = 'Glia'\n",
    "\n",
    "HPC_mse_neur2 = HPC_mse_neur[['Low', 'High']].melt()\n",
    "HPC_mse_neur2['cell type'] = 'Neurons'\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "4b83d46d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 300x150 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=[3,1.5])\n",
    "sns.boxplot(data=pd.concat((HPC_mse_neur2, HPC_mse_glia2)),\n",
    "            x='value', y='variable', hue='cell type',\n",
    "           palette=['#9DC3E6', '#F4B68C'])\n",
    "sns.despine()\n",
    "plt.xlabel('MSE between cons. and var. exons')\n",
    "plt.ylabel('PSI value')\n",
    "plt.legend(bbox_to_anchor=(1.1, 1.05))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "246331d8",
   "metadata": {},
   "outputs": [],
   "source": [
    "FC_mse_neur = pd.read_csv('../../PSI_prediction/Zenodo/Human/FC/RBP/mse_neur.csv', index_col=0)\n",
    "FC_mse_glia = pd.read_csv('../../PSI_prediction/Zenodo/Human/FC/RBP/mse_glia.csv', index_col=0)\n",
    "\n",
    "# Only keep exons with at least 50 exons in every category\n",
    "tk = (FC_mse_neur['Exons low var'] > 50) & (FC_mse_glia['Exons low var'] > 50) & (FC_mse_glia['Exons high var'] > 50) & (FC_mse_neur['Exons high var'] > 50)\n",
    "FC_mse_neur = FC_mse_neur[tk]\n",
    "FC_mse_glia = FC_mse_glia[tk]\n",
    "\n",
    "FC_mse_glia2 = FC_mse_glia[['Low', 'High']].melt()\n",
    "FC_mse_glia2['cell type'] = 'Glia'\n",
    "\n",
    "FC_mse_neur2 = FC_mse_neur[['Low', 'High']].melt()\n",
    "FC_mse_neur2['cell type'] = 'Neurons'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b6512488",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAT8AAADHCAYAAABvEFRrAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAAwwklEQVR4nO3deVxU5f4H8M8MMMMqIIgSwSCJsiSKkr9IEwENg0xcSsUF3K6WFuZSmCXpVan7sxtdU+uqFzS1LCFzyR3HJTXNRM0NCdRMvQgioMg6398f/ubIbDADgyzzfb9e83o5z3nOs8yMX845zznPIyIiAmOMmRhxUzeAMcaaAgc/xphJ4uDHGDNJHPwYYyaJgx9jzCRx8GOMmSQOfowxk8TBjzFmkjj4McZMEgc/JkhNTYVIJNL6mj179hNtS2lpKT766CPI5fJ6l3H16lWIRCKkpqYarV0NYYw+MeMxb+oGsOYnJSUFPj4+KmlPPfXUE21DaWkpFixYAADo16/fE627sbTGPrVkHPyYhmeffRZBQUF65a2srIRIJIK5Of+UdCEilJWVNXUzmBo+7WV6k8vlEIlE+PrrrzFr1iy4ublBKpUiOzsbAPCf//wH3bp1g6WlJdq2bYshQ4bg4sWLKmXExcXB1tYW2dnZiIyMhK2tLdzd3TFr1iyUl5cDeHS62q5dOwDAggULhFPvuLg4AEB2djbGjx8Pb29vWFtbw83NDYMGDcK5c+fq3bd79+5h1qxZ8PLyglQqhYuLCyIjI3Hp0iUhz927d/Hmm2/Czc0NEokEXl5emDdvntBuJZFIhOnTp+PLL7+Er68vpFIp1q5dW2ufAODKlSuIiYmBi4sLpFIpfH19sXz5cmF7WVkZAgMD0alTJxQVFQnpt2/fRocOHdCvXz9UV1fX+zMwOcTY/0tJSSEAdPz4caqsrFR5EREdOHCAAJCbmxsNHz6ctm7dStu3b6eCggJasmQJAaBRo0bRjh07aN26deTl5UX29vaUlZUl1BEbG0sSiYR8fX1p6dKltG/fPpo/fz6JRCJasGABERGVlZXRrl27CABNnDiRjh07RseOHaPs7GwiIjp48CDNmjWLNm/eTAcPHqQffviBoqOjycrKii5duiTUlZubSwAoJSWl1n4XFxeTv78/2djY0MKFC2n37t2UlpZG8fHxlJGRQUREDx8+pICAALKxsaGlS5fSnj176MMPPyRzc3OKjIxUKU/5GQUEBNDGjRspIyODMjMza+3T+fPnyd7enrp27Urr1q2jPXv20KxZs0gsFtNHH30klJ2VlUV2dnY0dOhQIiKqrq6msLAwcnFxoZs3b9bnazdZHPyYQBn8tL0qKyuF4Ne3b1+V/QoLC8nKykojCFy/fp2kUinFxMQIabGxsQSAvvvuO5W8kZGR1KVLF+H9nTt3CAAlJibW2e6qqiqqqKggb29veuedd4R0fYPfwoULCQDt3btXZ54vv/xSa7s/+eQTAkB79uwR0gCQvb093b17VyVvbX2KiIigp59+moqKilTSp0+fTpaWliplbdq0iQBQcnIyzZ8/n8RisUr9TD982ss0rFu3DidPnlR51bymN2zYMJX8x44dw8OHD1VO4QDA3d0dYWFh2L9/v0q6SCTCoEGDVNICAgJw7do1vdpXVVWFJUuWwM/PDxKJBObm5pBIJLhy5YrGabY+du7cic6dO6N///4682RkZMDGxgbDhw9XSVf2Wb2PYWFhcHR01Kv+srIy7N+/H0OGDIG1tTWqqqqEV2RkJMrKynD8+HEh/+uvv4433ngDc+bMwaJFi/D+++9jwIABevaWKfFVaqbB19e31gEPV1dXlfcFBQVa04FHo8R79+5VSbO2toalpaVKmlQq1XtQYObMmVi+fDnee+89hISEwNHREWKxGJMmTcLDhw/1KqOmO3fuwMPDo9Y8BQUF6NChA0QikUq6i4sLzM3Nhc9ASdtnUVvZVVVVWLZsGZYtW6Y1T35+vsr7CRMmYOXKlZBIJHj77bf1ros9xsGPGUw9ADg5OQEAbt26pZH35s2bcHZ2Nmr969evx7hx47BkyRKV9Pz8fDg4OBhcXrt27XDjxo1a8zg5OeGXX34BEan0Py8vD1VVVRp9VP+MauPo6AgzMzOMHTsW06ZN05qnY8eOwr8fPHiAsWPHonPnzvjvf/+LSZMm4ccff9S7PvYIn/ayBgsODoaVlRXWr1+vkn7jxg1kZGQgPDzc4DKlUikAaD2SE4lEwnalHTt24K+//jK4HgB4+eWXkZWVhYyMDJ15wsPDcf/+fWzZskUlfd26dcL2uujqk7W1NUJDQ3H69GkEBAQgKChI46X8AwMAU6dOxfXr15Geno41a9Zg69at+Oyzz/TtLvt/fOTHGszBwQEffvgh3n//fYwbNw6jRo1CQUEBFixYAEtLSyQmJhpcpp2dHWQyGX788UeEh4ejbdu2cHZ2hqenJ1555RWkpqbCx8cHAQEBOHXqFP73f/8XTz/9dL3aP2PGDGzatAmDBw9GQkICevXqhYcPH+LgwYN45ZVXEBoainHjxmH58uWIjY3F1atX0bVrVxw5cgRLlixBZGRkrdcL9enT559/jj59+uDFF1/EG2+8AU9PT5SUlCA7Oxvbtm0TAvPq1auxfv16pKSkwN/fH/7+/pg+fTree+899O7dG7169arXZ2CSmnrEhTUfytHekydPat2uHO39/vvvtW5fvXo1BQQEkEQiIXt7exo8eDCdP39eJU9sbCzZ2Nho7JuYmEjqP8d9+/ZRYGAgSaVSAkCxsbFE9Gh0eeLEieTi4kLW1tbUp08fOnz4MIWEhFBISIiwv76jvcoy4+PjycPDgywsLMjFxYWioqJUbp0pKCigqVOnkqurK5mbm5NMJqO5c+dSWVmZSlkAaNq0aVrr0dUnZXsnTJhAbm5uZGFhQe3ataMXXniBFi1aREREZ8+eJSsrK5V9iB7dGtSzZ0/y9PSkwsLCOvvKHhER8eptjDHTw9f8GGMmiYMfY8wkcfBjjJkkDn6MMZPEwY8xZpI4+DHGTBIHPx2ICMXFxeA7gRhrnTj46VBSUgJ7e3uUlJQ0dVMYY42Agx9jzCRx8GOMmSQOfowxk8SzuhgZEaGiokLvvJWVlQAACwsLg+aAAwCJRGLwPoyxRzj4GVlFRQXmzZv3ROpavHixxrx2jDH98GkvY8wk8ZFfIwp4OQ5iMwud26urKnFuVyoAoOvAOJiZ686rpKiuxNmdqUZqIWOmi4NfIxKbWegV0ADAzFz/vIyxhuPTXsaYSeIjPyOoOcJrSo/D1ew3jzyzloaP/IxAOcI7b9484dYVU1Cz3/re3sNYc8HBjzFmkjj4MaM6f/48Fi9ejJ07d2Lx4sU4f/68xraaaQ2tR1dZNbfXp15jtrW1a6mfFQc/ZjQVFRVIT09HYWEhMjIyUFhYiPT0dFRUVKhsU6YZox5tZdXcnpaWhrS0NIPqNWZbW7uW/Flx8GNGk5GRgeLiYgCPB36Ki4uRkZGhsk2ZZox6tJWlvt3Qeo3Z1tauJX9WPNprZDX/8jXGyG/NMpv6r2zN+gsKCpCRkaHRZyIS0pXbiAgHDhxAUFAQnJ2dDaozPz9fpR71stS3q7elrnrrKp891tI/Kw5+RlAzCCQlJQn/VlRXARYSo9alqK4S/r1gwQKjlt0QP/74o85gr1AoNNKICOnp6Zg8ebLet8go99GVPmnSJK3b9a23rvINaWtr1xo+Kz7tZUbxxx9/GHSkq1AokJWVhby8PL33ycvLQ1ZWlkYwVZZ16dIlrdv1rbeu8g1pa2vXGj4rPvIzAonk8dHd3LlzhaM/sZnxP96aZSYmJqrU/aRVVFQIR5+dOnUyKACKxWJ4e3vDxcVF7/pcXFzQuXNnZGdnq/ynU5bl4+Ojdbu+9dZVviFtbe1aw2fFR35GVjMYNcZhf80yJRIJpFJpk71q9vXVV1/V2V+xWKyxTSQSYciQIQZ9RiKRCEOHDtWaPmTIEIjFYq3b9a23rvKb+2nck9QaPisOfswonJycEBYWpjXIhYWFITw8XNgmEokQGhpar4vizs7OKvWol6W+Xb0tddVbV/nssZb+WXHwY0YTFhaGNm3aAHh8hNqmTRuEhYWpbFOmGaMebWWpbze0XmO2tbVryZ8VBz9mNBKJBEOHDoWjoyPCwsLg6OiIoUOHQiKRqGxTphmjHm1l1dw+bNgwDBs2zKB6jdnW1q4lf1YiMqVpSAxQXFwMe3t7FBUVCX/ZdFGf1eWDDz4AAHR/ZXKtc/RVV1Uic/sqvfJq26epp7HnWV1YS9asjvzi4uIQHR3d1M0wmEgkEgYBTCkAmGq/WevQrIIfY4w9KS0m+B08eBC9evWCVCqFq6srEhISUFX16GmHbdu2wcHBQbjfKDMzEyKRCHPmzBH2nzJlCkaNGtUobSMilJWVoaSkBPfv3xfSq8pLUVVZgeqqSp0vpdry1Hwpqk1nvkDGGlO97sL9+uuv8eWXXyI3NxfHjh2DTCZDcnIyOnbsiMGDBxu7jfjrr78QGRmJuLg4rFu3DpcuXcLkyZNhaWmJjz76CH379kVJSQlOnz6Nnj174uDBg3B2dsbBgweFMuRyOd555x2ddZSXl6O8vFx4r3xYWx8VFRXCdb6aft+7Qe8ylAsZMcaeDIOP/FauXImZM2ciMjIS9+7dQ3V1NQDAwcEBycnJxm4fAGDFihVwd3fHF198AR8fH0RHR2PBggX49NNPoVAoYG9vj+7du0MulwN4HOjOnDmDkpIS3L59G1lZWejXr5/OOpKSkmBvby+83N3dG6UvjLHmweAjv2XLlmHVqlWIjo7Gxx9/LKQHBQVh9uzZRm2c0sWLFxEcHKxyUb137964f/8+bty4AQ8PD/Tr1w9yuRwzZ87E4cOHsWjRIqSlpeHIkSO4d+8e2rdvDx8fH511zJ07FzNnzhTeFxcX1zsATurjidVHrgKo/RE0IhKmvbewsDB40KAl3VbAWHNjcPDLzc1FYGCgRrpUKsWDBw+M0ih1RKR1Bg7g8c20/fr1w5o1a3DmzBmIxWL4+fkhJCQEBw8eRGFhIUJCQmqtQzlqaQwWZo8PqJWPoOliaWlplDoZY4Yx+LS3Y8eOyMzM1EjfuXMn/Pz8jNEmDX5+fjh69KjKQ/NHjx6FnZ0d3NzcAEC47pecnIyQkBCIRCKEhIRALpdDLpfXGfwYY6bF4CO/OXPmYNq0aSgrKwMR4cSJE/jmm2+QlJSE1atXN7hBRUVFGsH1b3/7G5KTk/HWW29h+vTpuHz5MhITEzFz5kyIxY/it/K63/r16/H5558DeBQQX3vtNVRWVtZ6va8hat7oq08evhmYsebB4OA3fvx4VFVV4d1330VpaSliYmLg5uaGzz//HCNHjmxwg+RyucZpdWxsLH766SfMmTMH3bp1Q9u2bTFx4kSNEdbQ0FD89ttvQqBzdHSEn58fbt68CV9f3wa3TZua0zrVlmfevHkAmv6pDMbYIw16vC0/Px8KhaJFzN1lKH0fbysvLxcCm9IbIV5YeTAHwKNgB4CDH2PNTINm22wpU9c0Rzt37kRGRgYkEgliYmLg7+8vbDt//jy2bNmC6OhovdIZY4YzOPh17Nix1mtWOTk5DWqQKbh//76w8Et5eTk2b94Mb29vSCQSYSnAoqIipKen15nOGKsfg4PfjBkzVN5XVlbi9OnT2LVrl8rjZEy3tWvXqoxcl5SUICMjAwMHDtS6FGBt6Yyx+jE4+MXHx2tNX758OX799dcGN6g1qG15yezsbOTm5mrss3//fjzzzDNalwLUld5SlghkrDky2nx+OTk56N69u0HPxDZn+g54lJSUaIz21nzCQ52VlRUePnyoc1t5ebnGgjCWlpZ4+PChSlAVi8Xo1KlTi1gikLHmyGizumzevBlt27Y1VnGtlq7Ap9ymbSnA0tJSjVXRWtISgYw1Rwaf9gYGBqocaRARbt++jTt37mDFihVGbVxLoG3QwVz8+PNJTEwE8HiBcWMe+bWUJQIZa44MDn7qMy2LxWK0a9cO/fr1q3XiAFOivrxkTSNHjkRKSorWfcaNG4dVq1ZppI8dOxarVq1SCX4taYlAxpojg4Of8kiG1U+nTp3QsWNHjUGP8PBweHt7IywsDPv37xcmcwgNDdWZzoMdjNWfXtf8iouL9X6xusXGxqocsdnZ2QlL/ulaCrAlLxHIWHOkV/BzcHCAo6NjrS9lHlY3W1tbYbFnqVSK4cOHC6fHupYCbMlLBDLWHOl1q0vN6eDr0lqmjtL3Vhciwv3791Vud1F/tlf5hAbAs7ow1lzodc2vtQS0xiASieo8ClMe4THGmo96T2xQWlqK69evazzBEBAQ0OBGMcZYYzM4+N25cwfjx4/Hzp07tW5XLmhkyiqrFXVnYow1KYOf8JgxYwYKCwtx/PhxWFlZYdeuXVi7di28vb2xdevWxmhji6Pr0TbGWPNh8JFfRkYGfvzxRzz33HMQi8WQyWQYMGAA2rRpg6SkJERFRTVGOxljzKgMDn4PHjwQHqlq27Yt7ty5g86dO6Nr16747bffjN7AlkAikWDRokVal6HkW1IYa54MDn5dunTB5cuX4enpie7du+Orr76Cp6cnvvzyS7i6ujZGG5s9kUgES0tLXoaSsRakXpOZ3rp1C8CjR90iIiKwYcMGSCQSpKamGrt9jDHWKBo8n19paSkuXboEDw+PVvWsqb43OTPGWiaDR3vVn/awtrZGjx49WlXgY4y1fgYHvwEDBsDDwwMJCQn4/fffG6NNjDHW6AwOfjdv3sS7776Lw4cPIyAgAAEBAfjHP/6BGzduNEb7GGOsUTToml9ubi42btyIb775BpcuXULfvn2RkZFhzPY1Gb7mx1jr1uABj+rqauzcuRMffvghzp4922oeb+Pgx1jrVu+JDX7++Wds2LABmzdvRllZGV599VUsWbLEmG1rNYhIYwKI+paj7UbqloKn82LNicHB7/3338c333yDmzdvon///khOTkZ0dDSsra0bo32tQkVFBebNm9fUzWhyixcv5qm9WLNhcPCTy+WYPXs2RowYwbe3MMZaLIOD39GjRxujHSYj4OU4iM0s6rVvdVUlzu1KBQB0HRgHM/P6lfMkKaorcXZnalM3gzEN9b7mx+pHbGZhlKBlZm6cchgzVQbf58cYY60BH/kZQc3RXB7RZMbGv6/GwUd+RqAczZ03b55RbmlhrCb+fTUODn6MMRXnz5/H4sWLcf78+Seyn7HL0Jfep72Ojo56HW7fvXu3QQ1ijDWdiooKpKeno6ioCOnp6fD29tZrNvL67mfsMgyhd/BLTk5utEYwxpqHjIwMFBcXA3j0iGdGRgYGDhzYaPsZuwxD6B38YmNjG60RrYm2azI10xr4KHWLU7O/fL2qfp7U55afn4+MjAzhOyMiHDhwAEFBQbU+0FDf/YxdhqEaNNpbVlaGTZs24cGDBxgwYAC8vb2N1a4WpeaPc8GCBbXmVVRXARams6iRorpK+Hddnw2rW0VFRaM8IkhESE9P15k+efJkrZe96rufscuoD70HPObMmYP4+HjhfUVFBYKDgzF58mS8//77CAwMxLFjx4zeQMZY48vLy0NWVhYUCoVKukKhQFZWFvLy8oy6n7HLqA+9j/x27typMmvLhg0bcO3aNVy5cgUeHh6YMGECFi1ahB07djRKQ5uzmhdlExMTNS7SVlRUCEc9YjPTurWyZn+1fTasbjV/P431+bm4uKBz587Izs5WCUJisRje3t7CcrXG2s/YZdSH3v8Tr1+/Dj8/P+H9nj17MHz4cMhkMgBAfHw8IiMjjd/CFkYikdR6WmJqN6jW7G9dnw1rOiKRCEOHDsU//vEPjfQhQ4bo/N3Wdz9jl1Efep/2isVilYvXx48fx/PPPy+8d3BwQGFhoXFbxxh7YpydnREWFiYEG5FIhNDQ0DoHHOq7n7HLMJTewc/Hxwfbtm0D8OhGxOvXryM0NFTYfu3aNbRv3974LWSMPTFhYWHCzOVt2rRBWFhYo+5n7DIMYdCAR0JCAsLDwxEeHo7IyEh07NhR2P7TTz+hV69ejdJIxtiTIZFIMHToUDg6OmLo0KF6X2Os737GLsMQel/zGzZsGH766Sfs2LEDL730Et566y2V7dbW1njzzTeN3sDU1FTMmDED9+7d03ufuLg43Lt3D1u2bDF6exhr7fz9/eHv7//E9jN2GfoyaOixf//+6N+/v9ZtiYmJBleuK0jJ5XKEhoaisLAQI0aMaPYDKRKJBIsXLxb+zZgx8e+rceh92ltaWopp06bBzc0NLi4uiImJQX5+fmO2DQBgZWXVaEPdxiISiSCVSiGVSk1uNJc1Pv59NQ69g19iYiJSU1MRFRWFkSNHYu/evXjjjTcas20AHp32Ojg4qKQtWrQILi4usLOzw6RJk5CQkIDu3btr7Lt06VK4urrCyckJ06ZNE1Y+Y4wxvU9709PTsWbNGowcORIAMGbMGPTu3RvV1dUwMzNrtAaq27BhAxYvXowVK1agd+/e+Pbbb/Hpp5+qDL4AwIEDB+Dq6ooDBw4gOzsbI0aMQPfu3TF58uRGbZ+2ZSprvq8oLYZIbA6RmbnBf8UVVY+Dd3VVywjkiuqW0U5mevQOfn/++SdefPFF4X2vXr1gbm6Omzdvwt3dvd4N2L59O2xtbVXSalv4fNmyZZg4cSLGjx8PAJg/fz727NmD+/fvq+RzdHTEF198ATMzM/j4+CAqKgr79+/XGfzKy8tRXl4uvFfOLmGoupapvJCxqV7lqlMuZMQYqx+9T3urq6s1Lraam5ujqqpKxx76CQ0NRWZmpspr9erVOvNfvnxZ45YabbfY+Pv7qxyRurq61vqMYFJSEuzt7YVXQwI6Y6z50/vIj4gQFxen8nhSWVkZpk6dChsbGyFN2+wMtbGxsUGnTp1U0m7cuFHrPuqni9qmibKwUF3ZTCQSaTw4XdPcuXMxc+ZM4X1xcXGDA+AbIV6wMHv896WyWoGVB3O0bqtLzX3nzp0LGxubFnfxm0cqWXPSoPn8xowZY9TG6KNLly44ceIExo4dK6T9+uuvDS5XOZpmTBZmYkh0BLjattXF1taWn5FlrIH0Dn4pKSmN2Q69vfXWW5g8eTKCgoLwwgsvYNOmTTh79iy8vLyaummMsRakxc2vNHr0aOTk5GD27NkoKyvD66+/jri4OJw4caJJ2qO+rGBrwksmstZMRK1gXvUBAwagQ4cO+Prrr41WZnFxMezt7VFUVCQ8bK1NeXm5MLqrvAtf+f7tsE4qp7YV1Qr8KyNb67a61Nx38eLFT+S0V71vfKrNWpMWt3RlaWkp/vnPf+L8+fO4dOkSEhMTsW/fPl5jpBlQX3awrmUIn+QyhYypa3HBTyQS4aeffsKLL76Inj17Ytu2bUhLS9P5zDF7MpTLDhYWFiI9PR33799Xea/txu/atjPW2FrcNT8rKyvs27evqZvB1KgvO7h27dpalyF80ssUMqauxQW/5kz96MWYl1ObYglIfevRtuxgbm6usF19GcKmWKaQMXUc/BqotmUrqxQEYw0RVCkeB7+mWAJS15KJupYd1JVv0qRJTbJMIWPqWtw1P9a86Fp2UJ1yGcJLly41yTKFjKnjI78GUl+2Enh8ZGYuNt4RTM2yntQSkPosmahr2UF1ymUIfXx8mmSZQsbU8ZGfEUkkEpUgYczTN21LQDb2S58Aq1x2UJ98Q4YMgVgs1pq/sZcpZEwdBz/WYNqWHezYsaPOZQibYplCxtRx8GNGob7sYGxsbK3LED7pZQoZU8fBjxmF+rKDtra2tS5D+KSXKWRMHQ94MKNRX3awrmUIn+QyhYyp4+DXQOrLCramx7R4yUTWmnHwayDlsoKtUWvuG2N8zY8xZpL4yK8RVVYrdL5X32ZoWYyxhuHg14iUCw4Zuo0x1vj4tJcxZpJaxTT2jUHfaezV1Vz3Qtu2yspKAI+W1qzvo1y8ngZjDcenvUZW1wippaXlE2wNY0wXPu1ljJkkDn6MMZPEp706KC+FKteZYIw1T3Z2dvW6Bs7BT4eSkhIAgLu7exO3hDFWG0MHJZV4tFcHhUKBmzdv1vlXpbi4GO7u7vjzzz/r9QU0N9yf5q+19amh/eEjPyMTi8V4+umn9c7fpk2bVvFDVOL+NH+trU9Puj884MEYM0kc/BhjJomDXwNJpVIkJia2mqmfuD/NX2vrU1P1hwc8GGMmiY/8GGMmiYMfY8wkcfBjjJkkDn6MMZNk8sFvxYoV6NixIywtLdGzZ08cPny41vwHDx5Ez549YWlpCS8vL3z55ZcaedLS0uDn5wepVAo/Pz/88MMPDa63OfcnKSkJzz33HOzs7ODi4oLo6Ghcvny5xfanpqSkJIhEIsyYMaOhXRE0VZ/++usvjBkzBk5OTrC2tkb37t1x6tSpFtmfqqoqfPDBB+jYsSOsrKzg5eWFhQsXQqEwYLkHMmHffvstWVhY0KpVq+jChQsUHx9PNjY2dO3aNa35c3JyyNramuLj4+nChQu0atUqsrCwoM2bNwt5jh49SmZmZrRkyRK6ePEiLVmyhMzNzen48eP1rre59yciIoJSUlLo999/p8zMTIqKiiIPDw+6f/9+i+yP0okTJ8jT05MCAgIoPj6+QX1p6j7dvXuXZDIZxcXF0S+//EK5ubm0b98+ys7ObpH9WbRoETk5OdH27dspNzeXvv/+e7K1taXk5GS9227Swa9Xr140depUlTQfHx9KSEjQmv/dd98lHx8flbQpU6bQ888/L7x//fXXaeDAgSp5IiIiaOTIkfWuV19N1R91eXl5BIAOHjxoaBdUNGV/SkpKyNvbm/bu3UshISFGC35N1af33nuP+vTp09Dma2iq/kRFRdGECRNU8gwdOpTGjBmjd9tN9rS3oqICp06dwksvvaSS/tJLL+Ho0aNa9zl27JhG/oiICPz666/C9PS68ijLrE+9zbk/2hQVFQEA2rZta3A/lJq6P9OmTUNUVBT69+9f7z6oa8o+bd26FUFBQXjttdfg4uKCwMBArFq1qsX2p0+fPti/fz+ysrIAAGfOnMGRI0cQGRmpd/tNNvjl5+ejuroa7du3V0lv3749bt++rXWf27dva81fVVWF/Pz8WvMoy6xPvc25P+qICDNnzkSfPn3w7LPP1rc7Tdqfb7/9Fr/99huSkpLq3X5tmrJPOTk5WLlyJby9vbF7925MnToVb7/9NtatW9ci+/Pee+9h1KhR8PHxgYWFBQIDAzFjxgyMGjVK7/ab/Kwu6lPhEFGt0+Noy6+erk+Zhtarr6bqj9L06dNx9uxZHDlyxKB2G9K+xuzPn3/+ifj4eOzZs6fR1ltpiu9IoVAgKCgIS5YsAQAEBgbi/PnzWLlyJcaNG1e/juhZtz751dPrKnPTpk1Yv349Nm7cCH9/f2RmZmLGjBl46qmnEBsbq1e7TTb4OTs7w8zMTOMvVF5ensZfHaUOHTpozW9ubg4nJ6da8yjLrE+9zbk/Nb311lvYunUrDh06ZNB0YNo0VX9OnTqFvLw89OzZU9heXV2NQ4cO4YsvvkB5eTnMzMxaVJ8AwNXVFX5+fip5fH19kZaWVq++NHV/5syZg4SEBIwcORIA0LVrV1y7dg1JSUl6Bz+TPe2VSCTo2bMn9u7dq5K+d+9evPDCC1r3CQ4O1si/Z88eBAUFwcLCotY8yjLrU29z7g/w6K/y9OnTkZ6ejoyMDHTs2LHe/Wjq/oSHh+PcuXPIzMwUXkFBQRg9ejQyMzPrHfiask8A0Lt3b43bj7KysiCTyVpkf0pLSyEWq4YvMzMzvtVFX8ph+jVr1tCFCxdoxowZZGNjQ1evXiUiooSEBBo7dqyQXzlM/84779CFCxdozZo1GsP0P//8M5mZmdHHH39MFy9epI8//ljnrS666m1p/XnjjTfI3t6e5HI53bp1S3iVlpa2yP6oM+Zob1P16cSJE2Rubk6LFy+mK1eu0IYNG8ja2prWr1/fIvsTGxtLbm5uwq0u6enp5OzsTO+++67ebTfp4EdEtHz5cpLJZCSRSKhHjx4qt2fExsZSSEiISn65XE6BgYEkkUjI09OTVq5cqVHm999/T126dCELCwvy8fGhtLQ0g+ptaf0BoPWVkpLSIvujzpjBj6jp+rRt2zZ69tlnSSqVko+PD/373/9usf0pLi6m+Ph48vDwIEtLS/Ly8qJ58+ZReXm53u3mKa0YYybJZK/5McZMGwc/xphJ4uDHGDNJHPwYYyaJgx9jzCRx8GOMmSQOfowxk8TBr5ny9PREcnJyUzeDNYHU1FQ4ODg0dTNaPQ5+auLi4iASiTB16lSNbW+++SZEIhHi4uKEtLy8PEyZMgUeHh6QSqXo0KEDIiIicOzYMSGPp6cnRCKRxuvjjz9u1L589NFH6N69e6PWwVhLZbKzutTG3d0d3377LT777DNYWVkBAMrKyvDNN9/Aw8NDJe+wYcNQWVmJtWvXwsvLC//973+xf/9+3L17VyXfwoULMXnyZJU0Ozu7xu0IY3oiIlRXV8Pc3HRCAh/5adGjRw94eHggPT1dSEtPT4e7uzsCAwOFtHv37uHIkSP45JNPEBoaCplMhl69emHu3LmIiopSKdPOzg4dOnRQednY2NTajpKSEsTExMDW1hZPPfUUli1bprK9qKgIf/vb3+Di4oI2bdogLCwMZ86cAfDo1GnBggU4c+aMcKSZmpqKWbNmYdCgQUIZycnJEIlE2LFjh5DWpUsXfPXVV8L7lJQU+Pr6wtLSEj4+PlixYoVKO/766y+MGDECjo6OcHJywuDBg3H16lVhe1xcHKKjo7F06VK4urrCyckJ06ZNE2bu1UU5+7ClpSWcnZ0xdOhQYVthYSHGjRsHR0dHWFtb4+WXX8aVK1eE7cpTx927d8PX1xe2trYYOHAgbt26JeSRy+Xo1asXbGxs4ODggN69e+PatWu1tqmmXbt2oU+fPnBwcICTkxNeeeUV/PHHH8L2q1evQiQSIT09HaGhobC2tka3bt1UzgqUbfXw8IC1tTWGDBmCgoKCWusNDg5GQkKCStqdO3dgYWGBAwcOAADWr1+PoKAg4XcXExODvLw8lb6LRCLs3r0bQUFBkEqlei2iVdtv7s6dO+jQoYMwZyAA/PLLL5BIJNizZw+A5vG9CQx4ftkkxMbG0uDBg+mf//wnhYeHC+nh4eH02Wef0eDBgyk2NpaIiCorK8nW1pZmzJhBZWVlOsuUyWT02WefGdQOmUxGdnZ2lJSURJcvX6Z//etfZGZmRnv27CEiIoVCQb1796ZBgwbRyZMnKSsri2bNmkVOTk5UUFBApaWlNGvWLPL391eZZWXr1q1kb29P1dXVREQUHR1Nzs7ONGfOHCIiunXrFgGgixcvEhHRv//9b3J1daW0tDTKycmhtLQ0atu2LaWmphIR0YMHD8jb25smTJhAZ8+epQsXLlBMTAx16dJFeMg8NjaW2rRpQ1OnTqWLFy/Stm3byNrautYH67dv305mZmY0f/58unDhAmVmZtLixYuF7a+++ir5+vrSoUOHKDMzkyIiIqhTp05UUVFBREQpKSlkYWFB/fv3p5MnT9KpU6fI19eXYmJihO/O3t6eZs+eTdnZ2XThwgVKTU01aBGpzZs3U1paGmVlZdHp06dp0KBB1LVrV+Gzzc3NJQDk4+ND27dvp8uXL9Pw4cNJJpNRZWUlEREdP36cRCKR8D1//vnn5ODgQPb29jrrXbZsGXl4eJBCoVBJc3NzE+pes2YN/fTTT/THH3/QsWPH6Pnnn6eXX35ZyH/gwAECQAEBAbRnzx7Kzs6m/Pz8Wvtb12+OiGjHjh1kYWFBJ0+epJKSEurUqZPKpBDN4XtT4uCnRhn87ty5Q1KplHJzc+nq1atkaWlJd+7cUQl+RI/+Azg6OpKlpSW98MILNHfuXDpz5oxKmcoZL2xsbFReBw4c0NkOmUymsYjLiBEjhB/w/v37qU2bNhpB95lnnqGvvvqKiIgSExOpW7duKtvv3btHYrGYfv31V1IoFOTk5ERJSUn03HPPERHRxo0bqX379kJ+d3d32rhxo0oZf//73yk4OJiIHv0n69Kli8p/xPLycrKysqLdu3cLn6lMJqOqqiohz2uvvUYjRozQ2f/g4GAaPXq01m1ZWVkEgH7++WchLT8/n6ysrOi7774jokf/iQCorE62fPlyoW8FBQUEgORyuc42GEq5cNO5c+eI6HHwW716tZDn/PnzKn9cRo0apfV7ri345eXlkbm5OR06dEhICw4OFv6AaXPixAkCQCUlJUT0OPht2bJF7/7p85sjInrzzTepc+fONHr0aHr22Wfp4cOHRNT8vjc+7dXB2dkZUVFRWLt2LVJSUhAVFQVnZ2eNfMOGDcPNmzexdetWREREQC6Xo0ePHkhNTVXJN2fOHJUJMjMzM/E///M/tbYhODhY4/3FixcBPJpx+P79+3BycoKtra3wys3NVTn1Umdvb4/u3btDLpfj3LlzEIvFmDJlCs6cOYOSkhLI5XKEhIQAeHQa8+eff2LixIkqdSxatEio49SpU8jOzoadnZ2wvW3btigrK1Nph7+/v8pEoK6uriqnYeoyMzMRHh6uddvFixdhbm6u8vk5OTmhS5cuwucDANbW1njmmWe01tm2bVvExcUhIiICgwYNwueff65yaqWPP/74AzExMfDy8kKbNm2ESVyvX7+uki8gIEClDQCEdly8eFHr91ybdu3aYcCAAdiwYQMAIDc3F8eOHcPo0aOFPKdPn8bgwYMhk8lgZ2eHfv36aW1bUFCQvt3V+ze3dOlSVFVV4bvvvsOGDRuE5QCay/emZDpXN+thwoQJmD59OgBg+fLlOvNZWlpiwIABGDBgAObPn49JkyYhMTFRZVTY2dkZnTp1anCblOsYKBQKuLq6Qi6Xa+Sp6zaJfv36QS6XQyKRICQkBI6OjvD398fPP/8MuVwuLNCtnBV31apVGoFaGcgUCgV69uwp/EesqV27dsK/lbP01uxHbbPuKgeatCEds7CR2joP2uqsuW9KSgrefvtt7Nq1C5s2bcIHH3yAvXv34vnnn9dZd02DBg2Cu7s7Vq1ahaeeegoKhQLPPvssKioqVPLVbEfN76+2vtRl9OjRiI+Px7Jly4R1LLp16wYAePDgAV566SW89NJLWL9+Pdq1a4fr168jIiJCo211XXeuSd/fXE5ODm7evAmFQoFr164Jwb+5fG9KHPxqMXDgQOHHEhERofd+fn5+2LJlS4PrP378uMZ7Hx8fAI8GZW7fvg1zc3N4enpq3V8ikaC6ulojvV+/flizZg3Mzc2FpRlDQkLw7bffIisrSzjya9++Pdzc3JCTk6NyVFFTjx49sGnTJuECuLEEBARg//79GD9+vMY2Pz8/VFVV4ZdffhGmNi8oKEBWVhZ8fX0NqicwMBCBgYGYO3cugoODsXHjRr3+ExUUFODixYv46quv8OKLLwJAvRZt8vPz0/o91yU6OhpTpkzBrl27sHHjRowdO1bYdunSJeTn5+Pjjz+Gu7s7AODXX381uG3q9PnNVVRUYPTo0RgxYgR8fHwwceJEnDt3Du3bt28W35uKBp84tzLKa35KRUVFVFRUJLyvec0vPz+fQkND6euvv6YzZ85QTk4Offfdd9S+fXuVBZVlMhktXLhQZYr3W7duqZSrTiaTUZs2beiTTz6hy5cv0xdffEFmZma0a9cuInp08blPnz7UrVs32rVrF+Xm5tLPP/9M8+bNo5MnTxIR0YYNG8jGxoZOnz5Nd+7cEa7VKK/7mZmZ0e+//05ERFu2bCEzMzNq166dSjtWrVpFVlZWlJycTJcvX6azZ8/Sf/7zH/r000+J6PGAR79+/ejQoUOUk5NDcrmc3n77bfrzzz+1fqZERPHx8Roz/NZ04MABEovFwoDH2bNn6ZNPPlH5Hvz8/Ojw4cOUmZlJAwcO1Lhwrn7d7IcffiDlTz4nJ4cSEhLo6NGjdPXqVdq9eze1bduWVqxYQUREv/zyC3Xp0oVu3LihtX3V1dXk5OREY8aMoStXrtD+/fvpueeeIwD0ww8/ENHja36nT58W9issLCQAwvXeY8eOkUgkEr7nZcuW1TngoRQTE0PdunUjkUikcsE/Ly+PJBIJzZkzh/744w/68ccfqXPnziptUV7zKywsFPZbtmwZhYWF6axPn9/c7NmzydPTk4qKiqi6upr69u1LUVFRQhmN/b0ZgoOfGm3/UWuqGfzKysooISGBevToQfb29mRtbU1dunShDz74QGX9CplMpnWa9ylTpuisRyaT0YIFC+j1118na2trat++PSUnJ6vkKS4uprfeeoueeuopsrCwIHd3dxo9ejRdv35daN+wYcPIwcFBY1r5nj17Urt27YSBioKCAhKJRDR8+HCNtmzYsIG6d+9OEomEHB0dqW/fvpSeni5sv3XrFo0bN46cnZ1JKpWSl5cXTZ48WQju9Ql+RERpaWlCvc7OzjR06FBh2927d2ns2LFkb29PVlZWFBERQVlZWcL2uv4T3b59m6Kjo8nV1ZUkEgnJZDKaP3++MFqqDA65ubk627d3717y9fUlqVRKAQEBJJfLDQ5+RI8GjZ5++mmysrKiQYMG0dKlS/UKfjt27CAA1LdvX41tGzduJE9PT5JKpRQcHExbt26tM/glJiaSTCartc7afnMHDhwgc3NzOnz4sJD/2rVrZG9vLwSnxv7eDMHT2DPGTBKP9jLGTBIHP8aYSeLgxxgzSRz8GGMmiYMfY8wkcfBjjJkkDn6MMZPEwY8xZpI4+DHGTBIHP8aYSeLgxxgzSRz8GGMm6f8AVRh4GV4F3pQAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 300x150 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=[3,1.5])\n",
    "sns.boxplot(data=pd.concat((FC_mse_neur2, FC_mse_glia2)),\n",
    "            x='value', y='variable', hue='cell type',\n",
    "           palette=['#9DC3E6', '#F4B68C'])\n",
    "plt.legend([], [], frameon=False)\n",
    "sns.despine()\n",
    "plt.xlabel('MSE between cons. and var. exons')\n",
    "plt.ylabel('PSI value')\n",
    "plt.title('Frontal cortex')\n",
    "# plt.savefig('../../Evaluate models/Figures/boxplot_RBP_FC.pdf', bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "573c8465",
   "metadata": {},
   "source": [
    "### Check whether the difference between neurons and glia is significant"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "2c579717",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WilcoxonResult(statistic=216.0, pvalue=3.578079338965863e-08)\n",
      "WilcoxonResult(statistic=105.0, pvalue=4.0311466518183515e-10)\n",
      "WilcoxonResult(statistic=473.0, pvalue=8.982839066503265e-15)\n",
      "WilcoxonResult(statistic=192.0, pvalue=9.51518447580423e-18)\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import wilcoxon\n",
    "\n",
    "print(wilcoxon(FC_mse_glia2[FC_mse_glia2['variable'] == 'Low']['value'], FC_mse_neur2[FC_mse_neur2['variable'] == 'Low']['value']))\n",
    "print(wilcoxon(FC_mse_glia2[FC_mse_glia2['variable'] == 'High']['value'], FC_mse_neur2[FC_mse_neur2['variable'] == 'High']['value']))\n",
    "\n",
    "print(wilcoxon(HPC_mse_glia2[HPC_mse_glia2['variable'] == 'Low']['value'], HPC_mse_neur2[HPC_mse_neur2['variable'] == 'Low']['value']))\n",
    "print(wilcoxon(HPC_mse_glia2[HPC_mse_glia2['variable'] == 'High']['value'], HPC_mse_neur2[HPC_mse_neur2['variable'] == 'High']['value']))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "22b47976",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}