Pass_miRNAs_through_model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append(\"../../Chapter_3/\")\n",
    "\n",
    "import torch\n",
    "from Bio import SeqIO\n",
    "import yaml\n",
    "\n",
    "import pandas as pd\n",
    "import pickle\n",
    "\n",
    "from Utils.train import BlockInputShapeReformatter\n",
    "\n",
    "import seaborn as sbn\n",
    "import yaml"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "unrefined_model = torch.load(\"../Models/NonRefinedMainModel.torch\")\n",
    "refined_model = torch.load(\"../Models/TatRefinedMainModel.torch\")\n",
    "\n",
    "unrefined_model.eval()\n",
    "refined_model.eval();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"../Data/EncodedData/EncodedCARESPositivemiRNACombos.pickle\",'rb') as datafile:\n",
    "    positive_cares_mirna_combos = pickle.load(datafile)\n",
    "    \n",
    "with open(\"../Data/EncodedData/CARESPositivemiRNAComboIDs.pickle\",'rb') as datafile:\n",
    "    positive_combo_labels = pickle.load(datafile)\n",
    "    \n",
    "with open(\"../Data/EncodedData/EncodedCARESNegativemiRNACombos.pickle\",'rb') as datafile:\n",
    "    negative_cares_mirna_combos = pickle.load(datafile)\n",
    "    \n",
    "with open(\"../Data/EncodedData/CARESNegativemiRNAComboIDs.pickle\",'rb') as datafile:\n",
    "    negative_combo_labels = pickle.load(datafile)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "shape_reformatter = BlockInputShapeReformatter()\n",
    "BLOCKNAME = \"StackedRes\"\n",
    "\n",
    "cares_interacting_mirna_combos = [shape_reformatter.reformat_input(BLOCKNAME, positive_cares_mirna_combos[i]) for i in range(2)]\n",
    "cares_non_interacting_mirna_combos = [shape_reformatter.reformat_input(BLOCKNAME, negative_cares_mirna_combos[i]) for i in range(2)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Data is saved as (protein,RNA) order, but needs to be input as (RNA,protein) order\n",
    "unrefined_interacting_combo_predictions = unrefined_model(cares_interacting_mirna_combos[-1].cuda(), cares_interacting_mirna_combos[0].cuda()).detach().cpu().numpy().flatten()\n",
    "unrefined_noninteracting_combo_predictions = unrefined_model(cares_non_interacting_mirna_combos[-1].cuda(), cares_non_interacting_mirna_combos[0].cuda()).detach().cpu().numpy().flatten()\n",
    "\n",
    "refined_interacting_combo_predictions = refined_model(cares_interacting_mirna_combos[-1].cuda(), cares_interacting_mirna_combos[0].cuda()).detach().cpu().numpy().flatten()\n",
    "refined_noninteracting_combo_predictions = refined_model(cares_non_interacting_mirna_combos[-1].cuda(), cares_non_interacting_mirna_combos[0].cuda()).detach().cpu().numpy().flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_combo_df(combo_predictions, combo_labels):\n",
    "    cares_ids, mirna_ids = zip(*combo_labels)\n",
    "\n",
    "    df = pd.DataFrame([cares_ids, mirna_ids, combo_predictions], index = ['CARES_id','miRNA_name','prediction']).T\n",
    "    df = df.pivot(index = 'CARES_id', columns = 'miRNA_name', values = 'prediction')\n",
    "    df = df.astype(float)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "unrefined_interact_df = generate_combo_df(unrefined_interacting_combo_predictions, positive_combo_labels)\n",
    "unrefined_noninteract_df = generate_combo_df(unrefined_noninteracting_combo_predictions, negative_combo_labels)\n",
    "\n",
    "refined_interact_df = generate_combo_df(refined_interacting_combo_predictions, positive_combo_labels)\n",
    "refined_noninteract_df = generate_combo_df(refined_noninteracting_combo_predictions, negative_combo_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import re\n",
    "\n",
    "def isolate_patient_id_number(complete_id : str) -> int:\n",
    "    return int(re.sub('A0+(\\d+)-.*',r'\\1', complete_id))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x576 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig = plt.figure(figsize = (7,8))\n",
    "gs = fig.add_gridspec(3,2, height_ratios = [0.02, 0.29, 0.69], wspace = 0.05, hspace = 0.03)\n",
    "\n",
    "cbar = fig.add_subplot(gs[0,:])\n",
    "interact_heatmap = fig.add_subplot(gs[1,:])\n",
    "noninteract_heatmap = fig.add_subplot(gs[2,:])\n",
    "\n",
    "g = sbn.heatmap(unrefined_interact_df.T,\n",
    "                cmap = 'vlag', xticklabels=False, yticklabels=False,\n",
    "                vmin = 0, vmax = 1,\n",
    "                cbar_kws = {'orientation': 'horizontal'},\n",
    "                cbar_ax = cbar,\n",
    "                ax = interact_heatmap)\n",
    "\n",
    "g = sbn.heatmap(unrefined_noninteract_df.T,\n",
    "                cmap = 'vlag', xticklabels=False, yticklabels=False,\n",
    "                vmin = 0, vmax = 1,\n",
    "                cbar = False,\n",
    "                ax = noninteract_heatmap)\n",
    "\n",
    "cbar.xaxis.tick_top()\n",
    "cbar.set_title(\"Interaction Likelihood\")\n",
    "\n",
    "interact_heatmap.set_ylabel(\"WT interacting miRNAs\")\n",
    "interact_heatmap.set_xlabel(\"\")\n",
    "\n",
    "noninteract_heatmap.set_ylabel(\"WT non-interacting miRNAs\")\n",
    "\n",
    "#plt.tight_layout()\n",
    "noninteract_heatmap.set_xlabel(\"Participant Tat sequences\")\n",
    "\n",
    "plt.savefig(\"UnrefinedCARESInteractionPrediction.png\", bbox_inches = 'tight', dpi = 300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>CARES_id</th>\n",
       "      <th>A0001-R01</th>\n",
       "      <th>A0001-R03</th>\n",
       "      <th>A0001-R09</th>\n",
       "      <th>A0002-R02</th>\n",
       "      <th>A0002-R11</th>\n",
       "      <th>A0002-R14</th>\n",
       "      <th>A0003-R00</th>\n",
       "      <th>A0004-R10</th>\n",
       "      <th>A0005-R00</th>\n",
       "      <th>A0005-R01</th>\n",
       "      <th>...</th>\n",
       "      <th>B0076-R00</th>\n",
       "      <th>B0146-R00</th>\n",
       "      <th>B0147-R00</th>\n",
       "      <th>B0303-R02</th>\n",
       "      <th>B0325-R00</th>\n",
       "      <th>B0325-R01</th>\n",
       "      <th>B0485-R00</th>\n",
       "      <th>B0522-R00</th>\n",
       "      <th>B0522-R01</th>\n",
       "      <th>B0538-R01</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>miRNA_name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>hsa-let-7g</th>\n",
       "      <td>2.530119e-07</td>\n",
       "      <td>1.133406e-06</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.000033</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000040</td>\n",
       "      <td>0.000029</td>\n",
       "      <td>0.000126</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000182</td>\n",
       "      <td>0.000939</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.000030</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000041</td>\n",
       "      <td>0.000025</td>\n",
       "      <td>0.000025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hsa-let-7i</th>\n",
       "      <td>2.974811e-07</td>\n",
       "      <td>1.332612e-06</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000017</td>\n",
       "      <td>0.000038</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000047</td>\n",
       "      <td>0.000034</td>\n",
       "      <td>0.000148</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000214</td>\n",
       "      <td>0.001104</td>\n",
       "      <td>0.000017</td>\n",
       "      <td>0.000036</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000030</td>\n",
       "      <td>0.000029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hsa-miR-106</th>\n",
       "      <td>1.620080e-07</td>\n",
       "      <td>7.257374e-07</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000026</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000080</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000116</td>\n",
       "      <td>0.000601</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000026</td>\n",
       "      <td>0.000016</td>\n",
       "      <td>0.000016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hsa-miR-125a-3p</th>\n",
       "      <td>3.769550e-07</td>\n",
       "      <td>1.688628e-06</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>0.000049</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000060</td>\n",
       "      <td>0.000044</td>\n",
       "      <td>0.000187</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000271</td>\n",
       "      <td>0.001398</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>0.000061</td>\n",
       "      <td>0.000038</td>\n",
       "      <td>0.000037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hsa-miR-128</th>\n",
       "      <td>1.972798e-07</td>\n",
       "      <td>8.837441e-07</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000011</td>\n",
       "      <td>0.000025</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000031</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.000098</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000142</td>\n",
       "      <td>0.000732</td>\n",
       "      <td>0.000011</td>\n",
       "      <td>0.000024</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000032</td>\n",
       "      <td>0.000020</td>\n",
       "      <td>0.000019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hsa-miR-9</th>\n",
       "      <td>2.313806e-07</td>\n",
       "      <td>1.036504e-06</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>0.000030</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000037</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>0.000115</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000166</td>\n",
       "      <td>0.000859</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>0.000028</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000037</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.000023</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hsa-miR-95</th>\n",
       "      <td>1.125158e-07</td>\n",
       "      <td>5.040314e-07</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000018</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>0.000056</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000081</td>\n",
       "      <td>0.000418</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000018</td>\n",
       "      <td>0.000011</td>\n",
       "      <td>0.000011</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hsa-miR-96</th>\n",
       "      <td>1.393418e-07</td>\n",
       "      <td>6.241992e-07</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000018</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>0.000016</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000100</td>\n",
       "      <td>0.000517</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000017</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.000014</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hsa-miR-99a</th>\n",
       "      <td>1.550336e-07</td>\n",
       "      <td>6.944968e-07</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>0.000020</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000025</td>\n",
       "      <td>0.000018</td>\n",
       "      <td>0.000077</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000111</td>\n",
       "      <td>0.000576</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000025</td>\n",
       "      <td>0.000016</td>\n",
       "      <td>0.000015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hsa-miR-99b</th>\n",
       "      <td>1.156004e-07</td>\n",
       "      <td>5.178487e-07</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000018</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>0.000057</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000083</td>\n",
       "      <td>0.000429</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000011</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>341 rows × 296 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "CARES_id            A0001-R01     A0001-R03  A0001-R09  A0002-R02  A0002-R11  \\\n",
       "miRNA_name                                                                     \n",
       "hsa-let-7g       2.530119e-07  1.133406e-06   0.000002   0.000014   0.000033   \n",
       "hsa-let-7i       2.974811e-07  1.332612e-06   0.000003   0.000017   0.000038   \n",
       "hsa-miR-106      1.620080e-07  7.257374e-07   0.000002   0.000009   0.000021   \n",
       "hsa-miR-125a-3p  3.769550e-07  1.688628e-06   0.000004   0.000021   0.000049   \n",
       "hsa-miR-128      1.972798e-07  8.837441e-07   0.000002   0.000011   0.000025   \n",
       "...                       ...           ...        ...        ...        ...   \n",
       "hsa-miR-9        2.313806e-07  1.036504e-06   0.000002   0.000013   0.000030   \n",
       "hsa-miR-95       1.125158e-07  5.040314e-07   0.000001   0.000006   0.000014   \n",
       "hsa-miR-96       1.393418e-07  6.241992e-07   0.000001   0.000008   0.000018   \n",
       "hsa-miR-99a      1.550336e-07  6.944968e-07   0.000002   0.000009   0.000020   \n",
       "hsa-miR-99b      1.156004e-07  5.178487e-07   0.000001   0.000007   0.000015   \n",
       "\n",
       "CARES_id         A0002-R14  A0003-R00  A0004-R10  A0005-R00  A0005-R01  ...  \\\n",
       "miRNA_name                                                              ...   \n",
       "hsa-let-7g        0.000006   0.000004   0.000040   0.000029   0.000126  ...   \n",
       "hsa-let-7i        0.000007   0.000005   0.000047   0.000034   0.000148  ...   \n",
       "hsa-miR-106       0.000004   0.000003   0.000026   0.000019   0.000080  ...   \n",
       "hsa-miR-125a-3p   0.000008   0.000007   0.000060   0.000044   0.000187  ...   \n",
       "hsa-miR-128       0.000004   0.000003   0.000031   0.000023   0.000098  ...   \n",
       "...                    ...        ...        ...        ...        ...  ...   \n",
       "hsa-miR-9         0.000005   0.000004   0.000037   0.000027   0.000115  ...   \n",
       "hsa-miR-95        0.000003   0.000002   0.000018   0.000013   0.000056  ...   \n",
       "hsa-miR-96        0.000003   0.000002   0.000022   0.000016   0.000069  ...   \n",
       "hsa-miR-99a       0.000003   0.000003   0.000025   0.000018   0.000077  ...   \n",
       "hsa-miR-99b       0.000003   0.000002   0.000018   0.000013   0.000057  ...   \n",
       "\n",
       "CARES_id         B0076-R00  B0146-R00  B0147-R00  B0303-R02  B0325-R00  \\\n",
       "miRNA_name                                                               \n",
       "hsa-let-7g        0.000182   0.000939   0.000014   0.000030   0.000008   \n",
       "hsa-let-7i        0.000214   0.001104   0.000017   0.000036   0.000010   \n",
       "hsa-miR-106       0.000116   0.000601   0.000009   0.000019   0.000005   \n",
       "hsa-miR-125a-3p   0.000271   0.001398   0.000021   0.000045   0.000012   \n",
       "hsa-miR-128       0.000142   0.000732   0.000011   0.000024   0.000006   \n",
       "...                    ...        ...        ...        ...        ...   \n",
       "hsa-miR-9         0.000166   0.000859   0.000013   0.000028   0.000007   \n",
       "hsa-miR-95        0.000081   0.000418   0.000006   0.000014   0.000004   \n",
       "hsa-miR-96        0.000100   0.000517   0.000008   0.000017   0.000005   \n",
       "hsa-miR-99a       0.000111   0.000576   0.000009   0.000019   0.000005   \n",
       "hsa-miR-99b       0.000083   0.000429   0.000007   0.000014   0.000004   \n",
       "\n",
       "CARES_id         B0325-R01  B0485-R00  B0522-R00  B0522-R01  B0538-R01  \n",
       "miRNA_name                                                              \n",
       "hsa-let-7g        0.000013   0.000006   0.000041   0.000025   0.000025  \n",
       "hsa-let-7i        0.000015   0.000007   0.000048   0.000030   0.000029  \n",
       "hsa-miR-106       0.000008   0.000004   0.000026   0.000016   0.000016  \n",
       "hsa-miR-125a-3p   0.000019   0.000009   0.000061   0.000038   0.000037  \n",
       "hsa-miR-128       0.000010   0.000005   0.000032   0.000020   0.000019  \n",
       "...                    ...        ...        ...        ...        ...  \n",
       "hsa-miR-9         0.000012   0.000006   0.000037   0.000023   0.000023  \n",
       "hsa-miR-95        0.000006   0.000003   0.000018   0.000011   0.000011  \n",
       "hsa-miR-96        0.000007   0.000004   0.000022   0.000014   0.000014  \n",
       "hsa-miR-99a       0.000008   0.000004   0.000025   0.000016   0.000015  \n",
       "hsa-miR-99b       0.000006   0.000003   0.000019   0.000012   0.000011  \n",
       "\n",
       "[341 rows x 296 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_unrefined_results = pd.concat([unrefined_interact_df, unrefined_noninteract_df], axis = 1).T\n",
    "all_unrefined_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of excluive binders (mean > 0.99) : 0\n",
      "Number of exclusive non-binders (mean < 0.01) : 291\n",
      "Summed total: 291 / 296\n",
      "---\n",
      "Total highly variable interactions (STD ≥ 0.01): 5\n",
      "Mean and STD of variable interaction means: 0.478 ± 0.455\n",
      "Mean and STD of variable interaction STDs: 0.035 ± 0.03\n",
      "---\n"
     ]
    }
   ],
   "source": [
    "all_unrefined_described = all_unrefined_results.describe()\n",
    "\n",
    "unrefined_exclusive_binders = all_unrefined_described.loc['mean',:] > 0.9\n",
    "# (all_unrefined_results.describe().loc['max',:] - all_unrefined_results.describe().loc['min',:]) > 0.05\n",
    "unrefined_variable_binders = all_unrefined_described.loc['std',:] > 0.01\n",
    "\n",
    "unrefined_exclusive_binders = all_unrefined_described.loc['mean',:] >= 0.99\n",
    "unrefined_exclusive_non_binders = all_unrefined_described.loc['mean',:] <= 0.01\n",
    "print(f\"\"\"Number of excluive binders (mean > 0.99) : {sum(unrefined_exclusive_binders)}\n",
    "Number of exclusive non-binders (mean < 0.01) : {sum(unrefined_exclusive_non_binders)}\n",
    "Summed total: {sum(unrefined_exclusive_binders) + sum(unrefined_exclusive_non_binders)} / {all_unrefined_described.shape[1]}\"\"\")\n",
    "\n",
    "unrefined_highly_variable_interaction_means = all_unrefined_described.loc['mean',unrefined_variable_binders]\n",
    "unrefined_highly_variable_interaction_stds = all_unrefined_described.loc['std',unrefined_variable_binders]\n",
    "\n",
    "print(f\"\"\"---\n",
    "Total highly variable interactions (STD ≥ 0.01): {sum(unrefined_variable_binders)}\n",
    "Mean and STD of variable interaction means: {round(unrefined_highly_variable_interaction_means.mean(),3)} ± {round(unrefined_highly_variable_interaction_means.std(),3)}\n",
    "Mean and STD of variable interaction STDs: {round(unrefined_highly_variable_interaction_stds.mean(),3)} ± {round(unrefined_highly_variable_interaction_stds.std(),3)}\n",
    "---\"\"\")\n",
    "\n",
    "\n",
    "#all_unrefined_described.loc[:,unrefined_variable_binders].round(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of excluive binders (mean > 0.99): 51 / 296 (17.23%)\n",
      "Number of exclusive non-binders (mean < 0.01): 232 / 296 (78.38000000000001%)\n",
      "Summed total: 283 / 296 (95.61%)\n",
      "---\n",
      "Total highly variable interactions (STD ≥ 0.01): 14\n",
      "Mean and STD of variable interaction means: 0.469 ± 0.292\n",
      "Mean and STD of variable interaction STDs: 0.156 ± 0.073\n",
      "---\n"
     ]
    }
   ],
   "source": [
    "all_refined_results = pd.concat([refined_interact_df, refined_noninteract_df], axis = 1).T\n",
    "\n",
    "all_refined_described = all_refined_results.describe()\n",
    "\n",
    "refined_exclusive_binders = all_refined_results.describe().loc['mean',:] >= 0.99\n",
    "refined_exclusive_non_binders = all_refined_results.describe().loc['mean',:] <= 0.01\n",
    "\n",
    "print(f\"\"\"Number of excluive binders (mean > 0.99): {sum(refined_exclusive_binders)} / {all_refined_results.shape[1]} ({round(sum(refined_exclusive_binders) / all_refined_results.shape[1],4)*100}%)\n",
    "Number of exclusive non-binders (mean < 0.01): {sum(refined_exclusive_non_binders)} / {all_refined_results.shape[1]} ({round(sum(refined_exclusive_non_binders) / all_refined_results.shape[1],4)*100}%)\n",
    "Summed total: {sum(refined_exclusive_binders) + sum(refined_exclusive_non_binders)} / {all_refined_described.shape[1]} ({round((sum(refined_exclusive_binders) + sum(refined_exclusive_non_binders)) / all_refined_results.shape[1],4)*100}%)\n",
    "---\"\"\")\n",
    "\n",
    "refined_variable_binders = all_refined_results.describe().loc['std',:] > 0.01\n",
    "\n",
    "refined_highly_variable_interaction_means = all_refined_described.loc['mean',refined_variable_binders]\n",
    "refined_highly_variable_interaction_stds = all_refined_described.loc['std',refined_variable_binders]\n",
    "\n",
    "print(f\"\"\"Total highly variable interactions (STD ≥ 0.01): {sum(refined_variable_binders)}\n",
    "Mean and STD of variable interaction means: {round(refined_highly_variable_interaction_means.mean(),3)} ± {round(refined_highly_variable_interaction_means.std(),3)}\n",
    "Mean and STD of variable interaction STDs: {round(refined_highly_variable_interaction_stds.mean(),3)} ± {round(refined_highly_variable_interaction_stds.std(),3)}\n",
    "---\"\"\")\n",
    "\n",
    "#all_refined_described.loc[:,refined_variable_binders].round(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x576 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig = plt.figure(figsize = (7,8))\n",
    "gs = fig.add_gridspec(3,2, height_ratios = [0.02, 0.29, 0.69], wspace = 0.05, hspace = 0.03)\n",
    "\n",
    "cbar = fig.add_subplot(gs[0,:])\n",
    "interact_heatmap = fig.add_subplot(gs[1,:])\n",
    "noninteract_heatmap = fig.add_subplot(gs[2,:])\n",
    "\n",
    "g = sbn.heatmap(refined_interact_df.T,\n",
    "                cmap = 'vlag', xticklabels=False, yticklabels=False,\n",
    "                vmin = 0, vmax = 1,\n",
    "                cbar_kws = {'orientation': 'horizontal'},\n",
    "                cbar_ax = cbar,\n",
    "                ax = interact_heatmap)\n",
    "\n",
    "g = sbn.heatmap(refined_noninteract_df.T,\n",
    "                cmap = 'vlag', xticklabels=False, yticklabels=False,\n",
    "                vmin = 0, vmax = 1,\n",
    "                cbar = False,\n",
    "                ax = noninteract_heatmap)\n",
    "\n",
    "cbar.xaxis.tick_top()\n",
    "cbar.set_title(\"Interaction Likelihood\")\n",
    "\n",
    "interact_heatmap.set_ylabel(\"WT interacting miRNAs\")\n",
    "interact_heatmap.set_xlabel(\"\")\n",
    "\n",
    "noninteract_heatmap.set_ylabel(\"WT non-interacting miRNAs\")\n",
    "\n",
    "#plt.tight_layout()\n",
    "noninteract_heatmap.set_xlabel(\"Participant Tat sequences\")\n",
    "\n",
    "plt.savefig(\"RefinedCARESInteractionPrediction.png\", bbox_inches = 'tight', dpi = 300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'A0001-R01': 0.0,\n",
       " 'A0001-R03': 0.0,\n",
       " 'A0001-R09': 0.0,\n",
       " 'A0002-R02': 0.0,\n",
       " 'A0002-R11': 0.0,\n",
       " 'A0002-R14': 0.0,\n",
       " 'A0003-R00': 1.0,\n",
       " 'A0004-R10': 0.0,\n",
       " 'A0005-R00': 0.0,\n",
       " 'A0005-R01': 0.0,\n",
       " 'A0005-R04': 0.0,\n",
       " 'A0005-R05': 0.0,\n",
       " 'A0005-R07': 0.0,\n",
       " 'A0008-R00': 1.0,\n",
       " 'A0008-R01': 0.0,\n",
       " 'A0008-R07': 1.0,\n",
       " 'A0008-R10': 0.0,\n",
       " 'A0010-R00': 0.0,\n",
       " 'A0010-R08': 1.0,\n",
       " 'A0010-R09': 0.0,\n",
       " 'A0013-R00': 0.0,\n",
       " 'A0013-R01': 1.0,\n",
       " 'A0013-R05': 0.0,\n",
       " 'A0013-R06': 1.0,\n",
       " 'A0013-R10': 0.0,\n",
       " 'A0013-R11': 0.0,\n",
       " 'A0014-R00': 0.0,\n",
       " 'A0015-R03': 1.0,\n",
       " 'A0015-R06': 1.0,\n",
       " 'A0015-R08': 0.0,\n",
       " 'A0015-R10': 0.0,\n",
       " 'A0015-R11': 1.0,\n",
       " 'A0015-R13': 0.0,\n",
       " 'A0016-R00': 1.0,\n",
       " 'A0017-R00': 0.0,\n",
       " 'A0017-R02': 0.0,\n",
       " 'A0017-R03': 0.0,\n",
       " 'A0017-R04': 0.0,\n",
       " 'A0017-R05': 1.0,\n",
       " 'A0017-R06': 0.0,\n",
       " 'A0017-R07': 0.0,\n",
       " 'A0017-R11': 0.0,\n",
       " 'A0019-R01': 1.0,\n",
       " 'A0019-R02': 0.0,\n",
       " 'A0019-R03': 1.0,\n",
       " 'A0019-R04': 0.0,\n",
       " 'A0019-R05': 1.0,\n",
       " 'A0019-R08': 1.0,\n",
       " 'A0019-R09': 1.0,\n",
       " 'A0019-R11': 1.0,\n",
       " 'A0019-R12': 1.0,\n",
       " 'A0022-R00': 0.0,\n",
       " 'A0022-R01': 1.0,\n",
       " 'A0025-R05': 0.0,\n",
       " 'A0025-R07': 1.0,\n",
       " 'A0025-R09': 0.0,\n",
       " 'A0025-R11': 0.0,\n",
       " 'A0026-R01': 1.0,\n",
       " 'A0026-R09': 1.0,\n",
       " 'A0027-R01': 0.0,\n",
       " 'A0029-R00': 0.0,\n",
       " 'A0029-R03': 0.0,\n",
       " 'A0030-R00': 0.0,\n",
       " 'A0032-R06': 0.0,\n",
       " 'A0034-R04': 1.0,\n",
       " 'A0034-R05': 1.0,\n",
       " 'A0034-R06': 0.0,\n",
       " 'A0036-R05': 0.0,\n",
       " 'A0037-R09': 0.0,\n",
       " 'A0037-R10': 0.0,\n",
       " 'A0037-R11': 0.0,\n",
       " 'A0037-R12': 0.0,\n",
       " 'A0037-R13': 0.0,\n",
       " 'A0038-R08': 0.0,\n",
       " 'A0039-R00': 0.0,\n",
       " 'A0040-R05': 0.0,\n",
       " 'A0040-R06': 0.0,\n",
       " 'A0041-R00': 0.0,\n",
       " 'A0041-R04': 0.0,\n",
       " 'A0044-R00': 0.0,\n",
       " 'A0044-R10': 0.0,\n",
       " 'A0044-R12': 0.0,\n",
       " 'A0045-R03': 0.0,\n",
       " 'A0045-R04': 0.0,\n",
       " 'A0045-R05': 0.0,\n",
       " 'A0045-R06': 0.0,\n",
       " 'A0046-R00': 1.0,\n",
       " 'A0047-R05': 0.0,\n",
       " 'A0055-R00': 1.0,\n",
       " 'A0056-R06': 0.0,\n",
       " 'A0056-R07': 0.0,\n",
       " 'A0058-R02': 0.0,\n",
       " 'A0058-R03': 0.0,\n",
       " 'A0058-R05': 1.0,\n",
       " 'A0058-R06': 0.0,\n",
       " 'A0058-R07': 1.0,\n",
       " 'A0058-R09': 1.0,\n",
       " 'A0058-R10': 0.0,\n",
       " 'A0059-R02': 0.0,\n",
       " 'A0059-R08': 0.0,\n",
       " 'A0059-R09': 0.0,\n",
       " 'A0060-R00': 0.0,\n",
       " 'A0063-R01': 0.0,\n",
       " 'A0065-R00': 0.0,\n",
       " 'A0066-R00': 0.0,\n",
       " 'A0067-R01': 0.0,\n",
       " 'A0067-R05': 0.0,\n",
       " 'A0067-R06': 0.0,\n",
       " 'A0067-R11': 1.0,\n",
       " 'A0068-R03': 1.0,\n",
       " 'A0070-R00': 0.0,\n",
       " 'A0071-R00': 1.0,\n",
       " 'A0074-R00': 0.0,\n",
       " 'A0075-R07': 0.0,\n",
       " 'A0075-R09': 0.0,\n",
       " 'A0075-R10': 0.0,\n",
       " 'A0076-R00': 0.0,\n",
       " 'A0076-R04': 0.0,\n",
       " 'A0076-R05': 0.0,\n",
       " 'A0078-R00': 1.0,\n",
       " 'A0078-R01': 0.0,\n",
       " 'A0079-R00': 0.0,\n",
       " 'A0080-R01': 0.0,\n",
       " 'A0080-R02': 0.0,\n",
       " 'A0080-R03': 0.0,\n",
       " 'A0080-R04': 0.0,\n",
       " 'A0080-R05': 0.0,\n",
       " 'A0080-R06': 0.0,\n",
       " 'A0080-R07': 0.0,\n",
       " 'A0080-R11': 0.0,\n",
       " 'A0083-R01': 0.0,\n",
       " 'A0083-R06': 0.0,\n",
       " 'A0085-R00': 0.0,\n",
       " 'A0086-R04': 0.0,\n",
       " 'A0086-R05': 0.0,\n",
       " 'A0088-R00': 0.0,\n",
       " 'A0089-R00': 0.0,\n",
       " 'A0089-R02': 0.0,\n",
       " 'A0091-R04': 0.0,\n",
       " 'A0094-R00': 0.0,\n",
       " 'A0094-R02': 1.0,\n",
       " 'A0094-R03': 0.0,\n",
       " 'A0095-R00': 0.0,\n",
       " 'A0095-R01': 0.0,\n",
       " 'A0095-R04': 0.0,\n",
       " 'A0095-R06': 0.0,\n",
       " 'A0095-R09': 0.0,\n",
       " 'A0095-R10': 0.0,\n",
       " 'A0095-R11': 0.0,\n",
       " 'A0096-R01': 0.0,\n",
       " 'A0096-R07': 1.0,\n",
       " 'A0096-R08': 0.0,\n",
       " 'A0098-R00': 0.0,\n",
       " 'A0098-R01': 0.0,\n",
       " 'A0100-R08': 0.0,\n",
       " 'A0101-R00': 0.0,\n",
       " 'A0103-R00': 1.0,\n",
       " 'A0103-R06': 1.0,\n",
       " 'A0103-R07': 1.0,\n",
       " 'A0105-R01': 0.0,\n",
       " 'A0107-R00': 1.0,\n",
       " 'A0107-R01': 0.0,\n",
       " 'A0107-R04': 0.0,\n",
       " 'A0107-R06': 0.0,\n",
       " 'A0107-R07': 0.0,\n",
       " 'A0107-R08': 0.0,\n",
       " 'A0107-R09': 0.0,\n",
       " 'A0107-R10': 0.0,\n",
       " 'A0107-R11': 0.0,\n",
       " 'A0107-R12': 0.0,\n",
       " 'A0109-R00': 0.0,\n",
       " 'A0109-R11': 0.0,\n",
       " 'A0110-R01': 0.0,\n",
       " 'A0110-R02': 0.0,\n",
       " 'A0110-R04': 0.0,\n",
       " 'A0110-R06': 0.0,\n",
       " 'A0110-R10': 0.0,\n",
       " 'A0111-R05': 0.0,\n",
       " 'A0111-R07': 0.0,\n",
       " 'A0113-R11': 0.0,\n",
       " 'A0113-R12': 0.0,\n",
       " 'A0116-R00': 0.0,\n",
       " 'A0117-R01': 0.0,\n",
       " 'A0117-R11': 0.0,\n",
       " 'A0121-R00': 0.0,\n",
       " 'A0124-R02': 0.0,\n",
       " 'A0127-R00': 0.0,\n",
       " 'A0127-R08': 0.0,\n",
       " 'A0131-R01': 0.0,\n",
       " 'A0132-R06': 0.0,\n",
       " 'A0132-R11': 1.0,\n",
       " 'A0133-R05': 1.0,\n",
       " 'A0142-R03': 0.0,\n",
       " 'A0143-R06': 0.0,\n",
       " 'A0145-R08': 0.0,\n",
       " 'A0145-R11': 0.0,\n",
       " 'A0145-R12': 0.0,\n",
       " 'A0151-R02': 0.0,\n",
       " 'A0154-R00': 0.0,\n",
       " 'A0156-R00': 0.0,\n",
       " 'A0156-R06': 0.0,\n",
       " 'A0159-R00': 1.0,\n",
       " 'A0159-R01': 0.0,\n",
       " 'A0160-R00': 0.0,\n",
       " 'A0164-R02': 0.0,\n",
       " 'A0171-R00': 0.0,\n",
       " 'A0176-R00': 0.0,\n",
       " 'A0179-R02': 0.0,\n",
       " 'A0186-R00': 0.0,\n",
       " 'A0189-R00': 0.0,\n",
       " 'A0193-R00': 0.0,\n",
       " 'A0195-R00': 0.0,\n",
       " 'A0198-R00': 0.0,\n",
       " 'A0199-R00': 0.0,\n",
       " 'A0199-R02': 1.0,\n",
       " 'A0199-R03': 0.0,\n",
       " 'A0199-R04': 0.0,\n",
       " 'A0199-R07': 1.0,\n",
       " 'A0200-R00': 0.0,\n",
       " 'A0208-R05': 1.0,\n",
       " 'A0209-R07': 1.0,\n",
       " 'A0211-R00': 0.0,\n",
       " 'A0213-R04': 1.0,\n",
       " 'A0214-R00': 0.0,\n",
       " 'A0215-R09': 0.0,\n",
       " 'A0218-R04': 0.0,\n",
       " 'A0223-R04': 0.0,\n",
       " 'A0238-R00': 0.0,\n",
       " 'A0239-R05': 0.0,\n",
       " 'A0241-R01': 1.0,\n",
       " 'A0241-R03': 0.0,\n",
       " 'A0255-R02': 0.0,\n",
       " 'A0257-R05': 0.0,\n",
       " 'A0264-R01': 0.0,\n",
       " 'A0264-R03': 1.0,\n",
       " 'A0272-R01': 0.0,\n",
       " 'A0274-R06': 0.0,\n",
       " 'A0278-R05': 0.0,\n",
       " 'A0282-R01': 0.0,\n",
       " 'A0284-R07': 0.0,\n",
       " 'A0285-R04': 1.0,\n",
       " 'A0300-R02': 0.0,\n",
       " 'A0304-R01': 0.0,\n",
       " 'A0310-R04': 0.0,\n",
       " 'A0312-R00': 0.0,\n",
       " 'A0318-R03': 0.0,\n",
       " 'A0323-R08': 0.0,\n",
       " 'A0325-R03': 0.0,\n",
       " 'A0335-R03': 0.0,\n",
       " 'A0338-R01': 0.0,\n",
       " 'A0339-R00': 0.0,\n",
       " 'A0342-R03': 0.0,\n",
       " 'A0357-R04': 1.0,\n",
       " 'A0359-R07': 1.0,\n",
       " 'A0372-R00': 0.0,\n",
       " 'A0378-R00': 0.0,\n",
       " 'A0379-R06': 0.0,\n",
       " 'A0380-R03': 0.0,\n",
       " 'A0382-R00': 0.0,\n",
       " 'A0384-R03': 0.0,\n",
       " 'A0389-R07': 0.0,\n",
       " 'A0401-R03': 0.0,\n",
       " 'A0405-R00': 0.0,\n",
       " 'A0408-R02': 0.0,\n",
       " 'A0417-R01': 0.0,\n",
       " 'A0417-R03': 0.0,\n",
       " 'A0429-R04': 0.0,\n",
       " 'A0430-R01': 0.0,\n",
       " 'A0433-R04': 0.0,\n",
       " 'A0436-R01': 0.0,\n",
       " 'A0439-R01': 0.0,\n",
       " 'A0440-R02': 0.0,\n",
       " 'A0446-R04': 0.0,\n",
       " 'A0447-R01': 0.0,\n",
       " 'A0448-R04': 0.0,\n",
       " 'A0449-R01': 0.0,\n",
       " 'A0451-R03': 0.0,\n",
       " 'A0471-R02': 0.0,\n",
       " 'A0474-R03': 0.0,\n",
       " 'A0490-R02': 0.0,\n",
       " 'A0500-R03': 0.0,\n",
       " 'A0501-R02': 0.0,\n",
       " 'A0503-R04': 1.0,\n",
       " 'A0523-R03': 0.0,\n",
       " 'B0037-R01': 0.0,\n",
       " 'B0044-R02': 0.0,\n",
       " 'B0076-R00': 1.0,\n",
       " 'B0146-R00': 1.0,\n",
       " 'B0147-R00': 1.0,\n",
       " 'B0303-R02': 0.0,\n",
       " 'B0325-R00': 0.0,\n",
       " 'B0325-R01': 0.0,\n",
       " 'B0485-R00': 0.0,\n",
       " 'B0522-R00': 0.0,\n",
       " 'B0522-R01': 0.0,\n",
       " 'B0538-R01': 1.0}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "refined_interact_df.iloc[:,0].round().to_dict()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Because all Tats either interact with or do not interact with a model\n",
    "patient_maintains_interaction = { patient : {\"Maintains interaction\" : interact == 1, \"Maintains non-interaction\" : interact == 0} for patient, interact in refined_interact_df.iloc[:,0].round().to_dict().items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"../Data/CARESInteractionResults/Patient_maintains_WT_interaction.yml\",'w') as outfile:\n",
    "    yaml.dump(patient_maintains_interaction, outfile)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}