{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Minimal working example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "from mprod.dimensionality_reduction import TCAM\n",
    "from mprod import table2tensor\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns; sns.set_style('ticks')\n",
    "%matplotlib inline\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>s__Vagococcus_lutrae</th>\n",
       "      <th>s__Asaccharobacter_celatus</th>\n",
       "      <th>s__Megasphaera_elsdenii</th>\n",
       "      <th>s__Leuconostoc_carnosum</th>\n",
       "      <th>s__Streptococcus_agalactiae</th>\n",
       "      <th>s__Tyzzerella_nexilis</th>\n",
       "      <th>s__Akkermansia_muciniphila</th>\n",
       "      <th>s__Alistipes_timonensis</th>\n",
       "      <th>s__Peptostreptococcus_anaerobius</th>\n",
       "      <th>s__Streptococcus_anginosus_group</th>\n",
       "      <th>...</th>\n",
       "      <th>s__Clostridium_celatum</th>\n",
       "      <th>s__Fusobacterium_periodonticum</th>\n",
       "      <th>s__Acidaminococcus_intestini</th>\n",
       "      <th>s__Streptococcus_sobrinus</th>\n",
       "      <th>s__Anaerostipes_caccae</th>\n",
       "      <th>s__Enterococcus_faecium</th>\n",
       "      <th>s__Eubacterium_sp_CAG_180</th>\n",
       "      <th>s__Veillonella_dispar</th>\n",
       "      <th>s__Actinomyces_sp_S6_Spd3</th>\n",
       "      <th>s__Firmicutes_bacterium_CAG_238</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Participant</th>\n",
       "      <th>Phase</th>\n",
       "      <th>Group</th>\n",
       "      <th>Day</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 rowspan=\"5\" valign=\"top\">602</th>\n",
       "      <th rowspan=\"2\" valign=\"top\">BAS</th>\n",
       "      <th rowspan=\"2\" valign=\"top\">FMT</th>\n",
       "      <th>3</th>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>0.000051</td>\n",
       "      <td>0.018276</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>0.000198</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>...</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>2.247164e-04</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>0.005217</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>5.632903e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>0.000263</td>\n",
       "      <td>0.017956</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>0.000175</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>...</td>\n",
       "      <td>9.586594e-05</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>3.205194e-04</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>0.005515</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.263375e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">ABX</th>\n",
       "      <th rowspan=\"2\" valign=\"top\">FMT</th>\n",
       "      <th>10</th>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>2.271773e-03</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>1.000001e-06</td>\n",
       "      <td>1.000001e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000270</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>0.000209</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>1.298585e-03</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>5.046216e-04</td>\n",
       "      <td>0.001349</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>1.000000e-06</td>\n",
       "      <td>1.000000e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>INT</th>\n",
       "      <th>FMT</th>\n",
       "      <th>17</th>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>0.000628</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>0.000486</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>...</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>1.176117e-03</td>\n",
       "      <td>0.003147</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "      <td>9.999999e-07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 482 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                             s__Vagococcus_lutrae  s__Asaccharobacter_celatus  \\\n",
       "Participant Phase Group Day                                                     \n",
       "602         BAS   FMT   3            9.999999e-07                    0.000051   \n",
       "                        6            9.999999e-07                    0.000263   \n",
       "            ABX   FMT   10           1.000001e-06                    0.000001   \n",
       "                        13           1.000000e-06                    0.000012   \n",
       "            INT   FMT   17           9.999999e-07                    0.000027   \n",
       "\n",
       "                             s__Megasphaera_elsdenii  s__Leuconostoc_carnosum  \\\n",
       "Participant Phase Group Day                                                     \n",
       "602         BAS   FMT   3                   0.018276             9.999999e-07   \n",
       "                        6                   0.017956             9.999999e-07   \n",
       "            ABX   FMT   10                  0.000001             1.000001e-06   \n",
       "                        13                  0.000270             1.000000e-06   \n",
       "            INT   FMT   17                  0.000628             9.999999e-07   \n",
       "\n",
       "                             s__Streptococcus_agalactiae  \\\n",
       "Participant Phase Group Day                                \n",
       "602         BAS   FMT   3                   9.999999e-07   \n",
       "                        6                   9.999999e-07   \n",
       "            ABX   FMT   10                  1.000001e-06   \n",
       "                        13                  1.000000e-06   \n",
       "            INT   FMT   17                  9.999999e-07   \n",
       "\n",
       "                             s__Tyzzerella_nexilis  \\\n",
       "Participant Phase Group Day                          \n",
       "602         BAS   FMT   3             9.999999e-07   \n",
       "                        6             9.999999e-07   \n",
       "            ABX   FMT   10            1.000001e-06   \n",
       "                        13            1.000000e-06   \n",
       "            INT   FMT   17            9.999999e-07   \n",
       "\n",
       "                             s__Akkermansia_muciniphila  \\\n",
       "Participant Phase Group Day                               \n",
       "602         BAS   FMT   3                      0.000198   \n",
       "                        6                      0.000175   \n",
       "            ABX   FMT   10                     0.000001   \n",
       "                        13                     0.000209   \n",
       "            INT   FMT   17                     0.000486   \n",
       "\n",
       "                             s__Alistipes_timonensis  \\\n",
       "Participant Phase Group Day                            \n",
       "602         BAS   FMT   3               9.999999e-07   \n",
       "                        6               9.999999e-07   \n",
       "            ABX   FMT   10              1.000001e-06   \n",
       "                        13              1.000000e-06   \n",
       "            INT   FMT   17              9.999999e-07   \n",
       "\n",
       "                             s__Peptostreptococcus_anaerobius  \\\n",
       "Participant Phase Group Day                                     \n",
       "602         BAS   FMT   3                        9.999999e-07   \n",
       "                        6                        9.999999e-07   \n",
       "            ABX   FMT   10                       1.000001e-06   \n",
       "                        13                       1.000000e-06   \n",
       "            INT   FMT   17                       9.999999e-07   \n",
       "\n",
       "                             s__Streptococcus_anginosus_group  ...  \\\n",
       "Participant Phase Group Day                                    ...   \n",
       "602         BAS   FMT   3                        9.999999e-07  ...   \n",
       "                        6                        9.999999e-07  ...   \n",
       "            ABX   FMT   10                       2.271773e-03  ...   \n",
       "                        13                       1.298585e-03  ...   \n",
       "            INT   FMT   17                       9.999999e-07  ...   \n",
       "\n",
       "                             s__Clostridium_celatum  \\\n",
       "Participant Phase Group Day                           \n",
       "602         BAS   FMT   3              9.999999e-07   \n",
       "                        6              9.586594e-05   \n",
       "            ABX   FMT   10             1.000001e-06   \n",
       "                        13             1.000000e-06   \n",
       "            INT   FMT   17             9.999999e-07   \n",
       "\n",
       "                             s__Fusobacterium_periodonticum  \\\n",
       "Participant Phase Group Day                                   \n",
       "602         BAS   FMT   3                      9.999999e-07   \n",
       "                        6                      9.999999e-07   \n",
       "            ABX   FMT   10                     1.000001e-06   \n",
       "                        13                     1.000000e-06   \n",
       "            INT   FMT   17                     9.999999e-07   \n",
       "\n",
       "                             s__Acidaminococcus_intestini  \\\n",
       "Participant Phase Group Day                                 \n",
       "602         BAS   FMT   3                    2.247164e-04   \n",
       "                        6                    3.205194e-04   \n",
       "            ABX   FMT   10                   1.000001e-06   \n",
       "                        13                   1.000000e-06   \n",
       "            INT   FMT   17                   9.999999e-07   \n",
       "\n",
       "                             s__Streptococcus_sobrinus  \\\n",
       "Participant Phase Group Day                              \n",
       "602         BAS   FMT   3                 9.999999e-07   \n",
       "                        6                 9.999999e-07   \n",
       "            ABX   FMT   10                1.000001e-06   \n",
       "                        13                1.000000e-06   \n",
       "            INT   FMT   17                9.999999e-07   \n",
       "\n",
       "                             s__Anaerostipes_caccae  s__Enterococcus_faecium  \\\n",
       "Participant Phase Group Day                                                    \n",
       "602         BAS   FMT   3              9.999999e-07             9.999999e-07   \n",
       "                        6              9.999999e-07             9.999999e-07   \n",
       "            ABX   FMT   10             1.000001e-06             1.000001e-06   \n",
       "                        13             1.000000e-06             5.046216e-04   \n",
       "            INT   FMT   17             9.999999e-07             1.176117e-03   \n",
       "\n",
       "                             s__Eubacterium_sp_CAG_180  s__Veillonella_dispar  \\\n",
       "Participant Phase Group Day                                                     \n",
       "602         BAS   FMT   3                     0.005217           9.999999e-07   \n",
       "                        6                     0.005515           9.999999e-07   \n",
       "            ABX   FMT   10                    0.000001           1.000001e-06   \n",
       "                        13                    0.001349           1.000000e-06   \n",
       "            INT   FMT   17                    0.003147           9.999999e-07   \n",
       "\n",
       "                             s__Actinomyces_sp_S6_Spd3  \\\n",
       "Participant Phase Group Day                              \n",
       "602         BAS   FMT   3                 9.999999e-07   \n",
       "                        6                 9.999999e-07   \n",
       "            ABX   FMT   10                1.000001e-06   \n",
       "                        13                1.000000e-06   \n",
       "            INT   FMT   17                9.999999e-07   \n",
       "\n",
       "                             s__Firmicutes_bacterium_CAG_238  \n",
       "Participant Phase Group Day                                   \n",
       "602         BAS   FMT   3                       5.632903e-05  \n",
       "                        6                       9.263375e-05  \n",
       "            ABX   FMT   10                      1.000001e-06  \n",
       "                        13                      1.000000e-06  \n",
       "            INT   FMT   17                      9.999999e-07  \n",
       "\n",
       "[5 rows x 482 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>Participant</th>\n",
       "      <th>Phase</th>\n",
       "      <th>Group</th>\n",
       "      <th>Day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>602</td>\n",
       "      <td>BAS</td>\n",
       "      <td>FMT</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>602</td>\n",
       "      <td>BAS</td>\n",
       "      <td>FMT</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>602</td>\n",
       "      <td>ABX</td>\n",
       "      <td>FMT</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>602</td>\n",
       "      <td>ABX</td>\n",
       "      <td>FMT</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>602</td>\n",
       "      <td>INT</td>\n",
       "      <td>FMT</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Participant Phase Group  Day\n",
       "0          602   BAS   FMT    3\n",
       "1          602   BAS   FMT    6\n",
       "2          602   ABX   FMT   10\n",
       "3          602   ABX   FMT   13\n",
       "4          602   INT   FMT   17"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>Participant</th>\n",
       "      <th>Group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>602</td>\n",
       "      <td>FMT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>603</td>\n",
       "      <td>FMT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>604</td>\n",
       "      <td>FMT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>605</td>\n",
       "      <td>FMT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>606</td>\n",
       "      <td>FMT</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Participant Group\n",
       "0           602   FMT\n",
       "9           603   FMT\n",
       "18          604   FMT\n",
       "27          605   FMT\n",
       "36          606   FMT"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "file_path = \"./data/Suez2018.txt\"\n",
    "data_raw = pd.read_csv(file_path, index_col=[0,1,2,3]\n",
    "                       , dtype={'Participant':int,'Day':int})\n",
    "meta_time = data_raw.index.to_frame().reset_index(drop = True)\n",
    "meta = meta_time.drop(['Day','Phase'], axis=1).drop_duplicates()\n",
    "\n",
    "display(data_raw.head())\n",
    "display(meta_time.head())\n",
    "display(meta.head())\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data normalization and mode-mapping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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></th>\n",
       "      <th>s__Vagococcus_lutrae</th>\n",
       "      <th>s__Asaccharobacter_celatus</th>\n",
       "      <th>s__Megasphaera_elsdenii</th>\n",
       "      <th>s__Leuconostoc_carnosum</th>\n",
       "      <th>s__Streptococcus_agalactiae</th>\n",
       "      <th>s__Tyzzerella_nexilis</th>\n",
       "      <th>s__Akkermansia_muciniphila</th>\n",
       "      <th>s__Alistipes_timonensis</th>\n",
       "      <th>s__Peptostreptococcus_anaerobius</th>\n",
       "      <th>s__Streptococcus_anginosus_group</th>\n",
       "      <th>...</th>\n",
       "      <th>s__Clostridium_celatum</th>\n",
       "      <th>s__Fusobacterium_periodonticum</th>\n",
       "      <th>s__Acidaminococcus_intestini</th>\n",
       "      <th>s__Streptococcus_sobrinus</th>\n",
       "      <th>s__Anaerostipes_caccae</th>\n",
       "      <th>s__Enterococcus_faecium</th>\n",
       "      <th>s__Eubacterium_sp_CAG_180</th>\n",
       "      <th>s__Veillonella_dispar</th>\n",
       "      <th>s__Actinomyces_sp_S6_Spd3</th>\n",
       "      <th>s__Firmicutes_bacterium_CAG_238</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Participant</th>\n",
       "      <th>Day</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 rowspan=\"5\" valign=\"top\">602</th>\n",
       "      <th>3</th>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-1.623746</td>\n",
       "      <td>0.012707</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>0.086871</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.597918</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-0.278775</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-0.040705</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-8.872499e-09</td>\n",
       "      <td>-0.403001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>0.744599</td>\n",
       "      <td>-0.012820</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>-0.092439</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>...</td>\n",
       "      <td>0.985029</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>0.233532</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>0.039588</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>8.872499e-09</td>\n",
       "      <td>0.314658</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>-7.295634</td>\n",
       "      <td>-14.144966</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>-7.542312</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>1.114960e+01</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.597916</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>-8.090735</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>-12.389628</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>1.337567e-06</td>\n",
       "      <td>-6.218807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>-3.707767</td>\n",
       "      <td>-6.069695</td>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>0.164989</td>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>1.034272e+01</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.597917</td>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>-8.090736</td>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>8.979058e+00</td>\n",
       "      <td>-1.991681</td>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>7.558580e-07</td>\n",
       "      <td>-6.218808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>-2.555618</td>\n",
       "      <td>-4.850362</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>1.383432</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.597918</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>-8.090737</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>1.019982e+01</td>\n",
       "      <td>-0.769899</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>-1.975416e-08</td>\n",
       "      <td>-6.218808</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 482 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 s__Vagococcus_lutrae  s__Asaccharobacter_celatus  \\\n",
       "Participant Day                                                     \n",
       "602         3           -8.872499e-09                   -1.623746   \n",
       "            6            8.872499e-09                    0.744599   \n",
       "            10           1.337567e-06                   -7.295634   \n",
       "            13           7.558580e-07                   -3.707767   \n",
       "            17          -1.975416e-08                   -2.555618   \n",
       "\n",
       "                 s__Megasphaera_elsdenii  s__Leuconostoc_carnosum  \\\n",
       "Participant Day                                                     \n",
       "602         3                   0.012707            -8.872499e-09   \n",
       "            6                  -0.012820             8.872499e-09   \n",
       "            10                -14.144966             1.337567e-06   \n",
       "            13                 -6.069695             7.558580e-07   \n",
       "            17                 -4.850362            -1.975416e-08   \n",
       "\n",
       "                 s__Streptococcus_agalactiae  s__Tyzzerella_nexilis  \\\n",
       "Participant Day                                                       \n",
       "602         3                  -8.872499e-09          -8.872499e-09   \n",
       "            6                   8.872499e-09           8.872499e-09   \n",
       "            10                  1.337567e-06           1.337567e-06   \n",
       "            13                  7.558580e-07           7.558580e-07   \n",
       "            17                 -1.975416e-08          -1.975416e-08   \n",
       "\n",
       "                 s__Akkermansia_muciniphila  s__Alistipes_timonensis  \\\n",
       "Participant Day                                                        \n",
       "602         3                      0.086871            -8.872499e-09   \n",
       "            6                     -0.092439             8.872499e-09   \n",
       "            10                    -7.542312             1.337567e-06   \n",
       "            13                     0.164989             7.558580e-07   \n",
       "            17                     1.383432            -1.975416e-08   \n",
       "\n",
       "                 s__Peptostreptococcus_anaerobius  \\\n",
       "Participant Day                                     \n",
       "602         3                       -8.872499e-09   \n",
       "            6                        8.872499e-09   \n",
       "            10                       1.337567e-06   \n",
       "            13                       7.558580e-07   \n",
       "            17                      -1.975416e-08   \n",
       "\n",
       "                 s__Streptococcus_anginosus_group  ...  \\\n",
       "Participant Day                                    ...   \n",
       "602         3                       -8.872499e-09  ...   \n",
       "            6                        8.872499e-09  ...   \n",
       "            10                       1.114960e+01  ...   \n",
       "            13                       1.034272e+01  ...   \n",
       "            17                      -1.975416e-08  ...   \n",
       "\n",
       "                 s__Clostridium_celatum  s__Fusobacterium_periodonticum  \\\n",
       "Participant Day                                                           \n",
       "602         3                 -5.597918                   -8.872499e-09   \n",
       "            6                  0.985029                    8.872499e-09   \n",
       "            10                -5.597916                    1.337567e-06   \n",
       "            13                -5.597917                    7.558580e-07   \n",
       "            17                -5.597918                   -1.975416e-08   \n",
       "\n",
       "                 s__Acidaminococcus_intestini  s__Streptococcus_sobrinus  \\\n",
       "Participant Day                                                            \n",
       "602         3                       -0.278775              -8.872499e-09   \n",
       "            6                        0.233532               8.872499e-09   \n",
       "            10                      -8.090735               1.337567e-06   \n",
       "            13                      -8.090736               7.558580e-07   \n",
       "            17                      -8.090737              -1.975416e-08   \n",
       "\n",
       "                 s__Anaerostipes_caccae  s__Enterococcus_faecium  \\\n",
       "Participant Day                                                    \n",
       "602         3             -8.872499e-09            -8.872499e-09   \n",
       "            6              8.872499e-09             8.872499e-09   \n",
       "            10             1.337567e-06             1.337567e-06   \n",
       "            13             7.558580e-07             8.979058e+00   \n",
       "            17            -1.975416e-08             1.019982e+01   \n",
       "\n",
       "                 s__Eubacterium_sp_CAG_180  s__Veillonella_dispar  \\\n",
       "Participant Day                                                     \n",
       "602         3                    -0.040705          -8.872499e-09   \n",
       "            6                     0.039588           8.872499e-09   \n",
       "            10                  -12.389628           1.337567e-06   \n",
       "            13                   -1.991681           7.558580e-07   \n",
       "            17                   -0.769899          -1.975416e-08   \n",
       "\n",
       "                 s__Actinomyces_sp_S6_Spd3  s__Firmicutes_bacterium_CAG_238  \n",
       "Participant Day                                                              \n",
       "602         3                -8.872499e-09                        -0.403001  \n",
       "            6                 8.872499e-09                         0.314658  \n",
       "            10                1.337567e-06                        -6.218807  \n",
       "            13                7.558580e-07                        -6.218808  \n",
       "            17               -1.975416e-08                        -6.218808  \n",
       "\n",
       "[5 rows x 482 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_normalized = data_raw.groupby(level = 'Participant' )\\\n",
    "                    .apply(lambda x:np.log2(x/x.query(\"Phase == 'BAS'\").mean()))\n",
    "data_normalized = data_normalized.reset_index(level=['Phase','Group']\n",
    "                                              , drop = True)\n",
    "display(data_normalized.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(17, 482, 9)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{602: 0, 603: 1, 604: 2, 605: 3, 606: 4, 701: 5, 702: 6, 703: 7, 704: 8, 707: 9, 708: 10, 801: 11, 802: 12, 803: 13, 804: 14, 806: 15, 807: 16}\n",
      "{0: 602, 1: 603, 2: 604, 3: 605, 4: 606, 5: 701, 6: 702, 7: 703, 8: 704, 9: 707, 10: 708, 11: 801, 12: 802, 13: 803, 14: 804, 15: 806, 16: 807}\n"
     ]
    }
   ],
   "source": [
    "tensor_data, mode1_map, mode3_map = table2tensor(data_normalized)\n",
    "mode1_reverse_map = {val:k for k,val in mode1_map.items()}\n",
    "\n",
    "display(tensor_data.shape)\n",
    "print(mode1_map)\n",
    "print(mode1_reverse_map)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Transform"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The TCAM transformation is done by calling  `.fit` and `.transform` methods of a `TCAM` object.\n",
    "\n",
    "The `TCAM` class implements scikit-learn's transformer-mixin methods, so its API is rather similar to sklearn's `PCA`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tcam = TCAM(n_components=1.)    # will produce all components\n",
    "\n",
    "transformed_data = tcam.fit_transform(tensor_data)\n",
    "rounded_expvar = np.round(100*tcam.explained_variance_ratio_,2) # store explained variance ratios\n",
    "\n",
    "# cast the results data into a pandas dataframe, using the first mode identifiers (subjects)\n",
    "# as indices (row names)\n",
    "df_tca = pd.DataFrame(transformed_data).rename(index = mode1_reverse_map)\n",
    "df_tca.columns = [f'F{i}:{val}%' for i,val in enumerate(rounded_expvar, start = 1)]\n",
    "\n",
    "\n",
    "\n",
    "# append metadata to the factors dataframe\n",
    "df_plot = meta.merge(df_tca, left_on = 'Participant', right_index=True)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "f1, f2 = df_tca.columns[:2].tolist() # get the column names for the first 2 factors\n",
    "fig, ax = plt.subplots(figsize = [2,2], dpi = 500)\n",
    "sns.scatterplot(data = df_plot, x = f1, y = f2 , hue='Group', ax = ax, edgecolor = 'k')\n",
    "ax.legend(fontsize = 6, fancybox = False, framealpha = 1, edgecolor = 'k' )\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "celltoolbar": "Edit Metadata",
  "kernelspec": {
   "display_name": "mprod",
   "language": "python",
   "name": "mprod"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "417.5px"
   },
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}