feature_selection_with_variance.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "d526065f",
   "metadata": {},
   "source": [
    "# Feature selection with variance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "342beaa6",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import statistics\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import f_oneway, rankdata\n",
    "from sklearn.feature_selection import RFE\n",
    "from sklearn.feature_selection import mutual_info_classif\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from imblearn.over_sampling import SMOTE"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4fae360",
   "metadata": {},
   "source": [
    "## read in data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "242e963d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# read in file\n",
    "df = pd.read_csv(\"Brain_GSE50161.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c6eaa5fe",
   "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>samples</th>\n",
       "      <th>type</th>\n",
       "      <th>1007_s_at</th>\n",
       "      <th>1053_at</th>\n",
       "      <th>117_at</th>\n",
       "      <th>121_at</th>\n",
       "      <th>1255_g_at</th>\n",
       "      <th>1294_at</th>\n",
       "      <th>1316_at</th>\n",
       "      <th>1320_at</th>\n",
       "      <th>...</th>\n",
       "      <th>AFFX-r2-Ec-bioD-3_at</th>\n",
       "      <th>AFFX-r2-Ec-bioD-5_at</th>\n",
       "      <th>AFFX-r2-P1-cre-3_at</th>\n",
       "      <th>AFFX-r2-P1-cre-5_at</th>\n",
       "      <th>AFFX-ThrX-3_at</th>\n",
       "      <th>AFFX-ThrX-5_at</th>\n",
       "      <th>AFFX-ThrX-M_at</th>\n",
       "      <th>AFFX-TrpnX-3_at</th>\n",
       "      <th>AFFX-TrpnX-5_at</th>\n",
       "      <th>AFFX-TrpnX-M_at</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>834</td>\n",
       "      <td>ependymoma</td>\n",
       "      <td>12.498150</td>\n",
       "      <td>7.604868</td>\n",
       "      <td>6.880934</td>\n",
       "      <td>9.027128</td>\n",
       "      <td>4.176175</td>\n",
       "      <td>7.224920</td>\n",
       "      <td>6.085942</td>\n",
       "      <td>6.835999</td>\n",
       "      <td>...</td>\n",
       "      <td>9.979005</td>\n",
       "      <td>9.926470</td>\n",
       "      <td>12.719785</td>\n",
       "      <td>12.777792</td>\n",
       "      <td>5.403657</td>\n",
       "      <td>4.870548</td>\n",
       "      <td>4.047380</td>\n",
       "      <td>3.721936</td>\n",
       "      <td>4.516434</td>\n",
       "      <td>4.749940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>835</td>\n",
       "      <td>ependymoma</td>\n",
       "      <td>13.067436</td>\n",
       "      <td>7.998090</td>\n",
       "      <td>7.209076</td>\n",
       "      <td>9.723322</td>\n",
       "      <td>4.826126</td>\n",
       "      <td>7.539381</td>\n",
       "      <td>6.250962</td>\n",
       "      <td>8.012549</td>\n",
       "      <td>...</td>\n",
       "      <td>11.924749</td>\n",
       "      <td>11.215930</td>\n",
       "      <td>13.605662</td>\n",
       "      <td>13.401342</td>\n",
       "      <td>5.224555</td>\n",
       "      <td>4.895315</td>\n",
       "      <td>3.786437</td>\n",
       "      <td>3.564481</td>\n",
       "      <td>4.430891</td>\n",
       "      <td>4.491416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>836</td>\n",
       "      <td>ependymoma</td>\n",
       "      <td>13.068179</td>\n",
       "      <td>8.573674</td>\n",
       "      <td>8.647684</td>\n",
       "      <td>9.613002</td>\n",
       "      <td>4.396581</td>\n",
       "      <td>7.813101</td>\n",
       "      <td>6.007746</td>\n",
       "      <td>7.178156</td>\n",
       "      <td>...</td>\n",
       "      <td>12.154405</td>\n",
       "      <td>11.532460</td>\n",
       "      <td>13.764593</td>\n",
       "      <td>13.477800</td>\n",
       "      <td>5.303565</td>\n",
       "      <td>5.052184</td>\n",
       "      <td>4.005343</td>\n",
       "      <td>3.595382</td>\n",
       "      <td>4.563494</td>\n",
       "      <td>4.668827</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>837</td>\n",
       "      <td>ependymoma</td>\n",
       "      <td>12.456040</td>\n",
       "      <td>9.098977</td>\n",
       "      <td>6.628784</td>\n",
       "      <td>8.517677</td>\n",
       "      <td>4.154847</td>\n",
       "      <td>8.361843</td>\n",
       "      <td>6.596064</td>\n",
       "      <td>6.347285</td>\n",
       "      <td>...</td>\n",
       "      <td>11.969072</td>\n",
       "      <td>11.288801</td>\n",
       "      <td>13.600828</td>\n",
       "      <td>13.379029</td>\n",
       "      <td>4.953429</td>\n",
       "      <td>4.708371</td>\n",
       "      <td>3.892318</td>\n",
       "      <td>3.759429</td>\n",
       "      <td>4.748381</td>\n",
       "      <td>4.521275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>838</td>\n",
       "      <td>ependymoma</td>\n",
       "      <td>12.699958</td>\n",
       "      <td>8.800721</td>\n",
       "      <td>11.556188</td>\n",
       "      <td>9.166309</td>\n",
       "      <td>4.165891</td>\n",
       "      <td>7.923826</td>\n",
       "      <td>6.212754</td>\n",
       "      <td>6.866387</td>\n",
       "      <td>...</td>\n",
       "      <td>11.411701</td>\n",
       "      <td>11.169317</td>\n",
       "      <td>13.751442</td>\n",
       "      <td>13.803646</td>\n",
       "      <td>4.892677</td>\n",
       "      <td>4.773806</td>\n",
       "      <td>3.796856</td>\n",
       "      <td>3.577544</td>\n",
       "      <td>4.504385</td>\n",
       "      <td>4.541450</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>125</th>\n",
       "      <td>959</td>\n",
       "      <td>pilocytic_astrocytoma</td>\n",
       "      <td>12.658228</td>\n",
       "      <td>8.843270</td>\n",
       "      <td>7.672655</td>\n",
       "      <td>9.125912</td>\n",
       "      <td>5.495477</td>\n",
       "      <td>8.603892</td>\n",
       "      <td>7.747514</td>\n",
       "      <td>5.828978</td>\n",
       "      <td>...</td>\n",
       "      <td>13.170441</td>\n",
       "      <td>12.676080</td>\n",
       "      <td>14.124837</td>\n",
       "      <td>13.996436</td>\n",
       "      <td>4.913579</td>\n",
       "      <td>4.399176</td>\n",
       "      <td>3.878855</td>\n",
       "      <td>3.680103</td>\n",
       "      <td>4.726784</td>\n",
       "      <td>4.564637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>126</th>\n",
       "      <td>960</td>\n",
       "      <td>pilocytic_astrocytoma</td>\n",
       "      <td>12.812823</td>\n",
       "      <td>8.510550</td>\n",
       "      <td>8.729699</td>\n",
       "      <td>9.104402</td>\n",
       "      <td>3.967228</td>\n",
       "      <td>7.719089</td>\n",
       "      <td>7.092496</td>\n",
       "      <td>6.504812</td>\n",
       "      <td>...</td>\n",
       "      <td>13.040267</td>\n",
       "      <td>12.403316</td>\n",
       "      <td>13.978009</td>\n",
       "      <td>13.812916</td>\n",
       "      <td>5.189600</td>\n",
       "      <td>4.912618</td>\n",
       "      <td>3.764800</td>\n",
       "      <td>3.664920</td>\n",
       "      <td>4.628355</td>\n",
       "      <td>4.761351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>961</td>\n",
       "      <td>pilocytic_astrocytoma</td>\n",
       "      <td>12.706991</td>\n",
       "      <td>8.795721</td>\n",
       "      <td>7.772359</td>\n",
       "      <td>8.327273</td>\n",
       "      <td>6.329383</td>\n",
       "      <td>8.550471</td>\n",
       "      <td>6.613332</td>\n",
       "      <td>6.308945</td>\n",
       "      <td>...</td>\n",
       "      <td>12.825383</td>\n",
       "      <td>12.439265</td>\n",
       "      <td>14.328373</td>\n",
       "      <td>14.008693</td>\n",
       "      <td>4.931460</td>\n",
       "      <td>4.712895</td>\n",
       "      <td>3.913637</td>\n",
       "      <td>3.700964</td>\n",
       "      <td>4.764693</td>\n",
       "      <td>4.834952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>962</td>\n",
       "      <td>pilocytic_astrocytoma</td>\n",
       "      <td>12.684593</td>\n",
       "      <td>8.293938</td>\n",
       "      <td>7.228186</td>\n",
       "      <td>8.494428</td>\n",
       "      <td>6.049414</td>\n",
       "      <td>8.214729</td>\n",
       "      <td>7.287758</td>\n",
       "      <td>5.732710</td>\n",
       "      <td>...</td>\n",
       "      <td>13.116581</td>\n",
       "      <td>12.657967</td>\n",
       "      <td>14.390346</td>\n",
       "      <td>14.194904</td>\n",
       "      <td>4.871092</td>\n",
       "      <td>4.739400</td>\n",
       "      <td>3.782980</td>\n",
       "      <td>3.920363</td>\n",
       "      <td>4.665584</td>\n",
       "      <td>4.613326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129</th>\n",
       "      <td>963</td>\n",
       "      <td>pilocytic_astrocytoma</td>\n",
       "      <td>12.397722</td>\n",
       "      <td>8.843524</td>\n",
       "      <td>8.825100</td>\n",
       "      <td>8.551541</td>\n",
       "      <td>5.002072</td>\n",
       "      <td>8.547894</td>\n",
       "      <td>6.920827</td>\n",
       "      <td>5.738159</td>\n",
       "      <td>...</td>\n",
       "      <td>13.117180</td>\n",
       "      <td>12.699932</td>\n",
       "      <td>14.286779</td>\n",
       "      <td>14.088950</td>\n",
       "      <td>4.932377</td>\n",
       "      <td>4.700205</td>\n",
       "      <td>3.952850</td>\n",
       "      <td>3.678112</td>\n",
       "      <td>5.018501</td>\n",
       "      <td>4.704779</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>130 rows × 54677 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     samples                   type  1007_s_at   1053_at     117_at    121_at  \\\n",
       "0        834             ependymoma  12.498150  7.604868   6.880934  9.027128   \n",
       "1        835             ependymoma  13.067436  7.998090   7.209076  9.723322   \n",
       "2        836             ependymoma  13.068179  8.573674   8.647684  9.613002   \n",
       "3        837             ependymoma  12.456040  9.098977   6.628784  8.517677   \n",
       "4        838             ependymoma  12.699958  8.800721  11.556188  9.166309   \n",
       "..       ...                    ...        ...       ...        ...       ...   \n",
       "125      959  pilocytic_astrocytoma  12.658228  8.843270   7.672655  9.125912   \n",
       "126      960  pilocytic_astrocytoma  12.812823  8.510550   8.729699  9.104402   \n",
       "127      961  pilocytic_astrocytoma  12.706991  8.795721   7.772359  8.327273   \n",
       "128      962  pilocytic_astrocytoma  12.684593  8.293938   7.228186  8.494428   \n",
       "129      963  pilocytic_astrocytoma  12.397722  8.843524   8.825100  8.551541   \n",
       "\n",
       "     1255_g_at   1294_at   1316_at   1320_at  ...  AFFX-r2-Ec-bioD-3_at  \\\n",
       "0     4.176175  7.224920  6.085942  6.835999  ...              9.979005   \n",
       "1     4.826126  7.539381  6.250962  8.012549  ...             11.924749   \n",
       "2     4.396581  7.813101  6.007746  7.178156  ...             12.154405   \n",
       "3     4.154847  8.361843  6.596064  6.347285  ...             11.969072   \n",
       "4     4.165891  7.923826  6.212754  6.866387  ...             11.411701   \n",
       "..         ...       ...       ...       ...  ...                   ...   \n",
       "125   5.495477  8.603892  7.747514  5.828978  ...             13.170441   \n",
       "126   3.967228  7.719089  7.092496  6.504812  ...             13.040267   \n",
       "127   6.329383  8.550471  6.613332  6.308945  ...             12.825383   \n",
       "128   6.049414  8.214729  7.287758  5.732710  ...             13.116581   \n",
       "129   5.002072  8.547894  6.920827  5.738159  ...             13.117180   \n",
       "\n",
       "     AFFX-r2-Ec-bioD-5_at  AFFX-r2-P1-cre-3_at  AFFX-r2-P1-cre-5_at  \\\n",
       "0                9.926470            12.719785            12.777792   \n",
       "1               11.215930            13.605662            13.401342   \n",
       "2               11.532460            13.764593            13.477800   \n",
       "3               11.288801            13.600828            13.379029   \n",
       "4               11.169317            13.751442            13.803646   \n",
       "..                    ...                  ...                  ...   \n",
       "125             12.676080            14.124837            13.996436   \n",
       "126             12.403316            13.978009            13.812916   \n",
       "127             12.439265            14.328373            14.008693   \n",
       "128             12.657967            14.390346            14.194904   \n",
       "129             12.699932            14.286779            14.088950   \n",
       "\n",
       "     AFFX-ThrX-3_at  AFFX-ThrX-5_at  AFFX-ThrX-M_at  AFFX-TrpnX-3_at  \\\n",
       "0          5.403657        4.870548        4.047380         3.721936   \n",
       "1          5.224555        4.895315        3.786437         3.564481   \n",
       "2          5.303565        5.052184        4.005343         3.595382   \n",
       "3          4.953429        4.708371        3.892318         3.759429   \n",
       "4          4.892677        4.773806        3.796856         3.577544   \n",
       "..              ...             ...             ...              ...   \n",
       "125        4.913579        4.399176        3.878855         3.680103   \n",
       "126        5.189600        4.912618        3.764800         3.664920   \n",
       "127        4.931460        4.712895        3.913637         3.700964   \n",
       "128        4.871092        4.739400        3.782980         3.920363   \n",
       "129        4.932377        4.700205        3.952850         3.678112   \n",
       "\n",
       "     AFFX-TrpnX-5_at  AFFX-TrpnX-M_at  \n",
       "0           4.516434         4.749940  \n",
       "1           4.430891         4.491416  \n",
       "2           4.563494         4.668827  \n",
       "3           4.748381         4.521275  \n",
       "4           4.504385         4.541450  \n",
       "..               ...              ...  \n",
       "125         4.726784         4.564637  \n",
       "126         4.628355         4.761351  \n",
       "127         4.764693         4.834952  \n",
       "128         4.665584         4.613326  \n",
       "129         5.018501         4.704779  \n",
       "\n",
       "[130 rows x 54677 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "761c553b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Count the number of samples of five subtypes.\n",
    "sns.set(rc={'figure.figsize':(10,7)})\n",
    "cell_type = ['ependymoma', 'glioblastoma',' medulloblastoma', 'pilocytic_astrocytoma','normal']\n",
    "sns.set(font_scale = 1.3)\n",
    "def count_plot(label, title):\n",
    "    ax = sns.countplot(x=label)\n",
    "    for p in ax.patches:\n",
    "       ax.annotate('{:.1f}'.format(p.get_height()), (p.get_x()+0.25, p.get_height()+0.01))\n",
    "    \n",
    "    plt.title(title)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2d7e5b50",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "subtype_counts = count_plot(df[\"type\"], \"Five different subtype data counts\")\n",
    "subtype_counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c302ce7e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ependymoma               46\n",
      "glioblastoma             34\n",
      "medulloblastoma          22\n",
      "pilocytic_astrocytoma    15\n",
      "normal                   13\n",
      "Name: type, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# first column is the samples\n",
    "# second column is the types for brain cancers.\n",
    "print(df['type'].value_counts())\n",
    "# replace type to numbers\n",
    "df['type'] = df['type'].replace(['ependymoma', 'glioblastoma','medulloblastoma', 'pilocytic_astrocytoma','normal'],[0,1,2,3,4])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "20758745",
   "metadata": {},
   "source": [
    "## Filter Methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0aa69562",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000\n",
      "2000\n",
      "3000\n",
      "4000\n",
      "5000\n",
      "6000\n",
      "7000\n",
      "8000\n",
      "9000\n",
      "10000\n",
      "11000\n",
      "12000\n",
      "13000\n",
      "14000\n",
      "15000\n",
      "16000\n",
      "17000\n",
      "18000\n",
      "19000\n",
      "20000\n",
      "21000\n",
      "22000\n",
      "23000\n",
      "24000\n",
      "25000\n",
      "26000\n",
      "27000\n",
      "28000\n",
      "29000\n",
      "30000\n",
      "31000\n",
      "32000\n",
      "33000\n",
      "34000\n",
      "35000\n",
      "36000\n",
      "37000\n",
      "38000\n",
      "39000\n",
      "40000\n",
      "41000\n",
      "42000\n",
      "43000\n",
      "44000\n",
      "45000\n",
      "46000\n",
      "47000\n",
      "48000\n",
      "49000\n",
      "50000\n",
      "51000\n",
      "52000\n",
      "53000\n",
      "54000\n"
     ]
    }
   ],
   "source": [
    "# Use ANOVA to filter genes based on p-value\n",
    "\n",
    "stat_ANOVA = []\n",
    "\n",
    "# loop through each gene \n",
    "# to get gene expression values for a certain gene for each subtype\n",
    "# to get all gene expression values for each subtype (whole loop)\n",
    "for g in range(2, df.shape[1]):\n",
    "    \n",
    "    # subset type 0 (ependymoma) for gene g\n",
    "    # .iloc[:,g] means all rows and column g\n",
    "    type0 = np.array(df[df.type == 0].iloc[:,g])\n",
    "    \n",
    "    # subset type 1 (glioblastoma) for gene g\n",
    "    type1 = np.array(df[df.type == 1].iloc[:,g])\n",
    "    \n",
    "    # subset type 2 (medulloblastoma) for gene g\n",
    "    type2 = np.array(df[df.type == 2].iloc[:,g])\n",
    "    \n",
    "    # subset type 3 (pilocytic_astrocytoma) for gene g\n",
    "    type3 = np.array(df[df.type == 3].iloc[:,g])\n",
    "    \n",
    "    # subset type 4 (normal) for gene g\n",
    "    type4 = np.array(df[df.type == 4].iloc[:,g])\n",
    "    \n",
    "    # run 1-way between subjects ANOVA\n",
    "    # Returns: The computed F statistic of the test; The associated p-value from the F distribution.\n",
    "    result = f_oneway(type0, type1, type2, type3, type4)\n",
    "    \n",
    "    # append p-value\n",
    "    stat_ANOVA.append(result[1])\n",
    "        \n",
    "    if g%1000 == 0:\n",
    "        print(g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "639456a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# if the p value of the specific gene (between all subtypes) is less than 0.0001 then return the gene\n",
    "# first two colunms are sample and type\n",
    "stat_TF = [True, True]\n",
    "\n",
    "for i in stat_ANOVA:\n",
    "    if i <= 0.0001:\n",
    "        stat_TF.append(True)\n",
    "    else:\n",
    "        stat_TF.append(False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f0926a0f",
   "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>samples</th>\n",
       "      <th>type</th>\n",
       "      <th>1007_s_at</th>\n",
       "      <th>1053_at</th>\n",
       "      <th>117_at</th>\n",
       "      <th>1255_g_at</th>\n",
       "      <th>1294_at</th>\n",
       "      <th>1431_at</th>\n",
       "      <th>1438_at</th>\n",
       "      <th>1487_at</th>\n",
       "      <th>...</th>\n",
       "      <th>91920_at</th>\n",
       "      <th>AFFX-HSAC07/X00351_3_at</th>\n",
       "      <th>AFFX-HSAC07/X00351_5_at</th>\n",
       "      <th>AFFX-HSAC07/X00351_M_at</th>\n",
       "      <th>AFFX-HUMGAPDH/M33197_3_at</th>\n",
       "      <th>AFFX-HUMGAPDH/M33197_M_at</th>\n",
       "      <th>AFFX-HUMISGF3A/M97935_3_at</th>\n",
       "      <th>AFFX-HUMISGF3A/M97935_MB_at</th>\n",
       "      <th>AFFX-PheX-M_at</th>\n",
       "      <th>AFFX-r2-Bs-phe-M_at</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>834</td>\n",
       "      <td>0</td>\n",
       "      <td>12.498150</td>\n",
       "      <td>7.604868</td>\n",
       "      <td>6.880934</td>\n",
       "      <td>4.176175</td>\n",
       "      <td>7.224920</td>\n",
       "      <td>5.513410</td>\n",
       "      <td>7.541986</td>\n",
       "      <td>7.905140</td>\n",
       "      <td>...</td>\n",
       "      <td>10.956206</td>\n",
       "      <td>13.087353</td>\n",
       "      <td>12.786701</td>\n",
       "      <td>13.104597</td>\n",
       "      <td>13.027510</td>\n",
       "      <td>13.092592</td>\n",
       "      <td>9.104209</td>\n",
       "      <td>7.452986</td>\n",
       "      <td>3.882678</td>\n",
       "      <td>4.483687</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>835</td>\n",
       "      <td>0</td>\n",
       "      <td>13.067436</td>\n",
       "      <td>7.998090</td>\n",
       "      <td>7.209076</td>\n",
       "      <td>4.826126</td>\n",
       "      <td>7.539381</td>\n",
       "      <td>6.173106</td>\n",
       "      <td>8.317030</td>\n",
       "      <td>8.269136</td>\n",
       "      <td>...</td>\n",
       "      <td>10.255266</td>\n",
       "      <td>13.912946</td>\n",
       "      <td>13.532344</td>\n",
       "      <td>13.814658</td>\n",
       "      <td>14.237459</td>\n",
       "      <td>14.229670</td>\n",
       "      <td>10.411498</td>\n",
       "      <td>9.122687</td>\n",
       "      <td>3.744942</td>\n",
       "      <td>3.673009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>836</td>\n",
       "      <td>0</td>\n",
       "      <td>13.068179</td>\n",
       "      <td>8.573674</td>\n",
       "      <td>8.647684</td>\n",
       "      <td>4.396581</td>\n",
       "      <td>7.813101</td>\n",
       "      <td>6.323471</td>\n",
       "      <td>7.661874</td>\n",
       "      <td>8.327897</td>\n",
       "      <td>...</td>\n",
       "      <td>10.364209</td>\n",
       "      <td>14.378190</td>\n",
       "      <td>13.937428</td>\n",
       "      <td>14.248673</td>\n",
       "      <td>14.402930</td>\n",
       "      <td>14.447217</td>\n",
       "      <td>11.321259</td>\n",
       "      <td>9.822739</td>\n",
       "      <td>3.813011</td>\n",
       "      <td>3.541582</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>837</td>\n",
       "      <td>0</td>\n",
       "      <td>12.456040</td>\n",
       "      <td>9.098977</td>\n",
       "      <td>6.628784</td>\n",
       "      <td>4.154847</td>\n",
       "      <td>8.361843</td>\n",
       "      <td>6.008684</td>\n",
       "      <td>7.663790</td>\n",
       "      <td>7.730238</td>\n",
       "      <td>...</td>\n",
       "      <td>8.451971</td>\n",
       "      <td>13.755865</td>\n",
       "      <td>13.383430</td>\n",
       "      <td>13.660059</td>\n",
       "      <td>13.369370</td>\n",
       "      <td>13.457756</td>\n",
       "      <td>9.924395</td>\n",
       "      <td>8.028984</td>\n",
       "      <td>3.820466</td>\n",
       "      <td>3.593411</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>838</td>\n",
       "      <td>0</td>\n",
       "      <td>12.699958</td>\n",
       "      <td>8.800721</td>\n",
       "      <td>11.556188</td>\n",
       "      <td>4.165891</td>\n",
       "      <td>7.923826</td>\n",
       "      <td>5.279579</td>\n",
       "      <td>9.521956</td>\n",
       "      <td>7.581397</td>\n",
       "      <td>...</td>\n",
       "      <td>10.490535</td>\n",
       "      <td>14.077636</td>\n",
       "      <td>13.760155</td>\n",
       "      <td>14.042310</td>\n",
       "      <td>13.835195</td>\n",
       "      <td>14.029920</td>\n",
       "      <td>10.786574</td>\n",
       "      <td>9.086955</td>\n",
       "      <td>3.865816</td>\n",
       "      <td>4.355079</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>125</th>\n",
       "      <td>959</td>\n",
       "      <td>3</td>\n",
       "      <td>12.658228</td>\n",
       "      <td>8.843270</td>\n",
       "      <td>7.672655</td>\n",
       "      <td>5.495477</td>\n",
       "      <td>8.603892</td>\n",
       "      <td>4.967369</td>\n",
       "      <td>7.014209</td>\n",
       "      <td>8.666578</td>\n",
       "      <td>...</td>\n",
       "      <td>8.719159</td>\n",
       "      <td>13.967509</td>\n",
       "      <td>13.242701</td>\n",
       "      <td>13.713351</td>\n",
       "      <td>13.663155</td>\n",
       "      <td>13.609845</td>\n",
       "      <td>10.719271</td>\n",
       "      <td>8.359550</td>\n",
       "      <td>3.906140</td>\n",
       "      <td>3.892008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>126</th>\n",
       "      <td>960</td>\n",
       "      <td>3</td>\n",
       "      <td>12.812823</td>\n",
       "      <td>8.510550</td>\n",
       "      <td>8.729699</td>\n",
       "      <td>3.967228</td>\n",
       "      <td>7.719089</td>\n",
       "      <td>5.082896</td>\n",
       "      <td>7.981177</td>\n",
       "      <td>8.737005</td>\n",
       "      <td>...</td>\n",
       "      <td>9.022079</td>\n",
       "      <td>13.896949</td>\n",
       "      <td>13.232039</td>\n",
       "      <td>13.696252</td>\n",
       "      <td>14.382744</td>\n",
       "      <td>14.382647</td>\n",
       "      <td>10.303234</td>\n",
       "      <td>8.257289</td>\n",
       "      <td>4.002341</td>\n",
       "      <td>4.069697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>961</td>\n",
       "      <td>3</td>\n",
       "      <td>12.706991</td>\n",
       "      <td>8.795721</td>\n",
       "      <td>7.772359</td>\n",
       "      <td>6.329383</td>\n",
       "      <td>8.550471</td>\n",
       "      <td>4.919414</td>\n",
       "      <td>6.638493</td>\n",
       "      <td>8.724114</td>\n",
       "      <td>...</td>\n",
       "      <td>9.545324</td>\n",
       "      <td>13.823257</td>\n",
       "      <td>12.218713</td>\n",
       "      <td>13.097621</td>\n",
       "      <td>13.699895</td>\n",
       "      <td>13.556569</td>\n",
       "      <td>11.074307</td>\n",
       "      <td>8.822692</td>\n",
       "      <td>3.904377</td>\n",
       "      <td>3.685399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>962</td>\n",
       "      <td>3</td>\n",
       "      <td>12.684593</td>\n",
       "      <td>8.293938</td>\n",
       "      <td>7.228186</td>\n",
       "      <td>6.049414</td>\n",
       "      <td>8.214729</td>\n",
       "      <td>5.140635</td>\n",
       "      <td>7.225376</td>\n",
       "      <td>8.333919</td>\n",
       "      <td>...</td>\n",
       "      <td>10.455301</td>\n",
       "      <td>13.935221</td>\n",
       "      <td>13.133453</td>\n",
       "      <td>13.494138</td>\n",
       "      <td>13.499406</td>\n",
       "      <td>13.422512</td>\n",
       "      <td>10.848698</td>\n",
       "      <td>8.689571</td>\n",
       "      <td>3.809154</td>\n",
       "      <td>3.683920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129</th>\n",
       "      <td>963</td>\n",
       "      <td>3</td>\n",
       "      <td>12.397722</td>\n",
       "      <td>8.843524</td>\n",
       "      <td>8.825100</td>\n",
       "      <td>5.002072</td>\n",
       "      <td>8.547894</td>\n",
       "      <td>5.229208</td>\n",
       "      <td>6.839369</td>\n",
       "      <td>8.492447</td>\n",
       "      <td>...</td>\n",
       "      <td>8.478032</td>\n",
       "      <td>13.855325</td>\n",
       "      <td>13.000637</td>\n",
       "      <td>13.411613</td>\n",
       "      <td>13.939375</td>\n",
       "      <td>13.956265</td>\n",
       "      <td>10.665013</td>\n",
       "      <td>8.071967</td>\n",
       "      <td>3.797747</td>\n",
       "      <td>3.660039</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>130 rows × 21371 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     samples  type  1007_s_at   1053_at     117_at  1255_g_at   1294_at  \\\n",
       "0        834     0  12.498150  7.604868   6.880934   4.176175  7.224920   \n",
       "1        835     0  13.067436  7.998090   7.209076   4.826126  7.539381   \n",
       "2        836     0  13.068179  8.573674   8.647684   4.396581  7.813101   \n",
       "3        837     0  12.456040  9.098977   6.628784   4.154847  8.361843   \n",
       "4        838     0  12.699958  8.800721  11.556188   4.165891  7.923826   \n",
       "..       ...   ...        ...       ...        ...        ...       ...   \n",
       "125      959     3  12.658228  8.843270   7.672655   5.495477  8.603892   \n",
       "126      960     3  12.812823  8.510550   8.729699   3.967228  7.719089   \n",
       "127      961     3  12.706991  8.795721   7.772359   6.329383  8.550471   \n",
       "128      962     3  12.684593  8.293938   7.228186   6.049414  8.214729   \n",
       "129      963     3  12.397722  8.843524   8.825100   5.002072  8.547894   \n",
       "\n",
       "      1431_at   1438_at   1487_at  ...   91920_at  AFFX-HSAC07/X00351_3_at  \\\n",
       "0    5.513410  7.541986  7.905140  ...  10.956206                13.087353   \n",
       "1    6.173106  8.317030  8.269136  ...  10.255266                13.912946   \n",
       "2    6.323471  7.661874  8.327897  ...  10.364209                14.378190   \n",
       "3    6.008684  7.663790  7.730238  ...   8.451971                13.755865   \n",
       "4    5.279579  9.521956  7.581397  ...  10.490535                14.077636   \n",
       "..        ...       ...       ...  ...        ...                      ...   \n",
       "125  4.967369  7.014209  8.666578  ...   8.719159                13.967509   \n",
       "126  5.082896  7.981177  8.737005  ...   9.022079                13.896949   \n",
       "127  4.919414  6.638493  8.724114  ...   9.545324                13.823257   \n",
       "128  5.140635  7.225376  8.333919  ...  10.455301                13.935221   \n",
       "129  5.229208  6.839369  8.492447  ...   8.478032                13.855325   \n",
       "\n",
       "     AFFX-HSAC07/X00351_5_at  AFFX-HSAC07/X00351_M_at  \\\n",
       "0                  12.786701                13.104597   \n",
       "1                  13.532344                13.814658   \n",
       "2                  13.937428                14.248673   \n",
       "3                  13.383430                13.660059   \n",
       "4                  13.760155                14.042310   \n",
       "..                       ...                      ...   \n",
       "125                13.242701                13.713351   \n",
       "126                13.232039                13.696252   \n",
       "127                12.218713                13.097621   \n",
       "128                13.133453                13.494138   \n",
       "129                13.000637                13.411613   \n",
       "\n",
       "     AFFX-HUMGAPDH/M33197_3_at  AFFX-HUMGAPDH/M33197_M_at  \\\n",
       "0                    13.027510                  13.092592   \n",
       "1                    14.237459                  14.229670   \n",
       "2                    14.402930                  14.447217   \n",
       "3                    13.369370                  13.457756   \n",
       "4                    13.835195                  14.029920   \n",
       "..                         ...                        ...   \n",
       "125                  13.663155                  13.609845   \n",
       "126                  14.382744                  14.382647   \n",
       "127                  13.699895                  13.556569   \n",
       "128                  13.499406                  13.422512   \n",
       "129                  13.939375                  13.956265   \n",
       "\n",
       "     AFFX-HUMISGF3A/M97935_3_at  AFFX-HUMISGF3A/M97935_MB_at  AFFX-PheX-M_at  \\\n",
       "0                      9.104209                     7.452986        3.882678   \n",
       "1                     10.411498                     9.122687        3.744942   \n",
       "2                     11.321259                     9.822739        3.813011   \n",
       "3                      9.924395                     8.028984        3.820466   \n",
       "4                     10.786574                     9.086955        3.865816   \n",
       "..                          ...                          ...             ...   \n",
       "125                   10.719271                     8.359550        3.906140   \n",
       "126                   10.303234                     8.257289        4.002341   \n",
       "127                   11.074307                     8.822692        3.904377   \n",
       "128                   10.848698                     8.689571        3.809154   \n",
       "129                   10.665013                     8.071967        3.797747   \n",
       "\n",
       "     AFFX-r2-Bs-phe-M_at  \n",
       "0               4.483687  \n",
       "1               3.673009  \n",
       "2               3.541582  \n",
       "3               3.593411  \n",
       "4               4.355079  \n",
       "..                   ...  \n",
       "125             3.892008  \n",
       "126             4.069697  \n",
       "127             3.685399  \n",
       "128             3.683920  \n",
       "129             3.660039  \n",
       "\n",
       "[130 rows x 21371 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# stat_TF has the corresponding order with genes\n",
    "# if stat_TF = True, get the column through .loc\n",
    "df_ANOVA = df.loc[:,stat_TF]\n",
    "df_ANOVA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "18e8651e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000\n",
      "2000\n",
      "3000\n",
      "4000\n",
      "5000\n",
      "6000\n",
      "7000\n",
      "8000\n",
      "9000\n",
      "10000\n",
      "11000\n",
      "12000\n",
      "13000\n",
      "14000\n",
      "15000\n",
      "16000\n",
      "17000\n",
      "18000\n",
      "19000\n",
      "20000\n",
      "21000\n"
     ]
    }
   ],
   "source": [
    "# Use variance to filter\n",
    "stat_var = []\n",
    "\n",
    "# loop through each genes\n",
    "for g in range(2, df_ANOVA.shape[1]):\n",
    "    stat_var.append(statistics.variance(df_ANOVA.iloc[:,g]))\n",
    "    if g % 1000 == 0:\n",
    "        print(g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "9b1640c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5088812901520702"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.median(stat_var)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "53e1a58a",
   "metadata": {},
   "outputs": [],
   "source": [
    "var_TF = [True, True]\n",
    "for i in stat_var:\n",
    "    if i >= 0.5:\n",
    "        var_TF.append(True)\n",
    "    else:\n",
    "        var_TF.append(False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "20336257",
   "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>samples</th>\n",
       "      <th>type</th>\n",
       "      <th>1007_s_at</th>\n",
       "      <th>117_at</th>\n",
       "      <th>1255_g_at</th>\n",
       "      <th>1438_at</th>\n",
       "      <th>1552256_a_at</th>\n",
       "      <th>1552269_at</th>\n",
       "      <th>1552274_at</th>\n",
       "      <th>1552275_s_at</th>\n",
       "      <th>...</th>\n",
       "      <th>65588_at</th>\n",
       "      <th>823_at</th>\n",
       "      <th>87100_at</th>\n",
       "      <th>89977_at</th>\n",
       "      <th>90265_at</th>\n",
       "      <th>91816_f_at</th>\n",
       "      <th>91920_at</th>\n",
       "      <th>AFFX-HSAC07/X00351_5_at</th>\n",
       "      <th>AFFX-HUMISGF3A/M97935_3_at</th>\n",
       "      <th>AFFX-HUMISGF3A/M97935_MB_at</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>834</td>\n",
       "      <td>0</td>\n",
       "      <td>12.498150</td>\n",
       "      <td>6.880934</td>\n",
       "      <td>4.176175</td>\n",
       "      <td>7.541986</td>\n",
       "      <td>8.386617</td>\n",
       "      <td>8.222273</td>\n",
       "      <td>6.522746</td>\n",
       "      <td>7.031419</td>\n",
       "      <td>...</td>\n",
       "      <td>8.227121</td>\n",
       "      <td>7.232380</td>\n",
       "      <td>7.193072</td>\n",
       "      <td>7.166401</td>\n",
       "      <td>8.178087</td>\n",
       "      <td>7.762491</td>\n",
       "      <td>10.956206</td>\n",
       "      <td>12.786701</td>\n",
       "      <td>9.104209</td>\n",
       "      <td>7.452986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>835</td>\n",
       "      <td>0</td>\n",
       "      <td>13.067436</td>\n",
       "      <td>7.209076</td>\n",
       "      <td>4.826126</td>\n",
       "      <td>8.317030</td>\n",
       "      <td>9.208559</td>\n",
       "      <td>6.914568</td>\n",
       "      <td>7.805840</td>\n",
       "      <td>7.953560</td>\n",
       "      <td>...</td>\n",
       "      <td>9.021009</td>\n",
       "      <td>9.822285</td>\n",
       "      <td>8.391833</td>\n",
       "      <td>6.422113</td>\n",
       "      <td>10.667338</td>\n",
       "      <td>11.170625</td>\n",
       "      <td>10.255266</td>\n",
       "      <td>13.532344</td>\n",
       "      <td>10.411498</td>\n",
       "      <td>9.122687</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>836</td>\n",
       "      <td>0</td>\n",
       "      <td>13.068179</td>\n",
       "      <td>8.647684</td>\n",
       "      <td>4.396581</td>\n",
       "      <td>7.661874</td>\n",
       "      <td>9.344553</td>\n",
       "      <td>6.706525</td>\n",
       "      <td>7.702766</td>\n",
       "      <td>7.713712</td>\n",
       "      <td>...</td>\n",
       "      <td>8.395228</td>\n",
       "      <td>12.257851</td>\n",
       "      <td>8.582821</td>\n",
       "      <td>6.786965</td>\n",
       "      <td>9.574368</td>\n",
       "      <td>8.443621</td>\n",
       "      <td>10.364209</td>\n",
       "      <td>13.937428</td>\n",
       "      <td>11.321259</td>\n",
       "      <td>9.822739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>837</td>\n",
       "      <td>0</td>\n",
       "      <td>12.456040</td>\n",
       "      <td>6.628784</td>\n",
       "      <td>4.154847</td>\n",
       "      <td>7.663790</td>\n",
       "      <td>8.996408</td>\n",
       "      <td>8.985858</td>\n",
       "      <td>7.073578</td>\n",
       "      <td>7.603237</td>\n",
       "      <td>...</td>\n",
       "      <td>9.058105</td>\n",
       "      <td>8.526016</td>\n",
       "      <td>6.461401</td>\n",
       "      <td>6.104773</td>\n",
       "      <td>8.743637</td>\n",
       "      <td>9.597551</td>\n",
       "      <td>8.451971</td>\n",
       "      <td>13.383430</td>\n",
       "      <td>9.924395</td>\n",
       "      <td>8.028984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>838</td>\n",
       "      <td>0</td>\n",
       "      <td>12.699958</td>\n",
       "      <td>11.556188</td>\n",
       "      <td>4.165891</td>\n",
       "      <td>9.521956</td>\n",
       "      <td>10.040652</td>\n",
       "      <td>8.221752</td>\n",
       "      <td>7.916434</td>\n",
       "      <td>8.139826</td>\n",
       "      <td>...</td>\n",
       "      <td>8.123893</td>\n",
       "      <td>9.156358</td>\n",
       "      <td>7.462321</td>\n",
       "      <td>7.626676</td>\n",
       "      <td>8.123024</td>\n",
       "      <td>8.106023</td>\n",
       "      <td>10.490535</td>\n",
       "      <td>13.760155</td>\n",
       "      <td>10.786574</td>\n",
       "      <td>9.086955</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>125</th>\n",
       "      <td>959</td>\n",
       "      <td>3</td>\n",
       "      <td>12.658228</td>\n",
       "      <td>7.672655</td>\n",
       "      <td>5.495477</td>\n",
       "      <td>7.014209</td>\n",
       "      <td>10.233177</td>\n",
       "      <td>4.996330</td>\n",
       "      <td>7.350546</td>\n",
       "      <td>7.826477</td>\n",
       "      <td>...</td>\n",
       "      <td>8.665639</td>\n",
       "      <td>10.151682</td>\n",
       "      <td>8.639360</td>\n",
       "      <td>6.926578</td>\n",
       "      <td>8.559671</td>\n",
       "      <td>9.353849</td>\n",
       "      <td>8.719159</td>\n",
       "      <td>13.242701</td>\n",
       "      <td>10.719271</td>\n",
       "      <td>8.359550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>126</th>\n",
       "      <td>960</td>\n",
       "      <td>3</td>\n",
       "      <td>12.812823</td>\n",
       "      <td>8.729699</td>\n",
       "      <td>3.967228</td>\n",
       "      <td>7.981177</td>\n",
       "      <td>9.902292</td>\n",
       "      <td>5.379189</td>\n",
       "      <td>8.527913</td>\n",
       "      <td>8.150819</td>\n",
       "      <td>...</td>\n",
       "      <td>9.805503</td>\n",
       "      <td>10.226974</td>\n",
       "      <td>7.584308</td>\n",
       "      <td>6.476481</td>\n",
       "      <td>8.753927</td>\n",
       "      <td>9.382903</td>\n",
       "      <td>9.022079</td>\n",
       "      <td>13.232039</td>\n",
       "      <td>10.303234</td>\n",
       "      <td>8.257289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>961</td>\n",
       "      <td>3</td>\n",
       "      <td>12.706991</td>\n",
       "      <td>7.772359</td>\n",
       "      <td>6.329383</td>\n",
       "      <td>6.638493</td>\n",
       "      <td>9.638112</td>\n",
       "      <td>4.627095</td>\n",
       "      <td>7.797212</td>\n",
       "      <td>8.870719</td>\n",
       "      <td>...</td>\n",
       "      <td>8.518508</td>\n",
       "      <td>9.964082</td>\n",
       "      <td>8.383192</td>\n",
       "      <td>7.098009</td>\n",
       "      <td>8.466273</td>\n",
       "      <td>8.744321</td>\n",
       "      <td>9.545324</td>\n",
       "      <td>12.218713</td>\n",
       "      <td>11.074307</td>\n",
       "      <td>8.822692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>962</td>\n",
       "      <td>3</td>\n",
       "      <td>12.684593</td>\n",
       "      <td>7.228186</td>\n",
       "      <td>6.049414</td>\n",
       "      <td>7.225376</td>\n",
       "      <td>10.145374</td>\n",
       "      <td>4.917388</td>\n",
       "      <td>8.126196</td>\n",
       "      <td>8.580536</td>\n",
       "      <td>...</td>\n",
       "      <td>9.374640</td>\n",
       "      <td>9.843288</td>\n",
       "      <td>9.016851</td>\n",
       "      <td>8.037112</td>\n",
       "      <td>8.469799</td>\n",
       "      <td>8.441841</td>\n",
       "      <td>10.455301</td>\n",
       "      <td>13.133453</td>\n",
       "      <td>10.848698</td>\n",
       "      <td>8.689571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129</th>\n",
       "      <td>963</td>\n",
       "      <td>3</td>\n",
       "      <td>12.397722</td>\n",
       "      <td>8.825100</td>\n",
       "      <td>5.002072</td>\n",
       "      <td>6.839369</td>\n",
       "      <td>9.731481</td>\n",
       "      <td>4.435104</td>\n",
       "      <td>7.350742</td>\n",
       "      <td>7.187392</td>\n",
       "      <td>...</td>\n",
       "      <td>9.142607</td>\n",
       "      <td>10.071155</td>\n",
       "      <td>8.022076</td>\n",
       "      <td>7.307878</td>\n",
       "      <td>8.660479</td>\n",
       "      <td>8.800392</td>\n",
       "      <td>8.478032</td>\n",
       "      <td>13.000637</td>\n",
       "      <td>10.665013</td>\n",
       "      <td>8.071967</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>130 rows × 10850 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     samples  type  1007_s_at     117_at  1255_g_at   1438_at  1552256_a_at  \\\n",
       "0        834     0  12.498150   6.880934   4.176175  7.541986      8.386617   \n",
       "1        835     0  13.067436   7.209076   4.826126  8.317030      9.208559   \n",
       "2        836     0  13.068179   8.647684   4.396581  7.661874      9.344553   \n",
       "3        837     0  12.456040   6.628784   4.154847  7.663790      8.996408   \n",
       "4        838     0  12.699958  11.556188   4.165891  9.521956     10.040652   \n",
       "..       ...   ...        ...        ...        ...       ...           ...   \n",
       "125      959     3  12.658228   7.672655   5.495477  7.014209     10.233177   \n",
       "126      960     3  12.812823   8.729699   3.967228  7.981177      9.902292   \n",
       "127      961     3  12.706991   7.772359   6.329383  6.638493      9.638112   \n",
       "128      962     3  12.684593   7.228186   6.049414  7.225376     10.145374   \n",
       "129      963     3  12.397722   8.825100   5.002072  6.839369      9.731481   \n",
       "\n",
       "     1552269_at  1552274_at  1552275_s_at  ...  65588_at     823_at  87100_at  \\\n",
       "0      8.222273    6.522746      7.031419  ...  8.227121   7.232380  7.193072   \n",
       "1      6.914568    7.805840      7.953560  ...  9.021009   9.822285  8.391833   \n",
       "2      6.706525    7.702766      7.713712  ...  8.395228  12.257851  8.582821   \n",
       "3      8.985858    7.073578      7.603237  ...  9.058105   8.526016  6.461401   \n",
       "4      8.221752    7.916434      8.139826  ...  8.123893   9.156358  7.462321   \n",
       "..          ...         ...           ...  ...       ...        ...       ...   \n",
       "125    4.996330    7.350546      7.826477  ...  8.665639  10.151682  8.639360   \n",
       "126    5.379189    8.527913      8.150819  ...  9.805503  10.226974  7.584308   \n",
       "127    4.627095    7.797212      8.870719  ...  8.518508   9.964082  8.383192   \n",
       "128    4.917388    8.126196      8.580536  ...  9.374640   9.843288  9.016851   \n",
       "129    4.435104    7.350742      7.187392  ...  9.142607  10.071155  8.022076   \n",
       "\n",
       "     89977_at   90265_at  91816_f_at   91920_at  AFFX-HSAC07/X00351_5_at  \\\n",
       "0    7.166401   8.178087    7.762491  10.956206                12.786701   \n",
       "1    6.422113  10.667338   11.170625  10.255266                13.532344   \n",
       "2    6.786965   9.574368    8.443621  10.364209                13.937428   \n",
       "3    6.104773   8.743637    9.597551   8.451971                13.383430   \n",
       "4    7.626676   8.123024    8.106023  10.490535                13.760155   \n",
       "..        ...        ...         ...        ...                      ...   \n",
       "125  6.926578   8.559671    9.353849   8.719159                13.242701   \n",
       "126  6.476481   8.753927    9.382903   9.022079                13.232039   \n",
       "127  7.098009   8.466273    8.744321   9.545324                12.218713   \n",
       "128  8.037112   8.469799    8.441841  10.455301                13.133453   \n",
       "129  7.307878   8.660479    8.800392   8.478032                13.000637   \n",
       "\n",
       "     AFFX-HUMISGF3A/M97935_3_at  AFFX-HUMISGF3A/M97935_MB_at  \n",
       "0                      9.104209                     7.452986  \n",
       "1                     10.411498                     9.122687  \n",
       "2                     11.321259                     9.822739  \n",
       "3                      9.924395                     8.028984  \n",
       "4                     10.786574                     9.086955  \n",
       "..                          ...                          ...  \n",
       "125                   10.719271                     8.359550  \n",
       "126                   10.303234                     8.257289  \n",
       "127                   11.074307                     8.822692  \n",
       "128                   10.848698                     8.689571  \n",
       "129                   10.665013                     8.071967  \n",
       "\n",
       "[130 rows x 10850 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_var = df_ANOVA.loc[:,var_TF]\n",
    "df_var"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "52e5d6fb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "IG done\n"
     ]
    }
   ],
   "source": [
    "# Use Information Gain to filter\n",
    "\n",
    "# x is columns for genes\n",
    "# y is the cancer type\n",
    "X = np.array(df_var.iloc[:,2:])\n",
    "y = np.array(df_var.iloc[:,1])\n",
    "\n",
    "IG = mutual_info_classif(X,y,n_neighbors = 5)\n",
    "print(\"IG done\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "c6af515c",
   "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>samples</th>\n",
       "      <th>type</th>\n",
       "      <th>1438_at</th>\n",
       "      <th>1552274_at</th>\n",
       "      <th>1552275_s_at</th>\n",
       "      <th>1552301_a_at</th>\n",
       "      <th>1552309_a_at</th>\n",
       "      <th>1552312_a_at</th>\n",
       "      <th>1552316_a_at</th>\n",
       "      <th>1552343_s_at</th>\n",
       "      <th>...</th>\n",
       "      <th>53071_s_at</th>\n",
       "      <th>53991_at</th>\n",
       "      <th>55081_at</th>\n",
       "      <th>57163_at</th>\n",
       "      <th>59644_at</th>\n",
       "      <th>61297_at</th>\n",
       "      <th>65588_at</th>\n",
       "      <th>89977_at</th>\n",
       "      <th>AFFX-HSAC07/X00351_5_at</th>\n",
       "      <th>AFFX-HUMISGF3A/M97935_MB_at</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>834</td>\n",
       "      <td>0</td>\n",
       "      <td>7.541986</td>\n",
       "      <td>6.522746</td>\n",
       "      <td>7.031419</td>\n",
       "      <td>7.160071</td>\n",
       "      <td>6.998608</td>\n",
       "      <td>7.363721</td>\n",
       "      <td>5.409652</td>\n",
       "      <td>6.035313</td>\n",
       "      <td>...</td>\n",
       "      <td>8.183152</td>\n",
       "      <td>8.856580</td>\n",
       "      <td>6.916262</td>\n",
       "      <td>7.913747</td>\n",
       "      <td>6.543144</td>\n",
       "      <td>8.215540</td>\n",
       "      <td>8.227121</td>\n",
       "      <td>7.166401</td>\n",
       "      <td>12.786701</td>\n",
       "      <td>7.452986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>835</td>\n",
       "      <td>0</td>\n",
       "      <td>8.317030</td>\n",
       "      <td>7.805840</td>\n",
       "      <td>7.953560</td>\n",
       "      <td>6.529700</td>\n",
       "      <td>5.850427</td>\n",
       "      <td>9.629058</td>\n",
       "      <td>5.343544</td>\n",
       "      <td>6.244279</td>\n",
       "      <td>...</td>\n",
       "      <td>9.191431</td>\n",
       "      <td>6.764527</td>\n",
       "      <td>8.056402</td>\n",
       "      <td>8.941896</td>\n",
       "      <td>7.792050</td>\n",
       "      <td>9.819146</td>\n",
       "      <td>9.021009</td>\n",
       "      <td>6.422113</td>\n",
       "      <td>13.532344</td>\n",
       "      <td>9.122687</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>836</td>\n",
       "      <td>0</td>\n",
       "      <td>7.661874</td>\n",
       "      <td>7.702766</td>\n",
       "      <td>7.713712</td>\n",
       "      <td>8.191684</td>\n",
       "      <td>7.742723</td>\n",
       "      <td>7.989959</td>\n",
       "      <td>5.454159</td>\n",
       "      <td>6.211971</td>\n",
       "      <td>...</td>\n",
       "      <td>9.263549</td>\n",
       "      <td>8.960278</td>\n",
       "      <td>8.520088</td>\n",
       "      <td>9.040109</td>\n",
       "      <td>7.771230</td>\n",
       "      <td>8.715964</td>\n",
       "      <td>8.395228</td>\n",
       "      <td>6.786965</td>\n",
       "      <td>13.937428</td>\n",
       "      <td>9.822739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>837</td>\n",
       "      <td>0</td>\n",
       "      <td>7.663790</td>\n",
       "      <td>7.073578</td>\n",
       "      <td>7.603237</td>\n",
       "      <td>5.758467</td>\n",
       "      <td>7.417490</td>\n",
       "      <td>7.954845</td>\n",
       "      <td>6.253341</td>\n",
       "      <td>7.136300</td>\n",
       "      <td>...</td>\n",
       "      <td>10.339334</td>\n",
       "      <td>10.307893</td>\n",
       "      <td>10.024374</td>\n",
       "      <td>9.342572</td>\n",
       "      <td>8.635930</td>\n",
       "      <td>9.115695</td>\n",
       "      <td>9.058105</td>\n",
       "      <td>6.104773</td>\n",
       "      <td>13.383430</td>\n",
       "      <td>8.028984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>838</td>\n",
       "      <td>0</td>\n",
       "      <td>9.521956</td>\n",
       "      <td>7.916434</td>\n",
       "      <td>8.139826</td>\n",
       "      <td>7.042512</td>\n",
       "      <td>7.179180</td>\n",
       "      <td>8.816403</td>\n",
       "      <td>5.502234</td>\n",
       "      <td>6.502507</td>\n",
       "      <td>...</td>\n",
       "      <td>8.723818</td>\n",
       "      <td>9.719175</td>\n",
       "      <td>7.451403</td>\n",
       "      <td>8.676021</td>\n",
       "      <td>7.422708</td>\n",
       "      <td>8.689031</td>\n",
       "      <td>8.123893</td>\n",
       "      <td>7.626676</td>\n",
       "      <td>13.760155</td>\n",
       "      <td>9.086955</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>125</th>\n",
       "      <td>959</td>\n",
       "      <td>3</td>\n",
       "      <td>7.014209</td>\n",
       "      <td>7.350546</td>\n",
       "      <td>7.826477</td>\n",
       "      <td>6.133921</td>\n",
       "      <td>7.299499</td>\n",
       "      <td>6.570050</td>\n",
       "      <td>7.712112</td>\n",
       "      <td>7.448245</td>\n",
       "      <td>...</td>\n",
       "      <td>10.482143</td>\n",
       "      <td>9.400551</td>\n",
       "      <td>8.849441</td>\n",
       "      <td>9.790994</td>\n",
       "      <td>7.989509</td>\n",
       "      <td>9.151995</td>\n",
       "      <td>8.665639</td>\n",
       "      <td>6.926578</td>\n",
       "      <td>13.242701</td>\n",
       "      <td>8.359550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>126</th>\n",
       "      <td>960</td>\n",
       "      <td>3</td>\n",
       "      <td>7.981177</td>\n",
       "      <td>8.527913</td>\n",
       "      <td>8.150819</td>\n",
       "      <td>6.679918</td>\n",
       "      <td>7.388121</td>\n",
       "      <td>7.773944</td>\n",
       "      <td>6.707031</td>\n",
       "      <td>7.609170</td>\n",
       "      <td>...</td>\n",
       "      <td>9.311126</td>\n",
       "      <td>9.609665</td>\n",
       "      <td>8.187313</td>\n",
       "      <td>9.055682</td>\n",
       "      <td>7.156948</td>\n",
       "      <td>9.135004</td>\n",
       "      <td>9.805503</td>\n",
       "      <td>6.476481</td>\n",
       "      <td>13.232039</td>\n",
       "      <td>8.257289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>961</td>\n",
       "      <td>3</td>\n",
       "      <td>6.638493</td>\n",
       "      <td>7.797212</td>\n",
       "      <td>8.870719</td>\n",
       "      <td>6.233379</td>\n",
       "      <td>7.548870</td>\n",
       "      <td>7.181821</td>\n",
       "      <td>8.336774</td>\n",
       "      <td>7.059833</td>\n",
       "      <td>...</td>\n",
       "      <td>10.290373</td>\n",
       "      <td>9.719978</td>\n",
       "      <td>8.714027</td>\n",
       "      <td>9.361393</td>\n",
       "      <td>8.063106</td>\n",
       "      <td>9.554597</td>\n",
       "      <td>8.518508</td>\n",
       "      <td>7.098009</td>\n",
       "      <td>12.218713</td>\n",
       "      <td>8.822692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>962</td>\n",
       "      <td>3</td>\n",
       "      <td>7.225376</td>\n",
       "      <td>8.126196</td>\n",
       "      <td>8.580536</td>\n",
       "      <td>5.912674</td>\n",
       "      <td>7.800654</td>\n",
       "      <td>6.194283</td>\n",
       "      <td>7.951095</td>\n",
       "      <td>6.789710</td>\n",
       "      <td>...</td>\n",
       "      <td>9.652443</td>\n",
       "      <td>9.914592</td>\n",
       "      <td>8.153587</td>\n",
       "      <td>9.111973</td>\n",
       "      <td>7.787384</td>\n",
       "      <td>9.712491</td>\n",
       "      <td>9.374640</td>\n",
       "      <td>8.037112</td>\n",
       "      <td>13.133453</td>\n",
       "      <td>8.689571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129</th>\n",
       "      <td>963</td>\n",
       "      <td>3</td>\n",
       "      <td>6.839369</td>\n",
       "      <td>7.350742</td>\n",
       "      <td>7.187392</td>\n",
       "      <td>6.266402</td>\n",
       "      <td>7.318047</td>\n",
       "      <td>6.030525</td>\n",
       "      <td>6.937213</td>\n",
       "      <td>7.769401</td>\n",
       "      <td>...</td>\n",
       "      <td>9.969752</td>\n",
       "      <td>10.169648</td>\n",
       "      <td>8.711477</td>\n",
       "      <td>9.186671</td>\n",
       "      <td>8.260522</td>\n",
       "      <td>8.700461</td>\n",
       "      <td>9.142607</td>\n",
       "      <td>7.307878</td>\n",
       "      <td>13.000637</td>\n",
       "      <td>8.071967</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>130 rows × 3002 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     samples  type   1438_at  1552274_at  1552275_s_at  1552301_a_at  \\\n",
       "0        834     0  7.541986    6.522746      7.031419      7.160071   \n",
       "1        835     0  8.317030    7.805840      7.953560      6.529700   \n",
       "2        836     0  7.661874    7.702766      7.713712      8.191684   \n",
       "3        837     0  7.663790    7.073578      7.603237      5.758467   \n",
       "4        838     0  9.521956    7.916434      8.139826      7.042512   \n",
       "..       ...   ...       ...         ...           ...           ...   \n",
       "125      959     3  7.014209    7.350546      7.826477      6.133921   \n",
       "126      960     3  7.981177    8.527913      8.150819      6.679918   \n",
       "127      961     3  6.638493    7.797212      8.870719      6.233379   \n",
       "128      962     3  7.225376    8.126196      8.580536      5.912674   \n",
       "129      963     3  6.839369    7.350742      7.187392      6.266402   \n",
       "\n",
       "     1552309_a_at  1552312_a_at  1552316_a_at  1552343_s_at  ...  53071_s_at  \\\n",
       "0        6.998608      7.363721      5.409652      6.035313  ...    8.183152   \n",
       "1        5.850427      9.629058      5.343544      6.244279  ...    9.191431   \n",
       "2        7.742723      7.989959      5.454159      6.211971  ...    9.263549   \n",
       "3        7.417490      7.954845      6.253341      7.136300  ...   10.339334   \n",
       "4        7.179180      8.816403      5.502234      6.502507  ...    8.723818   \n",
       "..            ...           ...           ...           ...  ...         ...   \n",
       "125      7.299499      6.570050      7.712112      7.448245  ...   10.482143   \n",
       "126      7.388121      7.773944      6.707031      7.609170  ...    9.311126   \n",
       "127      7.548870      7.181821      8.336774      7.059833  ...   10.290373   \n",
       "128      7.800654      6.194283      7.951095      6.789710  ...    9.652443   \n",
       "129      7.318047      6.030525      6.937213      7.769401  ...    9.969752   \n",
       "\n",
       "      53991_at   55081_at  57163_at  59644_at  61297_at  65588_at  89977_at  \\\n",
       "0     8.856580   6.916262  7.913747  6.543144  8.215540  8.227121  7.166401   \n",
       "1     6.764527   8.056402  8.941896  7.792050  9.819146  9.021009  6.422113   \n",
       "2     8.960278   8.520088  9.040109  7.771230  8.715964  8.395228  6.786965   \n",
       "3    10.307893  10.024374  9.342572  8.635930  9.115695  9.058105  6.104773   \n",
       "4     9.719175   7.451403  8.676021  7.422708  8.689031  8.123893  7.626676   \n",
       "..         ...        ...       ...       ...       ...       ...       ...   \n",
       "125   9.400551   8.849441  9.790994  7.989509  9.151995  8.665639  6.926578   \n",
       "126   9.609665   8.187313  9.055682  7.156948  9.135004  9.805503  6.476481   \n",
       "127   9.719978   8.714027  9.361393  8.063106  9.554597  8.518508  7.098009   \n",
       "128   9.914592   8.153587  9.111973  7.787384  9.712491  9.374640  8.037112   \n",
       "129  10.169648   8.711477  9.186671  8.260522  8.700461  9.142607  7.307878   \n",
       "\n",
       "     AFFX-HSAC07/X00351_5_at  AFFX-HUMISGF3A/M97935_MB_at  \n",
       "0                  12.786701                     7.452986  \n",
       "1                  13.532344                     9.122687  \n",
       "2                  13.937428                     9.822739  \n",
       "3                  13.383430                     8.028984  \n",
       "4                  13.760155                     9.086955  \n",
       "..                       ...                          ...  \n",
       "125                13.242701                     8.359550  \n",
       "126                13.232039                     8.257289  \n",
       "127                12.218713                     8.822692  \n",
       "128                13.133453                     8.689571  \n",
       "129                13.000637                     8.071967  \n",
       "\n",
       "[130 rows x 3002 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Select top 3000 feature based on importance from Information Gain\n",
    "rank = rankdata(IG)\n",
    "rank_TF = [True, True]\n",
    "for i in rank:\n",
    "    if i <= 3000:\n",
    "        rank_TF.append(True)\n",
    "    else:\n",
    "        rank_TF.append(False)\n",
    "\n",
    "df_ig = df_var.loc[:,rank_TF]\n",
    "df_ig"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f69f424",
   "metadata": {},
   "source": [
    "## Wrapper Methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "efb73e05",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rfe done\n",
      "fit done\n"
     ]
    }
   ],
   "source": [
    "#x is columns for genes (features)\n",
    "#y is the cancer type\n",
    "X_new = df_ig.iloc[:,2:]\n",
    "y_new = df_ig.iloc[:,1]\n",
    "model = LinearRegression()\n",
    "rfe = RFE(model, n_features_to_select = 300)\n",
    "print(\"rfe done\")\n",
    "#Use RFE to select the top 300 features \n",
    "fit = rfe.fit(X_new, y_new)\n",
    "print('fit done')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "192dc28b",
   "metadata": {},
   "outputs": [],
   "source": [
    "RFE_results = []\n",
    "for i in range(X_new.shape[1]):\n",
    "    RFE_results.append(rfe.support_[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "ff08f125",
   "metadata": {},
   "outputs": [],
   "source": [
    "RFE_TF = [True, True]\n",
    "for i in RFE_results:\n",
    "    if i == True:\n",
    "        RFE_TF.append(True)\n",
    "    else:\n",
    "        RFE_TF.append(False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "891fc494",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_RFE = df_ig.loc[:,RFE_TF]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "b9f5aa55",
   "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>type</th>\n",
       "      <th>1552508_at</th>\n",
       "      <th>1552516_a_at</th>\n",
       "      <th>1552575_a_at</th>\n",
       "      <th>1552807_a_at</th>\n",
       "      <th>1553479_at</th>\n",
       "      <th>1553654_at</th>\n",
       "      <th>1554018_at</th>\n",
       "      <th>1554507_at</th>\n",
       "      <th>1554544_a_at</th>\n",
       "      <th>...</th>\n",
       "      <th>242134_at</th>\n",
       "      <th>242888_at</th>\n",
       "      <th>242931_at</th>\n",
       "      <th>242943_at</th>\n",
       "      <th>243584_at</th>\n",
       "      <th>244163_at</th>\n",
       "      <th>244493_at</th>\n",
       "      <th>244503_at</th>\n",
       "      <th>244751_at</th>\n",
       "      <th>53071_s_at</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>5.989892</td>\n",
       "      <td>8.660293</td>\n",
       "      <td>6.891804</td>\n",
       "      <td>7.924462</td>\n",
       "      <td>6.895976</td>\n",
       "      <td>4.827520</td>\n",
       "      <td>4.369720</td>\n",
       "      <td>4.015369</td>\n",
       "      <td>8.396550</td>\n",
       "      <td>...</td>\n",
       "      <td>7.226616</td>\n",
       "      <td>6.705961</td>\n",
       "      <td>8.573175</td>\n",
       "      <td>5.559319</td>\n",
       "      <td>6.382599</td>\n",
       "      <td>3.841604</td>\n",
       "      <td>3.423704</td>\n",
       "      <td>3.627346</td>\n",
       "      <td>6.001562</td>\n",
       "      <td>8.183152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>8.319513</td>\n",
       "      <td>9.980202</td>\n",
       "      <td>6.826210</td>\n",
       "      <td>7.493678</td>\n",
       "      <td>7.775287</td>\n",
       "      <td>6.873171</td>\n",
       "      <td>3.636735</td>\n",
       "      <td>3.204273</td>\n",
       "      <td>8.035201</td>\n",
       "      <td>...</td>\n",
       "      <td>8.765334</td>\n",
       "      <td>6.234970</td>\n",
       "      <td>6.750153</td>\n",
       "      <td>8.727043</td>\n",
       "      <td>6.476319</td>\n",
       "      <td>4.599455</td>\n",
       "      <td>3.214800</td>\n",
       "      <td>4.951005</td>\n",
       "      <td>5.518132</td>\n",
       "      <td>9.191431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>7.261455</td>\n",
       "      <td>8.212629</td>\n",
       "      <td>6.270775</td>\n",
       "      <td>8.045542</td>\n",
       "      <td>8.177457</td>\n",
       "      <td>4.041513</td>\n",
       "      <td>6.366196</td>\n",
       "      <td>4.369383</td>\n",
       "      <td>8.530794</td>\n",
       "      <td>...</td>\n",
       "      <td>6.562527</td>\n",
       "      <td>6.614429</td>\n",
       "      <td>8.047375</td>\n",
       "      <td>7.003226</td>\n",
       "      <td>6.268696</td>\n",
       "      <td>3.411819</td>\n",
       "      <td>3.078852</td>\n",
       "      <td>3.624622</td>\n",
       "      <td>5.927809</td>\n",
       "      <td>9.263549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>5.883377</td>\n",
       "      <td>6.236439</td>\n",
       "      <td>6.029005</td>\n",
       "      <td>8.071778</td>\n",
       "      <td>5.224647</td>\n",
       "      <td>4.879736</td>\n",
       "      <td>4.881586</td>\n",
       "      <td>3.394249</td>\n",
       "      <td>6.981226</td>\n",
       "      <td>...</td>\n",
       "      <td>6.083050</td>\n",
       "      <td>5.891748</td>\n",
       "      <td>5.999268</td>\n",
       "      <td>5.481434</td>\n",
       "      <td>6.353627</td>\n",
       "      <td>5.667723</td>\n",
       "      <td>3.429725</td>\n",
       "      <td>3.859270</td>\n",
       "      <td>5.952769</td>\n",
       "      <td>10.339334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>6.665956</td>\n",
       "      <td>8.035231</td>\n",
       "      <td>7.614323</td>\n",
       "      <td>7.692385</td>\n",
       "      <td>7.442830</td>\n",
       "      <td>5.849879</td>\n",
       "      <td>7.108214</td>\n",
       "      <td>4.431447</td>\n",
       "      <td>7.199892</td>\n",
       "      <td>...</td>\n",
       "      <td>8.006517</td>\n",
       "      <td>6.688329</td>\n",
       "      <td>6.308003</td>\n",
       "      <td>6.349058</td>\n",
       "      <td>6.409389</td>\n",
       "      <td>3.632523</td>\n",
       "      <td>3.294988</td>\n",
       "      <td>3.604101</td>\n",
       "      <td>5.653083</td>\n",
       "      <td>8.723818</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>125</th>\n",
       "      <td>3</td>\n",
       "      <td>7.563709</td>\n",
       "      <td>5.266555</td>\n",
       "      <td>6.658161</td>\n",
       "      <td>8.404671</td>\n",
       "      <td>6.951571</td>\n",
       "      <td>4.212751</td>\n",
       "      <td>6.812968</td>\n",
       "      <td>4.303145</td>\n",
       "      <td>8.041560</td>\n",
       "      <td>...</td>\n",
       "      <td>5.846897</td>\n",
       "      <td>6.589755</td>\n",
       "      <td>10.269227</td>\n",
       "      <td>7.748888</td>\n",
       "      <td>6.293694</td>\n",
       "      <td>8.436834</td>\n",
       "      <td>3.184355</td>\n",
       "      <td>5.008725</td>\n",
       "      <td>7.046908</td>\n",
       "      <td>10.482143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>126</th>\n",
       "      <td>3</td>\n",
       "      <td>10.704805</td>\n",
       "      <td>6.485646</td>\n",
       "      <td>6.094978</td>\n",
       "      <td>9.212460</td>\n",
       "      <td>9.218975</td>\n",
       "      <td>5.511615</td>\n",
       "      <td>6.541559</td>\n",
       "      <td>3.438830</td>\n",
       "      <td>8.264806</td>\n",
       "      <td>...</td>\n",
       "      <td>9.220842</td>\n",
       "      <td>6.866817</td>\n",
       "      <td>8.604499</td>\n",
       "      <td>7.580293</td>\n",
       "      <td>5.951765</td>\n",
       "      <td>4.060489</td>\n",
       "      <td>2.974504</td>\n",
       "      <td>4.354122</td>\n",
       "      <td>6.664799</td>\n",
       "      <td>9.311126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>3</td>\n",
       "      <td>6.450710</td>\n",
       "      <td>5.612284</td>\n",
       "      <td>5.766023</td>\n",
       "      <td>10.130702</td>\n",
       "      <td>7.201212</td>\n",
       "      <td>6.760974</td>\n",
       "      <td>8.377978</td>\n",
       "      <td>5.779441</td>\n",
       "      <td>8.114371</td>\n",
       "      <td>...</td>\n",
       "      <td>5.915235</td>\n",
       "      <td>6.767851</td>\n",
       "      <td>9.653469</td>\n",
       "      <td>6.974722</td>\n",
       "      <td>5.489107</td>\n",
       "      <td>6.878120</td>\n",
       "      <td>3.190270</td>\n",
       "      <td>4.028819</td>\n",
       "      <td>6.493181</td>\n",
       "      <td>10.290373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>3</td>\n",
       "      <td>7.456670</td>\n",
       "      <td>5.587341</td>\n",
       "      <td>5.885478</td>\n",
       "      <td>8.975262</td>\n",
       "      <td>7.278170</td>\n",
       "      <td>6.600490</td>\n",
       "      <td>8.601761</td>\n",
       "      <td>7.847652</td>\n",
       "      <td>9.493311</td>\n",
       "      <td>...</td>\n",
       "      <td>6.543639</td>\n",
       "      <td>6.514000</td>\n",
       "      <td>9.727747</td>\n",
       "      <td>6.188328</td>\n",
       "      <td>5.823914</td>\n",
       "      <td>3.935959</td>\n",
       "      <td>3.396742</td>\n",
       "      <td>4.461881</td>\n",
       "      <td>6.473156</td>\n",
       "      <td>9.652443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129</th>\n",
       "      <td>3</td>\n",
       "      <td>8.118853</td>\n",
       "      <td>5.633060</td>\n",
       "      <td>6.214792</td>\n",
       "      <td>9.423301</td>\n",
       "      <td>7.472433</td>\n",
       "      <td>4.514391</td>\n",
       "      <td>7.105446</td>\n",
       "      <td>6.637282</td>\n",
       "      <td>8.415955</td>\n",
       "      <td>...</td>\n",
       "      <td>5.923473</td>\n",
       "      <td>6.450032</td>\n",
       "      <td>9.062178</td>\n",
       "      <td>8.846653</td>\n",
       "      <td>5.697126</td>\n",
       "      <td>4.662247</td>\n",
       "      <td>2.993329</td>\n",
       "      <td>6.269305</td>\n",
       "      <td>6.678276</td>\n",
       "      <td>9.969752</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>130 rows × 301 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     type  1552508_at  1552516_a_at  1552575_a_at  1552807_a_at  1553479_at  \\\n",
       "0       0    5.989892      8.660293      6.891804      7.924462    6.895976   \n",
       "1       0    8.319513      9.980202      6.826210      7.493678    7.775287   \n",
       "2       0    7.261455      8.212629      6.270775      8.045542    8.177457   \n",
       "3       0    5.883377      6.236439      6.029005      8.071778    5.224647   \n",
       "4       0    6.665956      8.035231      7.614323      7.692385    7.442830   \n",
       "..    ...         ...           ...           ...           ...         ...   \n",
       "125     3    7.563709      5.266555      6.658161      8.404671    6.951571   \n",
       "126     3   10.704805      6.485646      6.094978      9.212460    9.218975   \n",
       "127     3    6.450710      5.612284      5.766023     10.130702    7.201212   \n",
       "128     3    7.456670      5.587341      5.885478      8.975262    7.278170   \n",
       "129     3    8.118853      5.633060      6.214792      9.423301    7.472433   \n",
       "\n",
       "     1553654_at  1554018_at  1554507_at  1554544_a_at  ...  242134_at  \\\n",
       "0      4.827520    4.369720    4.015369      8.396550  ...   7.226616   \n",
       "1      6.873171    3.636735    3.204273      8.035201  ...   8.765334   \n",
       "2      4.041513    6.366196    4.369383      8.530794  ...   6.562527   \n",
       "3      4.879736    4.881586    3.394249      6.981226  ...   6.083050   \n",
       "4      5.849879    7.108214    4.431447      7.199892  ...   8.006517   \n",
       "..          ...         ...         ...           ...  ...        ...   \n",
       "125    4.212751    6.812968    4.303145      8.041560  ...   5.846897   \n",
       "126    5.511615    6.541559    3.438830      8.264806  ...   9.220842   \n",
       "127    6.760974    8.377978    5.779441      8.114371  ...   5.915235   \n",
       "128    6.600490    8.601761    7.847652      9.493311  ...   6.543639   \n",
       "129    4.514391    7.105446    6.637282      8.415955  ...   5.923473   \n",
       "\n",
       "     242888_at  242931_at  242943_at  243584_at  244163_at  244493_at  \\\n",
       "0     6.705961   8.573175   5.559319   6.382599   3.841604   3.423704   \n",
       "1     6.234970   6.750153   8.727043   6.476319   4.599455   3.214800   \n",
       "2     6.614429   8.047375   7.003226   6.268696   3.411819   3.078852   \n",
       "3     5.891748   5.999268   5.481434   6.353627   5.667723   3.429725   \n",
       "4     6.688329   6.308003   6.349058   6.409389   3.632523   3.294988   \n",
       "..         ...        ...        ...        ...        ...        ...   \n",
       "125   6.589755  10.269227   7.748888   6.293694   8.436834   3.184355   \n",
       "126   6.866817   8.604499   7.580293   5.951765   4.060489   2.974504   \n",
       "127   6.767851   9.653469   6.974722   5.489107   6.878120   3.190270   \n",
       "128   6.514000   9.727747   6.188328   5.823914   3.935959   3.396742   \n",
       "129   6.450032   9.062178   8.846653   5.697126   4.662247   2.993329   \n",
       "\n",
       "     244503_at  244751_at  53071_s_at  \n",
       "0     3.627346   6.001562    8.183152  \n",
       "1     4.951005   5.518132    9.191431  \n",
       "2     3.624622   5.927809    9.263549  \n",
       "3     3.859270   5.952769   10.339334  \n",
       "4     3.604101   5.653083    8.723818  \n",
       "..         ...        ...         ...  \n",
       "125   5.008725   7.046908   10.482143  \n",
       "126   4.354122   6.664799    9.311126  \n",
       "127   4.028819   6.493181   10.290373  \n",
       "128   4.461881   6.473156    9.652443  \n",
       "129   6.269305   6.678276    9.969752  \n",
       "\n",
       "[130 rows x 301 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_RFE.iloc[:,1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "640aa66f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# store the final dataset after the feature selection is done\n",
    "df_RFE.iloc[:,1:].to_csv(\"df_w_var.csv\", index = False)"
   ]
  }
 ],
 "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}