DrJiang_CCLE_Metastasis_Primary_ROC.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/public/apps/python/3.7.0/lib/python3.7/site-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
      "  import pandas.util.testing as tm\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import sys\n",
    "import math\n",
    "from scipy.stats import pearsonr\n",
    "from scipy.stats import spearmanr\n",
    "from scipy.stats import ttest_ind\n",
    "import seaborn as sns\n",
    "from sklearn import metrics\n",
    "from sklearn.cluster import AgglomerativeClustering\n",
    "from sklearn.model_selection import train_test_split, cross_validate\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import RocCurveDisplay, roc_curve, auc, roc_auc_score\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.preprocessing import LabelEncoder, label_binarize, StandardScaler\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.inspection import permutation_importance\n",
    "from sklearn.naive_bayes import GaussianNB, MultinomialNB, BernoulliNB, CategoricalNB\n",
    "from sklearn.neural_network import MLPClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>primary_disease</th>\n",
       "      <th>primary_or_metastasis</th>\n",
       "      <th>ITGA1</th>\n",
       "      <th>ITGA10</th>\n",
       "      <th>ITGA11</th>\n",
       "      <th>ITGA2</th>\n",
       "      <th>ITGA2B</th>\n",
       "      <th>ITGA3</th>\n",
       "      <th>ITGA4</th>\n",
       "      <th>ITGA5</th>\n",
       "      <th>...</th>\n",
       "      <th>ITGB3</th>\n",
       "      <th>ITGB4</th>\n",
       "      <th>ITGB5</th>\n",
       "      <th>ITGB6</th>\n",
       "      <th>ITGB7</th>\n",
       "      <th>ITGB8</th>\n",
       "      <th>ITGBL1</th>\n",
       "      <th>CD44</th>\n",
       "      <th>ERBB2</th>\n",
       "      <th>EPCAM</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>stripped_cell_line_name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ZR7530</th>\n",
       "      <td>Breast Cancer</td>\n",
       "      <td>Metastasis</td>\n",
       "      <td>0.014355</td>\n",
       "      <td>2.046142</td>\n",
       "      <td>0.028569</td>\n",
       "      <td>1.604071</td>\n",
       "      <td>0.389567</td>\n",
       "      <td>4.553361</td>\n",
       "      <td>0.042644</td>\n",
       "      <td>2.592158</td>\n",
       "      <td>...</td>\n",
       "      <td>0.150560</td>\n",
       "      <td>4.606442</td>\n",
       "      <td>7.654708</td>\n",
       "      <td>3.066950</td>\n",
       "      <td>0.536053</td>\n",
       "      <td>0.056584</td>\n",
       "      <td>0.056584</td>\n",
       "      <td>5.469560</td>\n",
       "      <td>11.437388</td>\n",
       "      <td>7.190319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EFM192A</th>\n",
       "      <td>Breast Cancer</td>\n",
       "      <td>Metastasis</td>\n",
       "      <td>0.310340</td>\n",
       "      <td>0.526069</td>\n",
       "      <td>0.014355</td>\n",
       "      <td>1.310340</td>\n",
       "      <td>0.704872</td>\n",
       "      <td>5.718362</td>\n",
       "      <td>0.189034</td>\n",
       "      <td>3.397803</td>\n",
       "      <td>...</td>\n",
       "      <td>0.150560</td>\n",
       "      <td>4.785027</td>\n",
       "      <td>5.088735</td>\n",
       "      <td>1.084064</td>\n",
       "      <td>1.879706</td>\n",
       "      <td>0.321928</td>\n",
       "      <td>0.124328</td>\n",
       "      <td>4.010780</td>\n",
       "      <td>11.436035</td>\n",
       "      <td>7.317232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKBR3</th>\n",
       "      <td>Breast Cancer</td>\n",
       "      <td>Metastasis</td>\n",
       "      <td>0.526069</td>\n",
       "      <td>0.163499</td>\n",
       "      <td>0.084064</td>\n",
       "      <td>1.263034</td>\n",
       "      <td>0.659925</td>\n",
       "      <td>3.708187</td>\n",
       "      <td>0.124328</td>\n",
       "      <td>2.241840</td>\n",
       "      <td>...</td>\n",
       "      <td>0.070389</td>\n",
       "      <td>4.319040</td>\n",
       "      <td>7.204082</td>\n",
       "      <td>3.438293</td>\n",
       "      <td>0.214125</td>\n",
       "      <td>0.137504</td>\n",
       "      <td>0.056584</td>\n",
       "      <td>1.985500</td>\n",
       "      <td>10.844706</td>\n",
       "      <td>7.035294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UACC893</th>\n",
       "      <td>Breast Cancer</td>\n",
       "      <td>Primary</td>\n",
       "      <td>0.070389</td>\n",
       "      <td>0.475085</td>\n",
       "      <td>0.124328</td>\n",
       "      <td>2.906891</td>\n",
       "      <td>0.014355</td>\n",
       "      <td>5.942984</td>\n",
       "      <td>0.201634</td>\n",
       "      <td>2.778209</td>\n",
       "      <td>...</td>\n",
       "      <td>0.111031</td>\n",
       "      <td>6.836177</td>\n",
       "      <td>7.076816</td>\n",
       "      <td>5.200457</td>\n",
       "      <td>0.226509</td>\n",
       "      <td>0.070389</td>\n",
       "      <td>0.084064</td>\n",
       "      <td>3.035624</td>\n",
       "      <td>11.648582</td>\n",
       "      <td>9.178466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HCC38</th>\n",
       "      <td>Breast Cancer</td>\n",
       "      <td>Primary</td>\n",
       "      <td>1.250962</td>\n",
       "      <td>0.422233</td>\n",
       "      <td>0.389567</td>\n",
       "      <td>4.240314</td>\n",
       "      <td>2.370164</td>\n",
       "      <td>7.142720</td>\n",
       "      <td>4.936873</td>\n",
       "      <td>4.252476</td>\n",
       "      <td>...</td>\n",
       "      <td>5.000451</td>\n",
       "      <td>5.739308</td>\n",
       "      <td>7.382148</td>\n",
       "      <td>7.287989</td>\n",
       "      <td>0.757023</td>\n",
       "      <td>3.636915</td>\n",
       "      <td>1.851999</td>\n",
       "      <td>8.722193</td>\n",
       "      <td>7.115408</td>\n",
       "      <td>8.790739</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>ZR751</th>\n",
       "      <td>Breast Cancer</td>\n",
       "      <td>Metastasis</td>\n",
       "      <td>0.111031</td>\n",
       "      <td>1.077243</td>\n",
       "      <td>0.163499</td>\n",
       "      <td>2.704872</td>\n",
       "      <td>2.289834</td>\n",
       "      <td>5.074249</td>\n",
       "      <td>0.310340</td>\n",
       "      <td>1.726831</td>\n",
       "      <td>...</td>\n",
       "      <td>0.464668</td>\n",
       "      <td>4.779785</td>\n",
       "      <td>4.296457</td>\n",
       "      <td>2.257011</td>\n",
       "      <td>1.244887</td>\n",
       "      <td>0.536053</td>\n",
       "      <td>0.389567</td>\n",
       "      <td>4.678635</td>\n",
       "      <td>6.756623</td>\n",
       "      <td>7.933868</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BT549</th>\n",
       "      <td>Breast Cancer</td>\n",
       "      <td>Primary</td>\n",
       "      <td>0.124328</td>\n",
       "      <td>3.054848</td>\n",
       "      <td>0.464668</td>\n",
       "      <td>0.321928</td>\n",
       "      <td>0.641546</td>\n",
       "      <td>6.630522</td>\n",
       "      <td>3.420887</td>\n",
       "      <td>5.229588</td>\n",
       "      <td>...</td>\n",
       "      <td>0.855990</td>\n",
       "      <td>0.941106</td>\n",
       "      <td>5.025472</td>\n",
       "      <td>0.070389</td>\n",
       "      <td>0.070389</td>\n",
       "      <td>1.974529</td>\n",
       "      <td>2.028569</td>\n",
       "      <td>10.035486</td>\n",
       "      <td>4.135863</td>\n",
       "      <td>0.956057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COLO824</th>\n",
       "      <td>Breast Cancer</td>\n",
       "      <td>Metastasis</td>\n",
       "      <td>0.084064</td>\n",
       "      <td>1.454176</td>\n",
       "      <td>0.454176</td>\n",
       "      <td>3.058316</td>\n",
       "      <td>0.111031</td>\n",
       "      <td>2.127633</td>\n",
       "      <td>0.963474</td>\n",
       "      <td>0.823749</td>\n",
       "      <td>...</td>\n",
       "      <td>0.042644</td>\n",
       "      <td>4.133399</td>\n",
       "      <td>3.527321</td>\n",
       "      <td>0.097611</td>\n",
       "      <td>0.056584</td>\n",
       "      <td>4.228819</td>\n",
       "      <td>0.042644</td>\n",
       "      <td>9.311612</td>\n",
       "      <td>6.770036</td>\n",
       "      <td>9.385755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM159PT</th>\n",
       "      <td>Breast Cancer</td>\n",
       "      <td>Primary</td>\n",
       "      <td>1.316146</td>\n",
       "      <td>3.523562</td>\n",
       "      <td>1.521051</td>\n",
       "      <td>4.342697</td>\n",
       "      <td>0.124328</td>\n",
       "      <td>6.496814</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.727240</td>\n",
       "      <td>...</td>\n",
       "      <td>1.627607</td>\n",
       "      <td>5.734168</td>\n",
       "      <td>4.863443</td>\n",
       "      <td>0.111031</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.150560</td>\n",
       "      <td>0.137504</td>\n",
       "      <td>7.425929</td>\n",
       "      <td>4.188243</td>\n",
       "      <td>0.226509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MDAMB361</th>\n",
       "      <td>Breast Cancer</td>\n",
       "      <td>Metastasis</td>\n",
       "      <td>1.427606</td>\n",
       "      <td>1.244887</td>\n",
       "      <td>0.124328</td>\n",
       "      <td>3.056584</td>\n",
       "      <td>0.985500</td>\n",
       "      <td>7.544192</td>\n",
       "      <td>1.883621</td>\n",
       "      <td>4.950002</td>\n",
       "      <td>...</td>\n",
       "      <td>2.558268</td>\n",
       "      <td>6.971888</td>\n",
       "      <td>6.526225</td>\n",
       "      <td>4.221104</td>\n",
       "      <td>1.480265</td>\n",
       "      <td>0.704872</td>\n",
       "      <td>0.084064</td>\n",
       "      <td>7.838637</td>\n",
       "      <td>9.655799</td>\n",
       "      <td>8.472407</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>61 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                        primary_disease primary_or_metastasis     ITGA1  \\\n",
       "stripped_cell_line_name                                                   \n",
       "ZR7530                    Breast Cancer            Metastasis  0.014355   \n",
       "EFM192A                   Breast Cancer            Metastasis  0.310340   \n",
       "SKBR3                     Breast Cancer            Metastasis  0.526069   \n",
       "UACC893                   Breast Cancer               Primary  0.070389   \n",
       "HCC38                     Breast Cancer               Primary  1.250962   \n",
       "...                                 ...                   ...       ...   \n",
       "ZR751                     Breast Cancer            Metastasis  0.111031   \n",
       "BT549                     Breast Cancer               Primary  0.124328   \n",
       "COLO824                   Breast Cancer            Metastasis  0.084064   \n",
       "SUM159PT                  Breast Cancer               Primary  1.316146   \n",
       "MDAMB361                  Breast Cancer            Metastasis  1.427606   \n",
       "\n",
       "                           ITGA10    ITGA11     ITGA2    ITGA2B     ITGA3  \\\n",
       "stripped_cell_line_name                                                     \n",
       "ZR7530                   2.046142  0.028569  1.604071  0.389567  4.553361   \n",
       "EFM192A                  0.526069  0.014355  1.310340  0.704872  5.718362   \n",
       "SKBR3                    0.163499  0.084064  1.263034  0.659925  3.708187   \n",
       "UACC893                  0.475085  0.124328  2.906891  0.014355  5.942984   \n",
       "HCC38                    0.422233  0.389567  4.240314  2.370164  7.142720   \n",
       "...                           ...       ...       ...       ...       ...   \n",
       "ZR751                    1.077243  0.163499  2.704872  2.289834  5.074249   \n",
       "BT549                    3.054848  0.464668  0.321928  0.641546  6.630522   \n",
       "COLO824                  1.454176  0.454176  3.058316  0.111031  2.127633   \n",
       "SUM159PT                 3.523562  1.521051  4.342697  0.124328  6.496814   \n",
       "MDAMB361                 1.244887  0.124328  3.056584  0.985500  7.544192   \n",
       "\n",
       "                            ITGA4     ITGA5  ...     ITGB3     ITGB4  \\\n",
       "stripped_cell_line_name                      ...                       \n",
       "ZR7530                   0.042644  2.592158  ...  0.150560  4.606442   \n",
       "EFM192A                  0.189034  3.397803  ...  0.150560  4.785027   \n",
       "SKBR3                    0.124328  2.241840  ...  0.070389  4.319040   \n",
       "UACC893                  0.201634  2.778209  ...  0.111031  6.836177   \n",
       "HCC38                    4.936873  4.252476  ...  5.000451  5.739308   \n",
       "...                           ...       ...  ...       ...       ...   \n",
       "ZR751                    0.310340  1.726831  ...  0.464668  4.779785   \n",
       "BT549                    3.420887  5.229588  ...  0.855990  0.941106   \n",
       "COLO824                  0.963474  0.823749  ...  0.042644  4.133399   \n",
       "SUM159PT                 0.000000  6.727240  ...  1.627607  5.734168   \n",
       "MDAMB361                 1.883621  4.950002  ...  2.558268  6.971888   \n",
       "\n",
       "                            ITGB5     ITGB6     ITGB7     ITGB8    ITGBL1  \\\n",
       "stripped_cell_line_name                                                     \n",
       "ZR7530                   7.654708  3.066950  0.536053  0.056584  0.056584   \n",
       "EFM192A                  5.088735  1.084064  1.879706  0.321928  0.124328   \n",
       "SKBR3                    7.204082  3.438293  0.214125  0.137504  0.056584   \n",
       "UACC893                  7.076816  5.200457  0.226509  0.070389  0.084064   \n",
       "HCC38                    7.382148  7.287989  0.757023  3.636915  1.851999   \n",
       "...                           ...       ...       ...       ...       ...   \n",
       "ZR751                    4.296457  2.257011  1.244887  0.536053  0.389567   \n",
       "BT549                    5.025472  0.070389  0.070389  1.974529  2.028569   \n",
       "COLO824                  3.527321  0.097611  0.056584  4.228819  0.042644   \n",
       "SUM159PT                 4.863443  0.111031  0.000000  0.150560  0.137504   \n",
       "MDAMB361                 6.526225  4.221104  1.480265  0.704872  0.084064   \n",
       "\n",
       "                              CD44      ERBB2     EPCAM  \n",
       "stripped_cell_line_name                                  \n",
       "ZR7530                    5.469560  11.437388  7.190319  \n",
       "EFM192A                   4.010780  11.436035  7.317232  \n",
       "SKBR3                     1.985500  10.844706  7.035294  \n",
       "UACC893                   3.035624  11.648582  9.178466  \n",
       "HCC38                     8.722193   7.115408  8.790739  \n",
       "...                            ...        ...       ...  \n",
       "ZR751                     4.678635   6.756623  7.933868  \n",
       "BT549                    10.035486   4.135863  0.956057  \n",
       "COLO824                   9.311612   6.770036  9.385755  \n",
       "SUM159PT                  7.425929   4.188243  0.226509  \n",
       "MDAMB361                  7.838637   9.655799  8.472407  \n",
       "\n",
       "[61 rows x 32 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "integrin_rna = pd.read_csv('integrins_rna_primary_metastasis.csv', index_col=0)\n",
    "for i in range(3,33):\n",
    "    intg = integrin_rna.columns[i].split(' ')\n",
    "    integrin_rna.rename(columns={integrin_rna.columns[i]: intg[0]}, inplace=True)\n",
    "integrins_rna_breast = integrin_rna[integrin_rna[\"primary_disease\"]==\"Breast Cancer\"].copy()\n",
    "integrins_rna_breast.set_index('stripped_cell_line_name',inplace=True)\n",
    "integrins_rna_breast"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# identify top three features, ITGB8, ITGB6, ITGB1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['ITGB7' 'CD44' 'ITGA2' 'ITGA9' 'ITGA5' 'ITGA4' 'ITGAD' 'EPCAM' 'ITGAM'\n",
      " 'ITGA8' 'ITGAX' 'ITGA6' 'ITGAL' 'ITGAV' 'ITGA10' 'ITGA7' 'ITGA2B' 'ITGB5'\n",
      " 'ITGB2' 'ITGA1' 'ITGB3' 'ERBB2' 'ITGBL1' 'ITGA3' 'ITGA11' 'ITGAE' 'ITGB4'\n",
      " 'ITGB6' 'ITGB8' 'ITGB1'] [0.03790504 0.06860711 0.09015857 0.12728832 0.15999466 0.16494459\n",
      " 0.21225403 0.21518117 0.23052315 0.29004571 0.30860918 0.32029034\n",
      " 0.34585117 0.36980542 0.37649772 0.37881526 0.39016464 0.41183117\n",
      " 0.41969244 0.4424992  0.50317186 0.5768823  0.59641376 0.69864884\n",
      " 0.75840888 0.77937471 0.81581374 0.8264434  0.91237356 1.        ]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Relative Feature Importance')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "accuracy=[]\n",
    "X=integrins_rna_breast.iloc[:,2:]\n",
    "y=integrins_rna_breast['primary_or_metastasis']\n",
    "#Scale data\n",
    "scaler = StandardScaler()\n",
    "scaler.fit(X) \n",
    "X_scaled = pd.DataFrame(scaler.transform(X),columns = X.columns)\n",
    "#X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.3)\n",
    "X_train=X_scaled\n",
    "y_train=y\n",
    "X_test=X_scaled\n",
    "y_test=y\n",
    "classifier = LogisticRegression(random_state=0,multi_class='multinomial',solver='newton-cg')\n",
    "classifier.fit(X_train, y_train)\n",
    "y_pred = classifier.predict(X_test)\n",
    "#    print(\"The classifier Accuracy: \", accuracy_score(y_test,y_pred))\n",
    "accuracy.append(accuracy_score(y_test,y_pred))\n",
    "feature_importance = abs(classifier.coef_[0])\n",
    "feature_importance =feature_importance / feature_importance.max()\n",
    "sorted_idx = np.argsort(feature_importance)\n",
    "pos = np.arange(sorted_idx.shape[0]) + .5\n",
    "featfig = plt.figure(figsize=(10,8))\n",
    "featax = featfig.add_subplot(1, 1, 1)\n",
    "featax.barh(pos, feature_importance[sorted_idx], align='center')\n",
    "print(np.array(X.columns)[sorted_idx],feature_importance[sorted_idx])\n",
    "featax.set_yticks(pos)\n",
    "featax.set_yticklabels(np.array(X.columns)[sorted_idx], fontsize=10)\n",
    "featax.set_xlabel('Relative Feature Importance')\n",
    "#RocCurveDisplay.from_estimator(classifier, X_test, y_test, alpha=0.8)\n",
    "#plt.plot([0, 1], [0, 1], color='orange', lw=2, linestyle='--')\n",
    "#plt.tight_layout()   \n",
    "#plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rna_roc = ['ITGB8','ITGB6', 'ITGB1','all_three_combined','all_combined']\n",
    "fig,ax=plt.subplots(figsize=(8,6))\n",
    "for integrin in rna_roc:\n",
    "    if integrin=='all_three_combined':\n",
    "        integrin=['ITGB8','ITGB6','ITGB1']\n",
    "        accuracy=[]\n",
    "        X=integrins_rna_breast.iloc[:,2:][integrin]\n",
    "        y=integrins_rna_breast['primary_or_metastasis']\n",
    "        #Scale data\n",
    "        scaler = StandardScaler()\n",
    "        scaler.fit(X) \n",
    "        X_scaled = pd.DataFrame(scaler.transform(X),columns = X.columns)\n",
    "        #X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.3)\n",
    "        X_train=X_scaled\n",
    "        y_train=y\n",
    "        X_test=X_scaled\n",
    "        y_test=y\n",
    "        classifier = LogisticRegression(random_state=0,multi_class='multinomial',solver='newton-cg')\n",
    "        classifier.fit(X_train, y_train)\n",
    "        y_pred = classifier.predict(X_test)\n",
    "        #    print(\"The classifier Accuracy: \", accuracy_score(y_test,y_pred))\n",
    "        accuracy.append(accuracy_score(y_test,y_pred))\n",
    "        feature_importance = abs(classifier.coef_[0])\n",
    "        feature_importance =feature_importance / feature_importance.max()\n",
    "        sorted_idx = np.argsort(feature_importance)\n",
    "        pos = np.arange(sorted_idx.shape[0]) + .5\n",
    "        #featfig = plt.figure(figsize=(10,8))\n",
    "        #featax = featfig.add_subplot(1, 1, 1)\n",
    "        #featax.barh(pos, feature_importance[sorted_idx], align='center')\n",
    "        #print(np.array(X.columns)[sorted_idx],feature_importance[sorted_idx])\n",
    "        #featax.set_yticks(pos)\n",
    "        #featax.set_yticklabels(np.array(X.columns)[sorted_idx], fontsize=10)\n",
    "        #featax.set_xlabel('Relative Feature Importance')\n",
    "        RocCurveDisplay.from_estimator(classifier, X_test, y_test, alpha=0.8,name='combined',ax=ax)\n",
    "    elif integrin=='all_combined':\n",
    "        accuracy=[]\n",
    "        X=integrins_rna_breast.iloc[:,2:]\n",
    "        y=integrins_rna_breast['primary_or_metastasis']\n",
    "        #Scale data\n",
    "        scaler = StandardScaler()\n",
    "        scaler.fit(X) \n",
    "        X_scaled = pd.DataFrame(scaler.transform(X),columns = X.columns)\n",
    "        #X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.3)\n",
    "        X_train=X_scaled\n",
    "        y_train=y\n",
    "        X_test=X_scaled\n",
    "        y_test=y\n",
    "        classifier = LogisticRegression(random_state=0,multi_class='multinomial',solver='newton-cg')\n",
    "        classifier.fit(X_train, y_train)\n",
    "        y_pred = classifier.predict(X_test)\n",
    "        #    print(\"The classifier Accuracy: \", accuracy_score(y_test,y_pred))\n",
    "        accuracy.append(accuracy_score(y_test,y_pred))\n",
    "        feature_importance = abs(classifier.coef_[0])\n",
    "        feature_importance =feature_importance / feature_importance.max()\n",
    "        sorted_idx = np.argsort(feature_importance)\n",
    "        pos = np.arange(sorted_idx.shape[0]) + .5\n",
    "        #featfig = plt.figure(figsize=(10,8))\n",
    "        #featax = featfig.add_subplot(1, 1, 1)\n",
    "        #featax.barh(pos, feature_importance[sorted_idx], align='center')\n",
    "        #print(np.array(X.columns)[sorted_idx],feature_importance[sorted_idx])\n",
    "        #featax.set_yticks(pos)\n",
    "        #featax.set_yticklabels(np.array(X.columns)[sorted_idx], fontsize=10)\n",
    "        #featax.set_xlabel('Relative Feature Importance')\n",
    "        RocCurveDisplay.from_estimator(classifier, X_test, y_test, alpha=0.8,name='all_combined',ax=ax)        \n",
    "    else:\n",
    "        accuracy=[]\n",
    "        X=integrins_rna_breast.iloc[:,2:][[integrin]]\n",
    "        y=integrins_rna_breast['primary_or_metastasis']\n",
    "        #Scale data\n",
    "        scaler = StandardScaler()\n",
    "        scaler.fit(X) \n",
    "        X_scaled = pd.DataFrame(scaler.transform(X),columns = X.columns)\n",
    "        #X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.3)\n",
    "        X_train=X_scaled\n",
    "        y_train=y\n",
    "        X_test=X_scaled\n",
    "        y_test=y\n",
    "        classifier = LogisticRegression(random_state=0,multi_class='multinomial',solver='newton-cg')\n",
    "        classifier.fit(X_train, y_train)\n",
    "        y_pred = classifier.predict(X_test)\n",
    "        #    print(\"The classifier Accuracy: \", accuracy_score(y_test,y_pred))\n",
    "        accuracy.append(accuracy_score(y_test,y_pred))\n",
    "        feature_importance = abs(classifier.coef_[0])\n",
    "        feature_importance =feature_importance / feature_importance.max()\n",
    "        sorted_idx = np.argsort(feature_importance)\n",
    "        pos = np.arange(sorted_idx.shape[0]) + .5\n",
    "        #featfig = plt.figure(figsize=(10,8))\n",
    "        #featax = featfig.add_subplot(1, 1, 1)\n",
    "        #featax.barh(pos, feature_importance[sorted_idx], align='center')\n",
    "        #print(np.array(X.columns)[sorted_idx],feature_importance[sorted_idx])\n",
    "        #featax.set_yticks(pos)\n",
    "        #featax.set_yticklabels(np.array(X.columns)[sorted_idx], fontsize=10)\n",
    "        #featax.set_xlabel('Relative Feature Importance')\n",
    "        RocCurveDisplay.from_estimator(classifier, X_test, y_test, alpha=0.8,name=integrin,ax=ax)\n",
    "        \n",
    "ax.plot([0, 1], [0, 1], color='orange', lw=2, linestyle='--')\n",
    "ax.legend(fontsize=16)\n",
    "xlbl = ax.xaxis.get_label()\n",
    "ylbl = ax.yaxis.get_label()\n",
    "plt.setp(xlbl,fontsize=16)\n",
    "plt.setp(ylbl,fontsize=16)\n",
    "plt.setp(ax.get_xticklabels(),fontsize=14)\n",
    "plt.setp(ax.get_yticklabels(),fontsize=14)\n",
    "plt.tight_layout()   \n",
    "plt.savefig('top_features_roc_logistic_regression.jpg', dpi = 1000, bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rna_roc = integrins_rna_breast.iloc[:,2:].columns\n",
    "fig,ax=plt.subplots(figsize=(8,6))\n",
    "for integrin in rna_roc:\n",
    "    if integrin=='all_three_combined':\n",
    "        integrin=['ITGB8','ITGB6','ITGB1']\n",
    "        accuracy=[]\n",
    "        X=integrins_rna_breast.iloc[:,2:][integrin]\n",
    "        y=integrins_rna_breast['primary_or_metastasis']\n",
    "        #Scale data\n",
    "        scaler = StandardScaler()\n",
    "        scaler.fit(X) \n",
    "        X_scaled = pd.DataFrame(scaler.transform(X),columns = X.columns)\n",
    "        #X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.3)\n",
    "        X_train=X_scaled\n",
    "        y_train=y\n",
    "        X_test=X_scaled\n",
    "        y_test=y\n",
    "        classifier = LogisticRegression(random_state=0,multi_class='multinomial',solver='newton-cg')\n",
    "        classifier.fit(X_train, y_train)\n",
    "        y_pred = classifier.predict(X_test)\n",
    "        #    print(\"The classifier Accuracy: \", accuracy_score(y_test,y_pred))\n",
    "        accuracy.append(accuracy_score(y_test,y_pred))\n",
    "        feature_importance = abs(classifier.coef_[0])\n",
    "        feature_importance =feature_importance / feature_importance.max()\n",
    "        sorted_idx = np.argsort(feature_importance)\n",
    "        pos = np.arange(sorted_idx.shape[0]) + .5\n",
    "        #featfig = plt.figure(figsize=(10,8))\n",
    "        #featax = featfig.add_subplot(1, 1, 1)\n",
    "        #featax.barh(pos, feature_importance[sorted_idx], align='center')\n",
    "        #print(np.array(X.columns)[sorted_idx],feature_importance[sorted_idx])\n",
    "        #featax.set_yticks(pos)\n",
    "        #featax.set_yticklabels(np.array(X.columns)[sorted_idx], fontsize=10)\n",
    "        #featax.set_xlabel('Relative Feature Importance')\n",
    "        RocCurveDisplay.from_estimator(classifier, X_test, y_test, alpha=0.8,name='combined',ax=ax)\n",
    "    else:\n",
    "        accuracy=[]\n",
    "        X=integrins_rna_breast.iloc[:,2:][[integrin]]\n",
    "        y=integrins_rna_breast['primary_or_metastasis']\n",
    "        #Scale data\n",
    "        scaler = StandardScaler()\n",
    "        scaler.fit(X) \n",
    "        X_scaled = pd.DataFrame(scaler.transform(X),columns = X.columns)\n",
    "        #X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.3)\n",
    "        X_train=X_scaled\n",
    "        y_train=y\n",
    "        X_test=X_scaled\n",
    "        y_test=y\n",
    "        classifier = LogisticRegression(random_state=0,multi_class='multinomial',solver='newton-cg')\n",
    "        classifier.fit(X_train, y_train)\n",
    "        y_pred = classifier.predict(X_test)\n",
    "        #    print(\"The classifier Accuracy: \", accuracy_score(y_test,y_pred))\n",
    "        accuracy.append(accuracy_score(y_test,y_pred))\n",
    "        feature_importance = abs(classifier.coef_[0])\n",
    "        feature_importance =feature_importance / feature_importance.max()\n",
    "        sorted_idx = np.argsort(feature_importance)\n",
    "        pos = np.arange(sorted_idx.shape[0]) + .5\n",
    "        #featfig = plt.figure(figsize=(10,8))\n",
    "        #featax = featfig.add_subplot(1, 1, 1)\n",
    "        #featax.barh(pos, feature_importance[sorted_idx], align='center')\n",
    "        #print(np.array(X.columns)[sorted_idx],feature_importance[sorted_idx])\n",
    "        #featax.set_yticks(pos)\n",
    "        #featax.set_yticklabels(np.array(X.columns)[sorted_idx], fontsize=10)\n",
    "        #featax.set_xlabel('Relative Feature Importance')\n",
    "        RocCurveDisplay.from_estimator(classifier, X_test, y_test, alpha=0.8,name=integrin,ax=ax)\n",
    "        \n",
    "ax.plot([0, 1], [0, 1], color='orange', lw=2, linestyle='--')\n",
    "#ax.legend(fontsize=16)\n",
    "xlbl = ax.xaxis.get_label()\n",
    "ylbl = ax.yaxis.get_label()\n",
    "plt.setp(xlbl,fontsize=16)\n",
    "plt.setp(ylbl,fontsize=16)\n",
    "plt.setp(ax.get_xticklabels(),fontsize=14)\n",
    "plt.setp(ax.get_yticklabels(),fontsize=14)\n",
    "plt.tight_layout()   \n",
    "#plt.savefig('top_features_roc_logistic_regression.jpg', dpi = 1000, bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}