TumorClassification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Deep Learning for Tumor Classification using RNA expression and H50 signaling pathways"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We show how to train a NN model to predict tumor classes from RNA expression and H50 signaling pathways using TCGA samples.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The model achieved 0.96 accuracy predicting 33 different tumor types in a validation set of 3000 samples. This result exceed those reported in [1,2] The confussion matrix, shows that the model almost only failed to discriminate rectal from colon cancer and esophageal from stomach cancer, which are closely related hysthologically."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Confussion Matrix\n",
    "from IPython.display import Image\n",
    "Image(\"TumorClassification_RNAexp.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predicting tumor types \n",
    "The following section shows how to train a NN model to predict tumor classes from RNA expression and H50 signaling pathways using TCGA samples. \n",
    "To download the required data and generate the files from scratch see section [Data Preprocessing](#prep)\n",
    "This notebook is inspired in [@alenaharley article](https://medium.com/@alenaharley/tumor-classification-using-gene-expression-data-poking-at-a-problem-using-fast-ai-again-8633c2256c85) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "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>A1BG</th>\n",
       "      <th>A1BG-AS1</th>\n",
       "      <th>A1CF</th>\n",
       "      <th>A2M</th>\n",
       "      <th>A2M-AS1</th>\n",
       "      <th>A2ML1</th>\n",
       "      <th>A2MP1</th>\n",
       "      <th>A3GALT2</th>\n",
       "      <th>A4GALT</th>\n",
       "      <th>A4GNT</th>\n",
       "      <th>...</th>\n",
       "      <th>ZWINT</th>\n",
       "      <th>ZXDA</th>\n",
       "      <th>ZXDB</th>\n",
       "      <th>ZXDC</th>\n",
       "      <th>ZYG11A</th>\n",
       "      <th>ZYG11B</th>\n",
       "      <th>ZYX</th>\n",
       "      <th>ZZEF1</th>\n",
       "      <th>ZZZ3</th>\n",
       "      <th>cancercat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>14.511335</td>\n",
       "      <td>14.658286</td>\n",
       "      <td>7.818664</td>\n",
       "      <td>21.170343</td>\n",
       "      <td>16.323402</td>\n",
       "      <td>11.515712</td>\n",
       "      <td>10.927732</td>\n",
       "      <td>3.512874</td>\n",
       "      <td>16.375104</td>\n",
       "      <td>4.546174</td>\n",
       "      <td>...</td>\n",
       "      <td>16.803166</td>\n",
       "      <td>14.481343</td>\n",
       "      <td>16.262297</td>\n",
       "      <td>17.164771</td>\n",
       "      <td>11.034740</td>\n",
       "      <td>17.568689</td>\n",
       "      <td>19.878162</td>\n",
       "      <td>18.062571</td>\n",
       "      <td>17.549486</td>\n",
       "      <td>15.014002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.262990</td>\n",
       "      <td>1.781519</td>\n",
       "      <td>7.636277</td>\n",
       "      <td>1.819773</td>\n",
       "      <td>1.530948</td>\n",
       "      <td>5.025257</td>\n",
       "      <td>2.047451</td>\n",
       "      <td>7.174155</td>\n",
       "      <td>1.840625</td>\n",
       "      <td>7.976631</td>\n",
       "      <td>...</td>\n",
       "      <td>1.550544</td>\n",
       "      <td>1.295317</td>\n",
       "      <td>1.108982</td>\n",
       "      <td>0.948553</td>\n",
       "      <td>4.794702</td>\n",
       "      <td>1.120799</td>\n",
       "      <td>1.148660</td>\n",
       "      <td>1.065064</td>\n",
       "      <td>0.987241</td>\n",
       "      <td>9.301482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.906939</td>\n",
       "      <td>4.392386</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>10.540129</td>\n",
       "      <td>8.876520</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>6.988696</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>...</td>\n",
       "      <td>8.906894</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>7.584970</td>\n",
       "      <td>11.969387</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>10.388018</td>\n",
       "      <td>14.118455</td>\n",
       "      <td>12.988152</td>\n",
       "      <td>11.607331</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>13.115125</td>\n",
       "      <td>13.612897</td>\n",
       "      <td>6.247946</td>\n",
       "      <td>20.103902</td>\n",
       "      <td>15.396688</td>\n",
       "      <td>8.640249</td>\n",
       "      <td>10.006678</td>\n",
       "      <td>3.322072</td>\n",
       "      <td>15.161555</td>\n",
       "      <td>4.807406</td>\n",
       "      <td>...</td>\n",
       "      <td>15.698264</td>\n",
       "      <td>13.734075</td>\n",
       "      <td>15.639362</td>\n",
       "      <td>16.617780</td>\n",
       "      <td>9.012625</td>\n",
       "      <td>16.962933</td>\n",
       "      <td>19.163031</td>\n",
       "      <td>17.430391</td>\n",
       "      <td>16.973854</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>14.377075</td>\n",
       "      <td>14.614969</td>\n",
       "      <td>7.936644</td>\n",
       "      <td>21.240994</td>\n",
       "      <td>16.313299</td>\n",
       "      <td>10.918491</td>\n",
       "      <td>11.040290</td>\n",
       "      <td>6.584978</td>\n",
       "      <td>16.524121</td>\n",
       "      <td>7.584970</td>\n",
       "      <td>...</td>\n",
       "      <td>16.862601</td>\n",
       "      <td>14.602264</td>\n",
       "      <td>16.340676</td>\n",
       "      <td>17.227212</td>\n",
       "      <td>11.866893</td>\n",
       "      <td>17.655066</td>\n",
       "      <td>19.896538</td>\n",
       "      <td>18.149090</td>\n",
       "      <td>17.624513</td>\n",
       "      <td>14.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>15.611969</td>\n",
       "      <td>15.556147</td>\n",
       "      <td>11.956830</td>\n",
       "      <td>22.350001</td>\n",
       "      <td>17.271283</td>\n",
       "      <td>14.611140</td>\n",
       "      <td>12.092427</td>\n",
       "      <td>8.029291</td>\n",
       "      <td>17.722402</td>\n",
       "      <td>9.205794</td>\n",
       "      <td>...</td>\n",
       "      <td>17.988194</td>\n",
       "      <td>15.369069</td>\n",
       "      <td>16.953637</td>\n",
       "      <td>17.761895</td>\n",
       "      <td>14.115450</td>\n",
       "      <td>18.333242</td>\n",
       "      <td>20.566523</td>\n",
       "      <td>18.753259</td>\n",
       "      <td>18.193407</td>\n",
       "      <td>23.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>25.744343</td>\n",
       "      <td>23.457815</td>\n",
       "      <td>21.781484</td>\n",
       "      <td>27.526837</td>\n",
       "      <td>22.749258</td>\n",
       "      <td>25.852982</td>\n",
       "      <td>18.892860</td>\n",
       "      <td>14.192447</td>\n",
       "      <td>21.365289</td>\n",
       "      <td>20.369699</td>\n",
       "      <td>...</td>\n",
       "      <td>21.309852</td>\n",
       "      <td>17.978504</td>\n",
       "      <td>20.607729</td>\n",
       "      <td>20.899094</td>\n",
       "      <td>21.113483</td>\n",
       "      <td>22.043753</td>\n",
       "      <td>24.416677</td>\n",
       "      <td>22.150687</td>\n",
       "      <td>21.651597</td>\n",
       "      <td>32.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 25151 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               A1BG      A1BG-AS1          A1CF           A2M       A2M-AS1  \\\n",
       "count  11284.000000  11284.000000  11284.000000  11284.000000  11284.000000   \n",
       "mean      14.511335     14.658286      7.818664     21.170343     16.323402   \n",
       "std        2.262990      1.781519      7.636277      1.819773      1.530948   \n",
       "min        4.906939      4.392386     -9.965784     10.540129      8.876520   \n",
       "25%       13.115125     13.612897      6.247946     20.103902     15.396688   \n",
       "50%       14.377075     14.614969      7.936644     21.240994     16.313299   \n",
       "75%       15.611969     15.556147     11.956830     22.350001     17.271283   \n",
       "max       25.744343     23.457815     21.781484     27.526837     22.749258   \n",
       "\n",
       "              A2ML1         A2MP1       A3GALT2        A4GALT         A4GNT  \\\n",
       "count  11284.000000  11284.000000  11284.000000  11284.000000  11284.000000   \n",
       "mean      11.515712     10.927732      3.512874     16.375104      4.546174   \n",
       "std        5.025257      2.047451      7.174155      1.840625      7.976631   \n",
       "min       -9.965784     -9.965784     -9.965784      6.988696     -9.965784   \n",
       "25%        8.640249     10.006678      3.322072     15.161555      4.807406   \n",
       "50%       10.918491     11.040290      6.584978     16.524121      7.584970   \n",
       "75%       14.611140     12.092427      8.029291     17.722402      9.205794   \n",
       "max       25.852982     18.892860     14.192447     21.365289     20.369699   \n",
       "\n",
       "           ...              ZWINT          ZXDA          ZXDB          ZXDC  \\\n",
       "count      ...       11284.000000  11284.000000  11284.000000  11284.000000   \n",
       "mean       ...          16.803166     14.481343     16.262297     17.164771   \n",
       "std        ...           1.550544      1.295317      1.108982      0.948553   \n",
       "min        ...           8.906894     -9.965784      7.584970     11.969387   \n",
       "25%        ...          15.698264     13.734075     15.639362     16.617780   \n",
       "50%        ...          16.862601     14.602264     16.340676     17.227212   \n",
       "75%        ...          17.988194     15.369069     16.953637     17.761895   \n",
       "max        ...          21.309852     17.978504     20.607729     20.899094   \n",
       "\n",
       "             ZYG11A        ZYG11B           ZYX         ZZEF1          ZZZ3  \\\n",
       "count  11284.000000  11284.000000  11284.000000  11284.000000  11284.000000   \n",
       "mean      11.034740     17.568689     19.878162     18.062571     17.549486   \n",
       "std        4.794702      1.120799      1.148660      1.065064      0.987241   \n",
       "min       -9.965784     10.388018     14.118455     12.988152     11.607331   \n",
       "25%        9.012625     16.962933     19.163031     17.430391     16.973854   \n",
       "50%       11.866893     17.655066     19.896538     18.149090     17.624513   \n",
       "75%       14.115450     18.333242     20.566523     18.753259     18.193407   \n",
       "max       21.113483     22.043753     24.416677     22.150687     21.651597   \n",
       "\n",
       "          cancercat  \n",
       "count  11284.000000  \n",
       "mean      15.014002  \n",
       "std        9.301482  \n",
       "min        0.000000  \n",
       "25%        7.000000  \n",
       "50%       14.000000  \n",
       "75%       23.000000  \n",
       "max       32.000000  \n",
       "\n",
       "[8 rows x 25151 columns]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#This section requires preprocessed files. \n",
    "import pandas as pd\n",
    "H50 = pd.read_csv('h.all.v6.2.symbols.gmt', header = None, sep='\\t')\n",
    "df_categories = pd.read_pickle('RNAexp_H50cat_tumorTypes.pkl')\n",
    "df_categories.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Use FastAI \n",
    "from fastai import *\n",
    "from fastai.tabular import * \n",
    "\n",
    "#reserve a random sample of 3000 samples to validate the model\n",
    "valid_idx = np.random.choice(len(df_categories), 3000, replace=False)\n",
    "#Normalize continous variables\n",
    "procs = [FillMissing, Categorify, Normalize]\n",
    "\n",
    "#name of the dependent variable for the tumor classes to predict\n",
    "dep_var = 'cancertype'\n",
    "#name of categorical variables: 50 binary variables corresponding to the H50 pathways indicating if the pathway is activated or not in the tumor sample \n",
    "cat_names = H50.loc[:,0].values #in fastAI all other fields not in dep_var or cat_names are handled as continous variables\n",
    "#name of continous variables: 25K continous variables containing to the log2 of RNA expression for each gene in the sample\n",
    "cont_names = np.setdiff1d(list(df_categories.columns), list(cat_names) + [dep_var, 'cancercat'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Acute Myeloid Leukemia', 'Adrenocortical Carcinoma', 'Bladder Urothelial Carcinoma', 'Brain Lower Grade Glioma', 'Breast Invasive Carcinoma', 'Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma', 'Cholangiocarcinoma', 'Colon Adenocarcinoma', 'Esophageal Carcinoma', 'Glioblastoma Multiforme', 'Head and Neck Squamous Cell Carcinoma', 'Kidney Chromophobe', 'Kidney Renal Clear Cell Carcinoma', 'Kidney Renal Papillary Cell Carcinoma', 'Liver Hepatocellular Carcinoma', 'Lung Adenocarcinoma', 'Lung Squamous Cell Carcinoma', 'Lymphoid Neoplasm Diffuse Large B-cell Lymphoma', 'Mesothelioma', 'Ovarian Serous Cystadenocarcinoma', 'Pancreatic Adenocarcinoma', 'Pheochromocytoma and Paraganglioma', 'Prostate Adenocarcinoma', 'Rectum Adenocarcinoma', 'Sarcoma', 'Skin Cutaneous Melanoma', 'Stomach Adenocarcinoma', 'Testicular Germ Cell Tumors', 'Thymoma', 'Thyroid Carcinoma', 'Uterine Carcinosarcoma', 'Uterine Corpus Endometrial Carcinoma', 'Uveal Melanoma']\n"
     ]
    }
   ],
   "source": [
    "class_names = list(df_categories.cancertype.cat.categories)\n",
    "print(class_names)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### First Experiment: train a [model](#prep) using RNA exp + signaling pathways"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = TabularDataBunch.from_df('./', df_categories, \n",
    "                                dep_var=dep_var, valid_idx=valid_idx, procs=procs, \n",
    "                                cat_names=cat_names, cont_names=cont_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#print(data.train_ds.cont_names)  # `cont_names` defaults to: set(df)-set(cat_names)-{dep_var}\n",
    "\n",
    "learn = tabular_learner(data, layers=[1024,516,128,64], metrics=accuracy, ps=[.2,.2,.2,.2])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fd41776fe10>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.lr_find()\n",
    "learn.recorder.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total time: 00:49\n",
      "epoch  train_loss  valid_loss  accuracy\n",
      "1      1.071876    0.357942    0.909000  (00:04)\n",
      "2      0.403538    0.248604    0.914333  (00:04)\n",
      "3      0.305325    0.304598    0.914667  (00:04)\n",
      "4      0.238887    0.270671    0.912667  (00:04)\n",
      "5      0.152964    0.231212    0.934333  (00:04)\n",
      "6      0.115090    0.172455    0.944667  (00:04)\n",
      "7      0.083138    0.152248    0.951667  (00:04)\n",
      "8      0.040459    0.156743    0.954667  (00:04)\n",
      "9      0.018306    0.160180    0.959333  (00:04)\n",
      "10     0.012852    0.152014    0.963000  (00:04)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "learn.fit_one_cycle(10, 1e-2)\n",
    "learn.save('tumorClassifierFastAI')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = learn.get_preds()\n",
    "y_pred = np.argmax(y_pred[0].numpy(), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "y = data.valid_ds.y.items\n",
    "\n",
    "m = pd.DataFrame({'code': y, 'label': [x for x in data.valid_ds.y] })\n",
    "d = m.drop_duplicates(subset='code')\n",
    "d = d.sort_values(by=['code'])\n",
    "class_names = d.label.values\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Automatically created module for IPython interactive environment\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fd31cf9e320>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(__doc__)\n",
    "import sklearn\n",
    "import itertools\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn import svm, datasets\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "#np.set_printoptions(threshold=np.nan)\n",
    "from pylab import rcParams\n",
    "rcParams['figure.figsize'] = 15,15\n",
    "\n",
    "\n",
    "\n",
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "    \n",
    "    #print(cm)\n",
    "    \n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    fmt = '.2f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, format(cm[i, j], fmt),\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')\n",
    "    plt.xticks(rotation=90)\n",
    "    plt.tight_layout()\n",
    "\n",
    "\n",
    "# Compute confusion matrix\n",
    "cnf_matrix = confusion_matrix(y, y_pred)\n",
    "np.set_printoptions(precision=1)\n",
    "\n",
    "# Plot non-normalized confusion matrix\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names,\n",
    "                      title='Confusion matrix, without normalization')\n",
    "\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Breast Invasive Carcinoma                                           317\n",
       "Kidney Renal Clear Cell Carcinoma                                   171\n",
       "Prostate Adenocarcinoma                                             167\n",
       "Lung Adenocarcinoma                                                 166\n",
       "Uterine Corpus Endometrial Carcinoma                                160\n",
       "Head and Neck Squamous Cell Carcinoma                               150\n",
       "Brain Lower Grade Glioma                                            148\n",
       "Colon Adenocarcinoma                                                148\n",
       "Thyroid Carcinoma                                                   142\n",
       "Lung Squamous Cell Carcinoma                                        132\n",
       "Skin Cutaneous Melanoma                                             122\n",
       "Stomach Adenocarcinoma                                              114\n",
       "Ovarian Serous Cystadenocarcinoma                                   110\n",
       "Bladder Urothelial Carcinoma                                        107\n",
       "Liver Hepatocellular Carcinoma                                      107\n",
       "Kidney Renal Papillary Cell Carcinoma                               101\n",
       "Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma     78\n",
       "Sarcoma                                                              71\n",
       "Esophageal Carcinoma                                                 54\n",
       "Testicular Germ Cell Tumors                                          53\n",
       "Pheochromocytoma and Paraganglioma                                   51\n",
       "Pancreatic Adenocarcinoma                                            47\n",
       "Rectum Adenocarcinoma                                                47\n",
       "Glioblastoma Multiforme                                              46\n",
       "Acute Myeloid Leukemia                                               37\n",
       "Thymoma                                                              28\n",
       "Kidney Chromophobe                                                   23\n",
       "Mesothelioma                                                         23\n",
       "Uveal Melanoma                                                       23\n",
       "Lymphoid Neoplasm Diffuse Large B-cell Lymphoma                      17\n",
       "Uterine Carcinosarcoma                                               17\n",
       "Adrenocortical Carcinoma                                             14\n",
       "Cholangiocarcinoma                                                    9\n",
       "Name: cancertype, dtype: int64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "#Validation set\n",
    "df_categories.iloc[valid_idx.tolist(),df_categories.columns.get_loc('cancertype')].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Second Experiment: remove input from signaling pathways and use only RNA exp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = TabularDataBunch.from_df('./', df_categories, \n",
    "                                dep_var=dep_var, valid_idx=valid_idx, procs=procs, \n",
    "                                cat_names=[], cont_names=cont_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "#print(data.train_ds.cont_names)  # `cont_names` defaults to: set(df)-set(cat_names)-{dep_var}\n",
    "\n",
    "learn = tabular_learner(data, layers=[1024,516,128,64], metrics=accuracy, ps=[.2,.2,.2,.2])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fd344164828>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.lr_find()\n",
    "learn.recorder.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total time: 00:47\n",
      "epoch  train_loss  valid_loss  accuracy\n",
      "1      1.125185    0.393269    0.897000  (00:04)\n",
      "2      0.411790    0.272298    0.921333  (00:04)\n",
      "3      0.284616    0.282682    0.923333  (00:04)\n",
      "4      0.207195    0.240883    0.927333  (00:04)\n",
      "5      0.167336    0.229546    0.928333  (00:04)\n",
      "6      0.110502    0.243625    0.929000  (00:04)\n",
      "7      0.081649    0.156840    0.953000  (00:04)\n",
      "8      0.052377    0.143791    0.959333  (00:04)\n",
      "9      0.040370    0.140383    0.958333  (00:04)\n",
      "10     0.030391    0.142785    0.959000  (00:04)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "learn.fit_one_cycle(10, 1e-2)\n",
    "learn.save('tumorClassifierFastAI')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Topology \n",
    "<a id='model'></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TabularModel(\n",
       "  (embeds): ModuleList()\n",
       "  (emb_drop): Dropout(p=0.0)\n",
       "  (bn_cont): BatchNorm1d(25150, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "  (layers): Sequential(\n",
       "    (0): Linear(in_features=25150, out_features=1024, bias=True)\n",
       "    (1): ReLU(inplace)\n",
       "    (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (3): Dropout(p=0.2)\n",
       "    (4): Linear(in_features=1024, out_features=516, bias=True)\n",
       "    (5): ReLU(inplace)\n",
       "    (6): BatchNorm1d(516, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (7): Dropout(p=0.2)\n",
       "    (8): Linear(in_features=516, out_features=128, bias=True)\n",
       "    (9): ReLU(inplace)\n",
       "    (10): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (11): Dropout(p=0.2)\n",
       "    (12): Linear(in_features=128, out_features=64, bias=True)\n",
       "    (13): ReLU(inplace)\n",
       "    (14): BatchNorm1d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (15): Dropout(p=0.2)\n",
       "    (16): Linear(in_features=64, out_features=33, bias=True)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "learn.model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = learn.get_preds()\n",
    "y_pred = np.argmax(y_pred[0].numpy(), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = data.valid_ds.y.items\n",
    "\n",
    "m = pd.DataFrame({'code': y, 'label': [x for x in data.valid_ds.y] })\n",
    "d = m.drop_duplicates(subset='code')\n",
    "d = d.sort_values(by=['code'])\n",
    "class_names = d.label.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Automatically created module for IPython interactive environment\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fd1c5d5f240>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(__doc__)\n",
    "import sklearn\n",
    "import itertools\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn import svm, datasets\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "#np.set_printoptions(threshold=np.nan)\n",
    "from pylab import rcParams\n",
    "rcParams['figure.figsize'] = 15,15\n",
    "\n",
    "\n",
    "# Compute confusion matrix\n",
    "cnf_matrix = confusion_matrix(y, y_pred)\n",
    "np.set_printoptions(precision=1)\n",
    "\n",
    "# Plot non-normalized confusion matrix\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names,\n",
    "                      title='Confusion matrix, without normalization')\n",
    "\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(31,\n",
       " tensor(6),\n",
       " tensor([4.2717e-06, 2.3154e-06, 7.4061e-07, 1.9717e-05, 2.4829e-08, 4.3415e-07,\n",
       "         9.9986e-01, 5.7460e-10, 1.3946e-05, 9.6211e-08, 3.7760e-07, 8.5407e-08,\n",
       "         2.0124e-06, 6.8098e-06, 2.7512e-06, 1.5385e-07, 6.3460e-08, 7.1756e-09,\n",
       "         3.2403e-08, 1.6407e-05, 6.4813e-07, 1.3447e-07, 7.1733e-09, 1.3105e-07,\n",
       "         6.6687e-06, 5.8773e-07, 6.0080e-08, 6.9000e-07, 2.1876e-05, 3.3902e-06,\n",
       "         2.6645e-06, 1.4633e-07, 2.9564e-05]))"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "row = df_categories.iloc[52]\n",
    "learn.predict(row)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Uterine Corpus Endometrial Carcinoma'"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "row.cancertype"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Preprocessing and file generation \n",
    "<a id='prep'></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "%reload_ext autoreload\n",
    "%autoreload 2\n",
    "import h5py\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "meta\n",
      "data\n",
      "info\n",
      "auc\n",
      "(11284,)\n",
      "[b'3116748351' b'5069849036' b'4882401623' ... b'5742234142' b'6773612787'\n",
      " b'7613399024']\n",
      "bigwig_file\n",
      "(11284,)\n",
      "[b'3DFF72D2-F292-497E-ACE3-6FAA9C884205.bw'\n",
      " b'B1E54366-42B9-463C-8615-B34D52BD14DC.bw'\n",
      " b'473713F7-EB41-4F20-A37F-ACD209E3CB75.bw' ...\n",
      " b'76987B28-B56B-4A1F-B77C-1E08B8EFEF90.bw'\n",
      " b'33737781-8638-4FA2-AD4C-E888BB9343D8.bw'\n",
      " b'0BB05CA7-C6FF-42A1-919C-D801F471CBBD.bw']\n",
      "cancertype\n",
      "(11284,)\n",
      "[b'Liver Hepatocellular Carcinoma' b'Prostate Adenocarcinoma'\n",
      " b'Rectum Adenocarcinoma' ... b'Brain Lower Grade Glioma'\n",
      " b'Acute Myeloid Leukemia' b'Thyroid Carcinoma']\n",
      "gdc_annotations\n",
      "(11284,)\n",
      "[b'NULL' b'NULL' b'NULL' ... b'NULL'\n",
      " b'list(category = \"Alternate sample pipeline\", status = \"Approved\", entity_id = \"fff35c80-88cd-4923-80c1-0273ba5bed0f\", classification = \"Notification\", updated_datetime = \"2016-05-01T15:00:21.638875-05:00\", created_datetime = \"2012-11-13T12:37:22-05:00\", entity_submitter_id = \"TCGA-AB-2881\", notes = \"Biospecimens from this case belong to Batch 25 and were processed into analyte outside of the normal TCGA standardized laboratory pipeline.  Users should note that the resulting characterization data may exhibit features (e.g. batch effects) due to this difference.\", \\n    entity_type = \"case\", submitter_id = \"12189\", case_id = \"fff35c80-88cd-4923-80c1-0273ba5bed0f\", annotation_id = \"91e80142-dccb-546b-838d-03837dd1a30e\")'\n",
      " b'NULL']\n",
      "gdc_cases.annotations\n",
      "(11284,)\n",
      "[b'NULL' b'NULL' b'NULL' ... b'NULL'\n",
      " b'list(category = \"Alternate sample pipeline\", status = \"Approved\", classification = \"Notification\", updated_datetime = \"2016-05-01T15:00:21.638875-05:00\", created_datetime = \"2012-11-13T12:37:22-05:00\", notes = \"Biospecimens from this case belong to Batch 25 and were processed into analyte outside of the normal TCGA standardized laboratory pipeline.  Users should note that the resulting characterization data may exhibit features (e.g. batch effects) due to this difference.\", submitter_id = \"12189\", \\n    annotation_id = \"91e80142-dccb-546b-838d-03837dd1a30e\")'\n",
      " b'NULL']\n",
      "gdc_cases.case_id\n",
      "(11284,)\n",
      "[b'0004d251-3f70-4395-b175-c94c2f5b1b81'\n",
      " b'000d566c-96c7-4f1c-b36e-fa2222467983'\n",
      " b'0011a67b-1ba9-4a32-a6b8-7850759a38cf' ...\n",
      " b'fff304a2-113f-499d-a88c-9d3660c348d9'\n",
      " b'fff35c80-88cd-4923-80c1-0273ba5bed0f'\n",
      " b'fffdb1d9-58d1-425c-ac12-1e1e5f443bf7']\n",
      "gdc_cases.project.name\n",
      "(11284,)\n",
      "[b'Liver Hepatocellular Carcinoma' b'Prostate Adenocarcinoma'\n",
      " b'Rectum Adenocarcinoma' ... b'Brain Lower Grade Glioma'\n",
      " b'Acute Myeloid Leukemia' b'Thyroid Carcinoma']\n",
      "gdc_cases.project.primary_site\n",
      "(11284,)\n",
      "[b'Liver' b'Prostate' b'Colorectal' ... b'Brain' b'Bone Marrow' b'Thyroid']\n",
      "gdc_cases.project.project_id\n",
      "(11284,)\n",
      "[b'TCGA-LIHC' b'TCGA-PRAD' b'TCGA-READ' ... b'TCGA-LGG' b'TCGA-LAML'\n",
      " b'TCGA-THCA']\n",
      "gdc_cases.samples.is_ffpe\n",
      "(11284,)\n",
      "[b'FALSE' b'FALSE' b'FALSE' ... b'FALSE' b'FALSE' b'FALSE']\n",
      "gdc_cases.samples.portions.analytes.aliquots.aliquot_id\n",
      "(11284,)\n",
      "[b'6c3a4f09-1862-4fed-9977-759ec00a5aeb'\n",
      " b'9d5f0b9f-ff36-4a80-8472-412d0ec803c5'\n",
      " b'fbf67b9e-05a7-4280-a7b8-13363f1bd1c4' ...\n",
      " b'1d7d22de-0bfc-4357-8c5a-7289aa0d97b8'\n",
      " b'af389bfe-c693-4763-a719-8f74f265e6ae'\n",
      " b'e9eda4da-4b9d-4add-8229-985b27f5ac2c']\n",
      "gdc_cases.samples.portions.analytes.aliquots.annotations\n",
      "(11284,)\n",
      "[b'NULL' b'NULL' b'NULL' ... b'NULL' b'NULL' b'NULL']\n",
      "gdc_cases.samples.portions.analytes.aliquots.source_center\n",
      "(11284,)\n",
      "[b'23' b'23' b'22' ... b'23' b'23' b'23']\n",
      "gdc_cases.samples.portions.analytes.aliquots.submitter_id\n",
      "(11284,)\n",
      "[b'TCGA-DD-AAVP-01A-11R-A41C-07' b'TCGA-KK-A7B2-01A-12R-A32O-07'\n",
      " b'TCGA-DC-6158-01A-11R-1660-07' ... b'TCGA-RY-A847-01A-11R-A36H-07'\n",
      " b'TCGA-AB-2881-03A-01T-0735-13' b'TCGA-DJ-A4V0-01A-11R-A250-07']\n",
      "gdc_cases.samples.portions.analytes.analyte_id\n",
      "(11284,)\n",
      "[b'2246af47-8d57-49bc-8249-85b5dce8f42c'\n",
      " b'9e093595-180c-4a5c-940e-27ddea8374b3'\n",
      " b'867d8ba8-2261-4c57-9bc9-cb81f7226991' ...\n",
      " b'e2c8613e-f156-4ab7-af82-321dbdb9dc41'\n",
      " b'9cced397-aa3f-4451-a942-54528d5e8b1e'\n",
      " b'1624438d-5d9f-499d-84a1-14394d5228a1']\n",
      "gdc_cases.samples.portions.analytes.analyte_type\n",
      "(11284,)\n",
      "[b'RNA' b'RNA' b'RNA' ... b'RNA' b'Total RNA' b'RNA']\n",
      "gdc_cases.samples.portions.analytes.analyte_type_id\n",
      "(11284,)\n",
      "[b'R' b'R' b'R' ... b'R' b'T' b'R']\n",
      "gdc_cases.samples.portions.analytes.annotations\n",
      "(11284,)\n",
      "[b'NULL' b'NULL' b'NULL' ... b'NULL' b'NULL' b'NULL']\n",
      "gdc_cases.samples.portions.analytes.submitter_id\n",
      "(11284,)\n",
      "[b'TCGA-DD-AAVP-01A-11R' b'TCGA-KK-A7B2-01A-12R' b'TCGA-DC-6158-01A-11R'\n",
      " ... b'TCGA-RY-A847-01A-11R' b'TCGA-AB-2881-03A-01T'\n",
      " b'TCGA-DJ-A4V0-01A-11R']\n",
      "gdc_cases.samples.portions.creation_datetime\n",
      "(11284,)\n",
      "[b'1397088000' b'1377475200' b'1299196800' ... b'1382918400' b'0'\n",
      " b'1351641600']\n",
      "gdc_cases.samples.portions.portion_id\n",
      "(11284,)\n",
      "[b'671ac90c-63b4-450b-814b-797349f81f15'\n",
      " b'ab77d27b-5d3d-46d7-9b19-30aef7fc34d0'\n",
      " b'abc0ddaf-1842-4e7d-bca3-c54a91d54170' ...\n",
      " b'7a9f863a-d7fe-476a-93ff-3c9b95e597f7'\n",
      " b'71be866a-c0a1-447e-94de-ef2a85a891c7'\n",
      " b'e78cb405-c721-4a2c-a099-7c8a4eb5e46d']\n",
      "gdc_cases.samples.portions.portion_number\n",
      "(11284,)\n",
      "[b'11' b'12' b'11' ... b'11' b'01' b'11']\n",
      "gdc_cases.samples.portions.slides\n",
      "(11284,)\n",
      "[b'list(percent_tumor_nuclei = 80, percent_monocyte_infiltration = 1, percent_normal_cells = 0, percent_stromal_cells = 20, updated_datetime = \"2016-09-08T11:04:30.952291-05:00\", created_datetime = NA, percent_eosinophil_infiltration = NA, percent_lymphocyte_infiltration = 2, percent_neutrophil_infiltration = 1, state = NA, section_location = \"TOP\", percent_granulocyte_infiltration = NA, percent_necrosis = 0, slide_id = \"c1eab3d0-509c-411b-a23d-19b2515c781d\", percent_inflam_infiltration = NA, submitter_id = \"TCGA-DD-AAVP-01A-01-TS1\", \\n    number_proliferating_cells = NA, percent_tumor_cells = 80)'\n",
      " b'list(percent_tumor_nuclei = 80, percent_monocyte_infiltration = 2, percent_normal_cells = 0, percent_stromal_cells = 55, updated_datetime = \"2016-09-08T13:59:46.086986-05:00\", created_datetime = NA, percent_eosinophil_infiltration = NA, percent_lymphocyte_infiltration = 10, percent_neutrophil_infiltration = 2, state = NA, section_location = \"TOP\", percent_granulocyte_infiltration = NA, percent_necrosis = 1, slide_id = \"3f1e3102-294c-47f4-a1da-1d0f09abe9c1\", percent_inflam_infiltration = NA, submitter_id = \"TCGA-KK-A7B2-01A-01-TSA\", \\n    number_proliferating_cells = NA, percent_tumor_cells = 44)'\n",
      " b'list(percent_tumor_nuclei = c(70, 70), percent_monocyte_infiltration = c(NA, NA), percent_normal_cells = c(10, 8), percent_stromal_cells = c(20, 20), updated_datetime = c(\"2016-09-08T10:40:20.418685-05:00\", \"2016-09-08T10:40:20.418685-05:00\"), created_datetime = c(NA, NA), percent_eosinophil_infiltration = c(NA, NA), percent_lymphocyte_infiltration = c(NA, NA), percent_neutrophil_infiltration = c(NA, NA), state = c(NA, NA), section_location = c(\"TOP\", \"BOTTOM\"), percent_granulocyte_infiltration = c(NA, \\nNA), percent_necrosis = c(5, 5), slide_id = c(\"f5f1dac8-1047-478e-bc77-5c7ddd43566f\", \"b4e1dde1-d2b8-4dd2-b0c0-5bb927825869\"), percent_inflam_infiltration = c(NA, NA), submitter_id = c(\"TCGA-DC-6158-01A-01-TS1\", \"TCGA-DC-6158-01A-01-BS1\"), number_proliferating_cells = c(NA, NA), percent_tumor_cells = c(65, 67))'\n",
      " ...\n",
      " b'list(percent_tumor_nuclei = 75, percent_monocyte_infiltration = 0, percent_normal_cells = 10, percent_stromal_cells = 65, updated_datetime = \"2016-09-08T12:43:40.136159-05:00\", created_datetime = NA, percent_eosinophil_infiltration = NA, percent_lymphocyte_infiltration = 0, percent_neutrophil_infiltration = 0, state = NA, section_location = \"TOP\", percent_granulocyte_infiltration = NA, percent_necrosis = 0, slide_id = \"0d4e8888-4d3a-402f-8db8-495e9e366832\", percent_inflam_infiltration = NA, submitter_id = \"TCGA-RY-A847-01A-01-TS1\", \\n    number_proliferating_cells = NA, percent_tumor_cells = 25)'\n",
      " b'NULL'\n",
      " b'list(percent_tumor_nuclei = 75, percent_monocyte_infiltration = 1, percent_normal_cells = 2, percent_stromal_cells = 23, updated_datetime = \"2016-09-08T13:40:21.455089-05:00\", created_datetime = NA, percent_eosinophil_infiltration = NA, percent_lymphocyte_infiltration = 2, percent_neutrophil_infiltration = 0, state = NA, section_location = \"TOP\", percent_granulocyte_infiltration = NA, percent_necrosis = 0, slide_id = \"3fd33092-6128-4444-b9d3-0ac11627eff9\", percent_inflam_infiltration = NA, submitter_id = \"TCGA-DJ-A4V0-01A-01-TSA\", \\n    number_proliferating_cells = NA, percent_tumor_cells = 75)']\n",
      "gdc_cases.samples.portions.submitter_id\n",
      "(11284,)\n",
      "[b'TCGA-DD-AAVP-01A-11' b'TCGA-KK-A7B2-01A-12' b'TCGA-DC-6158-01A-11' ...\n",
      " b'TCGA-RY-A847-01A-11' b'TCGA-AB-2881-03A-01' b'TCGA-DJ-A4V0-01A-11']\n",
      "gdc_cases.samples.sample_id\n",
      "(11284,)\n",
      "[b'9259e9ee-7279-4b62-8512-509cb705029c'\n",
      " b'bccb7f9d-8dde-45b7-a50d-6c7233239ceb'\n",
      " b'03877fe0-b0d6-4800-be46-b45c70350e48' ...\n",
      " b'bf8e8426-dc44-40b0-8459-ad1dbaf9ba5a'\n",
      " b'421c391a-33c7-4a6d-aecb-9a1cb4e8a2d9'\n",
      " b'a418a2e8-6a8d-4e7d-8c43-f09b521e51d8']\n",
      "gdc_cases.samples.sample_type\n",
      "(11284,)\n",
      "[b'Primary Tumor' b'Primary Tumor' b'Primary Tumor' ... b'Primary Tumor'\n",
      " b'Primary Blood Derived Cancer - Peripheral Blood' b'Primary Tumor']\n",
      "gdc_cases.samples.sample_type_id\n",
      "(11284,)\n",
      "[b'01' b'01' b'01' ... b'01' b'03' b'01']\n",
      "gdc_cases.samples.submitter_id\n",
      "(11284,)\n",
      "[b'TCGA-DD-AAVP-01A' b'TCGA-KK-A7B2-01A' b'TCGA-DC-6158-01A' ...\n",
      " b'TCGA-RY-A847-01A' b'TCGA-AB-2881-03A' b'TCGA-DJ-A4V0-01A']\n",
      "gdc_cases.submitter_id\n",
      "(11284,)\n",
      "[b'TCGA-DD-AAVP' b'TCGA-KK-A7B2' b'TCGA-DC-6158' ... b'TCGA-RY-A847'\n",
      " b'TCGA-AB-2881' b'TCGA-DJ-A4V0']\n",
      "gdc_cases.tissue_source_site.bcr_id\n",
      "(11284,)\n",
      "[b'NCH' b'IGC' b'IGC' ... b'NCH' b'NCH' b'NCH']\n",
      "gdc_cases.tissue_source_site.name\n",
      "(11284,)\n",
      "[b'Mayo Clinic - Rochester' b'MD Anderson Cancer Center' b'MSKCC' ...\n",
      " b'University of California San Francisco' b'Washington University'\n",
      " b'Memorial Sloan Kettering']\n",
      "gdc_cases.tissue_source_site.project\n",
      "(11284,)\n",
      "[b'Liver hepatocellular carcinoma' b'Prostate adenocarcinoma'\n",
      " b'Rectum adenocarcinoma' ... b'Brain Lower Grade Glioma'\n",
      " b'Acute Myeloid Leukemia' b'Thyroid carcinoma']\n",
      "gdc_cases.tissue_source_site.tissue_source_site_id\n",
      "(11284,)\n",
      "[b'c2eb5c30-93cc-5a99-b9e4-f676e35ed7f8'\n",
      " b'c4a03d2d-b7e9-5506-9521-4df06091fb20'\n",
      " b'2ec37cf9-8bca-524f-9fb8-1241ab73fac7' ...\n",
      " b'02c87f8b-69ba-51fb-a99f-c2f22183b992'\n",
      " b'5b2bb0b1-686a-59fa-9f1c-960a8d701e83'\n",
      " b'612b370d-47ee-5e52-ab25-6770f395998f']\n",
      "gdc_cases.updated_datetime\n",
      "(11284,)\n",
      "[b'2016-09-08T11:04:30.952291-05:00' b'2016-09-08T13:59:46.086986-05:00'\n",
      " b'2016-09-08T10:40:20.418685-05:00' ...\n",
      " b'2016-09-08T12:43:40.136159-05:00' b'2016-09-08T13:13:12.360689-05:00'\n",
      " b'2016-09-08T13:40:21.455089-05:00']\n",
      "gdc_center.center_id\n",
      "(11284,)\n",
      "[b'ee7a85b3-8177-5d60-a10c-51180eb9009c'\n",
      " b'ee7a85b3-8177-5d60-a10c-51180eb9009c'\n",
      " b'ee7a85b3-8177-5d60-a10c-51180eb9009c' ...\n",
      " b'ee7a85b3-8177-5d60-a10c-51180eb9009c'\n",
      " b'6eba705a-0f00-5aa2-b1d0-04dbf62100cc'\n",
      " b'ee7a85b3-8177-5d60-a10c-51180eb9009c']\n",
      "gdc_center.code\n",
      "(11284,)\n",
      "[b'07' b'07' b'07' ... b'07' b'13' b'07']\n",
      "gdc_center.name\n",
      "(11284,)\n",
      "[b'University of North Carolina' b'University of North Carolina'\n",
      " b'University of North Carolina' ... b'University of North Carolina'\n",
      " b\"Canada's Michael Smith Genome Sciences Centre\"\n",
      " b'University of North Carolina']\n",
      "gdc_center.namespace\n",
      "(11284,)\n",
      "[b'unc.edu' b'unc.edu' b'unc.edu' ... b'unc.edu' b'bcgsc.ca' b'unc.edu']\n",
      "gdc_center.short_name\n",
      "(11284,)\n",
      "[b'UNC' b'UNC' b'UNC' ... b'UNC' b'BCGSC' b'UNC']\n",
      "gdc_data_format\n",
      "(11284,)\n",
      "[b'TARGZ' b'TARGZ' b'TARGZ' ... b'TARGZ' b'TAR' b'TARGZ']\n",
      "gdc_file_id\n",
      "(11284,)\n",
      "[b'3dff72d2-f292-497e-ace3-6faa9c884205'\n",
      " b'b1e54366-42b9-463c-8615-b34d52bd14dc'\n",
      " b'473713f7-eb41-4f20-a37f-acd209e3cb75' ...\n",
      " b'76987b28-b56b-4a1f-b77c-1e08b8efef90'\n",
      " b'33737781-8638-4fa2-ad4c-e888bb9343d8'\n",
      " b'0bb05ca7-c6ff-42a1-919c-d801f471cbbd']\n",
      "gdc_file_name\n",
      "(11284,)\n",
      "[b'UNCID_2660992.6c3a4f09-1862-4fed-9977-759ec00a5aeb.140826_UNC9-SN296_0433_BC5FWDACXX_5_GGCTAC.tar.gz'\n",
      " b'UNCID_2322574.9d5f0b9f-ff36-4a80-8472-412d0ec803c5.131021_UNC9-SN296_0409_BC2M6NACXX_4_ATCACG.tar.gz'\n",
      " b'UNCID_2185604.fbf67b9e-05a7-4280-a7b8-13363f1bd1c4.121211_UNC16-SN851_0199_BD1KLUACXX_3_AGTTCC.tar.gz'\n",
      " ...\n",
      " b'UNCID_2520250.1d7d22de-0bfc-4357-8c5a-7289aa0d97b8.140325_UNC15-SN850_0358_AC3LP7ACXX_3_ACAGTG.tar.gz'\n",
      " b'TCGA-AB-2881-03A-01T-0735-13_rnaseq_fastq.tar'\n",
      " b'UNCID_2184304.e9eda4da-4b9d-4add-8229-985b27f5ac2c.130214_UNC11-SN627_0281_AD1UHCACXX_7_ACTTGA.tar.gz']\n",
      "gdc_file_size\n",
      "(11284,)\n",
      "[b'2726695328' b'5351542755' b'4518948272' ... b'5653270016' b'5393807360'\n",
      " b'6577268060']\n",
      "gdc_md5sum\n",
      "(11284,)\n",
      "[b'9870f011553dcbea40fa35a43e4beac4' b'903fa5aa4470dfdf2b1b49fb63f33cc7'\n",
      " b'8bb79b9d0b7387fd617ac9a5cf9dc0f4' ...\n",
      " b'97e33728f3aa9e2f649f2cfdf58210eb' b'3fcf2c40f11d5143fb980012e5843d6d'\n",
      " b'c4073b2e1452bb5f8d7a731ef278e5f4']\n",
      "gdc_metadata_files.created_datetime.analysis\n",
      "(11284,)\n",
      "[b'2016-06-22T11:15:44.245302-05:00' b'2016-05-03T11:30:02.988971-05:00'\n",
      " b'2016-05-03T10:59:42.133373-05:00' ...\n",
      " b'2016-05-03T11:51:27.517228-05:00' b'2016-05-03T11:38:01.265665-05:00'\n",
      " b'2016-05-03T10:58:56.650557-05:00']\n",
      "gdc_metadata_files.file_id.analysis\n",
      "(11284,)\n",
      "[b'ebc35995-75c4-4dd6-b895-7153d04f27f9'\n",
      " b'9ae5c78f-8484-4edd-93a8-214a6d4de54e'\n",
      " b'29433d82-efbc-4d1f-9719-10a398da017d' ...\n",
      " b'473fa97a-9ea5-4758-82ee-524a0bceaca7'\n",
      " b'8d4bdcc3-232f-4b80-94f4-dd0bf1e44d65'\n",
      " b'd0e14c59-fa4c-47ba-96a9-80d6b5e18f10']\n",
      "gdc_metadata_files.file_id.experiment\n",
      "(11284,)\n",
      "[b'dbc3f41a-b742-4fb8-8648-69a99a215f78'\n",
      " b'2da96e5c-fb40-458c-8a16-b69248a67872'\n",
      " b'a8a92d7e-c2b5-418f-aa9c-490327c1ab77' ...\n",
      " b'1d3cbeee-5c1f-4daa-a7d6-5220ee370f4f'\n",
      " b'50d61ac3-6161-48be-9b97-888b4f027a87'\n",
      " b'2007eebf-67b9-411c-ac29-e0779a549e5e']\n",
      "gdc_metadata_files.file_id.run\n",
      "(11284,)\n",
      "[b'01bd9600-3510-49c7-a474-5e7e5fcc3414'\n",
      " b'47a46d53-de20-497d-b0d9-4b82f70b13a5'\n",
      " b'a5c90d99-90af-4c39-8b7a-d4a2b1613b1c' ...\n",
      " b'e80d0197-46f8-483e-9f18-df28ab8e823a'\n",
      " b'6b8b7430-898f-4cce-8189-19eb1094a2e6'\n",
      " b'764508a2-3d64-4a98-8d21-70de9e3309a4']\n",
      "gdc_metadata_files.file_name.analysis\n",
      "(11284,)\n",
      "[b'62dca2cf-5330-4d57-b322-ed57dc484550_analysis.xml'\n",
      " b'dc2e7732-07a8-42de-86b4-3dc5503c216e_analysis.xml'\n",
      " b'6c36a436-2ed6-49a0-a498-88a48503136e_analysis.xml' ...\n",
      " b'89ac6022-a529-4b81-93d9-fa851c8ee435_analysis.xml'\n",
      " b'44b39039-2cce-402c-bc94-671c306a8f08_analysis.xml'\n",
      " b'2c200aa9-7df8-4469-a421-db4334b7f338_analysis.xml']\n",
      "gdc_metadata_files.file_name.experiment\n",
      "(11284,)\n",
      "[b'62dca2cf-5330-4d57-b322-ed57dc484550_experiment.xml'\n",
      " b'dc2e7732-07a8-42de-86b4-3dc5503c216e_experiment.xml'\n",
      " b'6c36a436-2ed6-49a0-a498-88a48503136e_experiment.xml' ...\n",
      " b'89ac6022-a529-4b81-93d9-fa851c8ee435_experiment.xml'\n",
      " b'44b39039-2cce-402c-bc94-671c306a8f08_experiment.xml'\n",
      " b'2c200aa9-7df8-4469-a421-db4334b7f338_experiment.xml']\n",
      "gdc_metadata_files.file_name.run\n",
      "(11284,)\n",
      "[b'62dca2cf-5330-4d57-b322-ed57dc484550_run.xml'\n",
      " b'dc2e7732-07a8-42de-86b4-3dc5503c216e_run.xml'\n",
      " b'6c36a436-2ed6-49a0-a498-88a48503136e_run.xml' ...\n",
      " b'89ac6022-a529-4b81-93d9-fa851c8ee435_run.xml'\n",
      " b'44b39039-2cce-402c-bc94-671c306a8f08_run.xml'\n",
      " b'2c200aa9-7df8-4469-a421-db4334b7f338_run.xml']\n",
      "gdc_metadata_files.file_size.analysis\n",
      "(11284,)\n",
      "[b'2122' b'2122' b'2123' ... b'2124' b'2284' b'2123']\n",
      "gdc_metadata_files.file_size.experiment\n",
      "(11284,)\n",
      "[b'2628' b'2628' b'2431' ... b'2628' b'3938' b'2431']\n",
      "gdc_metadata_files.file_size.run\n",
      "(11284,)\n",
      "[b'354' b'354' b'354' ... b'354' b'3407' b'354']\n",
      "gdc_metadata_files.md5sum.analysis\n",
      "(11284,)\n",
      "[b'a71e168607d7a29685f654364a7631c5' b'f6549d14d438410af3212e514279a449'\n",
      " b'4b8e2d6dbdb5d83553b907995f482f24' ...\n",
      " b'26fd5c4a973e04c84d4a6b2ed8336b42' b'1ac10804b15fdb370987faa8446b9ee8'\n",
      " b'71b30d03d804238e8db28ceec332960b']\n",
      "gdc_metadata_files.md5sum.experiment\n",
      "(11284,)\n",
      "[b'7149a679ae2898df623b65d7807030dd' b'4e37fd2949af26989d268b44df3e4cdb'\n",
      " b'1bc2281ecb773b96f66f52c1279c10b3' ...\n",
      " b'6c83ad67e1bba307cc248624acd4d62b' b'88ef523ff48c6a35957b9ea8fdb098ad'\n",
      " b'b0d8a45d2090db49b0806056c915bf7f']\n",
      "gdc_metadata_files.md5sum.run\n",
      "(11284,)\n",
      "[b'ca0ec5262de5a6f211febe2f3ad74eca' b'fe4b20e0939b2640912fe846ee5e6b33'\n",
      " b'4c82b64deba5f35b8e29da5b90eb79ee' ...\n",
      " b'a6bfbc9862b3926cddafdc807c03b6ec' b'c4b519cfd319ab127627a608182565f8'\n",
      " b'0fe7555b28f656b22292e378a23c686e']\n",
      "gdc_metadata_files.updated_datetime.analysis\n",
      "(11284,)\n",
      "[b'2016-09-29T23:50:31.404039-05:00' b'2016-09-29T21:40:10.807547-05:00'\n",
      " b'2016-09-29T18:37:16.469086-05:00' ...\n",
      " b'2016-09-29T19:24:43.821317-05:00' b'2016-09-29T21:18:01.322684-05:00'\n",
      " b'2016-09-29T23:07:14.783439-05:00']\n",
      "gdc_metadata_files.updated_datetime.experiment\n",
      "(11284,)\n",
      "[b'2016-08-13T06:57:44.873718-05:00' b'2016-10-07T16:27:24.893743-05:00'\n",
      " b'2016-09-29T22:02:06.782908-05:00' ...\n",
      " b'2016-08-28T02:33:15.912268-05:00' b'2016-08-27T19:46:42.557267-05:00'\n",
      " b'2016-09-29T18:22:41.660257-05:00']\n",
      "gdc_metadata_files.updated_datetime.run\n",
      "(11284,)\n",
      "[b'2016-09-29T17:34:29.260368-05:00' b'2016-09-29T19:25:21.903235-05:00'\n",
      " b'2016-09-29T21:57:26.310154-05:00' ...\n",
      " b'2016-09-29T23:44:24.815358-05:00' b'2016-09-29T20:23:41.491977-05:00'\n",
      " b'2016-09-29T20:41:14.561790-05:00']\n",
      "gdc_platform\n",
      "(11284,)\n",
      "[b'Illumina HiSeq' b'Illumina HiSeq' b'Illumina HiSeq' ...\n",
      " b'Illumina HiSeq' b'Illumina GA' b'Illumina HiSeq']\n",
      "gdc_submitter_id\n",
      "(11284,)\n",
      "[b'62dca2cf-5330-4d57-b322-ed57dc484550'\n",
      " b'dc2e7732-07a8-42de-86b4-3dc5503c216e'\n",
      " b'6c36a436-2ed6-49a0-a498-88a48503136e' ...\n",
      " b'89ac6022-a529-4b81-93d9-fa851c8ee435'\n",
      " b'44b39039-2cce-402c-bc94-671c306a8f08'\n",
      " b'2c200aa9-7df8-4469-a421-db4334b7f338']\n",
      "gdc_updated_datetime\n",
      "(11284,)\n",
      "[b'2016-08-09T14:42:05.417980-05:00' b'2016-08-26T04:54:14.470936-05:00'\n",
      " b'2016-08-09T13:09:58.664649-05:00' ...\n",
      " b'2016-08-25T17:41:13.293379-05:00' b'2016-08-25T14:00:52.797929-05:00'\n",
      " b'2016-08-05T01:07:11.185164-05:00']\n",
      "genes\n",
      "(25150,)\n",
      "[b'A1BG' b'A1BG-AS1' b'A1CF' ... b'ZYX' b'ZZEF1' b'ZZZ3']\n",
      "mapped_read_count\n",
      "(11284,)\n",
      "[b'65175027' b'106227036' b'102592885' ... b'120236713' b'137419287'\n",
      " b'159646577']\n",
      "reads_downloaded\n",
      "(11284,)\n",
      "[b'65993288' b'108023394' b'103976586' ... b'122325822' b'155307816'\n",
      " b'161782584']\n",
      "sampleid\n",
      "(11284,)\n",
      "[b'3DFF72D2-F292-497E-ACE3-6FAA9C884205'\n",
      " b'B1E54366-42B9-463C-8615-B34D52BD14DC'\n",
      " b'473713F7-EB41-4F20-A37F-ACD209E3CB75' ...\n",
      " b'76987B28-B56B-4A1F-B77C-1E08B8EFEF90'\n",
      " b'33737781-8638-4FA2-AD4C-E888BB9343D8'\n",
      " b'0BB05CA7-C6FF-42A1-919C-D801F471CBBD']\n",
      "expression\n",
      "(11284, 25150)\n",
      "[[1454762  439935  316636 ...  302428   29863   44717]\n",
      " [  28886   23917     192 ... 1009426  156109  108371]\n",
      " [   4322    9719  154254 ... 1319876  348923  253112]\n",
      " ...\n",
      " [  10570   16608     220 ...  402715  307415   85944]\n",
      " [  28190   52035     157 ... 1888391  660036  369616]\n",
      " [  40426   40053      48 ...  793314  481256  313210]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "author\n",
      "(1,)\n",
      "[b'Alexander Lachmann']\n",
      "contact\n",
      "(1,)\n",
      "[b'alexander.lachmann@mssm.edu']\n",
      "countsource\n",
      "(1,)\n",
      "[b'recount2: https://jhubiostatistics.shinyapps.io/recount/']\n",
      "creationdate\n",
      "(1,)\n",
      "[b'10/9/2017']\n",
      "lab\n",
      "(1,)\n",
      "[b\"Ma'ayan Lab, Icahn School of Medicine\"]\n"
     ]
    }
   ],
   "source": [
    "#Importing TCGA RNA expression data downloaded from  https://amp.pharm.mssm.edu/archs4/download.html\n",
    "filename = \"tcga_matrix.h5\"\n",
    "\n",
    "h5 = h5py.File(filename,'r')\n",
    "for k in h5.keys():\n",
    "    print(k)\n",
    "\n",
    "\n",
    "meta = h5['meta']  \n",
    "dat = h5['data']  \n",
    "info = h5['info']  \n",
    "\n",
    "\n",
    "\n",
    "#Showing which type of data it contains    \n",
    "for s in meta.keys():\n",
    "    print(s)\n",
    "    print(meta[s].shape)\n",
    "    print(meta[s].value)\n",
    "    \n",
    "for s in dat.keys():\n",
    "    print(s)\n",
    "    print(dat[s].shape)\n",
    "    print(dat[s].value)    \n",
    "    \n",
    "for s in info.keys():\n",
    "    print(s)\n",
    "    print(info[s].shape)\n",
    "    print(info[s].value)    \n",
    "\n",
    "#h5.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>192</th>\n",
       "      <th>193</th>\n",
       "      <th>194</th>\n",
       "      <th>195</th>\n",
       "      <th>196</th>\n",
       "      <th>197</th>\n",
       "      <th>198</th>\n",
       "      <th>199</th>\n",
       "      <th>200</th>\n",
       "      <th>201</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>HALLMARK_TNFA_SIGNALING_VIA_NFKB</td>\n",
       "      <td>http://www.broadinstitute.org/gsea/msigdb/card...</td>\n",
       "      <td>JUNB</td>\n",
       "      <td>CXCL2</td>\n",
       "      <td>ATF3</td>\n",
       "      <td>NFKBIA</td>\n",
       "      <td>TNFAIP3</td>\n",
       "      <td>PTGS2</td>\n",
       "      <td>CXCL1</td>\n",
       "      <td>IER3</td>\n",
       "      <td>...</td>\n",
       "      <td>EIF1</td>\n",
       "      <td>BMP2</td>\n",
       "      <td>DUSP4</td>\n",
       "      <td>PDLIM5</td>\n",
       "      <td>ICOSLG</td>\n",
       "      <td>GFPT2</td>\n",
       "      <td>KLF2</td>\n",
       "      <td>TNC</td>\n",
       "      <td>SERPINB8</td>\n",
       "      <td>MXD1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>HALLMARK_HYPOXIA</td>\n",
       "      <td>http://www.broadinstitute.org/gsea/msigdb/card...</td>\n",
       "      <td>PGK1</td>\n",
       "      <td>PDK1</td>\n",
       "      <td>GBE1</td>\n",
       "      <td>PFKL</td>\n",
       "      <td>ALDOA</td>\n",
       "      <td>ENO2</td>\n",
       "      <td>PGM1</td>\n",
       "      <td>NDRG1</td>\n",
       "      <td>...</td>\n",
       "      <td>HDLBP</td>\n",
       "      <td>ILVBL</td>\n",
       "      <td>NCAN</td>\n",
       "      <td>TGM2</td>\n",
       "      <td>ETS1</td>\n",
       "      <td>HOXB9</td>\n",
       "      <td>SELENBP1</td>\n",
       "      <td>FOSL2</td>\n",
       "      <td>SULT2B1</td>\n",
       "      <td>TGFB3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>HALLMARK_CHOLESTEROL_HOMEOSTASIS</td>\n",
       "      <td>http://www.broadinstitute.org/gsea/msigdb/card...</td>\n",
       "      <td>FDPS</td>\n",
       "      <td>CYP51A1</td>\n",
       "      <td>IDI1</td>\n",
       "      <td>FDFT1</td>\n",
       "      <td>DHCR7</td>\n",
       "      <td>SQLE</td>\n",
       "      <td>HMGCS1</td>\n",
       "      <td>NSDHL</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>HALLMARK_MITOTIC_SPINDLE</td>\n",
       "      <td>http://www.broadinstitute.org/gsea/msigdb/card...</td>\n",
       "      <td>ARHGEF2</td>\n",
       "      <td>CLASP1</td>\n",
       "      <td>KIF11</td>\n",
       "      <td>KIF23</td>\n",
       "      <td>ALS2</td>\n",
       "      <td>ARF6</td>\n",
       "      <td>MYO9B</td>\n",
       "      <td>MYH9</td>\n",
       "      <td>...</td>\n",
       "      <td>RHOT2</td>\n",
       "      <td>SORBS2</td>\n",
       "      <td>CDC42EP1</td>\n",
       "      <td>VCL</td>\n",
       "      <td>CLIP1</td>\n",
       "      <td>STK38L</td>\n",
       "      <td>YWHAE</td>\n",
       "      <td>RAPGEF5</td>\n",
       "      <td>CEP72</td>\n",
       "      <td>CSNK1D</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>HALLMARK_WNT_BETA_CATENIN_SIGNALING</td>\n",
       "      <td>http://www.broadinstitute.org/gsea/msigdb/card...</td>\n",
       "      <td>MYC</td>\n",
       "      <td>CTNNB1</td>\n",
       "      <td>JAG2</td>\n",
       "      <td>NOTCH1</td>\n",
       "      <td>DLL1</td>\n",
       "      <td>AXIN2</td>\n",
       "      <td>PSEN2</td>\n",
       "      <td>FZD1</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 202 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   0    \\\n",
       "0     HALLMARK_TNFA_SIGNALING_VIA_NFKB   \n",
       "1                     HALLMARK_HYPOXIA   \n",
       "2     HALLMARK_CHOLESTEROL_HOMEOSTASIS   \n",
       "3             HALLMARK_MITOTIC_SPINDLE   \n",
       "4  HALLMARK_WNT_BETA_CATENIN_SIGNALING   \n",
       "\n",
       "                                                 1        2        3      4    \\\n",
       "0  http://www.broadinstitute.org/gsea/msigdb/card...     JUNB    CXCL2   ATF3   \n",
       "1  http://www.broadinstitute.org/gsea/msigdb/card...     PGK1     PDK1   GBE1   \n",
       "2  http://www.broadinstitute.org/gsea/msigdb/card...     FDPS  CYP51A1   IDI1   \n",
       "3  http://www.broadinstitute.org/gsea/msigdb/card...  ARHGEF2   CLASP1  KIF11   \n",
       "4  http://www.broadinstitute.org/gsea/msigdb/card...      MYC   CTNNB1   JAG2   \n",
       "\n",
       "      5        6      7       8      9     ...      192     193       194  \\\n",
       "0  NFKBIA  TNFAIP3  PTGS2   CXCL1   IER3   ...     EIF1    BMP2     DUSP4   \n",
       "1    PFKL    ALDOA   ENO2    PGM1  NDRG1   ...    HDLBP   ILVBL      NCAN   \n",
       "2   FDFT1    DHCR7   SQLE  HMGCS1  NSDHL   ...      NaN     NaN       NaN   \n",
       "3   KIF23     ALS2   ARF6   MYO9B   MYH9   ...    RHOT2  SORBS2  CDC42EP1   \n",
       "4  NOTCH1     DLL1  AXIN2   PSEN2   FZD1   ...      NaN     NaN       NaN   \n",
       "\n",
       "      195     196     197       198      199       200     201  \n",
       "0  PDLIM5  ICOSLG   GFPT2      KLF2      TNC  SERPINB8    MXD1  \n",
       "1    TGM2    ETS1   HOXB9  SELENBP1    FOSL2   SULT2B1   TGFB3  \n",
       "2     NaN     NaN     NaN       NaN      NaN       NaN     NaN  \n",
       "3     VCL   CLIP1  STK38L     YWHAE  RAPGEF5     CEP72  CSNK1D  \n",
       "4     NaN     NaN     NaN       NaN      NaN       NaN     NaN  \n",
       "\n",
       "[5 rows x 202 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Importing pathway enrichment analysis for 50 Cancer Hallmark pathways http://software.broadinstitute.org/gsea/msigdb/collections.jsp#H\n",
    "#In gmt format each gene set is a row and is described by a name, a description, and the genes in the gene set\n",
    "import pandas as pd\n",
    "\n",
    "H50 = pd.read_csv('h.all.v6.2.symbols.gmt', header = None, sep='\\t')\n",
    "H50.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Add H50 gen sets as a categorical variable "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'The number of unique genes involved in the 50 selected signaling pathways is 4387.'"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Different genes involved in 50 selected signaling pathways \n",
    "genePath_set = pd.unique(H50.loc[:,2:].values.ravel('K'))\n",
    "\n",
    "f'The number of unique genes involved in the 50 selected signaling pathways is {genePath_set.size}.'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'H50 contains 50 signaling pathways composed of a set of 202 genes'"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f'H50 contains {H50.shape[0]} signaling pathways composed of a set of {H50.shape[1]} genes'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DescribeResult(nobs=11284, minmax=(array([ 4.90693868,  4.39238612, -9.96578428, ..., 14.11845455,\n",
       "       12.98815228, 11.60733078]), array([25.74434345, 23.45781508, 21.78148379, ..., 24.41667745,\n",
       "       22.1506869 , 21.65159744])), mean=array([14.51133518, 14.65828565,  7.81866386, ..., 19.87816157,\n",
       "       18.0625713 , 17.54948585]), variance=array([ 5.12112323,  3.17381026, 58.31272768, ...,  1.31941998,\n",
       "        1.13436237,  0.97464454]), skewness=array([ 1.34799737,  1.15626794, -0.97460423, ..., -0.09400296,\n",
       "       -0.34315363, -0.48017077]), kurtosis=array([4.18004613, 4.2873423 , 0.95231487, ..., 0.81602377, 0.62193682,\n",
       "       0.94856403]))"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#if not exists it saves a file 'log2expression.npy' with the log2 transformation of RNA expression \n",
    "import os.path\n",
    "def f(x):\n",
    "    #return log2(TPM+0.001)\n",
    "    return np.log2(x +0.001 ) \n",
    "\n",
    "l = None\n",
    "if not os.path.exists('log2expression.npy'):\n",
    "    #log2(TPM+0.001) expression values computed per gene\n",
    "    l = dat['expression'].value\n",
    "    l = np.vectorize(f)(l)\n",
    "    np.save('log2expression.npy', l)\n",
    "else: \n",
    "    l = np.load('log2expression.npy')\n",
    "\n",
    "from scipy import stats\n",
    "stats.describe(l)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>A1BG</th>\n",
       "      <th>A1BG-AS1</th>\n",
       "      <th>A1CF</th>\n",
       "      <th>A2M</th>\n",
       "      <th>A2M-AS1</th>\n",
       "      <th>A2ML1</th>\n",
       "      <th>A2MP1</th>\n",
       "      <th>A3GALT2</th>\n",
       "      <th>A4GALT</th>\n",
       "      <th>A4GNT</th>\n",
       "      <th>...</th>\n",
       "      <th>ZWILCH</th>\n",
       "      <th>ZWINT</th>\n",
       "      <th>ZXDA</th>\n",
       "      <th>ZXDB</th>\n",
       "      <th>ZXDC</th>\n",
       "      <th>ZYG11A</th>\n",
       "      <th>ZYG11B</th>\n",
       "      <th>ZYX</th>\n",
       "      <th>ZZEF1</th>\n",
       "      <th>ZZZ3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "      <td>11284.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>14.511335</td>\n",
       "      <td>14.658286</td>\n",
       "      <td>7.818664</td>\n",
       "      <td>21.170343</td>\n",
       "      <td>16.323402</td>\n",
       "      <td>11.515712</td>\n",
       "      <td>10.927732</td>\n",
       "      <td>3.512874</td>\n",
       "      <td>16.375104</td>\n",
       "      <td>4.546174</td>\n",
       "      <td>...</td>\n",
       "      <td>16.317508</td>\n",
       "      <td>16.803166</td>\n",
       "      <td>14.481343</td>\n",
       "      <td>16.262297</td>\n",
       "      <td>17.164771</td>\n",
       "      <td>11.034740</td>\n",
       "      <td>17.568689</td>\n",
       "      <td>19.878162</td>\n",
       "      <td>18.062571</td>\n",
       "      <td>17.549486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.262990</td>\n",
       "      <td>1.781519</td>\n",
       "      <td>7.636277</td>\n",
       "      <td>1.819773</td>\n",
       "      <td>1.530948</td>\n",
       "      <td>5.025257</td>\n",
       "      <td>2.047451</td>\n",
       "      <td>7.174155</td>\n",
       "      <td>1.840625</td>\n",
       "      <td>7.976631</td>\n",
       "      <td>...</td>\n",
       "      <td>1.238517</td>\n",
       "      <td>1.550544</td>\n",
       "      <td>1.295317</td>\n",
       "      <td>1.108982</td>\n",
       "      <td>0.948553</td>\n",
       "      <td>4.794702</td>\n",
       "      <td>1.120799</td>\n",
       "      <td>1.148660</td>\n",
       "      <td>1.065064</td>\n",
       "      <td>0.987241</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.906939</td>\n",
       "      <td>4.392386</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>10.540129</td>\n",
       "      <td>8.876520</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>6.988696</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>...</td>\n",
       "      <td>9.238407</td>\n",
       "      <td>8.906894</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>7.584970</td>\n",
       "      <td>11.969387</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>10.388018</td>\n",
       "      <td>14.118455</td>\n",
       "      <td>12.988152</td>\n",
       "      <td>11.607331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>13.115125</td>\n",
       "      <td>13.612897</td>\n",
       "      <td>6.247946</td>\n",
       "      <td>20.103902</td>\n",
       "      <td>15.396688</td>\n",
       "      <td>8.640249</td>\n",
       "      <td>10.006678</td>\n",
       "      <td>3.322072</td>\n",
       "      <td>15.161555</td>\n",
       "      <td>4.807406</td>\n",
       "      <td>...</td>\n",
       "      <td>15.450841</td>\n",
       "      <td>15.698264</td>\n",
       "      <td>13.734075</td>\n",
       "      <td>15.639362</td>\n",
       "      <td>16.617780</td>\n",
       "      <td>9.012625</td>\n",
       "      <td>16.962933</td>\n",
       "      <td>19.163031</td>\n",
       "      <td>17.430391</td>\n",
       "      <td>16.973854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>14.377075</td>\n",
       "      <td>14.614969</td>\n",
       "      <td>7.936644</td>\n",
       "      <td>21.240994</td>\n",
       "      <td>16.313299</td>\n",
       "      <td>10.918491</td>\n",
       "      <td>11.040290</td>\n",
       "      <td>6.584978</td>\n",
       "      <td>16.524121</td>\n",
       "      <td>7.584970</td>\n",
       "      <td>...</td>\n",
       "      <td>16.299789</td>\n",
       "      <td>16.862601</td>\n",
       "      <td>14.602264</td>\n",
       "      <td>16.340676</td>\n",
       "      <td>17.227212</td>\n",
       "      <td>11.866893</td>\n",
       "      <td>17.655066</td>\n",
       "      <td>19.896538</td>\n",
       "      <td>18.149090</td>\n",
       "      <td>17.624513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>15.611969</td>\n",
       "      <td>15.556147</td>\n",
       "      <td>11.956830</td>\n",
       "      <td>22.350001</td>\n",
       "      <td>17.271283</td>\n",
       "      <td>14.611140</td>\n",
       "      <td>12.092427</td>\n",
       "      <td>8.029291</td>\n",
       "      <td>17.722402</td>\n",
       "      <td>9.205794</td>\n",
       "      <td>...</td>\n",
       "      <td>17.199718</td>\n",
       "      <td>17.988194</td>\n",
       "      <td>15.369069</td>\n",
       "      <td>16.953637</td>\n",
       "      <td>17.761895</td>\n",
       "      <td>14.115450</td>\n",
       "      <td>18.333242</td>\n",
       "      <td>20.566523</td>\n",
       "      <td>18.753259</td>\n",
       "      <td>18.193407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>25.744343</td>\n",
       "      <td>23.457815</td>\n",
       "      <td>21.781484</td>\n",
       "      <td>27.526837</td>\n",
       "      <td>22.749258</td>\n",
       "      <td>25.852982</td>\n",
       "      <td>18.892860</td>\n",
       "      <td>14.192447</td>\n",
       "      <td>21.365289</td>\n",
       "      <td>20.369699</td>\n",
       "      <td>...</td>\n",
       "      <td>20.524677</td>\n",
       "      <td>21.309852</td>\n",
       "      <td>17.978504</td>\n",
       "      <td>20.607729</td>\n",
       "      <td>20.899094</td>\n",
       "      <td>21.113483</td>\n",
       "      <td>22.043753</td>\n",
       "      <td>24.416677</td>\n",
       "      <td>22.150687</td>\n",
       "      <td>21.651597</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 25150 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               A1BG      A1BG-AS1          A1CF           A2M       A2M-AS1  \\\n",
       "count  11284.000000  11284.000000  11284.000000  11284.000000  11284.000000   \n",
       "mean      14.511335     14.658286      7.818664     21.170343     16.323402   \n",
       "std        2.262990      1.781519      7.636277      1.819773      1.530948   \n",
       "min        4.906939      4.392386     -9.965784     10.540129      8.876520   \n",
       "25%       13.115125     13.612897      6.247946     20.103902     15.396688   \n",
       "50%       14.377075     14.614969      7.936644     21.240994     16.313299   \n",
       "75%       15.611969     15.556147     11.956830     22.350001     17.271283   \n",
       "max       25.744343     23.457815     21.781484     27.526837     22.749258   \n",
       "\n",
       "              A2ML1         A2MP1       A3GALT2        A4GALT         A4GNT  \\\n",
       "count  11284.000000  11284.000000  11284.000000  11284.000000  11284.000000   \n",
       "mean      11.515712     10.927732      3.512874     16.375104      4.546174   \n",
       "std        5.025257      2.047451      7.174155      1.840625      7.976631   \n",
       "min       -9.965784     -9.965784     -9.965784      6.988696     -9.965784   \n",
       "25%        8.640249     10.006678      3.322072     15.161555      4.807406   \n",
       "50%       10.918491     11.040290      6.584978     16.524121      7.584970   \n",
       "75%       14.611140     12.092427      8.029291     17.722402      9.205794   \n",
       "max       25.852982     18.892860     14.192447     21.365289     20.369699   \n",
       "\n",
       "           ...             ZWILCH         ZWINT          ZXDA          ZXDB  \\\n",
       "count      ...       11284.000000  11284.000000  11284.000000  11284.000000   \n",
       "mean       ...          16.317508     16.803166     14.481343     16.262297   \n",
       "std        ...           1.238517      1.550544      1.295317      1.108982   \n",
       "min        ...           9.238407      8.906894     -9.965784      7.584970   \n",
       "25%        ...          15.450841     15.698264     13.734075     15.639362   \n",
       "50%        ...          16.299789     16.862601     14.602264     16.340676   \n",
       "75%        ...          17.199718     17.988194     15.369069     16.953637   \n",
       "max        ...          20.524677     21.309852     17.978504     20.607729   \n",
       "\n",
       "               ZXDC        ZYG11A        ZYG11B           ZYX         ZZEF1  \\\n",
       "count  11284.000000  11284.000000  11284.000000  11284.000000  11284.000000   \n",
       "mean      17.164771     11.034740     17.568689     19.878162     18.062571   \n",
       "std        0.948553      4.794702      1.120799      1.148660      1.065064   \n",
       "min       11.969387     -9.965784     10.388018     14.118455     12.988152   \n",
       "25%       16.617780      9.012625     16.962933     19.163031     17.430391   \n",
       "50%       17.227212     11.866893     17.655066     19.896538     18.149090   \n",
       "75%       17.761895     14.115450     18.333242     20.566523     18.753259   \n",
       "max       20.899094     21.113483     22.043753     24.416677     22.150687   \n",
       "\n",
       "               ZZZ3  \n",
       "count  11284.000000  \n",
       "mean      17.549486  \n",
       "std        0.987241  \n",
       "min       11.607331  \n",
       "25%       16.973854  \n",
       "50%       17.624513  \n",
       "75%       18.193407  \n",
       "max       21.651597  \n",
       "\n",
       "[8 rows x 25150 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#associate gene names with RNA expression\n",
    "def f(x):\n",
    "    return x.decode(\"utf-8\")\n",
    "genes = np.vectorize(f)(meta['genes'].value)\n",
    "df = pd.DataFrame(l, columns = genes)\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['JUNB' 'PGK1' 'FDPS' ... 'DYRK3' 'TPH1' 'P2RY4']\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'RNA exp gene matrix is reduced to the 4304 genes involved in pathways'"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#One hot 50 signaling pathways for each instance\n",
    "#1. Select only genes involved in pathways\n",
    "print(genePath_set)\n",
    "genePath_I_geneRNA= set(df.columns).intersection(set(genePath_set))\n",
    "df_s = df[list(genePath_I_geneRNA)]\n",
    "f'RNA exp gene matrix is reduced to the {df_s.shape[1]} genes involved in pathways'\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#2. Add result of pathway enrichment analysis for 50 Cancer Hallmark pathways as categorical variables: \n",
    "#Array of 50 boolean\n",
    "#If all genes involved on each signaling pathway from H50 are RNA log2 expressed > 0 then this pathway is assigned True in the sample\n",
    "def f(row):\n",
    "    a = []\n",
    "    for index, p in H50.iterrows():\n",
    "        #print(p)\n",
    "        #exclude from pathway genes 'p' those that are not in genePath_I_geneRNA (Total 83)\n",
    "        p_s = set(p[2:]).intersection(genePath_I_geneRNA)\n",
    "        rna = row[list(p_s)]\n",
    "        #print (rna)\n",
    "        a.append(np.all(rna.values > 0))\n",
    "    return np.array(a)\n",
    "df_s['H50_onehot'] = df_s.apply(f, axis=1)\n",
    "#Merge H50 onehot with the full RNA expression matrix\n",
    "df['H50_onehot'] = df_s['H50_onehot'] \n",
    "df.to_pickle(\"RNAexp_H50onehot.pkl\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>A1BG</th>\n",
       "      <th>A1BG-AS1</th>\n",
       "      <th>A1CF</th>\n",
       "      <th>A2M</th>\n",
       "      <th>A2M-AS1</th>\n",
       "      <th>A2ML1</th>\n",
       "      <th>A2MP1</th>\n",
       "      <th>A3GALT2</th>\n",
       "      <th>A4GALT</th>\n",
       "      <th>A4GNT</th>\n",
       "      <th>...</th>\n",
       "      <th>ZXDA</th>\n",
       "      <th>ZXDB</th>\n",
       "      <th>ZXDC</th>\n",
       "      <th>ZYG11A</th>\n",
       "      <th>ZYG11B</th>\n",
       "      <th>ZYX</th>\n",
       "      <th>ZZEF1</th>\n",
       "      <th>ZZZ3</th>\n",
       "      <th>H50_onehot</th>\n",
       "      <th>cancertype</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>20.472352</td>\n",
       "      <td>18.746931</td>\n",
       "      <td>18.272466</td>\n",
       "      <td>19.290201</td>\n",
       "      <td>15.257498</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>9.299210</td>\n",
       "      <td>5.643885</td>\n",
       "      <td>13.595724</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>...</td>\n",
       "      <td>11.805341</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>15.517669</td>\n",
       "      <td>9.278452</td>\n",
       "      <td>15.699383</td>\n",
       "      <td>18.206232</td>\n",
       "      <td>14.866072</td>\n",
       "      <td>15.448536</td>\n",
       "      <td>[False, False, False, True, False, False, Fals...</td>\n",
       "      <td>Liver Hepatocellular Carcinoma</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>14.818083</td>\n",
       "      <td>14.545749</td>\n",
       "      <td>7.584970</td>\n",
       "      <td>21.170981</td>\n",
       "      <td>17.082887</td>\n",
       "      <td>13.392988</td>\n",
       "      <td>10.586841</td>\n",
       "      <td>3.322072</td>\n",
       "      <td>17.862692</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>...</td>\n",
       "      <td>13.460072</td>\n",
       "      <td>15.752668</td>\n",
       "      <td>16.611414</td>\n",
       "      <td>12.744834</td>\n",
       "      <td>16.470659</td>\n",
       "      <td>19.945104</td>\n",
       "      <td>17.252194</td>\n",
       "      <td>16.725619</td>\n",
       "      <td>[True, False, True, True, True, True, False, T...</td>\n",
       "      <td>Prostate Adenocarcinoma</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>12.077484</td>\n",
       "      <td>13.246592</td>\n",
       "      <td>17.234948</td>\n",
       "      <td>20.949892</td>\n",
       "      <td>14.822969</td>\n",
       "      <td>6.700454</td>\n",
       "      <td>10.472692</td>\n",
       "      <td>7.321937</td>\n",
       "      <td>16.977783</td>\n",
       "      <td>7.044405</td>\n",
       "      <td>...</td>\n",
       "      <td>15.637785</td>\n",
       "      <td>17.493347</td>\n",
       "      <td>16.861135</td>\n",
       "      <td>8.839207</td>\n",
       "      <td>17.636320</td>\n",
       "      <td>20.331971</td>\n",
       "      <td>18.412549</td>\n",
       "      <td>17.949416</td>\n",
       "      <td>[True, False, True, True, True, True, False, T...</td>\n",
       "      <td>Rectum Adenocarcinoma</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>24.457885</td>\n",
       "      <td>22.684448</td>\n",
       "      <td>19.907039</td>\n",
       "      <td>23.860669</td>\n",
       "      <td>19.778088</td>\n",
       "      <td>7.918869</td>\n",
       "      <td>13.952559</td>\n",
       "      <td>1.000721</td>\n",
       "      <td>16.081317</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>...</td>\n",
       "      <td>12.749870</td>\n",
       "      <td>15.258603</td>\n",
       "      <td>16.945592</td>\n",
       "      <td>12.561527</td>\n",
       "      <td>18.267911</td>\n",
       "      <td>16.341345</td>\n",
       "      <td>17.340346</td>\n",
       "      <td>17.422426</td>\n",
       "      <td>[False, False, True, True, True, True, False, ...</td>\n",
       "      <td>Liver Hepatocellular Carcinoma</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11.676398</td>\n",
       "      <td>12.256504</td>\n",
       "      <td>8.076821</td>\n",
       "      <td>19.457004</td>\n",
       "      <td>14.529736</td>\n",
       "      <td>17.318905</td>\n",
       "      <td>8.209458</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>17.581149</td>\n",
       "      <td>-9.965784</td>\n",
       "      <td>...</td>\n",
       "      <td>13.757911</td>\n",
       "      <td>15.446534</td>\n",
       "      <td>17.096242</td>\n",
       "      <td>12.186734</td>\n",
       "      <td>17.651276</td>\n",
       "      <td>20.299427</td>\n",
       "      <td>17.006589</td>\n",
       "      <td>17.689343</td>\n",
       "      <td>[False, False, True, True, False, True, False,...</td>\n",
       "      <td>Bladder Urothelial Carcinoma</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25152 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        A1BG   A1BG-AS1       A1CF        A2M    A2M-AS1      A2ML1  \\\n",
       "0  20.472352  18.746931  18.272466  19.290201  15.257498  -9.965784   \n",
       "1  14.818083  14.545749   7.584970  21.170981  17.082887  13.392988   \n",
       "2  12.077484  13.246592  17.234948  20.949892  14.822969   6.700454   \n",
       "3  24.457885  22.684448  19.907039  23.860669  19.778088   7.918869   \n",
       "4  11.676398  12.256504   8.076821  19.457004  14.529736  17.318905   \n",
       "\n",
       "       A2MP1   A3GALT2     A4GALT     A4GNT               ...                \\\n",
       "0   9.299210  5.643885  13.595724 -9.965784               ...                 \n",
       "1  10.586841  3.322072  17.862692 -9.965784               ...                 \n",
       "2  10.472692  7.321937  16.977783  7.044405               ...                 \n",
       "3  13.952559  1.000721  16.081317 -9.965784               ...                 \n",
       "4   8.209458 -9.965784  17.581149 -9.965784               ...                 \n",
       "\n",
       "        ZXDA       ZXDB       ZXDC     ZYG11A     ZYG11B        ZYX  \\\n",
       "0  11.805341  13.000000  15.517669   9.278452  15.699383  18.206232   \n",
       "1  13.460072  15.752668  16.611414  12.744834  16.470659  19.945104   \n",
       "2  15.637785  17.493347  16.861135   8.839207  17.636320  20.331971   \n",
       "3  12.749870  15.258603  16.945592  12.561527  18.267911  16.341345   \n",
       "4  13.757911  15.446534  17.096242  12.186734  17.651276  20.299427   \n",
       "\n",
       "       ZZEF1       ZZZ3                                         H50_onehot  \\\n",
       "0  14.866072  15.448536  [False, False, False, True, False, False, Fals...   \n",
       "1  17.252194  16.725619  [True, False, True, True, True, True, False, T...   \n",
       "2  18.412549  17.949416  [True, False, True, True, True, True, False, T...   \n",
       "3  17.340346  17.422426  [False, False, True, True, True, True, False, ...   \n",
       "4  17.006589  17.689343  [False, False, True, True, False, True, False,...   \n",
       "\n",
       "                       cancertype  \n",
       "0  Liver Hepatocellular Carcinoma  \n",
       "1         Prostate Adenocarcinoma  \n",
       "2           Rectum Adenocarcinoma  \n",
       "3  Liver Hepatocellular Carcinoma  \n",
       "4    Bladder Urothelial Carcinoma  \n",
       "\n",
       "[5 rows x 25152 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Add the ground truth: tumor type for each instance\n",
    "#Load processed data\n",
    "df = pd.read_pickle('RNAexp_H50onehot.pkl')\n",
    "\n",
    "df.loc[:,'cancertype']= pd.Series(np.vectorize(f)(meta['cancertype'].value))\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(11284, 25202)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Breast Invasive Carcinoma                                           1246\n",
       "Kidney Renal Clear Cell Carcinoma                                    616\n",
       "Lung Adenocarcinoma                                                  601\n",
       "Uterine Corpus Endometrial Carcinoma                                 589\n",
       "Thyroid Carcinoma                                                    572\n",
       "Prostate Adenocarcinoma                                              558\n",
       "Lung Squamous Cell Carcinoma                                         555\n",
       "Head and Neck Squamous Cell Carcinoma                                548\n",
       "Colon Adenocarcinoma                                                 546\n",
       "Brain Lower Grade Glioma                                             532\n",
       "Skin Cutaneous Melanoma                                              473\n",
       "Stomach Adenocarcinoma                                               453\n",
       "Bladder Urothelial Carcinoma                                         433\n",
       "Ovarian Serous Cystadenocarcinoma                                    430\n",
       "Liver Hepatocellular Carcinoma                                       424\n",
       "Kidney Renal Papillary Cell Carcinoma                                323\n",
       "Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma     309\n",
       "Sarcoma                                                              265\n",
       "Esophageal Carcinoma                                                 198\n",
       "Pheochromocytoma and Paraganglioma                                   187\n",
       "Pancreatic Adenocarcinoma                                            183\n",
       "Rectum Adenocarcinoma                                                177\n",
       "Glioblastoma Multiforme                                              175\n",
       "Testicular Germ Cell Tumors                                          156\n",
       "Acute Myeloid Leukemia                                               126\n",
       "Thymoma                                                              122\n",
       "Kidney Chromophobe                                                    91\n",
       "Mesothelioma                                                          87\n",
       "Uveal Melanoma                                                        80\n",
       "Adrenocortical Carcinoma                                              79\n",
       "Uterine Carcinosarcoma                                                57\n",
       "Lymphoid Neoplasm Diffuse Large B-cell Lymphoma                       48\n",
       "Cholangiocarcinoma                                                    45\n",
       "Name: cancertype, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "H50_categories = pd.DataFrame(df.H50_onehot.tolist(), columns = H50.loc[:,0].values)\n",
    "df_categories = df.join(H50_categories)\n",
    "print(df_categories.shape)\n",
    "df_categories = df_categories.drop(columns=['H50_onehot'])\n",
    "df_categories['cancertype'] = pd.Categorical(df_categories.cancertype)\n",
    "df_categories['cancercat'] = df_categories.cancertype.cat.codes #convert cancer types to codes\n",
    "df_categories.to_pickle(\"RNAexp_H50cat_tumorTypes.pkl\")\n",
    "df_categories['cancertype'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "References:\n",
    "1. Yuan, Yuchen, et al. \"DeepGene: an advanced cancer type classifier based on deep learning and somatic point mutations.\" BMC bioinformatics 17.17 (2016): 476.\n",
    "2. https://medium.com/@alenaharley/the-mystery-of-the-origin-cancer-type-classification-using-fast-ai-libray-212eaf8d3f4e "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python (gen)",
   "language": "python",
   "name": "gen"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}