malaria_report.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Open Malaria Predictive Model Report\n",
    "#### Ángela López del Río\n",
    "\n",
    "Taking profit of a competition from the Open Source Malaria project (https://github.com/OpenSourceMalaria/OSM_To_Do_List/issues/421), which provides a collection of molecules and their activity against the malaria parasite's ion pump, PfATP4, a machine learning-based QSAR model has been developed. This model aims to predict which molecules will block this target. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%run 'malaria_functions.ipynb'\n",
    "\n",
    "from __future__ import absolute_import, division, print_function, unicode_literals\n",
    "import os\n",
    "import sys\n",
    "from xgboost import plot_importance\n",
    "from matplotlib import pyplot\n",
    "from sklearn.feature_selection import SelectFromModel\n",
    "from sklearn.metrics import accuracy_score, roc_curve, auc\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Firstly, the dataset of active and inactive compounds for the target PfATP4 provided in the OSM Competition is loaded. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of instances of each level of ION_ACTIVITY \n",
      " 0.0    395\n",
      "1.0     55\n",
      "0.5      5\n",
      "Name: ION_ACTIVITY, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "#Loading data with compounds IDs, SMILES and information about their activity\n",
    "osm_list = reading_csv()\n",
    "osm_cleaned = data_preproc(osm_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since the Potency vs Parasite (EC50) data of most of the molecules is provided, this value is plotted against the PfATP4 activity label to check if there is a correlation between the whole cell assay results and the activity against this specific ion. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACScAAAiuCAYAAAD9vmTHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAwBD/AULyYeICqW7HAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1057847d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Boxplotting to check how pEC50 correlates to ION_ACTIVITY indexes\n",
    "fig = osm_cleaned.boxplot('pEC50', by='ION_ACTIVITY')\n",
    "fig.get_figure().suptitle(\"\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As it can be seen in the plot, the followed behaviour is not as expected: the whole cell assay response doesn't seem to be related to the ion activity label. \n",
    "\n",
    "However, if we separate the Series 4 compounds (those in whose design and synthesis is currently focused the OSM project), these two values correlate as expected:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of instances of each level of ION_ACTIVITY \n",
      " 1.0    23\n",
      "0.0    17\n",
      "0.5     4\n",
      "Name: ION_ACTIVITY, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACTEAAAiuCAYAAADXDpVoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10eed5c50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Separating Series 4 compounds\n",
    "osm_series4 = series4_cmp(osm_cleaned)\n",
    "\n",
    "# Boxplotting to check how pEC50 correlates to ION_ACTIVITY indexes only taking into account Series 4 indexes\n",
    "fig = osm_series4.boxplot('pEC50', by='ION_ACTIVITY')\n",
    "fig.get_figure().suptitle(\"\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pybel and RDkit descriptors of OSM molecules are computed, resulting in 215 descriptors for each compound."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 455 entries, 0 to 454\n",
      "Columns: 215 entries, HBA1 to fr_urea\n",
      "dtypes: float64(108), int64(104), object(3)\n",
      "memory usage: 767.8+ KB\n"
     ]
    }
   ],
   "source": [
    "#Getting RDkit descriptors\n",
    "osm_rdkit = rdkit_desc(osm_cleaned)\n",
    "#Getting Pybel descriptors\n",
    "osm_pyb = pybel_desc(osm_cleaned)\n",
    "#merging descriptors\n",
    "pyb_rdk = merge_dfs(osm_pyb, osm_rdkit)\n",
    "\n",
    "pyb_rdk.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The provided dataset contains ion activity information of 395 actives and 55 inactives molecules. Additionally, 5 molecules appear to be partially active, but since such a small number of samples has no statistically power, it will be removed from further steps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Converting dataframe to ndarray\n",
    "npdesc, nplab = df_to_ndarrays(pyb_rdk)\n",
    "# Scaling data\n",
    "npdesc = preprocessing.scale(npdesc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data is randomly divided in training set (70%) and test set (30%)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training dataset shape Counter({0.0: 281, 1.0: 34})\n",
      "Test dataset shape Counter({0.0: 114, 1.0: 21})\n"
     ]
    }
   ],
   "source": [
    "x_train_s, x_test, y_train_s, y_test = splitting_data(npdesc, nplab, 0.3)\n",
    "print(\"Training dataset shape {}\".format(Counter(y_train_s)))\n",
    "print(\"Test dataset shape {}\".format(Counter(y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "At this point, different classification models were tuned, trained and evaluated. The performance of different classification algorithms (Logistic Regression Model, Support Vector Machine, Random Forests, K-Nearest Neighbours, Extreme Gradient Boosting) was assessed, with and without applying a previous feature selection based on the importance of the variables and with and without applying an oversampling method to cover the imbalance between the inactives and actives. \n",
    "\n",
    "The preliminary first results showed that the **Extreme Gradient Boosting**\\* (XGBoost) presented a best performance, so the final model was built and exhaustively tuned using it. This algorithm dominates structured or tabular datasets on classification and regression predictive modeling problems, and it is the go-to method for competition winners on the Kaggle competitive data science platform.\n",
    "\n",
    "Gradient boosting is basically about \"boosting\" many weak predictive models into a strong one, in the form of ensemble of weak models. To build the strong model, we need to find a good way to \"combine\" weak models. In the most popular boosting algorithm, weak models are trained in an adaptive way: a weak model is trained using data samples drawn according to some weight distribution; then, the weight of samples that are misclassified by the model is increased and the weight of samples that are classified correctly is decreased. The next weak model is trained using samples drawn according to the updated weight distribution.\n",
    "\n",
    "In this way, the algorithm always trains models using data samples that are \"difficult\" to learn in previous rounds, which results an ensemble of models that are good at learning different \"parts\" of training data.\n",
    "\n",
    "\\* <sub>Chen, T., & Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System. https://doi.org/10.1145/2939672.2939785</sub>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 338,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/angela/virtualenvs/malaria/lib/python2.7/site-packages/sklearn/model_selection/_search.py:667: DeprecationWarning: The grid_scores_ attribute was deprecated in version 0.18 in favor of the more elaborate cv_results_ attribute. The grid_scores_ attribute will not be available from 0.20\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[mean: 0.86857, std: 0.04982, params: {u'max_depth': 3, u'min_child_weight': 1}, mean: 0.83821, std: 0.10305, params: {u'max_depth': 3, u'min_child_weight': 3}, mean: 0.82549, std: 0.09078, params: {u'max_depth': 3, u'min_child_weight': 5}, mean: 0.89394, std: 0.02957, params: {u'max_depth': 5, u'min_child_weight': 1}, mean: 0.83711, std: 0.10260, params: {u'max_depth': 5, u'min_child_weight': 3}, mean: 0.82812, std: 0.09415, params: {u'max_depth': 5, u'min_child_weight': 5}, mean: 0.88721, std: 0.01406, params: {u'max_depth': 7, u'min_child_weight': 1}, mean: 0.83711, std: 0.10260, params: {u'max_depth': 7, u'min_child_weight': 3}, mean: 0.82812, std: 0.09415, params: {u'max_depth': 7, u'min_child_weight': 5}, mean: 0.88721, std: 0.01406, params: {u'max_depth': 9, u'min_child_weight': 1}, mean: 0.83711, std: 0.10260, params: {u'max_depth': 9, u'min_child_weight': 3}, mean: 0.82812, std: 0.09415, params: {u'max_depth': 9, u'min_child_weight': 5}] {u'max_depth': 5, u'min_child_weight': 1} 0.893940207662\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/angela/virtualenvs/malaria/lib/python2.7/site-packages/sklearn/model_selection/_search.py:667: DeprecationWarning: The grid_scores_ attribute was deprecated in version 0.18 in favor of the more elaborate cv_results_ attribute. The grid_scores_ attribute will not be available from 0.20\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[mean: 0.89394, std: 0.02957, params: {u'gamma': 0.0}, mean: 0.88791, std: 0.01990, params: {u'gamma': 0.1}, mean: 0.89613, std: 0.02278, params: {u'gamma': 0.2}, mean: 0.88825, std: 0.01446, params: {u'gamma': 0.3}, mean: 0.88396, std: 0.02309, params: {u'gamma': 0.4}] {u'gamma': 0.2} 0.896128714644\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/angela/virtualenvs/malaria/lib/python2.7/site-packages/sklearn/model_selection/_search.py:667: DeprecationWarning: The grid_scores_ attribute was deprecated in version 0.18 in favor of the more elaborate cv_results_ attribute. The grid_scores_ attribute will not be available from 0.20\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[mean: 0.87271, std: 0.03473, params: {u'subsample': 0.6, u'colsample_bytree': 0.6}, mean: 0.89003, std: 0.03241, params: {u'subsample': 0.7, u'colsample_bytree': 0.6}, mean: 0.88774, std: 0.02665, params: {u'subsample': 0.8, u'colsample_bytree': 0.6}, mean: 0.87591, std: 0.04539, params: {u'subsample': 0.9, u'colsample_bytree': 0.6}, mean: 0.87674, std: 0.04067, params: {u'subsample': 0.6, u'colsample_bytree': 0.7}, mean: 0.87669, std: 0.04870, params: {u'subsample': 0.7, u'colsample_bytree': 0.7}, mean: 0.87923, std: 0.03392, params: {u'subsample': 0.8, u'colsample_bytree': 0.7}, mean: 0.87922, std: 0.04459, params: {u'subsample': 0.9, u'colsample_bytree': 0.7}, mean: 0.87690, std: 0.03694, params: {u'subsample': 0.6, u'colsample_bytree': 0.8}, mean: 0.87792, std: 0.04472, params: {u'subsample': 0.7, u'colsample_bytree': 0.8}, mean: 0.89236, std: 0.02442, params: {u'subsample': 0.8, u'colsample_bytree': 0.8}, mean: 0.87538, std: 0.04270, params: {u'subsample': 0.9, u'colsample_bytree': 0.8}, mean: 0.87314, std: 0.03515, params: {u'subsample': 0.6, u'colsample_bytree': 0.9}, mean: 0.88231, std: 0.04017, params: {u'subsample': 0.7, u'colsample_bytree': 0.9}, mean: 0.88955, std: 0.03405, params: {u'subsample': 0.8, u'colsample_bytree': 0.9}, mean: 0.86826, std: 0.05194, params: {u'subsample': 0.9, u'colsample_bytree': 0.9}] {u'subsample': 0.8, u'colsample_bytree': 0.8} 0.892355889724\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/angela/virtualenvs/malaria/lib/python2.7/site-packages/sklearn/model_selection/_search.py:667: DeprecationWarning: The grid_scores_ attribute was deprecated in version 0.18 in favor of the more elaborate cv_results_ attribute. The grid_scores_ attribute will not be available from 0.20\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[mean: 0.87682, std: 0.05468, params: {u'subsample': 0.75, u'colsample_bytree': 0.75}, mean: 0.88374, std: 0.02390, params: {u'subsample': 0.8, u'colsample_bytree': 0.75}, mean: 0.87347, std: 0.04924, params: {u'subsample': 0.85, u'colsample_bytree': 0.75}, mean: 0.87527, std: 0.03811, params: {u'subsample': 0.9, u'colsample_bytree': 0.75}, mean: 0.87494, std: 0.04645, params: {u'subsample': 0.75, u'colsample_bytree': 0.8}, mean: 0.89236, std: 0.02442, params: {u'subsample': 0.8, u'colsample_bytree': 0.8}, mean: 0.87607, std: 0.03262, params: {u'subsample': 0.85, u'colsample_bytree': 0.8}, mean: 0.87538, std: 0.04270, params: {u'subsample': 0.9, u'colsample_bytree': 0.8}, mean: 0.87861, std: 0.03686, params: {u'subsample': 0.75, u'colsample_bytree': 0.85}, mean: 0.87769, std: 0.04905, params: {u'subsample': 0.8, u'colsample_bytree': 0.85}, mean: 0.87501, std: 0.03839, params: {u'subsample': 0.85, u'colsample_bytree': 0.85}, mean: 0.86971, std: 0.04168, params: {u'subsample': 0.9, u'colsample_bytree': 0.85}, mean: 0.86358, std: 0.06042, params: {u'subsample': 0.75, u'colsample_bytree': 0.9}, mean: 0.88955, std: 0.03405, params: {u'subsample': 0.8, u'colsample_bytree': 0.9}, mean: 0.87327, std: 0.02915, params: {u'subsample': 0.85, u'colsample_bytree': 0.9}, mean: 0.86826, std: 0.05194, params: {u'subsample': 0.9, u'colsample_bytree': 0.9}] {u'subsample': 0.8, u'colsample_bytree': 0.8} 0.892355889724\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/angela/virtualenvs/malaria/lib/python2.7/site-packages/sklearn/model_selection/_search.py:667: DeprecationWarning: The grid_scores_ attribute was deprecated in version 0.18 in favor of the more elaborate cv_results_ attribute. The grid_scores_ attribute will not be available from 0.20\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[mean: 0.89236, std: 0.02442, params: {u'reg_alpha': 1e-05}, mean: 0.88211, std: 0.03421, params: {u'reg_alpha': 0.01}, mean: 0.88798, std: 0.02462, params: {u'reg_alpha': 0.1}, mean: 0.88622, std: 0.02256, params: {u'reg_alpha': 1}, mean: 0.50000, std: 0.00000, params: {u'reg_alpha': 100}] {u'reg_alpha': 1e-05} 0.892355889724\n",
      "[mean: 0.88912, std: 0.02473, params: {u'n_estimators': 100, u'learning_rate': 0.0001}, mean: 0.88585, std: 0.02643, params: {u'n_estimators': 200, u'learning_rate': 0.0001}, mean: 0.88562, std: 0.02849, params: {u'n_estimators': 300, u'learning_rate': 0.0001}, mean: 0.88562, std: 0.02849, params: {u'n_estimators': 400, u'learning_rate': 0.0001}, mean: 0.88709, std: 0.02283, params: {u'n_estimators': 500, u'learning_rate': 0.0001}, mean: 0.88609, std: 0.02585, params: {u'n_estimators': 100, u'learning_rate': 0.001}, mean: 0.88296, std: 0.02589, params: {u'n_estimators': 200, u'learning_rate': 0.001}, mean: 0.88198, std: 0.02933, params: {u'n_estimators': 300, u'learning_rate': 0.001}, mean: 0.88249, std: 0.03019, params: {u'n_estimators': 400, u'learning_rate': 0.001}, mean: 0.88507, std: 0.02592, params: {u'n_estimators': 500, u'learning_rate': 0.001}, mean: 0.88418, std: 0.03262, params: {u'n_estimators': 100, u'learning_rate': 0.01}, mean: 0.89299, std: 0.02557, params: {u'n_estimators': 200, u'learning_rate': 0.01}, mean: 0.89350, std: 0.02604, params: {u'n_estimators': 300, u'learning_rate': 0.01}, mean: 0.89309, std: 0.02850, params: {u'n_estimators': 400, u'learning_rate': 0.01}, mean: 0.89077, std: 0.02474, params: {u'n_estimators': 500, u'learning_rate': 0.01}, mean: 0.89415, std: 0.02335, params: {u'n_estimators': 100, u'learning_rate': 0.1}, mean: 0.89147, std: 0.02927, params: {u'n_estimators': 200, u'learning_rate': 0.1}, mean: 0.88934, std: 0.02834, params: {u'n_estimators': 300, u'learning_rate': 0.1}, mean: 0.88685, std: 0.03411, params: {u'n_estimators': 400, u'learning_rate': 0.1}, mean: 0.88736, std: 0.03422, params: {u'n_estimators': 500, u'learning_rate': 0.1}, mean: 0.87172, std: 0.04151, params: {u'n_estimators': 100, u'learning_rate': 0.2}, mean: 0.87035, std: 0.05036, params: {u'n_estimators': 200, u'learning_rate': 0.2}, mean: 0.86835, std: 0.05531, params: {u'n_estimators': 300, u'learning_rate': 0.2}, mean: 0.86845, std: 0.05736, params: {u'n_estimators': 400, u'learning_rate': 0.2}, mean: 0.86846, std: 0.05818, params: {u'n_estimators': 500, u'learning_rate': 0.2}, mean: 0.85508, std: 0.06979, params: {u'n_estimators': 100, u'learning_rate': 0.3}, mean: 0.84973, std: 0.08109, params: {u'n_estimators': 200, u'learning_rate': 0.3}, mean: 0.85037, std: 0.08579, params: {u'n_estimators': 300, u'learning_rate': 0.3}, mean: 0.85098, std: 0.08744, params: {u'n_estimators': 400, u'learning_rate': 0.3}, mean: 0.85157, std: 0.08757, params: {u'n_estimators': 500, u'learning_rate': 0.3}] {u'n_estimators': 100, u'learning_rate': 0.1} 0.894146079484\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/angela/virtualenvs/malaria/lib/python2.7/site-packages/sklearn/model_selection/_search.py:667: DeprecationWarning: The grid_scores_ attribute was deprecated in version 0.18 in favor of the more elaborate cv_results_ attribute. The grid_scores_ attribute will not be available from 0.20\n",
      "  DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "optim_model = xgboost_tuning(x_train_s, y_train_s);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 339,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, colsample_bylevel=1, colsample_bytree=0.8,\n",
       "       gamma=0.2, learning_rate=0.1, max_delta_step=0, max_depth=5,\n",
       "       min_child_weight=1, missing=None, n_estimators=100, nthread=-1,\n",
       "       objective=u'binary:logistic', reg_alpha=1e-05, reg_lambda=1,\n",
       "       scale_pos_weight=1, seed=27, silent=True, subsample=0.8)"
      ]
     },
     "execution_count": 339,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "optim_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "optim_model =XGBClassifier(base_score=0.5, colsample_bylevel=1, colsample_bytree=0.8,\n",
    "       gamma=0.2, learning_rate=0.1, max_delta_step=0, max_depth=5,\n",
    "       min_child_weight=1, missing=None, n_estimators=100, nthread=-1,\n",
    "       objective=u'binary:logistic', reg_alpha=1e-05, reg_lambda=1,\n",
    "       scale_pos_weight=281/34, seed=27, silent=True, subsample=0.8)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The classification model is built from the set defined as training, with the optimal parameters obtained from tuning. \n",
    "Down below, the model report is shown, along with its performance on the set of compounds defined as test set. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Model Report\n",
      "Accuracy (train set): 1\n",
      "AUC Score (Train): 1.000000\n",
      "Accuracy (test set): 0.8963\n",
      "AUC Score (Test): 0.860067\n",
      "Confusion matrix:\n",
      "[[111   3]\n",
      " [ 11  10]]\n",
      "Detailed classification report:\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "        0.0       0.91      0.97      0.94       114\n",
      "        1.0       0.77      0.48      0.59        21\n",
      "\n",
      "avg / total       0.89      0.90      0.89       135\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjUAAAHBCAYAAACR/koKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10557ddd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#building the predictive model\n",
    "cv_folds=5 \n",
    "early_stopping_rounds=50\n",
    "alg = optim_model\n",
    "# loading numpy array into dMatrix\n",
    "xgb_train = xgb.DMatrix( x_train_s, label=y_train_s)\n",
    "xgb_test = xgb.DMatrix( x_test, label=y_test)\n",
    "    \n",
    "#Train cross validation\n",
    "xgb_param = alg.get_xgb_params()\n",
    "cvresult = xgb.cv(xgb_param, xgb_train, num_boost_round=alg.get_params()['n_estimators'], nfold=cv_folds,\n",
    "                      metrics = ['logloss'], early_stopping_rounds=early_stopping_rounds)\n",
    "alg.set_params(n_estimators=cvresult.shape[0])\n",
    "    \n",
    "#Fit the algorithm on the data\n",
    "alg.fit(x_train_s, y_train_s, eval_metric='auc')\n",
    "#alg.fit(x_train, y_train, eval_metric='logloss')\n",
    "    \n",
    "#Predict in training set\n",
    "y_trpred = alg.predict(x_train_s)\n",
    "y_trpredprob = alg.predict_proba(x_train_s)[:,1]\n",
    "    \n",
    "#Predict in test set\n",
    "y_pred = alg.predict(x_test)\n",
    "y_predprob = alg.predict_proba(x_test)[:,1]\n",
    "    \n",
    "#Print model report:\n",
    "print (\"\\nModel Report\")\n",
    "    \n",
    "print (\"Accuracy (train set): %.4g\" % metrics.accuracy_score(y_train_s, y_trpred))\n",
    "print (\"AUC Score (Train): %f\" % metrics.roc_auc_score(y_train_s, y_trpredprob))\n",
    "    \n",
    "print (\"Accuracy (test set): %.4g\" % metrics.accuracy_score(y_test, y_pred))\n",
    "print (\"AUC Score (Test): %f\" % metrics.roc_auc_score(y_test, y_predprob))    \n",
    "    \n",
    "#parte añadida que puede petar\n",
    "################################################################################\n",
    "print(\"Confusion matrix:\")\n",
    "print (metrics.confusion_matrix(y_test, y_pred))\n",
    "print(\"Detailed classification report:\")\n",
    "print (metrics.classification_report(y_test, y_pred))\n",
    "                    \n",
    "# Computing ROC curve\n",
    "fpr, tpr, _ = metrics.roc_curve(y_test, y_predprob)\n",
    "\n",
    "# df = pd.DataFrame(dict(fpr=fpr, tpr=tpr))\n",
    "# x = ggplot(df, aes(x='fpr', y='tpr')) + \\\n",
    "#     geom_line() + \\\n",
    "#     geom_abline(linetype='dashed') + \\\n",
    "#     scale_x_continuous(limits=(0,1)) + \\\n",
    "#     scale_y_continuous(limits=(0,1)) + \\\n",
    "#     xlab('FPR') + \\\n",
    "#     ylab('TPR') + \\\n",
    "#     theme_gray() + \\\n",
    "#     ggtitle(\"ROC Curve w/ AUC=%s\" % str(metrics.auc(fpr,tpr)))\n",
    "\n",
    "plt.figure(figsize=(9,7))\n",
    "lw = 2\n",
    "plt.plot(fpr, tpr, lw=lw)\n",
    "#         lw=lw, label='ROC curve (area = %0.2f)' % roc_auc[2])\n",
    "            \n",
    "plt.plot([0, 1], [0, 1], lw=1, linestyle='--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate', fontsize = 14)\n",
    "plt.ylabel('True Positive Rate', fontsize = 14)\n",
    "#plt.title('Receiver operating characteristic example')\n",
    "plt.title(\"ROC Curve w/ AUC=%s\" % str(metrics.auc(fpr,tpr)), fontsize = 18)\n",
    "#plt.legend(loc=\"lower right\")\n",
    "plt.savefig('roc_BUENA2.eps')\n",
    "plt.show()\n",
    "\n",
    "\n",
    "#ggsave(x, filename=\"malaria_roc.png\", dpi=600 )\n",
    "#print(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As it is shown in the previous block, both accuracy and AUC curve show the predictive power of this model.\n",
    "\n",
    "The importance of the descriptors in the xgboost model was calculated, in order to gain insight in the performance of the classificator. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1155a4ad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.savefig('roc_BUENA.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 342,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACPUAAAi7CAYAAAA0z3dbAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDzsvEzP+0b8BAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11820b890>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot feature importance\n",
    "plot_importance(alg)\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EState_VSA4\n",
      "FractionCSP3\n",
      "PEOE_VSA7\n",
      "MaxAbsEStateIndex\n",
      "BalabanJ\n",
      "EState_VSA2\n",
      "MinEStateIndex\n",
      "Chi0n\n",
      "TPSA_x\n",
      "MP\n",
      "EState_VSA8\n",
      "SMR_VSA1\n",
      "PEOE_VSA3\n",
      "Chi1n\n",
      "MolMR\n"
     ]
    }
   ],
   "source": [
    "#Creating a map of variables to identify the most relevant ones\n",
    "#Names of the variables\n",
    "vars = pyb_rdk.drop(['ION_ACTIVITY', 'SMILES', 'MOLECULES'], axis=1).columns\n",
    "mapping_vars = {}\n",
    "xgnames = [j + str(i) for i, j in zip(range(len(vars)), [\"f\"] * len(vars))]\n",
    "mapping_vars = dict(zip(xgnames, vars))\n",
    "#Most relevant descriptors\n",
    "for i in ['f34', 'f41', 'f89', 'f50', 'f15','f32', 'f56', 'f18', 'f6', 'f3', 'f38', 'f93',  'f85', 'f21', 'f59' ]:\n",
    "    print(mapping_vars[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The descriptors shown above are the 15 most relevant features for the classification model constructed.\n",
    "\n",
    "Taking this into account, we try to fit the model imposing a threshold of importance (this is, allowing only the descriptors over a certain threshold of importance to be variables of the predictive model)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Thresh=0.000, n=211, Accuracy: 89.63%\n",
      "[[111   3]\n",
      " [ 11  10]]\n",
      "Thresh=0.005, n=64, Accuracy: 89.63%\n",
      "[[109   5]\n",
      " [  9  12]]\n",
      "Thresh=0.010, n=37, Accuracy: 88.15%\n",
      "[[109   5]\n",
      " [ 11  10]]\n",
      "Thresh=0.015, n=19, Accuracy: 88.15%\n",
      "[[107   7]\n",
      " [  9  12]]\n",
      "Thresh=0.020, n=11, Accuracy: 88.89%\n",
      "[[108   6]\n",
      " [  9  12]]\n",
      "Thresh=0.025, n=7, Accuracy: 85.93%\n",
      "[[107   7]\n",
      " [ 12   9]]\n"
     ]
    }
   ],
   "source": [
    "# Fit model using each importance as a threshold\n",
    "thresholds = np.arange(0.0, 0.030, 0.005)\n",
    "for thresh in thresholds:\n",
    "    # select features using threshold\n",
    "    selection = SelectFromModel(alg, threshold=thresh, prefit=True)\n",
    "    select_X_train = selection.transform(x_train_s)\n",
    "    # train model\n",
    "    #selection_model = XGBClassifier()\n",
    "    selection_model=XGBClassifier(base_score=0.5, colsample_bylevel=1, colsample_bytree=0.8,\n",
    "       gamma=0.2, learning_rate=0.1, max_delta_step=0, max_depth=5,\n",
    "       min_child_weight=1, missing=None, n_estimators=100, nthread=-1,\n",
    "       objective=u'binary:logistic', reg_alpha=1e-05, reg_lambda=1,\n",
    "       scale_pos_weight=281/34, seed=27, silent=True, subsample=0.8)\n",
    "    selection_model.fit(select_X_train, y_train_s)\n",
    "    # eval model\n",
    "    select_X_test = selection.transform(x_test)\n",
    "    y_pred = selection_model.predict(select_X_test)\n",
    "    predictions = [round(value) for value in y_pred]\n",
    "    accuracy = accuracy_score(y_test, predictions)\n",
    "    print(\"Thresh=%.3f, n=%d, Accuracy: %.2f%%\" % (thresh, select_X_train.shape[1], accuracy*100.0))\n",
    "    print (metrics.confusion_matrix(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can check that only includying in the model the 64 (threshold = 0.005) most relevant features, we already obtain the best accuracy of our model, so we re-construct the model only from these features to avoid overfitting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Thresh=0.010, n=37, Accuracy: 88.15%\n",
      "[[109   5]\n",
      " [ 11  10]]\n"
     ]
    }
   ],
   "source": [
    "#Definitive model\n",
    "\n",
    "#thresh = 0.005\n",
    "thresh = 0.010\n",
    "# select features using threshold\n",
    "selection = SelectFromModel(alg, threshold=thresh, prefit=True)\n",
    "select_X_train = selection.transform(x_train_s)\n",
    "# train model\n",
    "selection_model = XGBClassifier(base_score=0.5, colsample_bylevel=1, colsample_bytree=0.8,\n",
    "       gamma=0.2, learning_rate=0.1, max_delta_step=0, max_depth=5,\n",
    "       min_child_weight=1, missing=None, n_estimators=100, nthread=-1,\n",
    "       objective=u'binary:logistic', reg_alpha=1e-05, reg_lambda=1,\n",
    "       scale_pos_weight=281/34, seed=27, silent=True, subsample=0.8)\n",
    "selection_model.fit(select_X_train, y_train_s)\n",
    "# eval model\n",
    "select_X_test = selection.transform(x_test)\n",
    "y_pred = selection_model.predict(select_X_test)\n",
    "predictions = [round(value) for value in y_pred]\n",
    "accuracy = accuracy_score(y_test, predictions)\n",
    "print(\"Thresh=%.3f, n=%d, Accuracy: %.2f%%\" % (thresh, select_X_train.shape[1], accuracy*100.0))\n",
    "print (metrics.confusion_matrix(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prediction on the set of marine molecules\n",
    "\n",
    "Now the predictive model will be apply to predict the activity against malaria of the 27 marine molecules provided."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NUMBER</th>\n",
       "      <th>S</th>\n",
       "      <th>SMI</th>\n",
       "      <th>SEQUENCE</th>\n",
       "      <th>SMILES</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>#1</td>\n",
       "      <td>S</td>\n",
       "      <td>3,5-dibromo-4-hydroxybenzaldehyde.smi</td>\n",
       "      <td>SXRHGLQCOLNZPT-UHFFFAOYSA-N</td>\n",
       "      <td>O=Cc1cc(Br)c(O)c(Br)c1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>#2</td>\n",
       "      <td>S</td>\n",
       "      <td>Alcyopterosin-A.smi</td>\n",
       "      <td>AOZZRIFMYYWXLE-UHFFFAOYSA-N</td>\n",
       "      <td>Cc1cc2c(c(C)c1CCCl)CC(C)(C)C2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>#3</td>\n",
       "      <td>S</td>\n",
       "      <td>Antarcticoside-D.smi</td>\n",
       "      <td>SFEIVGCFQJAPTI-DCOKEWDMSA-N</td>\n",
       "      <td>COC1C(O)COC(OC[C@@H](C)C/C=C/C(C)[C@H]2C[C@@H]...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>#4</td>\n",
       "      <td>S</td>\n",
       "      <td>Aplicyanin-A.smi</td>\n",
       "      <td>VVONRGQMZKFFRP-UHFFFAOYSA-N</td>\n",
       "      <td>N=C1NCCC(c2c[nH]c3ccc(Br)cc23)N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>#5</td>\n",
       "      <td>S</td>\n",
       "      <td>Asterasterol-C.smi</td>\n",
       "      <td>KDUVNRKZGYTTJS-GACDVITBSA-N</td>\n",
       "      <td>CC(C/C=C/[C@@H](C)[C@H]1CC[C@H]2C3=CC(=O)[C@H]...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>#7</td>\n",
       "      <td>S</td>\n",
       "      <td>Austrodoral.smi</td>\n",
       "      <td>CNTXEQHESHSVFZ-UBHSHLNASA-N</td>\n",
       "      <td>CC1(C)CCC[C@@]2(C)[C@H]1CC[C@@]2(C)C=O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>#9</td>\n",
       "      <td>S</td>\n",
       "      <td>Caminatal.smi</td>\n",
       "      <td>JXMWXFSVMMJDHJ-IKERNBKFSA-N</td>\n",
       "      <td>CC(=O)O[C@@H]1C[C@@](C)(C=O)[C@H](CCC(=O)c2ccc...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>#10</td>\n",
       "      <td>S</td>\n",
       "      <td>Dendrillin.smi</td>\n",
       "      <td>APHPHCFXAWJFAA-AOGFDJAGSA-N</td>\n",
       "      <td>CC(=O)OC1OC(OC(C)=O)C2C1CC[C@H](C1(C)CCCC(C)(C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>#11</td>\n",
       "      <td>S</td>\n",
       "      <td>Dendrinolide.smi</td>\n",
       "      <td>IMIAJJXGLPJMII-CEXWTWQISA-N</td>\n",
       "      <td>CC(=O)O[C@H]1OC=C2CC[C@H](/C(C)=C3/CCCC3(C)C)C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>#12</td>\n",
       "      <td>S</td>\n",
       "      <td>Discorhabdin-B.smi</td>\n",
       "      <td>SVKKMXJIWIOCJC-VOJFVSQTSA-N</td>\n",
       "      <td>O=C1C=C2S[C@H]3C[C@]2(C=C1Br)C1=C(N3)C(=O)c2[n...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>#13</td>\n",
       "      <td>S</td>\n",
       "      <td>Flabellatene-A.smi</td>\n",
       "      <td>IVLQHBCLTSYTPN-BEFAXECRSA-N</td>\n",
       "      <td>C/C1=C/C=C(\\C(C)C)C(=O)[C@@H](O)[C@H](C)CCC(=O...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>#14</td>\n",
       "      <td>S</td>\n",
       "      <td>Focardin.smi</td>\n",
       "      <td>CHAVWXSUVDIUSK-VELITEBHSA-N</td>\n",
       "      <td>CC(C)=CC/C=C(/C)[C@H]1CCC2CCC=C3CO[C@H](O)C3[C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>#15</td>\n",
       "      <td>S</td>\n",
       "      <td>Fucostanol.smi</td>\n",
       "      <td>LGJMUZUPVCAVPU-HRJGVYIJSA-N</td>\n",
       "      <td>CC[C@H](CC[C@@H](C)[C@H]1CC[C@H]2[C@@H]3CC[C@H...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>#16</td>\n",
       "      <td>S</td>\n",
       "      <td>Hodgsonal.smi</td>\n",
       "      <td>FGOJRPSZZSBOHD-XPYUAMARSA-N</td>\n",
       "      <td>CC(=O)O[C@H]1CC(C)(C)[C@@H]2CC=C(C(C)=O)[C@H](...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>#18</td>\n",
       "      <td>S</td>\n",
       "      <td>Membranolide.smi</td>\n",
       "      <td>YWRLUOURIMRHED-ASSNKEHSSA-N</td>\n",
       "      <td>COC(=O)[C@H](C)c1c([C@@]2(C)CCCC(C)(C)C2)ccc2c...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>#19</td>\n",
       "      <td>S</td>\n",
       "      <td>Meridianin-A.smi</td>\n",
       "      <td>SZOKTSDFOPVLFD-UHFFFAOYSA-N</td>\n",
       "      <td>Nc1nccc(-c2c[nH]c3cccc(O)c23)n1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>#20</td>\n",
       "      <td>S</td>\n",
       "      <td>Palmerolide-A.smi</td>\n",
       "      <td>HEOKDDVDVGNHMR-ABVIYVDXSA-N</td>\n",
       "      <td>CC(C)=CC(=O)N/C=C/C(C)=C/[C@@H](C)[C@H]1C/C(C)...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>#22</td>\n",
       "      <td>S</td>\n",
       "      <td>Polyrhaphin-A.smi</td>\n",
       "      <td>QFFIZAQCLHVQCV-DIGFTJECSA-N</td>\n",
       "      <td>C=C1CCCC(C)(C)[C@@H]2CC[C@@](C)([C@H](COC(C)=O...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>#23</td>\n",
       "      <td>S</td>\n",
       "      <td>Psammopemmin-A.smi</td>\n",
       "      <td>CXLPBBKWDIYIPK-UHFFFAOYSA-N</td>\n",
       "      <td>Nc1nc(Br)ncc1-c1c[nH]c2cccc(O)c12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>#24</td>\n",
       "      <td>S</td>\n",
       "      <td>Pteroenone.smi</td>\n",
       "      <td>DVSSWRPHWSGOFM-GXTWGEPZSA-N</td>\n",
       "      <td>C/C=C(C)/C=C(\\C)[C@@H](O)[C@@H](C)C(=O)CCC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>#25</td>\n",
       "      <td>S</td>\n",
       "      <td>Rossinone-A.smi</td>\n",
       "      <td>RLXJEEFFCLVVPH-HXUWFJFHSA-N</td>\n",
       "      <td>CC(C)=C[C@@H](O)C(=O)/C(C)=C/CC/C(C)=C/Cc1cc(O...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>#26</td>\n",
       "      <td>S</td>\n",
       "      <td>Spongothymidine.smi</td>\n",
       "      <td>DWRXFEITVBNRMK-JAGXHNFQSA-N</td>\n",
       "      <td>Cc1cn([C@@H]2O[C@H](CO)[C@@H](O)[C@@H]2O)c(=O)...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>#27</td>\n",
       "      <td>S</td>\n",
       "      <td>Spongouridine.smi</td>\n",
       "      <td>DRTQHJPVMGBUCF-CCXZUQQUSA-N</td>\n",
       "      <td>O=c1ccn([C@@H]2O[C@H](CO)[C@@H](O)[C@@H]2O)c(=...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>#28</td>\n",
       "      <td>S</td>\n",
       "      <td>Suberiphenol.smi</td>\n",
       "      <td>CCXWTNLPPHXPFG-IWRPRKENSA-N</td>\n",
       "      <td>CC(=O)O[C@@H]1C[C@@]2(C)CC(c3ccc(C)c(O)c3)=CC[...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>#29</td>\n",
       "      <td>S</td>\n",
       "      <td>Suberitenone-A.smi</td>\n",
       "      <td>JANWLMMGERLQDI-MMKNOJKJSA-N</td>\n",
       "      <td>CC(=O)O[C@@H]1C[C@@]2(C)C=C([C@H]3CC(=O)C(C)=C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>#30</td>\n",
       "      <td>S</td>\n",
       "      <td>Thelepin.smi</td>\n",
       "      <td>LNKWAVLAOSVMOE-UHFFFAOYSA-N</td>\n",
       "      <td>O=C1C(Br)=CC2(C=C1Br)Cc1cc(CO)cc(Br)c1O2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>#31</td>\n",
       "      <td>S</td>\n",
       "      <td>Variolin-A.smi</td>\n",
       "      <td>DVFGBCFKUZPOMS-UHFFFAOYSA-O</td>\n",
       "      <td>C[n+]1cc(O)c(-c2c3c(O)ccnc3n3c(N)nccc23)nc1N</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   NUMBER  S                                    SMI  \\\n",
       "0      #1  S  3,5-dibromo-4-hydroxybenzaldehyde.smi   \n",
       "1      #2  S                    Alcyopterosin-A.smi   \n",
       "2      #3  S                   Antarcticoside-D.smi   \n",
       "3      #4  S                       Aplicyanin-A.smi   \n",
       "4      #5  S                     Asterasterol-C.smi   \n",
       "5      #7  S                        Austrodoral.smi   \n",
       "6      #9  S                          Caminatal.smi   \n",
       "7     #10  S                         Dendrillin.smi   \n",
       "8     #11  S                       Dendrinolide.smi   \n",
       "9     #12  S                     Discorhabdin-B.smi   \n",
       "10    #13  S                     Flabellatene-A.smi   \n",
       "11    #14  S                           Focardin.smi   \n",
       "12    #15  S                         Fucostanol.smi   \n",
       "13    #16  S                          Hodgsonal.smi   \n",
       "14    #18  S                       Membranolide.smi   \n",
       "15    #19  S                       Meridianin-A.smi   \n",
       "16    #20  S                      Palmerolide-A.smi   \n",
       "17    #22  S                      Polyrhaphin-A.smi   \n",
       "18    #23  S                     Psammopemmin-A.smi   \n",
       "19    #24  S                         Pteroenone.smi   \n",
       "20    #25  S                        Rossinone-A.smi   \n",
       "21    #26  S                    Spongothymidine.smi   \n",
       "22    #27  S                      Spongouridine.smi   \n",
       "23    #28  S                       Suberiphenol.smi   \n",
       "24    #29  S                     Suberitenone-A.smi   \n",
       "25    #30  S                           Thelepin.smi   \n",
       "26    #31  S                         Variolin-A.smi   \n",
       "\n",
       "                       SEQUENCE  \\\n",
       "0   SXRHGLQCOLNZPT-UHFFFAOYSA-N   \n",
       "1   AOZZRIFMYYWXLE-UHFFFAOYSA-N   \n",
       "2   SFEIVGCFQJAPTI-DCOKEWDMSA-N   \n",
       "3   VVONRGQMZKFFRP-UHFFFAOYSA-N   \n",
       "4   KDUVNRKZGYTTJS-GACDVITBSA-N   \n",
       "5   CNTXEQHESHSVFZ-UBHSHLNASA-N   \n",
       "6   JXMWXFSVMMJDHJ-IKERNBKFSA-N   \n",
       "7   APHPHCFXAWJFAA-AOGFDJAGSA-N   \n",
       "8   IMIAJJXGLPJMII-CEXWTWQISA-N   \n",
       "9   SVKKMXJIWIOCJC-VOJFVSQTSA-N   \n",
       "10  IVLQHBCLTSYTPN-BEFAXECRSA-N   \n",
       "11  CHAVWXSUVDIUSK-VELITEBHSA-N   \n",
       "12  LGJMUZUPVCAVPU-HRJGVYIJSA-N   \n",
       "13  FGOJRPSZZSBOHD-XPYUAMARSA-N   \n",
       "14  YWRLUOURIMRHED-ASSNKEHSSA-N   \n",
       "15  SZOKTSDFOPVLFD-UHFFFAOYSA-N   \n",
       "16  HEOKDDVDVGNHMR-ABVIYVDXSA-N   \n",
       "17  QFFIZAQCLHVQCV-DIGFTJECSA-N   \n",
       "18  CXLPBBKWDIYIPK-UHFFFAOYSA-N   \n",
       "19  DVSSWRPHWSGOFM-GXTWGEPZSA-N   \n",
       "20  RLXJEEFFCLVVPH-HXUWFJFHSA-N   \n",
       "21  DWRXFEITVBNRMK-JAGXHNFQSA-N   \n",
       "22  DRTQHJPVMGBUCF-CCXZUQQUSA-N   \n",
       "23  CCXWTNLPPHXPFG-IWRPRKENSA-N   \n",
       "24  JANWLMMGERLQDI-MMKNOJKJSA-N   \n",
       "25  LNKWAVLAOSVMOE-UHFFFAOYSA-N   \n",
       "26  DVFGBCFKUZPOMS-UHFFFAOYSA-O   \n",
       "\n",
       "                                               SMILES  \n",
       "0                              O=Cc1cc(Br)c(O)c(Br)c1  \n",
       "1                       Cc1cc2c(c(C)c1CCCl)CC(C)(C)C2  \n",
       "2   COC1C(O)COC(OC[C@@H](C)C/C=C/C(C)[C@H]2C[C@@H]...  \n",
       "3                    N=C1NCCC(c2c[nH]c3ccc(Br)cc23)N1  \n",
       "4   CC(C/C=C/[C@@H](C)[C@H]1CC[C@H]2C3=CC(=O)[C@H]...  \n",
       "5              CC1(C)CCC[C@@]2(C)[C@H]1CC[C@@]2(C)C=O  \n",
       "6   CC(=O)O[C@@H]1C[C@@](C)(C=O)[C@H](CCC(=O)c2ccc...  \n",
       "7   CC(=O)OC1OC(OC(C)=O)C2C1CC[C@H](C1(C)CCCC(C)(C...  \n",
       "8   CC(=O)O[C@H]1OC=C2CC[C@H](/C(C)=C3/CCCC3(C)C)C...  \n",
       "9   O=C1C=C2S[C@H]3C[C@]2(C=C1Br)C1=C(N3)C(=O)c2[n...  \n",
       "10  C/C1=C/C=C(\\C(C)C)C(=O)[C@@H](O)[C@H](C)CCC(=O...  \n",
       "11  CC(C)=CC/C=C(/C)[C@H]1CCC2CCC=C3CO[C@H](O)C3[C...  \n",
       "12  CC[C@H](CC[C@@H](C)[C@H]1CC[C@H]2[C@@H]3CC[C@H...  \n",
       "13  CC(=O)O[C@H]1CC(C)(C)[C@@H]2CC=C(C(C)=O)[C@H](...  \n",
       "14  COC(=O)[C@H](C)c1c([C@@]2(C)CCCC(C)(C)C2)ccc2c...  \n",
       "15                    Nc1nccc(-c2c[nH]c3cccc(O)c23)n1  \n",
       "16  CC(C)=CC(=O)N/C=C/C(C)=C/[C@@H](C)[C@H]1C/C(C)...  \n",
       "17  C=C1CCCC(C)(C)[C@@H]2CC[C@@](C)([C@H](COC(C)=O...  \n",
       "18                  Nc1nc(Br)ncc1-c1c[nH]c2cccc(O)c12  \n",
       "19         C/C=C(C)/C=C(\\C)[C@@H](O)[C@@H](C)C(=O)CCC  \n",
       "20  CC(C)=C[C@@H](O)C(=O)/C(C)=C/CC/C(C)=C/Cc1cc(O...  \n",
       "21  Cc1cn([C@@H]2O[C@H](CO)[C@@H](O)[C@@H]2O)c(=O)...  \n",
       "22  O=c1ccn([C@@H]2O[C@H](CO)[C@@H](O)[C@@H]2O)c(=...  \n",
       "23  CC(=O)O[C@@H]1C[C@@]2(C)CC(c3ccc(C)c(O)c3)=CC[...  \n",
       "24  CC(=O)O[C@@H]1C[C@@]2(C)C=C([C@H]3CC(=O)C(C)=C...  \n",
       "25           O=C1C(Br)=CC2(C=C1Br)Cc1cc(CO)cc(Br)c1O2  \n",
       "26       C[n+]1cc(O)c(-c2c3c(O)ccnc3n3c(N)nccc23)nc1N  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Loading marine molecules list\n",
    "mm_list = pd.read_csv(\"marine_molecules.csv\", index_col=None, decimal=\",\", na_values=\"ND\")\n",
    "mm_list"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As with the OSM dataset, RDKit and Pybel descriptors are obtained."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 27 entries, 0 to 26\n",
      "Columns: 198 entries, BalabanJ to fr_urea\n",
      "dtypes: float64(92), int64(104), object(2)\n",
      "memory usage: 41.8+ KB\n"
     ]
    }
   ],
   "source": [
    "# Getting RDKit descriptors\n",
    "descs = [x[0] for x in Chem.Descriptors.descList]\n",
    "col_r = ['SMILES'] + descs\n",
    "rddf = pd.DataFrame(columns=col_r, index=mm_list.index.values)\n",
    "rddf = rddf.fillna(0)\n",
    "rddf.loc[:, \"SMILES\"] = mm_list.loc[:, \"SMILES\"]\n",
    "mrdk_list = smiles_to_rdkit(rddf.loc[:, \"SMILES\"])\n",
    "rddf = rddf.assign(MOLECULES=mrdk_list)\n",
    "list_descriptors = []\n",
    "mdc = MolecularDescriptorCalculator(descs)\n",
    "for index, row in rddf.iterrows():\n",
    "    results = mdc.CalcDescriptors(row[\"MOLECULES\"])\n",
    "    descriptors = dict(zip(descs, results))\n",
    "    descriptors.update({'SMILES': row[\"SMILES\"], 'MOLECULES': row[\"MOLECULES\"]})\n",
    "    list_descriptors.append(descriptors)\n",
    "\n",
    "rdkit_df_mm = pd.DataFrame(list_descriptors)\n",
    "rdkit_df_mm.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 27 entries, 0 to 26\n",
      "Data columns (total 16 columns):\n",
      "HBA1      27 non-null float64\n",
      "HBA2      27 non-null float64\n",
      "HBD       27 non-null float64\n",
      "MP        27 non-null float64\n",
      "MR        27 non-null float64\n",
      "MW        27 non-null float64\n",
      "SMILES    27 non-null object\n",
      "TPSA      27 non-null float64\n",
      "abonds    27 non-null float64\n",
      "atoms     27 non-null float64\n",
      "bonds     27 non-null float64\n",
      "dbonds    27 non-null float64\n",
      "logP      27 non-null float64\n",
      "nF        27 non-null float64\n",
      "sbonds    27 non-null float64\n",
      "tbonds    27 non-null float64\n",
      "dtypes: float64(15), object(1)\n",
      "memory usage: 3.4+ KB\n"
     ]
    }
   ],
   "source": [
    "#Getting Pybel descriptors\n",
    "descs = pybel.descs\n",
    "col_r = ['SMILES']+ descs\n",
    "pydf = pd.DataFrame(columns=col_r, index=mm_list.index.values)\n",
    "pydf = pydf.fillna(0)\n",
    "pydf.loc[:, \"SMILES\"] = mm_list.loc[:, \"SMILES\"]\n",
    "mpy_list = smiles_to_pybel(pydf.loc[:, \"SMILES\"])\n",
    "pydf = pydf.assign(MOLECULES=mpy_list)\n",
    "list_descriptors = []\n",
    "for index, row in pydf.iterrows():\n",
    "    results = row[\"MOLECULES\"].calcdesc()\n",
    "    results.update({'SMILES': row[\"SMILES\"],'MOLECULES': row[\"MOLECULES\"]})\n",
    "    list_descriptors.append(results)\n",
    "\n",
    "pybel_df_mm = pd.DataFrame(list_descriptors)\n",
    "pybel_df_mm = pybel_df_mm.dropna(axis=1, how = 'all')\n",
    "pybel_df_mm = pybel_df_mm.drop(['MOLECULES'], axis=1)\n",
    "\n",
    "pybel_df_mm.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 27 entries, 0 to 26\n",
      "Columns: 213 entries, BalabanJ to tbonds\n",
      "dtypes: float64(107), int64(104), object(2)\n",
      "memory usage: 45.1+ KB\n"
     ]
    }
   ],
   "source": [
    "pyb_rdk_mm = pd.merge(rdkit_df_mm, pybel_df_mm, how='inner', on='SMILES')\n",
    "# Checking duplicated rows\n",
    "pyb_rdk_mm = pyb_rdk_mm.drop_duplicates()\n",
    "pyb_rdk_mm.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Converting descriptors dataframe to array and pre-processing\n",
    "mm_smiles = pyb_rdk_mm.loc[:,\"SMILES\"]\n",
    "mm_array = pyb_rdk_mm.drop(['SMILES', 'MOLECULES'], axis=1).values\n",
    "# Scaling data\n",
    "mm_array = preprocessing.scale(mm_array)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We apply the model to predict marine molecules activity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    27\n",
       "dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Predicting marine molecules active against malaria from feature selection model\n",
    "#Removing descriptors under the importance threshold\n",
    "select_X_mm = selection.transform(mm_array)\n",
    "\n",
    "y_pred_mm = selection_model.predict(select_X_mm)\n",
    "mm_pred_df = pd.Series(data=y_pred_mm)\n",
    "y_predprob_mm = selection_model.predict_proba(select_X_mm)[:,1]\n",
    "mm_predprob_df = pd.Series(data=y_predprob_mm)\n",
    "\n",
    "#Preparing results\n",
    "mm_results = pd.concat([mm_list, mm_pred_df, mm_predprob_df], axis=1)\n",
    "mm_results.columns = ['NUMBER', '#', 'NAME', 'INCHI_KEY','SMILES', 'ACTIVITY', 'ACTIVITY_PROBABILITY']\n",
    "mm_pred_df.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "None of the marine molecules seems to be active against the parasite ion's pump."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "mm_results.sort_values('ACTIVITY_PROBABILITY', ascending=False).to_csv(\"mm_activity_sorted2.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prediction on the set of DrugBank database\n",
    "\n",
    "Now the predictive model will be apply to predict the activity against malaria of the 7176 molecules available in the Drugbank Database. Data was filtered, removing those molecules whose molecular properties were beyond the standard limits, and those whose SMILES field referenced to a related molecule (_RelatedMol:_ name of a molecule), remaining 6537 molecules."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 6537 entries, 0 to 6536\n",
      "Data columns (total 5 columns):\n",
      "NUMBER      6537 non-null object\n",
      "S           6537 non-null object\n",
      "NAME        6537 non-null object\n",
      "SEQUENCE    6537 non-null object\n",
      "SMILES      6537 non-null object\n",
      "dtypes: object(5)\n",
      "memory usage: 255.4+ KB\n"
     ]
    }
   ],
   "source": [
    "drugbank_list = pd.read_csv(\"DrugBank_list.csv\", index_col=None, decimal=\",\", na_values=\"ND\", sep=\",\")\n",
    "#Dropping NAs from the list (not valid molecules)\n",
    "drugbank_mod = drugbank_list.dropna()\n",
    "drugbank_mod2 = drugbank_mod[~drugbank_mod['SMILES'].str.contains('RelatedMol:')]\n",
    "drugbank_mod2 = drugbank_mod2.reset_index(drop=True)\n",
    "drugbank_mod2.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "RDKit and Pybel descriptors of the molecules are computed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 6537 entries, 0 to 6536\n",
      "Columns: 198 entries, BalabanJ to fr_urea\n",
      "dtypes: float64(92), int64(104), object(2)\n",
      "memory usage: 9.9+ MB\n"
     ]
    }
   ],
   "source": [
    "# Getting RDKit descriptors\n",
    "descs = [x[0] for x in Chem.Descriptors.descList]\n",
    "col_r = ['SMILES'] + descs\n",
    "rddf = pd.DataFrame(columns=col_r, index=drugbank_mod2.index.values)\n",
    "rddf = rddf.fillna(0)\n",
    "rddf.loc[:, \"SMILES\"] = drugbank_mod2.loc[:, \"SMILES\"]\n",
    "mrdk_list = smiles_to_rdkit(rddf.loc[:, \"SMILES\"])\n",
    "rddf = rddf.assign(MOLECULES=mrdk_list)\n",
    "list_descriptors = []\n",
    "mdc = MolecularDescriptorCalculator(descs)\n",
    "for index, row in rddf.iterrows():\n",
    "    #print(\"Processing molecule\", row[\"SMILES\"])\n",
    "    results = mdc.CalcDescriptors(row[\"MOLECULES\"])\n",
    "    descriptors = dict(zip(descs, results))\n",
    "    descriptors.update({'SMILES': row[\"SMILES\"], 'MOLECULES': row[\"MOLECULES\"]})\n",
    "    list_descriptors.append(descriptors)\n",
    "\n",
    "rdkit_df_db = pd.DataFrame(list_descriptors)\n",
    "rdkit_df_db.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 6537 entries, 0 to 6536\n",
      "Data columns (total 16 columns):\n",
      "HBA1      6537 non-null float64\n",
      "HBA2      6537 non-null float64\n",
      "HBD       6537 non-null float64\n",
      "MP        6537 non-null float64\n",
      "MR        6537 non-null float64\n",
      "MW        6537 non-null float64\n",
      "SMILES    6537 non-null object\n",
      "TPSA      6537 non-null float64\n",
      "abonds    6537 non-null float64\n",
      "atoms     6537 non-null float64\n",
      "bonds     6537 non-null float64\n",
      "dbonds    6537 non-null float64\n",
      "logP      6537 non-null float64\n",
      "nF        6537 non-null float64\n",
      "sbonds    6537 non-null float64\n",
      "tbonds    6537 non-null float64\n",
      "dtypes: float64(15), object(1)\n",
      "memory usage: 817.2+ KB\n"
     ]
    }
   ],
   "source": [
    "#Getting Pybel descriptors\n",
    "descs = pybel.descs\n",
    "col_r = ['SMILES']+ descs\n",
    "pydf = pd.DataFrame(columns=col_r, index=drugbank_mod2.index.values)\n",
    "pydf = pydf.fillna(0)\n",
    "pydf.loc[:, \"SMILES\"] = drugbank_mod2.loc[:, \"SMILES\"]\n",
    "mpy_list = smiles_to_pybel(pydf.loc[:, \"SMILES\"])\n",
    "pydf = pydf.assign(MOLECULES=mpy_list)\n",
    "list_descriptors = []\n",
    "for index, row in pydf.iterrows():\n",
    "    #print(\"Processing molecule\", row[\"SMILES\"])\n",
    "    results = row[\"MOLECULES\"].calcdesc()\n",
    "    results.update({'SMILES': row[\"SMILES\"],'MOLECULES': row[\"MOLECULES\"]})\n",
    "    list_descriptors.append(results)\n",
    "\n",
    "pybel_df_db = pd.DataFrame(list_descriptors)\n",
    "pybel_df_db = pybel_df_db.dropna(axis=1, how = 'all')\n",
    "pybel_df_db = pybel_df_db.drop(['MOLECULES'], axis=1)\n",
    "pybel_df_db.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 6534 entries, 0 to 6533\n",
      "Columns: 213 entries, BalabanJ to tbonds\n",
      "dtypes: float64(107), int64(104), object(2)\n",
      "memory usage: 10.6+ MB\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 6534 entries, 0 to 6533\n",
      "Data columns (total 5 columns):\n",
      "NUMBER      6534 non-null object\n",
      "S           6534 non-null object\n",
      "NAME        6534 non-null object\n",
      "SEQUENCE    6534 non-null object\n",
      "SMILES      6534 non-null object\n",
      "dtypes: object(5)\n",
      "memory usage: 255.3+ KB\n"
     ]
    }
   ],
   "source": [
    "pyb_rdk_db = pd.merge(rdkit_df_db, pybel_df_db, how='inner', on='SMILES')\n",
    "# Checking duplicated rows\n",
    "pyb_rdk_db = pyb_rdk_db.drop_duplicates()\n",
    "\n",
    "\n",
    "#Check NaN rows and remove\n",
    "nan_rows = pyb_rdk_db.index[pyb_rdk_db.isnull().any(axis=1)]\n",
    "pyb_rdk_db_post = pyb_rdk_db[~pyb_rdk_db.isnull().any(axis=1)]\n",
    "pyb_rdk_db_post = pyb_rdk_db_post.reset_index(drop=True)\n",
    "\n",
    "#Removing these rows in the original list\n",
    "drugbank_mod3 = drugbank_mod2[~pyb_rdk_db.isnull().any(axis=1)]\n",
    "drugbank_mod3 = drugbank_mod3.reset_index(drop=True)\n",
    "pyb_rdk_db_post.info()\n",
    "drugbank_mod3.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Further pre-processing of the data is performed, in order to get rid of NaNs of Infinite values. 6534 molecules remained after this process."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Converting descriptors dataframe to array and pre-processing\n",
    "drugbank_smiles = pyb_rdk_db_post.loc[:,\"SMILES\"]\n",
    "drugbank_array = pyb_rdk_db_post.drop(['SMILES', 'MOLECULES'], axis=1).values\n",
    "\n",
    "# Scaling data\n",
    "drugbank_a_pre = preprocessing.scale(drugbank_array)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we predict the activity against malaria of the DrugBank molecules list from the model built after xgboost-based feature selection. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    6403\n",
       "1.0     131\n",
       "dtype: int64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Predicting DrugBank molecules active against malaria with feature selected model\n",
    "select_drugbank = selection.transform(drugbank_a_pre)\n",
    "y_pred_db = selection_model.predict(select_drugbank)\n",
    "\n",
    "y_predprob_db = selection_model.predict_proba(select_drugbank)[:,1]\n",
    "db_pred_df = pd.Series(data=y_pred_db)\n",
    "db_predprob_df = pd.Series(data=y_predprob_db)\n",
    "\n",
    "drugbank_results = pd.concat([drugbank_mod3, db_pred_df, db_predprob_df], axis=1)\n",
    "drugbank_results.columns = ['NUMBER', 'S', 'NAME', 'INCHI_KEY','SMILES', 'ACTIVITY_LABEL', 'ACTIVITY_PROBABILITY']\n",
    "#drugbank_results.to_csv(\"drugbank_activity.csv\")\n",
    "db_pred_df.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "96 out of the 6534 molecules available from DrugBank are pointed as active against the PfATP4. The csv with the results is attached. \n",
    "\n",
    "The algorithm has labeled molecules with a probability > 0.5 as ACTIVE (1.0) and with probability < 0.5 as INACTIVE (0.0). Therefore, in the file molecules are sorted in descending order of activity probability, being those in the top of the list the most probable to be active against PfATP4."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "drugbank_results.sort_values('ACTIVITY_PROBABILITY', ascending=False).to_csv(\"drugbank_activity_sorted2.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "malaria",
   "language": "python",
   "name": "malaria"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}