Model_evalutation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-15T14:14:19.988037600Z",
     "start_time": "2023-05-15T14:14:19.737088900Z"
    },
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package punkt to\n",
      "[nltk_data]     /Users/geremiapompei/nltk_data...\n",
      "[nltk_data]   Package punkt is already up-to-date!\n",
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     /Users/geremiapompei/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    }
   ],
   "source": [
    "from Utils import normalize_meshId,preprocess_text\n",
    "from pandas import read_csv\n",
    "from Data_retrieval import load_dataset\n",
    "from Evalutation import graphType, model_evaluation, create_knowledge_graph,find_diseases_precisionz, pValues_DDAs\n",
    "from collections import Counter\n",
    "from numpy import array\n",
    "\n",
    "from itertools import product\n",
    "from tqdm.notebook import tqdm\n",
    "tqdm.pandas()\n",
    "\n",
    "%reload_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-15T14:14:24.584943100Z",
     "start_time": "2023-05-15T14:14:21.190563700Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df_texts, df_entities = load_dataset()\n",
    "\n",
    "df_entities = df_entities[~df_entities['prob'].isna()]\n",
    "df_entities = df_entities[df_entities['prob'] > 0.8]\n",
    "df_entities = df_entities[['id', 'pmid', 'mention', 'obj', 'span_begin', 'span_end']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "## Test set creation\n",
    "\n",
    "To test the knowledge graph is used the SCMFDD dataset, there are three kinds of files: diseases, drugs and diseases-drugs associations.\n",
    "\n",
    "The dataframe `diseases-drug` associations contains:\n",
    "- drug_id: id of drug\n",
    "- drug_name: plain name of drug\n",
    "- disease_id: id of disease\n",
    "- disease_name: plain name of disease"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-15T14:14:24.892352Z",
     "start_time": "2023-05-15T14:14:24.584943100Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "SCMFF_DIR = 'dataset/SCMFDD/SCMFDD-L/'\n",
    "\n",
    "scmff_dda = read_csv(f'{SCMFF_DIR}drug-disease association.csv')\n",
    "scmff_drugs = read_csv(f'{SCMFF_DIR}drug.csv')\n",
    "scmff_diseases = read_csv(f'{SCMFF_DIR}disease.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "Add column `source` that is `true` for each disease of `df_entities` that is also inside the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-15T14:18:53.144650300Z",
     "start_time": "2023-05-15T14:18:52.676470400Z"
    },
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "64555"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scmff_diseases_labels = scmff_diseases['disease_id'].unique()\n",
    "union_diseases = list(filter(lambda x: normalize_meshId(x) in scmff_diseases_labels,\n",
    "                             df_entities[df_entities['obj'] == 'disease']['id'].unique()))\n",
    "\n",
    "df_entities['source'] = (df_entities['obj'] == 'disease') & (df_entities['id'].isin(union_diseases))\n",
    "\n",
    "sum(df_entities['source'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-15T14:18:54.456457100Z",
     "start_time": "2023-05-15T14:18:54.156759Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "scmff_dda['drug_id'] = scmff_dda['drug_id'].apply(normalize_meshId)\n",
    "scmff_dda['disease_id'] = scmff_dda['disease_id'].apply(normalize_meshId)\n",
    "\n",
    "normalized_union_diseases = map(normalize_meshId, union_diseases)\n",
    "scmff_dda = scmff_dda[scmff_dda['disease_id'].isin(normalized_union_diseases)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-15T14:18:56.715788900Z",
     "start_time": "2023-05-15T14:18:56.110356800Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "test_set = {\n",
    "    (\n",
    "        v['drug_id'],\n",
    "        v['disease_id']\n",
    "    ): True for v in scmff_dda[['drug_id', 'disease_id']].to_dict(orient='records')\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-15T14:18:57.491632500Z",
     "start_time": "2023-05-15T14:18:57.226128100Z"
    },
    "collapsed": false
   },
   "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>drug_id</th>\n",
       "      <th>drug_name</th>\n",
       "      <th>disease_id</th>\n",
       "      <th>disease_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C074619</td>\n",
       "      <td>bivalirudin</td>\n",
       "      <td>D006470</td>\n",
       "      <td>Hemorrhage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C074619</td>\n",
       "      <td>bivalirudin</td>\n",
       "      <td>D007249</td>\n",
       "      <td>Inflammation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>C074619</td>\n",
       "      <td>bivalirudin</td>\n",
       "      <td>D007511</td>\n",
       "      <td>Ischemia</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C074619</td>\n",
       "      <td>bivalirudin</td>\n",
       "      <td>D013921</td>\n",
       "      <td>Thrombocytopenia</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>C074619</td>\n",
       "      <td>bivalirudin</td>\n",
       "      <td>D013927</td>\n",
       "      <td>Thrombosis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49212</th>\n",
       "      <td>D013747</td>\n",
       "      <td>Tetrabenazine</td>\n",
       "      <td>D017109</td>\n",
       "      <td>Akathisia, Drug-Induced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49213</th>\n",
       "      <td>D013747</td>\n",
       "      <td>Tetrabenazine</td>\n",
       "      <td>D018476</td>\n",
       "      <td>Hypokinesia</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49214</th>\n",
       "      <td>D013747</td>\n",
       "      <td>Tetrabenazine</td>\n",
       "      <td>D020734</td>\n",
       "      <td>Parkinsonian Disorders</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49215</th>\n",
       "      <td>D013747</td>\n",
       "      <td>Tetrabenazine</td>\n",
       "      <td>D020820</td>\n",
       "      <td>Dyskinesias</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49216</th>\n",
       "      <td>D013747</td>\n",
       "      <td>Tetrabenazine</td>\n",
       "      <td>D020821</td>\n",
       "      <td>Dystonic Disorders</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>34947 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       drug_id      drug_name disease_id             disease_name\n",
       "1      C074619    bivalirudin    D006470               Hemorrhage\n",
       "2      C074619    bivalirudin    D007249             Inflammation\n",
       "3      C074619    bivalirudin    D007511                 Ischemia\n",
       "4      C074619    bivalirudin    D013921         Thrombocytopenia\n",
       "5      C074619    bivalirudin    D013927               Thrombosis\n",
       "...        ...            ...        ...                      ...\n",
       "49212  D013747  Tetrabenazine    D017109  Akathisia, Drug-Induced\n",
       "49213  D013747  Tetrabenazine    D018476              Hypokinesia\n",
       "49214  D013747  Tetrabenazine    D020734   Parkinsonian Disorders\n",
       "49215  D013747  Tetrabenazine    D020820              Dyskinesias\n",
       "49216  D013747  Tetrabenazine    D020821       Dystonic Disorders\n",
       "\n",
       "[34947 rows x 4 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scmff_dda"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# Model evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "09bf469b7bb549498f32b507064a21eb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/9996 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "id_to_wuid = {id: f'unique_{i}' for i, id in enumerate(df_entities['id'].unique())}\n",
    "df_entities['wuid'] = df_entities['id'].apply(lambda x: id_to_wuid[x])\n",
    "texts = df_texts.progress_apply(preprocess_text(df_entities,id_to_wuid), axis=1).tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "Configuration to try"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-14T10:22:48.315424900Z",
     "start_time": "2023-05-14T10:22:48.015938700Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "coc_configs = dict(\n",
    "    k=list(range(5, 41, 5))\n",
    ")\n",
    "\n",
    "w2v_configs = dict(\n",
    "    learning_rate=[0.1],    # Learning rate to training the model\n",
    "    epochs=[100],           # Number of epochs\n",
    "    vector_size=[100, 500], # Latent space dimension\n",
    "    sg=[False, True],       # False->CBOW (Continuous bag of words) True->skip-gram\n",
    "    window=[5, 10],         # contex window dimension\n",
    "    min_count=[2, 5],       # threshold for the frequency\n",
    ")\n",
    "\n",
    "configs = [\n",
    "    *[(graphType.COOCCURRENCES, dict(zip(coc_configs, t))) for t in product(*coc_configs.values())],\n",
    "    *[(graphType.WORD2VEC, dict(zip(w2v_configs, t))) for t in product(*w2v_configs.values())],\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "Automatically try all configurations and rank them by precision achieved"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "ranking = model_evaluation(configs=configs,\n",
    "                     ts_set=test_set,\n",
    "                     texts=texts,\n",
    "                     df_entities=df_entities,\n",
    "                     max_diseases=100,\n",
    "                     top_z=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for precision, config in ranking:\n",
    "    print(precision, config)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "### We found that the best configuration is with Co-occurrences with k=10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-14T10:22:48.454526500Z",
     "start_time": "2023-05-14T10:22:48.315424900Z"
    },
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'k': 10}"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "configs[1][1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-14T10:24:39.042122100Z",
     "start_time": "2023-05-14T10:22:48.454526500Z"
    },
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Co_occurrencesGraph populate_adj_matrix: 100%|██████████| 9898/9898 [00:44<00:00, 220.39it/s]\n"
     ]
    }
   ],
   "source": [
    "# precision, config = ranking[1]\n",
    "graph = create_knowledge_graph(configs[1], df_entities, test_set)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "## Evaluation A"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-14T10:28:26.394814100Z",
     "start_time": "2023-05-14T10:24:39.042122100Z"
    },
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "test_drug_disease: 100%|██████████| 100/100 [01:35<00:00,  1.04it/s]\n"
     ]
    }
   ],
   "source": [
    "matches = find_diseases_precisionz(graph, top_z=10, test_set=test_set, max_diseases=100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "Describe the precision achieved for 100 diseases"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-14T10:31:37.962076400Z",
     "start_time": "2023-05-14T10:31:37.387022500Z"
    },
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Counter({0.0: 71, 0.1: 27, 0.2: 1, 0.4: 1, 0.3: 0})\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from Evalutation import plot_hist\n",
    "%reload_ext autoreload\n",
    "%autoreload 2\n",
    "t = Counter(matches)\n",
    "t[.3] = 0\n",
    "print(t)\n",
    "t = dict(sorted(t.items(), key=lambda item: item[0]))\n",
    "x = array(list(t.keys())) * 10\n",
    "y = array(list(t.values()))\n",
    "\n",
    "plot_hist(x,y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "## Evaluation B"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-13T13:22:25.048341Z",
     "start_time": "2023-05-13T13:22:24.353022900Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# create a dictionary of DDA from test_set\n",
    "disease_drugs = {}\n",
    "for (drug, disease), _ in test_set.items():\n",
    "    if disease not in disease_drugs:\n",
    "        disease_drugs[disease] = [drug]\n",
    "    else:\n",
    "        disease_drugs[disease].append(drug)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "pValues = pValues_DDAs(graph=graph,\n",
    "                       ddas=disease_drugs,\n",
    "                       max_diseases=10,\n",
    "                       n_samples=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-15T14:41:40.073942700Z",
     "start_time": "2023-05-15T14:41:39.824964200Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "pValues = [0.4474150974725506,\n",
    " 0.7758916519397002,\n",
    " 0.42578987277497826,\n",
    " 0.6834724817423341,\n",
    " 0.44755628041210505,\n",
    " 0.2856933831503558,\n",
    " 0.00020957884158454972,\n",
    " 0.024716017356946285,\n",
    " 0.1359102710650009,\n",
    " 0.8359753942553461,\n",
    " 0.4708784117318766,\n",
    " 0.15633057892315078,\n",
    " 0.0417535284967034,\n",
    " 0.2770658573997668,\n",
    " 0.7696008559247212,\n",
    " 0.47669944424089467,\n",
    " 0.213685989598322]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-05-15T07:45:20.694167700Z",
     "start_time": "2023-05-15T07:45:19.585001500Z"
    },
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.figure(figsize=(8, 4))\n",
    "ax = plt.axes()\n",
    "ax.set_facecolor(\"#ececec\")\n",
    "ax.set_axisbelow(True)\n",
    "ax.yaxis.grid(color='white')\n",
    "ax.xaxis.grid(color='white')\n",
    "plt.bar(range(len(pValues)), pValues, color=\"#2985e5\", edgecolor=\"black\")\n",
    "plt.xticks(range(len(pValues)), [*range(1,len(pValues)+1)])\n",
    "plt.xlabel(\"First 17 disease in the KG and TS\")\n",
    "plt.ylabel(\"p-value\")\n",
    "plt.title(\"P-value for first 17 disease\")\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}