DrugMiningSVD.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Drug Mining using SVD Recommendation Strategy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Drug Disease Relationship"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Drug name</th>\n",
       "      <th>Disease name</th>\n",
       "      <th>Reported Effect</th>\n",
       "      <th>Unnamed: 3</th>\n",
       "      <th>Unnamed: 4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3-keto-desogestrel</td>\n",
       "      <td>Amnesia</td>\n",
       "      <td>side effect</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3-keto-desogestrel</td>\n",
       "      <td>Anemia, Hemolytic, Autoimmune</td>\n",
       "      <td>side effect</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3-keto-desogestrel</td>\n",
       "      <td>Arrhythmia, Sinus</td>\n",
       "      <td>side effect</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3-keto-desogestrel</td>\n",
       "      <td>Bronchopneumonia</td>\n",
       "      <td>side effect</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3-keto-desogestrel</td>\n",
       "      <td>Burkitt Lymphoma</td>\n",
       "      <td>side effect</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Drug name                   Disease name Reported Effect  \\\n",
       "0  3-keto-desogestrel                        Amnesia     side effect   \n",
       "1  3-keto-desogestrel  Anemia, Hemolytic, Autoimmune     side effect   \n",
       "2  3-keto-desogestrel              Arrhythmia, Sinus     side effect   \n",
       "3  3-keto-desogestrel               Bronchopneumonia     side effect   \n",
       "4  3-keto-desogestrel               Burkitt Lymphoma     side effect   \n",
       "\n",
       "   Unnamed: 3  Unnamed: 4  \n",
       "0         NaN         NaN  \n",
       "1         NaN         NaN  \n",
       "2         NaN         NaN  \n",
       "3         NaN         NaN  \n",
       "4         NaN         NaN  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "init_drug_disease_relation = pd.read_csv(\"RawData/KnownDrugDiseaseRelation.csv\")\n",
    "init_drug_disease_relation.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Drug Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Drug name</th>\n",
       "      <th>Drugbank id</th>\n",
       "      <th>Mesh id</th>\n",
       "      <th>CAS id</th>\n",
       "      <th>Unnamed: 4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3-keto-desogestrel</td>\n",
       "      <td>DB00294</td>\n",
       "      <td>C044815</td>\n",
       "      <td>54048-10-1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4-Aminopyridine</td>\n",
       "      <td>DB06637</td>\n",
       "      <td>D015761</td>\n",
       "      <td>504-24-5</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4-hydroxybutyric acid</td>\n",
       "      <td>DB01440</td>\n",
       "      <td>C111420</td>\n",
       "      <td>591-81-1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5,6,7,8-tetrahydrobiopterin</td>\n",
       "      <td>DB00360</td>\n",
       "      <td>C003402</td>\n",
       "      <td>17528-72-2</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6-Mercaptopurine</td>\n",
       "      <td>DB01033</td>\n",
       "      <td>D015122</td>\n",
       "      <td>50-44-2</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Drug name Drugbank id  Mesh id      CAS id  Unnamed: 4\n",
       "0           3-keto-desogestrel     DB00294  C044815  54048-10-1         NaN\n",
       "1              4-Aminopyridine     DB06637  D015761    504-24-5         NaN\n",
       "2        4-hydroxybutyric acid     DB01440  C111420    591-81-1         NaN\n",
       "3  5,6,7,8-tetrahydrobiopterin     DB00360  C003402  17528-72-2         NaN\n",
       "4             6-Mercaptopurine     DB01033  D015122     50-44-2         NaN"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drug = pd.read_csv(\"RawData/Drug.csv\")\n",
    "drug.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Disease Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Disease name</th>\n",
       "      <th>Mesh id</th>\n",
       "      <th>UMLS id</th>\n",
       "      <th>Unnamed: 3</th>\n",
       "      <th>Unnamed: 4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Abdomen, Acute</td>\n",
       "      <td>D000006</td>\n",
       "      <td>C0000727</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Abdominal Neoplasms</td>\n",
       "      <td>D000008</td>\n",
       "      <td>C0000735</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Abdominal Pain</td>\n",
       "      <td>D015746</td>\n",
       "      <td>C0000737</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Abnormalities, Drug-Induced</td>\n",
       "      <td>D000014</td>\n",
       "      <td>C0000771</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Abortion, Spontaneous</td>\n",
       "      <td>D000022</td>\n",
       "      <td>C0000786</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Disease name  Mesh id   UMLS id  Unnamed: 3  Unnamed: 4\n",
       "0               Abdomen, Acute  D000006  C0000727         NaN         NaN\n",
       "1          Abdominal Neoplasms  D000008  C0000735         NaN         NaN\n",
       "2               Abdominal Pain  D015746  C0000737         NaN         NaN\n",
       "3  Abnormalities, Drug-Induced  D000014  C0000771         NaN         NaN\n",
       "4        Abortion, Spontaneous  D000022  C0000786         NaN         NaN"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "disease = pd.read_csv(\"RawData/Disease.csv\")\n",
    "disease.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocess Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- The goal of the preprocessing is to create a relationship matrix between drug_id and disease_id\n",
    "- If there is a positive effect, the relationship is 1\n",
    "- If there is a negative effect, the relationship is -1\n",
    "- If there is unknown effect, the relationship is 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def getDrugbankId(drugName):\n",
    "    try:\n",
    "        return drug[drug['Drug name']==drugName]['Drugbank id'].values[0]\n",
    "    except:\n",
    "        return \"Drug Not Found\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def getDiseaseMeshId(diseaseName):\n",
    "    try:\n",
    "        return disease[disease['Disease name']==diseaseName]['Mesh id'].values[0]\n",
    "    except:\n",
    "        return \"Disease Not Found\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "init_drug_disease_relation['Drug bank Id'] = init_drug_disease_relation['Drug name'].apply(lambda x: getDrugbankId(x))\n",
    "init_drug_disease_relation['Disease Mesh Id'] = init_drug_disease_relation['Disease name'].apply(lambda x: getDiseaseMeshId(x))\n",
    "init_drug_disease_relation = init_drug_disease_relation.drop(['Drug name', 'Disease name', 'Unnamed: 3', 'Unnamed: 4'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>Reported Effect</th>\n",
       "      <th>Drug bank Id</th>\n",
       "      <th>Disease Mesh Id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>side effect</td>\n",
       "      <td>DB00294</td>\n",
       "      <td>D000647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>side effect</td>\n",
       "      <td>DB00294</td>\n",
       "      <td>D000744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>side effect</td>\n",
       "      <td>DB00294</td>\n",
       "      <td>D001146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>side effect</td>\n",
       "      <td>DB00294</td>\n",
       "      <td>D001996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>side effect</td>\n",
       "      <td>DB00294</td>\n",
       "      <td>D002051</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Reported Effect Drug bank Id Disease Mesh Id\n",
       "0     side effect      DB00294         D000647\n",
       "1     side effect      DB00294         D000744\n",
       "2     side effect      DB00294         D001146\n",
       "3     side effect      DB00294         D001996\n",
       "4     side effect      DB00294         D002051"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "init_drug_disease_relation.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "init_drug_disease_relation_matrix = pd.DataFrame()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(0, len(init_drug_disease_relation)):\n",
    "    line = init_drug_disease_relation.iloc[i]\n",
    "    drug_id = line['Drug bank Id']\n",
    "    disease_id = line['Disease Mesh Id']\n",
    "    value = line['Reported Effect']\n",
    "    #init_drug_disease_relation_matrix[disease_id][drug_id] = \n",
    "    if drug_id != 'Drug Not Found' and disease_id != 'Disease Not Found':\n",
    "        if value == 'therapeutic effect\t':\n",
    "            init_drug_disease_relation_matrix.at[drug_id, disease_id] = 1\n",
    "        if value == 'side effect':\n",
    "            init_drug_disease_relation_matrix.at[drug_id, disease_id] = -1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>D000647</th>\n",
       "      <th>D000744</th>\n",
       "      <th>D001146</th>\n",
       "      <th>D001996</th>\n",
       "      <th>D002051</th>\n",
       "      <th>D002446</th>\n",
       "      <th>D002532</th>\n",
       "      <th>D002544</th>\n",
       "      <th>D002971</th>\n",
       "      <th>D003128</th>\n",
       "      <th>...</th>\n",
       "      <th>D000690</th>\n",
       "      <th>D003920</th>\n",
       "      <th>D007414</th>\n",
       "      <th>D014811</th>\n",
       "      <th>D019873</th>\n",
       "      <th>D010378</th>\n",
       "      <th>D009459</th>\n",
       "      <th>D016736</th>\n",
       "      <th>D006562</th>\n",
       "      <th>D020138</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DB00294</th>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DB00360</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DB01033</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DB01267</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DB00284</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 563 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         D000647  D000744  D001146  D001996  D002051  D002446  D002532  \\\n",
       "DB00294     -1.0     -1.0     -1.0     -1.0     -1.0     -1.0     -1.0   \n",
       "DB00360      0.0      0.0      0.0      0.0      0.0      0.0      0.0   \n",
       "DB01033      0.0     -1.0      0.0      0.0      0.0     -1.0      0.0   \n",
       "DB01267      0.0      0.0     -1.0      0.0     -1.0      0.0      0.0   \n",
       "DB00284      0.0     -1.0      0.0     -1.0      0.0      0.0     -1.0   \n",
       "\n",
       "         D002544  D002971  D003128  ...  D000690  D003920  D007414  D014811  \\\n",
       "DB00294     -1.0     -1.0     -1.0  ...      0.0      0.0      0.0      0.0   \n",
       "DB00360     -1.0      0.0      0.0  ...      0.0      0.0      0.0      0.0   \n",
       "DB01033      0.0      0.0      0.0  ...      0.0      0.0      0.0      0.0   \n",
       "DB01267      0.0      0.0      0.0  ...      0.0      0.0      0.0      0.0   \n",
       "DB00284      0.0      0.0      0.0  ...      0.0      0.0      0.0      0.0   \n",
       "\n",
       "         D019873  D010378  D009459  D016736  D006562  D020138  \n",
       "DB00294      0.0      0.0      0.0      0.0      0.0      0.0  \n",
       "DB00360      0.0      0.0      0.0      0.0      0.0      0.0  \n",
       "DB01033      0.0      0.0      0.0      0.0      0.0      0.0  \n",
       "DB01267      0.0      0.0      0.0      0.0      0.0      0.0  \n",
       "DB00284      0.0      0.0      0.0      0.0      0.0      0.0  \n",
       "\n",
       "[5 rows x 563 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "init_drug_disease_relation_matrix = init_drug_disease_relation_matrix.fillna(0)\n",
    "init_drug_disease_relation_matrix.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "init_drug_disease_relation_matrix.to_csv(\"init_drug_disease_relation_matrix.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Start SVD Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "u, s, vh = np.linalg.svd(init_drug_disease_relation_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((654, 654), (563,), (563, 563))"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "u.shape, s.shape, vh.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "singular_value_number = s.shape[0]\n",
    "np.allclose(init_drug_disease_relation_matrix, np.dot(u[:,:singular_value_number] * s, vh))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "67"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "singular_value_number = s.shape[0]\n",
    "keep_rate = 0.12\n",
    "keep_number = (int)(singular_value_number * keep_rate)\n",
    "keep_number"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>D000647</th>\n",
       "      <th>D000744</th>\n",
       "      <th>D001146</th>\n",
       "      <th>D001996</th>\n",
       "      <th>D002051</th>\n",
       "      <th>D002446</th>\n",
       "      <th>D002532</th>\n",
       "      <th>D002544</th>\n",
       "      <th>D002971</th>\n",
       "      <th>D003128</th>\n",
       "      <th>...</th>\n",
       "      <th>D000690</th>\n",
       "      <th>D003920</th>\n",
       "      <th>D007414</th>\n",
       "      <th>D014811</th>\n",
       "      <th>D019873</th>\n",
       "      <th>D010378</th>\n",
       "      <th>D009459</th>\n",
       "      <th>D016736</th>\n",
       "      <th>D006562</th>\n",
       "      <th>D020138</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DB00294</th>\n",
       "      <td>-0.278317</td>\n",
       "      <td>-0.681843</td>\n",
       "      <td>-0.331201</td>\n",
       "      <td>-0.378043</td>\n",
       "      <td>-0.605217</td>\n",
       "      <td>-0.590330</td>\n",
       "      <td>-0.507194</td>\n",
       "      <td>-0.506194</td>\n",
       "      <td>-0.148257</td>\n",
       "      <td>-0.523923</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.013416</td>\n",
       "      <td>-0.029666</td>\n",
       "      <td>-0.013416</td>\n",
       "      <td>-0.006398</td>\n",
       "      <td>-0.006398</td>\n",
       "      <td>-0.007973</td>\n",
       "      <td>-0.005830</td>\n",
       "      <td>0.013271</td>\n",
       "      <td>-0.000168</td>\n",
       "      <td>-0.003231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DB00360</th>\n",
       "      <td>-0.056903</td>\n",
       "      <td>0.161720</td>\n",
       "      <td>-0.013412</td>\n",
       "      <td>-0.129826</td>\n",
       "      <td>-0.098111</td>\n",
       "      <td>-0.063728</td>\n",
       "      <td>-0.026327</td>\n",
       "      <td>-0.541821</td>\n",
       "      <td>0.005956</td>\n",
       "      <td>-0.351780</td>\n",
       "      <td>...</td>\n",
       "      <td>0.002503</td>\n",
       "      <td>-0.030056</td>\n",
       "      <td>0.002503</td>\n",
       "      <td>0.005982</td>\n",
       "      <td>0.005982</td>\n",
       "      <td>0.008859</td>\n",
       "      <td>0.014114</td>\n",
       "      <td>0.002103</td>\n",
       "      <td>-0.003994</td>\n",
       "      <td>-0.012690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DB01033</th>\n",
       "      <td>0.030332</td>\n",
       "      <td>-0.820692</td>\n",
       "      <td>-0.246105</td>\n",
       "      <td>-0.154090</td>\n",
       "      <td>-0.165664</td>\n",
       "      <td>-0.745996</td>\n",
       "      <td>-0.026524</td>\n",
       "      <td>-0.053840</td>\n",
       "      <td>-0.148550</td>\n",
       "      <td>-0.448419</td>\n",
       "      <td>...</td>\n",
       "      <td>0.050883</td>\n",
       "      <td>0.088423</td>\n",
       "      <td>0.050883</td>\n",
       "      <td>-0.019104</td>\n",
       "      <td>-0.019104</td>\n",
       "      <td>0.004752</td>\n",
       "      <td>0.002638</td>\n",
       "      <td>-0.041895</td>\n",
       "      <td>0.002363</td>\n",
       "      <td>-0.025212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DB01267</th>\n",
       "      <td>-0.077903</td>\n",
       "      <td>-0.046394</td>\n",
       "      <td>-0.375154</td>\n",
       "      <td>0.065177</td>\n",
       "      <td>-0.606484</td>\n",
       "      <td>-0.185595</td>\n",
       "      <td>0.027812</td>\n",
       "      <td>-0.262429</td>\n",
       "      <td>-0.013846</td>\n",
       "      <td>0.096221</td>\n",
       "      <td>...</td>\n",
       "      <td>0.002919</td>\n",
       "      <td>0.002784</td>\n",
       "      <td>0.002919</td>\n",
       "      <td>-0.000031</td>\n",
       "      <td>-0.000031</td>\n",
       "      <td>-0.018208</td>\n",
       "      <td>-0.003885</td>\n",
       "      <td>0.005870</td>\n",
       "      <td>0.004736</td>\n",
       "      <td>-0.009811</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DB00284</th>\n",
       "      <td>-0.386872</td>\n",
       "      <td>-0.468148</td>\n",
       "      <td>-0.215554</td>\n",
       "      <td>-0.622434</td>\n",
       "      <td>-0.090623</td>\n",
       "      <td>0.020544</td>\n",
       "      <td>-0.480147</td>\n",
       "      <td>0.085725</td>\n",
       "      <td>-0.018399</td>\n",
       "      <td>-0.214187</td>\n",
       "      <td>...</td>\n",
       "      <td>0.020377</td>\n",
       "      <td>0.017784</td>\n",
       "      <td>0.020377</td>\n",
       "      <td>0.013496</td>\n",
       "      <td>0.013496</td>\n",
       "      <td>-0.007980</td>\n",
       "      <td>-0.003913</td>\n",
       "      <td>0.018722</td>\n",
       "      <td>-0.007007</td>\n",
       "      <td>-0.000587</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 563 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          D000647   D000744   D001146   D001996   D002051   D002446   D002532  \\\n",
       "DB00294 -0.278317 -0.681843 -0.331201 -0.378043 -0.605217 -0.590330 -0.507194   \n",
       "DB00360 -0.056903  0.161720 -0.013412 -0.129826 -0.098111 -0.063728 -0.026327   \n",
       "DB01033  0.030332 -0.820692 -0.246105 -0.154090 -0.165664 -0.745996 -0.026524   \n",
       "DB01267 -0.077903 -0.046394 -0.375154  0.065177 -0.606484 -0.185595  0.027812   \n",
       "DB00284 -0.386872 -0.468148 -0.215554 -0.622434 -0.090623  0.020544 -0.480147   \n",
       "\n",
       "          D002544   D002971   D003128  ...   D000690   D003920   D007414  \\\n",
       "DB00294 -0.506194 -0.148257 -0.523923  ... -0.013416 -0.029666 -0.013416   \n",
       "DB00360 -0.541821  0.005956 -0.351780  ...  0.002503 -0.030056  0.002503   \n",
       "DB01033 -0.053840 -0.148550 -0.448419  ...  0.050883  0.088423  0.050883   \n",
       "DB01267 -0.262429 -0.013846  0.096221  ...  0.002919  0.002784  0.002919   \n",
       "DB00284  0.085725 -0.018399 -0.214187  ...  0.020377  0.017784  0.020377   \n",
       "\n",
       "          D014811   D019873   D010378   D009459   D016736   D006562   D020138  \n",
       "DB00294 -0.006398 -0.006398 -0.007973 -0.005830  0.013271 -0.000168 -0.003231  \n",
       "DB00360  0.005982  0.005982  0.008859  0.014114  0.002103 -0.003994 -0.012690  \n",
       "DB01033 -0.019104 -0.019104  0.004752  0.002638 -0.041895  0.002363 -0.025212  \n",
       "DB01267 -0.000031 -0.000031 -0.018208 -0.003885  0.005870  0.004736 -0.009811  \n",
       "DB00284  0.013496  0.013496 -0.007980 -0.003913  0.018722 -0.007007 -0.000587  \n",
       "\n",
       "[5 rows x 563 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reconstruct_drug_disease_relation = init_drug_disease_relation_matrix.copy()\n",
    "reconstruct_drug_disease_relation.values[:,:] = np.dot(u[:,:keep_number] * s[:keep_number], vh[:keep_number,:])\n",
    "reconstruct_drug_disease_relation.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use Genetics Prediction to Valid our SVD Predicting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>Drug Drugbank id</th>\n",
       "      <th>Disease Mesh ID</th>\n",
       "      <th>Medical genetics based prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DB00106</td>\n",
       "      <td>C537919</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DB00106</td>\n",
       "      <td>D007006</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DB01193</td>\n",
       "      <td>D006333</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DB01418</td>\n",
       "      <td>C564741</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DB00819</td>\n",
       "      <td>D010022</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Drug Drugbank id Disease Mesh ID  Medical genetics based prediction\n",
       "0          DB00106         C537919                                 -1\n",
       "1          DB00106         D007006                                 -1\n",
       "2          DB01193         D006333                                 -1\n",
       "3          DB01418         C564741                                 -1\n",
       "4          DB00819         D010022                                 -1"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validation = pd.read_csv(\"RawData/MedicalGeneticsPrediction.csv\")\n",
    "validation = validation.drop([\"Drug name\", \"Disease Name\", \"Drug action\", \"Function\"], axis=1)\n",
    "validation = validation.replace(\"side effect\", -1)\n",
    "validation = validation.replace(\"therapeutic effect\", 1)\n",
    "validation.to_csv(\"validation.csv\")\n",
    "validation.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "def check_svd_prediction(drug_id, disease_id):\n",
    "    try:\n",
    "        return reconstruct_drug_disease_relation[disease_id][drug_id]\n",
    "    except:\n",
    "        return \"Prediction Not Found\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "for index, line in validation.iterrows():\n",
    "    drug_id = line['Drug Drugbank id']\n",
    "    disease_id = line['Disease Mesh ID']\n",
    "    svd_predict = check_svd_prediction(drug_id, disease_id)\n",
    "    validation.loc[index, 'Raw SVD Predict'] = svd_predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "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>Drug Drugbank id</th>\n",
       "      <th>Disease Mesh ID</th>\n",
       "      <th>Medical genetics based prediction</th>\n",
       "      <th>Raw SVD Predict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DB01193</td>\n",
       "      <td>D006333</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.0369814</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>DB00459</td>\n",
       "      <td>D015473</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.0433578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>DB00210</td>\n",
       "      <td>D015473</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.164438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>DB00404</td>\n",
       "      <td>D004832</td>\n",
       "      <td>1</td>\n",
       "      <td>0.332956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>DB00182</td>\n",
       "      <td>D020734</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.27192</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Drug Drugbank id Disease Mesh ID  Medical genetics based prediction  \\\n",
       "2           DB01193         D006333                                 -1   \n",
       "7           DB00459         D015473                                 -1   \n",
       "11          DB00210         D015473                                 -1   \n",
       "21          DB00404         D004832                                  1   \n",
       "37          DB00182         D020734                                 -1   \n",
       "\n",
       "   Raw SVD Predict  \n",
       "2       -0.0369814  \n",
       "7       -0.0433578  \n",
       "11       -0.164438  \n",
       "21        0.332956  \n",
       "37        -0.27192  "
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validation = validation[validation['Raw SVD Predict']!='Prediction Not Found']\n",
    "validation.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Process SVD Prediction, negative to -1, positive to 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "def process_raw_predict(m):\n",
    "    if m<0:\n",
    "        return -1\n",
    "    elif m==0:\n",
    "        return 0\n",
    "    else:\n",
    "        return 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "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>Drug Drugbank id</th>\n",
       "      <th>Disease Mesh ID</th>\n",
       "      <th>Medical genetics based prediction</th>\n",
       "      <th>Raw SVD Predict</th>\n",
       "      <th>SVD Predict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DB01193</td>\n",
       "      <td>D006333</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.0369814</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>DB00459</td>\n",
       "      <td>D015473</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.0433578</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>DB00210</td>\n",
       "      <td>D015473</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.164438</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>DB00404</td>\n",
       "      <td>D004832</td>\n",
       "      <td>1</td>\n",
       "      <td>0.332956</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>DB00182</td>\n",
       "      <td>D020734</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.27192</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Drug Drugbank id Disease Mesh ID  Medical genetics based prediction  \\\n",
       "2           DB01193         D006333                                 -1   \n",
       "7           DB00459         D015473                                 -1   \n",
       "11          DB00210         D015473                                 -1   \n",
       "21          DB00404         D004832                                  1   \n",
       "37          DB00182         D020734                                 -1   \n",
       "\n",
       "   Raw SVD Predict  SVD Predict  \n",
       "2       -0.0369814           -1  \n",
       "7       -0.0433578           -1  \n",
       "11       -0.164438           -1  \n",
       "21        0.332956            1  \n",
       "37        -0.27192           -1  "
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validation['SVD Predict'] = validation['Raw SVD Predict'].apply(lambda x: process_raw_predict(x))\n",
    "validation.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Statistics on the rate of SVD Prediction that also occurs in Medical Genetics Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "for index, line in validation.iterrows():\n",
    "    medical_prediction = line['Medical genetics based prediction']\n",
    "    svd_prediction = line['SVD Predict']\n",
    "    result = 0\n",
    "    if medical_prediction == svd_prediction:\n",
    "        result = 1\n",
    "    \n",
    "    validation.loc[index, 'Prediction Right'] = result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "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>Drug Drugbank id</th>\n",
       "      <th>Disease Mesh ID</th>\n",
       "      <th>Medical genetics based prediction</th>\n",
       "      <th>Raw SVD Predict</th>\n",
       "      <th>SVD Predict</th>\n",
       "      <th>Prediction Right</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>DB00564</td>\n",
       "      <td>D014693</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.106189</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>130</th>\n",
       "      <td>DB00521</td>\n",
       "      <td>D006333</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.081127</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>DB00475</td>\n",
       "      <td>D004832</td>\n",
       "      <td>1</td>\n",
       "      <td>0.248697</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>171</th>\n",
       "      <td>DB00882</td>\n",
       "      <td>D001943</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.2096</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>173</th>\n",
       "      <td>DB00882</td>\n",
       "      <td>D008881</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.166484</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185</th>\n",
       "      <td>DB00907</td>\n",
       "      <td>D014693</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.0317429</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>187</th>\n",
       "      <td>DB00924</td>\n",
       "      <td>D012559</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.0196695</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>189</th>\n",
       "      <td>DB00434</td>\n",
       "      <td>D012559</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.00505454</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>204</th>\n",
       "      <td>DB01254</td>\n",
       "      <td>D046152</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.013152</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205</th>\n",
       "      <td>DB01254</td>\n",
       "      <td>D009373</td>\n",
       "      <td>1</td>\n",
       "      <td>0.00820425</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Drug Drugbank id Disease Mesh ID  Medical genetics based prediction  \\\n",
       "124          DB00564         D014693                                 -1   \n",
       "130          DB00521         D006333                                 -1   \n",
       "136          DB00475         D004832                                  1   \n",
       "171          DB00882         D001943                                 -1   \n",
       "173          DB00882         D008881                                  1   \n",
       "185          DB00907         D014693                                 -1   \n",
       "187          DB00924         D012559                                 -1   \n",
       "189          DB00434         D012559                                 -1   \n",
       "204          DB01254         D046152                                  1   \n",
       "205          DB01254         D009373                                  1   \n",
       "\n",
       "    Raw SVD Predict  SVD Predict  Prediction Right  \n",
       "124       -0.106189           -1               1.0  \n",
       "130       -0.081127           -1               1.0  \n",
       "136        0.248697            1               1.0  \n",
       "171         -0.2096           -1               1.0  \n",
       "173       -0.166484           -1               0.0  \n",
       "185       0.0317429            1               0.0  \n",
       "187      -0.0196695           -1               1.0  \n",
       "189     -0.00505454           -1               1.0  \n",
       "204       -0.013152           -1               0.0  \n",
       "205      0.00820425            1               1.0  "
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validation[10:20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "62.88659793814433% in SVD Prediction occcurs in Medical Genetics Prediction\n"
     ]
    }
   ],
   "source": [
    "len_validation = len(validation)\n",
    "len_predict_right = len(validation[validation['Prediction Right']==1])\n",
    "right_rate = len_predict_right / len_validation\n",
    "print(f\"{right_rate*100}% in SVD Prediction occcurs in Medical Genetics Prediction\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}