FAIR_assessment_ChEMBL.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "toc-hr-collapsed": false
   },
   "source": [
    "# FAIR maturity indicators for nanosafety"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "by [Ammar Ammar](https://github.com/ammar257ammar), [Serena Bonaretti](https://sbonaretti.github.io) and [Egon Willighagen](https://egonw.github.io), Maastricht University"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc-hr-collapsed": true
   },
   "source": [
    "---\n",
    "### Use case"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook template considers an example use cases of data retrieval in the nanosafety field. For each case, we\n",
    "- Execute data retrieval via API \n",
    "- Evaluate the FAIRness of the dataset\n",
    "\n",
    "This template is derived from [this Jupyter notebook](https://github.com/sbonaretti/FAIR_metrics/blob/master/code/FAIR_assessment_2.ipynb) which has two example use cases."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The use-cases is:\n",
    "\n",
    "\n",
    "<table>\n",
    "<colgroup>\n",
    "<col style=\"width: 20%\" />\n",
    "<col style=\"width: 80%\" /> \n",
    "</colgroup>\n",
    "    \n",
    "  <tr>\n",
    "    <th></th>\n",
    "    <th>TiO<sub>2</sub></th>\n",
    "  </tr>\n",
    "    \n",
    "  <tr>\n",
    "    <th>Research question</th>\n",
    "    <th>What does the database tell us about the safety of TiO<sub>2</sub>?</th>\n",
    "  </tr>\n",
    "  \n",
    "  <tr>\n",
    "    <th>Dataset</th>\n",
    "    <th>ChEMBL</th>\n",
    "  </tr>\n",
    "  \n",
    "  <tr>\n",
    "    <th>Repository</th>\n",
    "      <th><a href=\"https://www.ebi.ac.uk/chembl/\">ebi.ac.uk/chembl</a></th>\n",
    "  </tr>\n",
    "\n",
    "</table>\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Information about repository:\n",
    "- **Repository: ebi.ac.uk/chembl**:\n",
    "  - API: \n",
    "    - Documentation at https://www.ebi.ac.uk/chembl/api/data/docs\n",
    "    - The API to query the repository is a rest API: https://www.ebi.ac.uk/chembl/api/data/document/search\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc-hr-collapsed": true
   },
   "source": [
    "### Calculating FAIR metrics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[Prerequisites:](#prerequisites)  \n",
    "- [Imports](#imports)  \n",
    "- [Variables](#variables)  \n",
    "- [Functions](#functions)  \n",
    "\n",
    "[Retrieving datasets:](#retrieval)  \n",
    "- [Case: TEMPLATE: REPOSITORY](#Repository)\n",
    "\n",
    "[Retrieving repositories in www.re3data.org](#Retrieving_repositories_in_www.re3data.org)\n",
    "\n",
    "\n",
    "[FAIR assessment:](#metrics) \n",
    "- [F1. (Meta)data are assigned a globally unique and persistent identifier](#F1) \n",
    "- [F2. Data are described with rich metadata](#F2) \n",
    "- [F3. Metadata clearly and explicitly include the identifier of the data it describes](#F3)  \n",
    "- [F4. (Meta)data are registered or indexed in a searchable resource](#F4) \n",
    "- [A1. (Meta)data are retrievable by their identifier using a standardized communications protocol](#A1)    \n",
    "- [A1.1 The protocol is open, free, and universally implementable](#A11)   \n",
    "- [A1.2 The protocol allows for an authentication and authorization procedure, where necessary](#A12)   \n",
    "- [A2. Metadata are accessible, even when the data are no longer available](#A2)  \n",
    "- [I1. (Meta)data use a formal, accessible, shared, and broadly applicable language for knowledge representation](#I1) \n",
    "- [I2. Metadata use vocabularies that follow FAIR principles](#I2)  \n",
    "- [I3. (Meta)data include qualified references to other (meta)data](#I3)   \n",
    "- [R1.1. (Meta)data are released with a clear and accessible data usage license](#R11)  \n",
    "- [R1.2. (Meta)data are associated with detailed provenance](#R12)  \n",
    "- [R1.3. (Meta)data meet domain-relevant community standards](#R13)\n",
    "\n",
    "[FAIR assessment visualization](#visualization)\n",
    "\n",
    "[Dependencies](#dependencies)\n",
    "\n",
    "[Supplementary information](#supplementary_information)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"prerequisites\"></a>\n",
    "### Prerequisites"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"imports\"></a>\n",
    "#### Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import python packages\n",
    "import os\n",
    "import pandas as pd\n",
    "import requests\n",
    "import time\n",
    "import xml.etree.ElementTree as ET\n",
    "import json\n",
    "from io import StringIO\n",
    "import urllib.parse\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import R for visualization using ggplot2 - comment if used in mybinder, uncomment if used locally\n",
    "#os.environ['R_HOME'] = '/Library/Frameworks/R.framework/Resources' # put here your own R directory "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext rpy2.ipython"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"variables\"></a>\n",
    "#### Variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# re3Data API\n",
    "re3data_repos    = \"https://www.re3data.org/api/v1/repositories\"\n",
    "re3data_repo     = \"https://www.re3data.org/api/v1/repository\"\n",
    "re3data_schema   = '{http://www.re3data.org/schema/2-2}' # in the metadata xml"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# repository variables\n",
    "repository_name  = []\n",
    "repository_api   = []\n",
    "dataset_download = []\n",
    "\n",
    "# TEMPLATE: REPOSITORY\n",
    "repository_name.append(\"ChEMBL\")\n",
    "repository_api.append(\"https://www.ebi.ac.uk/chembl/api/data/document/search?format=json&q=\")\n",
    "dataset_download.append(\"https://www.ebi.ac.uk/chembl/api/data/document/search?format=json&q=\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# case names\n",
    "case_names        = []\n",
    "case_names.append(\"titanium_dioxide\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# search keywords\n",
    "keywords = []\n",
    "\n",
    "# TEMPLATE: REPOSITORY\n",
    "kw = []\n",
    "kw.append(\"titanium dioxide\")\n",
    "keywords.append(kw)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# output variables used for FAIR assessment\n",
    "\n",
    "request_status        = []\n",
    "metadata_xml          = [] \n",
    "metadata_list         = [] # list of lists. Dimension: n_of_cases x 1. \n",
    "                           # Inside each cell, list of dimension 2 x N. First row is tags, second row is text of the xml output containing metadata\n",
    "re3data_repo_root     = []\n",
    "re3data_metadata_list = [] # same structure as metadata_list\n",
    "n_of_found_keywords   = []\n",
    "dataset_title         = []\n",
    "\n",
    "F1  = []\n",
    "F2  = []\n",
    "F3  = []\n",
    "F4  = []\n",
    "A1  = []\n",
    "A11 = []\n",
    "A12 = []\n",
    "A2  = []\n",
    "I1  = []\n",
    "I2  = []\n",
    "I3  = []\n",
    "R1  = []\n",
    "R11 = []\n",
    "R12 = []\n",
    "R13 = []"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"functions\"></a>\n",
    "#### Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "def find_repo_in_re3_data (repository_name):\n",
    "    \"\"\"\n",
    "    It retrieves metadata about the input repository in re3data.org\n",
    "    \"\"\"\n",
    "    \n",
    "    # contact the re3data\n",
    "    re3data_r = requests.get(re3data_repos)\n",
    "    if re3data_r == None:\n",
    "        return None\n",
    "    \n",
    "    # put output string in a structured xml tree\n",
    "    re3data_root = ET.fromstring(re3data_r.text)\n",
    "\n",
    "    # look for the wanted repository\n",
    "    flag = 0\n",
    "    for i in range (0, len(re3data_root)):\n",
    "        if re3data_root[i][1].text == repository_name: # first index is a <repository> group, \n",
    "                                                       # second index is for the tags inside (<id>, <name>, <link>)\n",
    "            flag = 1\n",
    "            repository_index = i\n",
    "\n",
    "    if flag == 1:   \n",
    "        \n",
    "        re3data_metadata_address = str(re3data_root[repository_index][2].attrib[\"href\"])\n",
    "        #print (\"Index in list: \" + str(repository_index))\n",
    "        #print (\"Tag          : \" + str(re3data_root[repository_index][1].tag))\n",
    "        #print (\"Text         : \" + str(re3data_root[repository_index][1].text))\n",
    "        #print (\"Tag          : \" + str(re3data_root[repository_index][2].tag))\n",
    "        #print (\"Attribute    : \" + str(re3data_root[repository_index][2].attrib)) # attributes do not have a closing tag\n",
    "\n",
    "    else: \n",
    "        print (\"Repository not found\")\n",
    "        return None\n",
    "        \n",
    "    return re3data_metadata_address"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "def xml_to_list(root): \n",
    "\n",
    "    tag_list = []\n",
    "    text_list = []\n",
    "        \n",
    "    for elem in root.iter():\n",
    "        tag_list.append(elem.tag)\n",
    "        text = elem.text\n",
    "        if text:\n",
    "            text_list.append(text.strip())\n",
    "        else:\n",
    "            text_list.append(\"\")\n",
    "            text = \"\"\n",
    "        #print(\"We are adding element: \"+elem.tag + \" - with text: \"+text)\n",
    "    \n",
    "    #print(\"numbers: \" + str(len(tag_list)) + \"-\" + str(len(text_list)))\n",
    "    xml_list = [ tag_list, text_list ]\n",
    "    \n",
    "    return xml_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "def json_to_list(root): \n",
    "\n",
    "    tag_list = []\n",
    "    text_list = []\n",
    "           \n",
    "    for attribute, value in root.items():\n",
    "        tag_list.append(attribute)\n",
    "        text_list.append(value)\n",
    "    \n",
    "    #print(\"numbers: \" + str(len(tag_list)) + \"-\" + str(len(text_list)))\n",
    "    json_list = [ tag_list, text_list ]\n",
    "    \n",
    "    return json_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "def find_list_in_list(long_list, short_list):\n",
    "    \"\"\"\n",
    "    For each items of short_list, it looks in the long_list\n",
    "    It uses a \"for\" loop because it has to look also for parts of string\n",
    "    Inputs: \n",
    "     long_list: long list of strings\n",
    "     short_list: short list of strings\n",
    "    Outputs:\n",
    "     found: elements of short_list found in long_list\n",
    "     index: index of found elements in long_list   \n",
    "    \"\"\"\n",
    "    \n",
    "    found = []\n",
    "    index = []\n",
    "    for i in range(0, len(long_list)):\n",
    "        for j in range(0, len(short_list)):\n",
    "            \n",
    "            #print(\"compare: \"+long_list[i].lower() + \" TO \" + short_list[j].lower())\n",
    "            if str(long_list[i]).lower().find(str(short_list[j]).lower()) != -1 :\n",
    "                found.append(str(short_list[j]))\n",
    "                index.append(i)\n",
    "    return found, index"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"retrieval\"></a>\n",
    "### Retrieving datasets "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"Repository\"></a>\n",
    "#### Case: TEMPLATE: REPOSITORY"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "case_id = 0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The below test assumes a certain API with parameters. For example, the below code uses `keywords`\n",
    "as API parameter to pass keywords. The template code likely needs updating."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "https://www.ebi.ac.uk/chembl/api/data/document/search?format=json&q=titanium%20dioxide\n",
      "-> The time to retrieve the dataset was 0.33 seconds\n",
      "-> Request status 200 \n",
      "number of samples found: 20\n",
      "\n",
      "\n"
     ]
    },
    {
     "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>document_chembl_id</th>\n",
       "      <th>pubmed_id</th>\n",
       "      <th>authors</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CHEMBL1157425</td>\n",
       "      <td>18512988</td>\n",
       "      <td>Choquenet B, Couteau C, Paparis E, Coiffard LJ.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CHEMBL2346584</td>\n",
       "      <td>23434418</td>\n",
       "      <td>Sangshetti JN, Dharmadhikari PP, Chouthe RS, F...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CHEMBL1132950</td>\n",
       "      <td>11020280</td>\n",
       "      <td>Caruso F, Rossi M, Tanski J, Sartori R, Sarieg...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CHEMBL1130769</td>\n",
       "      <td>9871614</td>\n",
       "      <td>Kuang R, Venkataraman R, Ruan S, Groutas WC.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CHEMBL1133239</td>\n",
       "      <td>10915061</td>\n",
       "      <td>Ma J, Weiss E, Kyle DE, Ziffer H.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>CHEMBL1122467</td>\n",
       "      <td>6827542</td>\n",
       "      <td>Chapman JM, Cocolas GH, Hall IH.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>CHEMBL1137628</td>\n",
       "      <td>17933534</td>\n",
       "      <td>Chen CP, Zeiger AR, Wickstrom E.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>CHEMBL1132905</td>\n",
       "      <td>10866388</td>\n",
       "      <td>Leclercq F, Dubertret C, Pitard B, Scherman D,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>CHEMBL1127661</td>\n",
       "      <td>8126711</td>\n",
       "      <td>Benedetti-Doctorovich V, Burgess EM, Lambropou...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>CHEMBL1144360</td>\n",
       "      <td>15633996</td>\n",
       "      <td>Cerecetto H, González M, Lavaggi ML, Azqueta A...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>CHEMBL1137602</td>\n",
       "      <td>17919904</td>\n",
       "      <td>Tanaka H, Shishido Y.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>CHEMBL1149783</td>\n",
       "      <td>17566609</td>\n",
       "      <td>Salmi C, Loncle C, Vidal N, Letourneux Y, Brun...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>CHEMBL1152690</td>\n",
       "      <td>19875298</td>\n",
       "      <td>Pearson MS, Floquet N, Bello C, Vogel P, Plant...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>CHEMBL1132052</td>\n",
       "      <td>10201825</td>\n",
       "      <td>Yu KL, Civiello R, Roberts DG, Seiler SM, Mean...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>CHEMBL1123814</td>\n",
       "      <td>3039137</td>\n",
       "      <td>Micetich RG, Maiti SN, Spevak P, Hall TW, Yama...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>CHEMBL1133508</td>\n",
       "      <td>11063619</td>\n",
       "      <td>Ma J, Katz E, Kyle DE, Ziffer H.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>CHEMBL1140288</td>\n",
       "      <td>15974581</td>\n",
       "      <td>Nguyen Van Nhien A, Tomassi C, Len C, Marco-Co...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>CHEMBL1140456</td>\n",
       "      <td>17826998</td>\n",
       "      <td>Lory PM, Estrella-Jimenez ME, Shashack MJ, Lok...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>CHEMBL1137713</td>\n",
       "      <td>18835067</td>\n",
       "      <td>Barazarte A, Lobo G, Gamboa N, Rodrigues JR, C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>CHEMBL1255519</td>\n",
       "      <td>20728353</td>\n",
       "      <td>Watanabe T, Kurata I, Hayashi C, Igarashi M, S...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   document_chembl_id  pubmed_id  \\\n",
       "0       CHEMBL1157425   18512988   \n",
       "1       CHEMBL2346584   23434418   \n",
       "2       CHEMBL1132950   11020280   \n",
       "3       CHEMBL1130769    9871614   \n",
       "4       CHEMBL1133239   10915061   \n",
       "5       CHEMBL1122467    6827542   \n",
       "6       CHEMBL1137628   17933534   \n",
       "7       CHEMBL1132905   10866388   \n",
       "8       CHEMBL1127661    8126711   \n",
       "9       CHEMBL1144360   15633996   \n",
       "10      CHEMBL1137602   17919904   \n",
       "11      CHEMBL1149783   17566609   \n",
       "12      CHEMBL1152690   19875298   \n",
       "13      CHEMBL1132052   10201825   \n",
       "14      CHEMBL1123814    3039137   \n",
       "15      CHEMBL1133508   11063619   \n",
       "16      CHEMBL1140288   15974581   \n",
       "17      CHEMBL1140456   17826998   \n",
       "18      CHEMBL1137713   18835067   \n",
       "19      CHEMBL1255519   20728353   \n",
       "\n",
       "                                              authors  \n",
       "0     Choquenet B, Couteau C, Paparis E, Coiffard LJ.  \n",
       "1   Sangshetti JN, Dharmadhikari PP, Chouthe RS, F...  \n",
       "2   Caruso F, Rossi M, Tanski J, Sartori R, Sarieg...  \n",
       "3        Kuang R, Venkataraman R, Ruan S, Groutas WC.  \n",
       "4                   Ma J, Weiss E, Kyle DE, Ziffer H.  \n",
       "5                    Chapman JM, Cocolas GH, Hall IH.  \n",
       "6                    Chen CP, Zeiger AR, Wickstrom E.  \n",
       "7   Leclercq F, Dubertret C, Pitard B, Scherman D,...  \n",
       "8   Benedetti-Doctorovich V, Burgess EM, Lambropou...  \n",
       "9   Cerecetto H, González M, Lavaggi ML, Azqueta A...  \n",
       "10                              Tanaka H, Shishido Y.  \n",
       "11  Salmi C, Loncle C, Vidal N, Letourneux Y, Brun...  \n",
       "12  Pearson MS, Floquet N, Bello C, Vogel P, Plant...  \n",
       "13  Yu KL, Civiello R, Roberts DG, Seiler SM, Mean...  \n",
       "14  Micetich RG, Maiti SN, Spevak P, Hall TW, Yama...  \n",
       "15                   Ma J, Katz E, Kyle DE, Ziffer H.  \n",
       "16  Nguyen Van Nhien A, Tomassi C, Len C, Marco-Co...  \n",
       "17  Lory PM, Estrella-Jimenez ME, Shashack MJ, Lok...  \n",
       "18  Barazarte A, Lobo G, Gamboa N, Rodrigues JR, C...  \n",
       "19  Watanabe T, Kurata I, Hayashi C, Igarashi M, S...  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting dataset\n",
    "start_time = time.time()\n",
    "print(repository_api[case_id]+urllib.parse.quote(keywords[case_id][0]))\n",
    "r_repo = requests.get(repository_api[case_id]+urllib.parse.quote(keywords[case_id][0]))\n",
    "\n",
    "print (\"-> The time to retrieve the dataset was %.2f seconds\" % ((time.time() - start_time)))\n",
    "print (\"-> Request status %.3s \" % (r_repo.status_code))\n",
    "\n",
    "# add request status\n",
    "request_status.append(r_repo.status_code)\n",
    "\n",
    "# convert answer from string to json\n",
    "results = json.loads(r_repo.text)\n",
    "\n",
    "results_all = results['documents']\n",
    "\n",
    "results_df = pd.DataFrame(results['documents'], columns=[\"document_chembl_id\", \"pubmed_id\", \"authors\"])\n",
    "\n",
    "print(\"number of samples found: \" + str (len(results['documents'])) + \"\\n\\n\")\n",
    "\n",
    "results_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHEMBL1157425\n"
     ]
    }
   ],
   "source": [
    "# look for experiment_id \n",
    "experiment_id = []\n",
    "\n",
    "experiment_id = results_df['document_chembl_id'].apply(str).tolist()\n",
    "\n",
    "print (experiment_id[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get metadata of dataset 1\n",
    "#experiment_id[0] = '86507520'\n",
    "\n",
    "metadata_list = []\n",
    "\n",
    "metadata_xml.append(results_all[0])\n",
    "metadata_list.append(json_to_list(results_all[0]))\n",
    "\n",
    "#print (pd.DataFrame(metadata_list).head(1000))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2, 18)\n",
      "                                                   0  \\\n",
      "0  [abstract, authors, doc_type, document_chembl_...   \n",
      "\n",
      "                                                   1  \n",
      "0  [Given that flavonoids are known for their ult...  \n"
     ]
    }
   ],
   "source": [
    "\n",
    "print(pd.DataFrame(metadata_list[0]).shape)\n",
    "\n",
    "print(pd.DataFrame(metadata_list))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"Retrieving_repositories_in_www.re3data.org\"></a>\n",
    "### Retrieving repositories in www.re3data.org"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- ChEMBL\n",
      "-> The time to retrieve the information was 2.56 seconds\n"
     ]
    }
   ],
   "source": [
    "start_time = time.time()\n",
    "\n",
    "for case_id in range (0, len(repository_name)):  \n",
    "    \n",
    "    print (\"- \" + repository_name[case_id])\n",
    "        \n",
    "    # extract repository metadata\n",
    "    re3data_metadata_address = find_repo_in_re3_data (repository_name[case_id])\n",
    "\n",
    "    if re3data_metadata_address != None:\n",
    "        # extract repository number \n",
    "        repository_number = re3data_metadata_address[re3data_metadata_address.rfind('/') : len(re3data_metadata_address)]\n",
    "\n",
    "        # metadata_url\n",
    "        metadata_url      = re3data_repo + str(repository_number)\n",
    "\n",
    "        # access the repository metadata \n",
    "        re3data_repo_r = requests.get(metadata_url)\n",
    "\n",
    "        # put the output string in a structured xml tree\n",
    "        re3data_repo_root.append(ET.fromstring(re3data_repo_r.text)) # saved if needed for pretty_print_xml visualization\n",
    "    \n",
    "        # put the output string in a list of lists\n",
    "        re3data_metadata_list.append(xml_to_list(re3data_repo_root[case_id]))\n",
    "       \n",
    "print (\"-> The time to retrieve the information was %.2f seconds\" % ((time.time() - start_time)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"metrics\"></a>\n",
    "### FAIR assessment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"F1\"></a>\n",
    "#### F1. (Meta)data are assigned a globally unique and persistent identifier\n",
    "We used www.re3data.org to check whether the repository provides a globally unique and persistent identifier.  \n",
    "Because it is not possible to retrieve this information via API, we manually checked the \"doi\" icon at the top-right of each repository page. \n",
    "We manually assigned: \n",
    "- `1` when the hovered icon stated: *The research data repository uses DOI to make its provided data persistent, unique and citable* \n",
    "- `0` when the hovered icon stated: *The research data repository does not use a persistent identifier system*  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> F1 are: [1]\n"
     ]
    }
   ],
   "source": [
    "\n",
    "F1.append(1) # information retrieved at: https://www.re3data.org/repository/r3d100010574\n",
    "\n",
    "# Print out F1 values\n",
    "print (\"--> F1 are: \" + str(F1) )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"F2\"></a>\n",
    "#### F2. Data are described with rich metadata \n",
    "\n",
    "Metadata contain the keywords used in the query"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18\n",
      "- ChEMBL\n",
      "    ['titanium dioxide']\n",
      "    Found 1 keywords\n",
      "--> F2: [1.0]\n"
     ]
    }
   ],
   "source": [
    "if (len(metadata_list) == 0):\n",
    "    F2.append(0)\n",
    "    print (\"    No registered with re3data.org\")\n",
    "else:\n",
    "    for case_id in range (0, len(repository_name)):\n",
    "    \n",
    "        print(len(metadata_list[case_id][1]))\n",
    "        # look for keywords in metadata text \n",
    "        output, index = find_list_in_list(metadata_list[case_id][1], keywords[case_id]) # metadata_list[0] is tags, metadata_list[1] is text\n",
    "    \n",
    "        # get unique found keywords (one keyword can be present more than once\n",
    "        found = list(set(output) ) \n",
    "\n",
    "        # print out found keywords\n",
    "        print (\"- \" + repository_name[case_id])\n",
    "        print (\"    \" + str(found))\n",
    "    \n",
    "        # n of keywords found - used in R1\n",
    "        n_of_found_keywords.append(len(found))\n",
    "\n",
    "        # assign F2\n",
    "        if len(found) == 0: # none of the search keywords are in the metafile data\n",
    "            F2.append(0)\n",
    "            print (\"    No keyword found\")\n",
    "        else:\n",
    "            F2.append(len(found)/len(keywords[case_id]))\n",
    "            print (\"    Found \" + str(len(found)) + \" keywords\")\n",
    "\n",
    "# print F2\n",
    "print (\"--> F2: \" + str(F2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"F3\"></a>\n",
    "#### F3. Metadata clearly and explicitly include the identifier of the data it describes  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- ChEMBL\n",
      "id: CHEMBL1157425\n",
      "--> F3: [1]\n"
     ]
    }
   ],
   "source": [
    "if (len(metadata_list) == 0):\n",
    "    F2.append(0)\n",
    "    print (\"    No registered with re3data.org\")\n",
    "else:\n",
    "    for case_id in range (0, len(repository_name)):\n",
    "    \n",
    "        # look for keywords in metadata tag\n",
    "        output, index = find_list_in_list(metadata_list[case_id][0], [\"id\"]) # metadata_list[0] is tags, metadata_list[1] is text\n",
    "        print (\"- \" + repository_name[case_id])\n",
    "    \n",
    "        # both in Parkinson_AE and NBIA_GEO metadata there are various \"id\" tags. The one that refers to the datset is the first (=> index[0])\n",
    "        print (\"id: \" + metadata_list[case_id][1][index[0]])\n",
    "    \n",
    "        F3.append(1)\n",
    "          \n",
    "print (\"--> F3: \" + str(F3) )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"F4\"></a>\n",
    "#### F4. (Meta)data are registered or indexed in a searchable resource\n",
    "\n",
    "Manual search for the dataset title in [Google Dataset Search](https://toolbox.google.com/datasetsearch)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Title to be searched on Google Dataset: \n",
      "ChEMBL\n",
      "--> F4: [1]\n"
     ]
    }
   ],
   "source": [
    "# get dataset titles\n",
    "\n",
    "# TEMPLATE: REPOSITORY\n",
    "case_id = 0\n",
    "if (len(metadata_list) == 0):\n",
    "    print (\"    No registered with re3data.org\")\n",
    "else:\n",
    "    output, index = find_list_in_list(metadata_list[case_id][0], [\"title\"]) # metadata_list[0] is tags, metadata_list[1] is text\n",
    "    dataset_title.append(metadata_list[case_id][1][index[0]]) # there are 4 tags \"name\". the dataset title is the first (=> index[0])\n",
    "        \n",
    "    # print out\n",
    "    print (\"Title to be searched on Google Dataset: \")\n",
    "    print (\"ChEMBL\")\n",
    "    \n",
    "# manually find dataset in Google Dataset Search\n",
    "F4.append(1)    # Parkinson_AE: https://toolbox.google.com/datasetsearch/search?query=Transcription%20profiling%20by%20array%20of%20human%20frontal%20cortex%20(area%208)%20from%20Parkinson%20disease%20cases%20and%20controls&docid=eYq%2F%2Bx3Qkyz18DYbAAAAAA%3D%3D\n",
    "\n",
    "print (\"--> F4: \" + str(F4) )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"A1\"></a>\n",
    "#### A1. Metadata are retrievable by their identifier using a standardized communications protocol   \n",
    "\n",
    "In this implementation we retrieve information only via HTTP, so we check whether the request response is 200 (= site responding)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> A1: [1]\n"
     ]
    }
   ],
   "source": [
    "# for all use cases\n",
    "for i in range(0, len(request_status)):\n",
    "    if request_status[i] == 200:\n",
    "        A1.append(1) \n",
    "    else:\n",
    "        A1.append(0)  \n",
    "\n",
    "print (\"--> A1: \" + str(A1)) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"A11\"></a>\n",
    "#### A1.1 The protocol is open, free, and universally implementable\n",
    "\n",
    "In this implementation we retrieve information only via HTTP, so if A1=1 then A11=1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> A11: [1]\n"
     ]
    }
   ],
   "source": [
    "# for all use cases\n",
    "for case_id in range (0, len(A1)):\n",
    "    if A1[case_id] == 1:\n",
    "        A11.append(1)\n",
    "    else:\n",
    "        A11.append(None)\n",
    "        \n",
    "print (\"--> A11: \" + str(A11)) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"A12\"></a>\n",
    "#### A1.2 The protocol allows for an authentication and authorization procedure, where necessary\n",
    "In this implementation we retrieve information only via HTTP, so if A1=1 then A11=1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> A12: [1]\n"
     ]
    }
   ],
   "source": [
    "# for all use cases\n",
    "for case_id in range (0, len(A1)):\n",
    "    if A1[case_id] == 1:\n",
    "        A12.append(1)\n",
    "    else:\n",
    "        A12.append(None)\n",
    "        \n",
    "print (\"--> A12: \" + str(A12)) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "source": [
    "<a name=\"A2\"></a>\n",
    "#### A2. Metadata are accessible, even when the data are no longer available  \n",
    "\n",
    "We look for `data availability policy` in the repository metadata extracted from www.re3data.org"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- ChEMBL\n",
      "    Policy not found\n",
      "-> The time to retrieve the information was 0.00 seconds\n",
      "--> A2: [0]\n"
     ]
    }
   ],
   "source": [
    "start_time = time.time()\n",
    "\n",
    "for case_id in range (0, len(repository_name)):  \n",
    "    \n",
    "    print (\"- \" + repository_name[case_id])\n",
    "            \n",
    "    # find the policy\n",
    "    if len(re3data_metadata_list) == 0:\n",
    "        A2.append(0)\n",
    "    else:\n",
    "        output, index = find_list_in_list(re3data_metadata_list[case_id][1], [\"data availability  policy\"])\n",
    "        if len(index) != 0: \n",
    "            print (\"    Policy at: \" + re3data_metadata_list[case_id][1][index[0]+1]) # \n",
    "            A2.append(1)\n",
    "        else:\n",
    "            print (\"    Policy not found\")\n",
    "            A2.append(0)\n",
    "        \n",
    "print (\"-> The time to retrieve the information was %.2f seconds\" % ((time.time() - start_time)))\n",
    "\n",
    "print (\"--> A2: \" + str(A2)) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"I1\"></a>\n",
    "#### I1. Metadata use a formal, accessible, shared, and broadly applicable language for knowledge representation \n",
    "\n",
    "We manually check the metadata format"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> I1: [1]\n"
     ]
    }
   ],
   "source": [
    "I1.append(1) # Parkinson_AE  : XML (https://www.ebi.ac.uk/arrayexpress/help/programmatic_access.html#Format_XML_results)\n",
    "\n",
    "print (\"--> I1: \" + str(I1) )    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"I2\"></a>\n",
    "#### I2. Metadata use vocabularies that follow FAIR principles\n",
    "We did not assess this principle, so we assign `None` to the metric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> I2: [-0.001]\n"
     ]
    }
   ],
   "source": [
    "# Parkinson_AE\n",
    "I2.append(-0.001) \n",
    "\n",
    "print (\"--> I2: \" + str(I2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"I3\"></a>\n",
    "#### I3. (Meta)data include qualified references to other (meta)data  \n",
    "Manually checked if there are references in metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> I3: [1]\n"
     ]
    }
   ],
   "source": [
    "# Parkinson_AE\n",
    "I3.append(1)\n",
    "\n",
    "print (\"--> I3: \" + str(I3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"R1\"></a>\n",
    "#### R1 (meta)data are richly described with a plurality of accurate and relevant attributes \n",
    "Check if there are more metadata than the keywords used in F2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of metadata excluding keywords: 17\n",
      "--> R1: [1]\n"
     ]
    }
   ],
   "source": [
    "for case_id in range (0, len(repository_name)):\n",
    "    \n",
    "    if (len(metadata_list) > 0):\n",
    "        total_number_of_metadata = len(metadata_list[case_id][0])\n",
    "        remaining_metadata = total_number_of_metadata - n_of_found_keywords[case_id]\n",
    "    \n",
    "        if remaining_metadata > 0:\n",
    "            R1.append(1)\n",
    "            print (\"Number of metadata excluding keywords: \" + str(remaining_metadata))\n",
    "        else: \n",
    "            R1.append(0)\n",
    "    else:\n",
    "        R1.append(0)\n",
    "\n",
    "print (\"--> R1: \" + str(R1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"R11\"></a>\n",
    "#### R1.1. Metadata are released with a clear and accessible data usage license \n",
    "\n",
    "We look for `datalicensename` and `datalicenseurl` in the repository metadata extracted from www.re3data.org"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- ChEMBL\n",
      "    License name: CC\n",
      "    License url : https://creativecommons.org/licenses/by-sa/3.0/\n",
      "    License name: other\n",
      "    License url : https://www.ebi.ac.uk/about\n",
      "--> R11: [1]\n"
     ]
    }
   ],
   "source": [
    "for case_id in range (0, len(repository_name)):\n",
    "\n",
    "    print (\"- \" + repository_name[case_id])\n",
    "   \n",
    "    if (len(re3data_metadata_list) > 0):\n",
    "        # find the policy and its url\n",
    "        output_name, index_name = find_list_in_list(re3data_metadata_list[case_id][0], [\"datalicensename\"])\n",
    "        output_url , index_url  = find_list_in_list(re3data_metadata_list[case_id][0], [\"datalicenseurl\"])\n",
    "        for i in range(0, len(index_name)): \n",
    "            print (\"    License name: \" + re3data_metadata_list[case_id][1][index_name[i]])\n",
    "            print (\"    License url : \" + re3data_metadata_list[case_id][1][index_url[i]])\n",
    "    \n",
    "        if output_url != []:\n",
    "            R11.append(1) \n",
    "        else: \n",
    "            R11.append(0)\n",
    "    else:\n",
    "        R11.append(0)\n",
    "\n",
    "print (\"--> R11: \" + str(R11)) \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"R12\"></a>\n",
    "#### R1.2. (Meta)data are associated with detailed provenance  \n",
    "\n",
    "We check whether the dataset metadata contain `authors`, `email` of at least one author, and `title` of related publication "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- ChEMBL\n",
      "    authors: Choquenet B, Couteau C, Paparis E, Coiffard LJ.\n",
      "    title: Journal of natural products.\n",
      "--> R12: [0.5]\n"
     ]
    }
   ],
   "source": [
    "for case_id in range (0, len(repository_name)):\n",
    "    \n",
    "    print (\"- \" + repository_name[case_id])\n",
    "    \n",
    "    # counter \n",
    "    provenance = 0\n",
    "    \n",
    "    if len(metadata_list) > 0:\n",
    "        # author names\n",
    "        output, index = find_list_in_list(metadata_list[case_id][0], [\"authors\"]) # metadata_list[0] is tags, metadata_list[1] is text\n",
    "        if index != []:\n",
    "            print (\"    authors: \" + metadata_list[case_id][1][index[0]])\n",
    "            provenance = provenance + 1\n",
    "    \n",
    "        # author email\n",
    "        output, index = find_list_in_list(metadata_list[case_id][0], [\"email\"]) # metadata_list[0] is tags, metadata_list[1] is text\n",
    "        if len(index) != 0:\n",
    "            print (\"    email: \" + metadata_list[case_id][1][index[0]])\n",
    "            provenance = provenance + 1\n",
    "    \n",
    "        # publication title\n",
    "        output, index = find_list_in_list(metadata_list[case_id][0], [\"title\"]) # metadata_list[0] is tags, metadata_list[1] is text\n",
    "        if index != []:\n",
    "            print (\"    title: \" + metadata_list[case_id][1][index[0]])\n",
    "            provenance = provenance + 1\n",
    "    \n",
    "        # calculate R12\n",
    "        if provenance == 3:\n",
    "            R12.append(1)\n",
    "        elif provenance == 0: \n",
    "            R12.append(0)\n",
    "        else:\n",
    "            R12.append(0.5)\n",
    "    else:\n",
    "        R12.append(0)\n",
    "\n",
    "print (\"--> R12: \" + str(R12))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"R13\"></a>\n",
    "#### R1.3. (Meta)data meet domain-relevant community standards\n",
    "We do not evaluate this metric because of current lack of community standards. We assign `None` to the metric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> R13: [-0.001]\n"
     ]
    }
   ],
   "source": [
    "# Parkinson_AE\n",
    "R13.append(-0.001)\n",
    "\n",
    "print (\"--> R13: \" + str(R13))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"visualization\"></a>\n",
    "### FAIR assessment visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "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>TiO2_ChEMBL</th>\n",
       "      <th>principles</th>\n",
       "      <th>metric_names</th>\n",
       "      <th>retrieval_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>F1</th>\n",
       "      <td>1.000</td>\n",
       "      <td>F</td>\n",
       "      <td>F1</td>\n",
       "      <td>manual</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>F2</th>\n",
       "      <td>1.000</td>\n",
       "      <td>F</td>\n",
       "      <td>F2</td>\n",
       "      <td>automatic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>F3</th>\n",
       "      <td>1.000</td>\n",
       "      <td>F</td>\n",
       "      <td>F3</td>\n",
       "      <td>automatic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>F4</th>\n",
       "      <td>1.000</td>\n",
       "      <td>F</td>\n",
       "      <td>F4</td>\n",
       "      <td>manual</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1</th>\n",
       "      <td>1.000</td>\n",
       "      <td>A</td>\n",
       "      <td>A1</td>\n",
       "      <td>automatic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A11</th>\n",
       "      <td>1.000</td>\n",
       "      <td>A</td>\n",
       "      <td>A11</td>\n",
       "      <td>automatic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A12</th>\n",
       "      <td>1.000</td>\n",
       "      <td>A</td>\n",
       "      <td>A12</td>\n",
       "      <td>automatic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2</th>\n",
       "      <td>0.000</td>\n",
       "      <td>A</td>\n",
       "      <td>A2</td>\n",
       "      <td>automatic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>I1</th>\n",
       "      <td>1.000</td>\n",
       "      <td>I</td>\n",
       "      <td>I1</td>\n",
       "      <td>manual</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>I2</th>\n",
       "      <td>-0.001</td>\n",
       "      <td>I</td>\n",
       "      <td>I2</td>\n",
       "      <td>N/A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>I3</th>\n",
       "      <td>1.000</td>\n",
       "      <td>I</td>\n",
       "      <td>I3</td>\n",
       "      <td>manual</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R1</th>\n",
       "      <td>1.000</td>\n",
       "      <td>R</td>\n",
       "      <td>R1</td>\n",
       "      <td>automatic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R11</th>\n",
       "      <td>1.000</td>\n",
       "      <td>R</td>\n",
       "      <td>R11</td>\n",
       "      <td>automatic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R12</th>\n",
       "      <td>0.500</td>\n",
       "      <td>R</td>\n",
       "      <td>R12</td>\n",
       "      <td>automatic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R13</th>\n",
       "      <td>-0.001</td>\n",
       "      <td>R</td>\n",
       "      <td>R13</td>\n",
       "      <td>N/A</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     TiO2_ChEMBL principles metric_names retrieval_type\n",
       "F1         1.000          F           F1         manual\n",
       "F2         1.000          F           F2      automatic\n",
       "F3         1.000          F           F3      automatic\n",
       "F4         1.000          F           F4         manual\n",
       "A1         1.000          A           A1      automatic\n",
       "A11        1.000          A          A11      automatic\n",
       "A12        1.000          A          A12      automatic\n",
       "A2         0.000          A           A2      automatic\n",
       "I1         1.000          I           I1         manual\n",
       "I2        -0.001          I           I2            N/A\n",
       "I3         1.000          I           I3         manual\n",
       "R1         1.000          R           R1      automatic\n",
       "R11        1.000          R          R11      automatic\n",
       "R12        0.500          R          R12      automatic\n",
       "R13       -0.001          R          R13            N/A"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "# create empty dataframe \n",
    "df = pd.DataFrame()\n",
    "\n",
    "data = {\n",
    "        \"F1\"    : F1,\n",
    "        \"F2\"    : F2,\n",
    "        \"F3\"    : F3,\n",
    "        \"F4\"    : F4,\n",
    "        \"A1\"    : A1,\n",
    "        \"A11\"   : A11,\n",
    "        \"A12\"   : A12, \n",
    "        \"A2\"    : A2,\n",
    "        \"I1\"    : I1,\n",
    "        \"I2\"    : I2,\n",
    "        \"I3\"    : I3,\n",
    "        \"R1\"    : R1,\n",
    "        \"R11\"   : R11,   \n",
    "        \"R12\"   : R12,     \n",
    "        \"R13\"   : R13,\n",
    "\n",
    "\n",
    "       }\n",
    "\n",
    "# create dataframe\n",
    "df = pd.DataFrame (data)\n",
    "\n",
    "# transpose for plotting (to be retranspose in plotting)\n",
    "df = df.T\n",
    "\n",
    "# change column names\n",
    "df.columns = ['TiO2_ChEMBL']\n",
    "\n",
    "\n",
    "\n",
    "# add criteria for plotting\n",
    "principles         = [\"F\",      \"F\",         \"F\",         \"F\",      \"A\",         \"A\",         \"A\",         \"A\",         \"I\",       \"I\",             \"I\",      \"R\",         \"R\",         \"R\"          ,\"R\"]\n",
    "metric_names       = [\"F1\",     \"F2\",        \"F3\",        \"F4\",     \"A1\",        \"A11\",       \"A12\",       \"A2\",        \"I1\",      \"I2\",            \"I3\",     \"R1\",        \"R11\",       \"R12\"        ,\"R13\" ]\n",
    "retrieval_type     = [\"manual\", \"automatic\", \"automatic\", \"manual\", \"automatic\", \"automatic\", \"automatic\", \"automatic\", \"manual\",  \"N/A\",  \"manual\", \"automatic\", \"automatic\", \"automatic\" , \"N/A\" ]\n",
    "df[\"principles\"]     = principles\n",
    "df[\"metric_names\"]   = metric_names\n",
    "df[\"retrieval_type\"] = retrieval_type\n",
    "\n",
    "# display\n",
    "pd.set_option(\"display.max_rows\",df.shape[0])\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "R[write to console]: Saving 6.5 x 3.5 in image\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R -i df -w 6.5 -h 3.5 --units in -r 300 #-i df \n",
    "\n",
    "library(ggplot2)\n",
    "\n",
    "# variables\n",
    "y_values   <- seq(1, nrow(df))\n",
    "y_values   <- sort(y_values, decreasing = TRUE)\n",
    "n_of_cases <- ncol(df)-3\n",
    "#n_of_cases <- 4\n",
    "case_names <- rev(colnames(df)[1:n_of_cases])\n",
    "# font size\n",
    "title_text_size <- 12\n",
    "axis_text_size  <- 9\n",
    "tick_text_size  <- 8\n",
    "\n",
    "# initiate graph\n",
    "pl <- ggplot (data=df, aes(x=x_values1, y=y_values, color=principles, shape=retrieval_type))\n",
    "\n",
    "# add cases in for loop\n",
    "for (i in (1:n_of_cases)){\n",
    "\n",
    "    # create a new dataframe with what needed for the current repository\n",
    "    # (if one keeps indexes in geom_point, ggplot plots only the last very line - see https://stackoverflow.com/questions/18444620/set-layers-in-ggplot2-via-loop)\n",
    "    gg.data <- data.frame(current_df=df[i], x_values=rep(i, nrow(df)), principles=df$principles, retrieval_type=df$retrieval_type)\n",
    "   \n",
    "    # change specific column name (e.g.DANSEASY) to a generic name\n",
    "    colnames(gg.data)[1]<-\"dataset\"\n",
    "    \n",
    "    pl <- pl + \n",
    "          geom_point(data=gg.data, aes(x=x_values, y=y_values, size=dataset, color=principles, shape=retrieval_type)) + # add size, color, and shape\n",
    "          geom_point(data=gg.data, aes(x=x_values, y=y_values, size=dataset, fill=principles, shape=retrieval_type), alpha=0.5, show.legend = FALSE) # add fill but do not show the legend (it would be the same as for color)\n",
    "}\n",
    "\n",
    "# add layers to plot\n",
    "pl <- pl +\n",
    "        # change shapes to squares, circles, and crosses\n",
    "        scale_shape_manual(values = c(21,23,4)) +\n",
    "        # background\n",
    "        theme_bw() +  \n",
    "        # flip x and y\n",
    "        coord_flip()  +\n",
    "        # axis ticks\n",
    "        scale_x_reverse(limits=c(2.5,0.5), breaks=seq(n_of_cases,1,-1),  labels=c(colnames(df)[n_of_cases:1])) +  \n",
    "        scale_y_reverse(limits=c(15, 1),   breaks=seq(15,1,-1), labels=df$metric_names) + \n",
    "        theme(axis.text.x = element_text(size = tick_text_size)) + \n",
    "        theme(axis.text.y = element_text(size = tick_text_size)) +\n",
    "        # axis text\n",
    "        xlab (\"Use cases\") +\n",
    "        ylab (\"FAIR guidelines\") +\n",
    "        theme(text = element_text( size = axis_text_size)) + \n",
    "        # change colors for lines\n",
    "        scale_color_manual(values=c(\"#e31a1c\", \"#1f78b4\", \"#33a02c\", \"#ff7f00\") ,\n",
    "                           labels = c(\"F\", \"A\", \"I\", \"R\"),\n",
    "                           breaks = c(\"F\", \"A\", \"I\", \"R\")\n",
    "                          ) +  \n",
    "        # change colors for fills\n",
    "        scale_fill_manual(values=c(\"#e31a1c\", \"#1f78b4\", \"#33a02c\", \"#ff7f00\")) +\n",
    "        # title\n",
    "        ggtitle(\"FAIR maturity indicators\") +\n",
    "        theme(plot.title = element_text(size = title_text_size, hjust = 0.5))  +\n",
    "        # legend - colors\n",
    "        theme(legend.text = element_text(size=tick_text_size), legend.key.size = unit(0.2, \"cm\")) + \n",
    "        labs(color=\"\") +                             # no legend title   \n",
    "        labs(size=\"\")  +                             # no legend title \n",
    "        labs(shape=\"\") +                             # no legend title \n",
    "        theme (legend.spacing.x = unit(0.1, \"cm\"),   # reduce horizontal space between legends\n",
    "               legend.box = \"horizontal\",            # put legends next to each other\n",
    "               legend.direction = \"vertical\",        # \n",
    "               legend.position = \"bottom\",           # under the graph\n",
    "               legend.key.size = unit(1.3, 'lines'), # space better legend items\n",
    "               legend.margin=margin(c(1,7,1,1))) +   # margin around each legend \n",
    "        guides(color = guide_legend(order=2),        # change legend order\n",
    "               size  = guide_legend(order=1),\n",
    "               shape = guide_legend(order=3))\n",
    "        \n",
    "\n",
    "# display\n",
    "ggsave(\"chembl.png\", plot = pl)\n",
    "\n",
    "pl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%R \n",
    "# requires the package \"svglite\" to save as svg\n",
    "#ggsave(\"outcome_comparison.svg\", plot = last_plot(), width=6.5, height=3.5, dpi=300) # comment if used in mybinder, uncomment if used locally "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"dependencies\"></a>\n",
    "## Dependencies "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%reload_ext watermark\n",
    "%watermark -v -m -p bs4,os,pandas,requests,xml.etree.ElementTree,io\n",
    "print (\"\")\n",
    "%watermark -u -n -t -z"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%R\n",
    "sessionInfo()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%watermark -u -n -t -z"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"supplementary_information\"></a>\n",
    "## Supplementary information     "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Metadata of the dataset in xml:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for case_id in range (0, 2):\n",
    "\n",
    "    print (\"--> \" + repository_name[case_id] + \" :\")\n",
    "    pretty_print_xml(metadata_xml[case_id])\n",
    "    print (\" \")\n",
    "    print (\"---------------------------------------------------------------\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Metadata of the dataset in lists (no nesting):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for case_id in range (0, 2):\n",
    "\n",
    "    print (\"--> \" + repository_name[case_id] + \" :\")\n",
    "    \n",
    "    for a in range(0,len(metadata_list[case_id][0])):\n",
    "        print (metadata_list[case_id][0][a] + \": \" + metadata_list[case_id][1][a])\n",
    "    print (\" \")\n",
    "    print (\"---------------------------------------------------------------\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Metadata of the repositories in xml (no nesting):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for case_id in range (0, 2):\n",
    "\n",
    "    print (\"--> \" + repository_name[case_id] + \" :\")\n",
    "    pretty_print_xml(re3data_repo_root[case_id])\n",
    "    print (\" \")\n",
    "    print (\"---------------------------------------------------------------\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Metadata of the repositories in lists (no nesting):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for case_id in range (0, len(repository_name)):\n",
    "\n",
    "    print (\"--> \" + repository_name[case_id] + \" :\")\n",
    "    \n",
    "    \n",
    "    for a in range(0,len(re3data_metadata_list[case_id][0])):\n",
    "        print (re3data_metadata_list[case_id][0][a] + \": \" + re3data_metadata_list[case_id][1][a])\n",
    "    print (\" \")\n",
    "    print (\"---------------------------------------------------------------\")"
   ]
  }
 ],
 "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"
  },
  "toc-autonumbering": false,
  "toc-showcode": false,
  "toc-showmarkdowntxt": false,
  "toc-showtags": true
 },
 "nbformat": 4,
 "nbformat_minor": 4
}