j_10_VirtualScreening.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "10c59d90",
      "metadata": {
        "id": "10c59d90"
      },
      "source": [
        "# 가상 스크리닝\n",
        "\n",
        "\n",
        "- 어떤 조건을 만족하는 분자를 찾아내는 작업을 화학 실험을 수행하지 않고 머신러닝 모델로 찾는 것을 말한다\n",
        "- 여기서는 예제로 ERK2 단백질을 억제하는 분자를 선별하는 모델을 만든다\n",
        " - Extracellular signal-regulated kinase (다수의 생화학적 신호에 관여하고 세포의 증식 분화 전사조절에 관여)\n",
        " - 비소세포성 폐암 non-small cell lung cancer과 흑색종(피부암)에 대해 임상시험중\n",
        "\n",
        "- 그래프 컨볼류션 모델을 사용한다"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "6cdaff62",
      "metadata": {
        "id": "6cdaff62"
      },
      "source": [
        "# import"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "id": "3a89036f",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3a89036f",
        "outputId": "66859b45-454b-484d-e19f-54c762fcfc54"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: DeepChem in /usr/local/lib/python3.10/dist-packages (2.8.0)\n",
            "Requirement already satisfied: joblib in /usr/local/lib/python3.10/dist-packages (from DeepChem) (1.4.2)\n",
            "Requirement already satisfied: numpy>=1.21 in /usr/local/lib/python3.10/dist-packages (from DeepChem) (1.26.4)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (from DeepChem) (2.1.4)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.10/dist-packages (from DeepChem) (1.5.2)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from DeepChem) (1.13.3)\n",
            "Requirement already satisfied: scipy>=1.10.1 in /usr/local/lib/python3.10/dist-packages (from DeepChem) (1.13.1)\n",
            "Requirement already satisfied: rdkit in /usr/local/lib/python3.10/dist-packages (from DeepChem) (2024.3.5)\n",
            "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.10/dist-packages (from pandas->DeepChem) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas->DeepChem) (2024.2)\n",
            "Requirement already satisfied: tzdata>=2022.1 in /usr/local/lib/python3.10/dist-packages (from pandas->DeepChem) (2024.1)\n",
            "Requirement already satisfied: Pillow in /usr/local/lib/python3.10/dist-packages (from rdkit->DeepChem) (10.4.0)\n",
            "Requirement already satisfied: threadpoolctl>=3.1.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn->DeepChem) (3.5.0)\n",
            "Requirement already satisfied: mpmath<1.4,>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from sympy->DeepChem) (1.3.0)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.2->pandas->DeepChem) (1.16.0)\n",
            "Requirement already satisfied: tensorflow==2.9 in /usr/local/lib/python3.10/dist-packages (2.9.0)\n",
            "Requirement already satisfied: absl-py>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (1.4.0)\n",
            "Requirement already satisfied: astunparse>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (1.6.3)\n",
            "Requirement already satisfied: flatbuffers<2,>=1.12 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (1.12)\n",
            "Requirement already satisfied: gast<=0.4.0,>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (0.4.0)\n",
            "Requirement already satisfied: google-pasta>=0.1.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (0.2.0)\n",
            "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (1.64.1)\n",
            "Requirement already satisfied: h5py>=2.9.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (3.11.0)\n",
            "Requirement already satisfied: keras<2.10.0,>=2.9.0rc0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (2.9.0)\n",
            "Requirement already satisfied: keras-preprocessing>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (1.1.2)\n",
            "Requirement already satisfied: libclang>=13.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (18.1.1)\n",
            "Requirement already satisfied: numpy>=1.20 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (1.26.4)\n",
            "Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (3.3.0)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (24.1)\n",
            "Requirement already satisfied: protobuf>=3.9.2 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (3.19.6)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (71.0.4)\n",
            "Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (1.16.0)\n",
            "Requirement already satisfied: tensorboard<2.10,>=2.9 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (2.9.1)\n",
            "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (0.37.1)\n",
            "Requirement already satisfied: tensorflow-estimator<2.10.0,>=2.9.0rc0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (2.9.0)\n",
            "Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (2.4.0)\n",
            "Requirement already satisfied: typing-extensions>=3.6.6 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (4.12.2)\n",
            "Requirement already satisfied: wrapt>=1.11.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.9) (1.16.0)\n",
            "Requirement already satisfied: wheel<1.0,>=0.23.0 in /usr/local/lib/python3.10/dist-packages (from astunparse>=1.6.0->tensorflow==2.9) (0.44.0)\n",
            "Requirement already satisfied: google-auth<3,>=1.6.3 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.10,>=2.9->tensorflow==2.9) (2.27.0)\n",
            "Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.10,>=2.9->tensorflow==2.9) (0.4.6)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.10,>=2.9->tensorflow==2.9) (3.7)\n",
            "Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.10,>=2.9->tensorflow==2.9) (2.32.3)\n",
            "Requirement already satisfied: tensorboard-data-server<0.7.0,>=0.6.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.10,>=2.9->tensorflow==2.9) (0.6.1)\n",
            "Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.10,>=2.9->tensorflow==2.9) (1.8.1)\n",
            "Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.10,>=2.9->tensorflow==2.9) (3.0.4)\n",
            "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.10,>=2.9->tensorflow==2.9) (5.5.0)\n",
            "Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.10,>=2.9->tensorflow==2.9) (0.4.1)\n",
            "Requirement already satisfied: rsa<5,>=3.1.4 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.10,>=2.9->tensorflow==2.9) (4.9)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.10,>=2.9->tensorflow==2.9) (1.3.1)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.10,>=2.9->tensorflow==2.9) (3.3.2)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.10,>=2.9->tensorflow==2.9) (3.10)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.10,>=2.9->tensorflow==2.9) (2.2.3)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.10,>=2.9->tensorflow==2.9) (2024.8.30)\n",
            "Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.10/dist-packages (from werkzeug>=1.0.1->tensorboard<2.10,>=2.9->tensorflow==2.9) (2.1.5)\n",
            "Requirement already satisfied: pyasn1<0.7.0,>=0.4.6 in /usr/local/lib/python3.10/dist-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.10,>=2.9->tensorflow==2.9) (0.6.1)\n",
            "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.10/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.10,>=2.9->tensorflow==2.9) (3.2.2)\n"
          ]
        }
      ],
      "source": [
        "!pip install DeepChem\n",
        "!pip install tensorflow==2.9"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "id": "77060ccf",
      "metadata": {
        "id": "77060ccf",
        "outputId": "2f56b466-3794-48a9-cee3-6e7d82bf50a1",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:deepchem.feat.molecule_featurizers.rdkit_descriptors:No normalization for SPS. Feature removed!\n",
            "WARNING:deepchem.feat.molecule_featurizers.rdkit_descriptors:No normalization for AvgIpc. Feature removed!\n",
            "WARNING:deepchem.models.torch_models:Skipped loading modules with pytorch-geometric dependency, missing a dependency. No module named 'torch_geometric'\n",
            "WARNING:deepchem.models:Skipped loading modules with pytorch-geometric dependency, missing a dependency. cannot import name 'DMPNN' from 'deepchem.models.torch_models' (/usr/local/lib/python3.10/dist-packages/deepchem/models/torch_models/__init__.py)\n",
            "WARNING:deepchem.models:Skipped loading modules with pytorch-lightning dependency, missing a dependency. No module named 'lightning'\n",
            "WARNING:deepchem.models:Skipped loading some Jax models, missing a dependency. No module named 'haiku'\n"
          ]
        }
      ],
      "source": [
        "import deepchem as dc\n",
        "from rdkit import Chem\n",
        "from rdkit.Chem import AllChem\n",
        "from rdkit.Chem import Draw\n",
        "from rdkit.Chem.Draw import IPythonConsole\n",
        "from rdkit.Chem import PandasTools\n",
        "from rdkit.Chem import Descriptors\n",
        "from rdkit.Chem import rdmolops\n",
        "from rdkit import DataStructs\n",
        "from rdkit.ML.Cluster import Butina\n",
        "from rdkit.Chem import rdMolDescriptors as rdmd\n",
        "from deepchem.models import GraphConvModel\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "id": "9f7985a0",
      "metadata": {
        "id": "9f7985a0"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import sys\n",
        "import seaborn as sns\n",
        "%config InlineBackend.figure_format = 'retina'\n",
        "from tqdm.auto import tqdm\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "6fba2c6e",
      "metadata": {
        "id": "6fba2c6e"
      },
      "source": [
        "# 데이터\n",
        "\n",
        "- DUD-E 데이터베이스에서 활성 및 비활성 분자 데이터를 다운받아 사용한다\n",
        "- 분자 데이터는 SMILES 로 표현되어 있으며 이를 데이터프레임에 불러온다\n",
        "- AddMoleculeColumnToFrame()를 사용해 SMILES에 해당하는 Mol 객체를 데이터프레임에 추가할 수 있다"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "id": "eff258f6",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 161
        },
        "id": "eff258f6",
        "outputId": "38bbd17d-25f0-4ee0-d185-6f230769af9a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(79, 3)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                   0       1             2\n",
              "0  Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCC...  168691  CHEMBL318804\n",
              "1  C[C@@]12[C@@H]([C@@H](CC(O1)n3c4ccccc4c5c3c6n2...   86358     CHEMBL162\n",
              "2  Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...  575087  CHEMBL576683"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8b9346fc-2b67-4466-8b45-89cfcf344e46\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCC...</td>\n",
              "      <td>168691</td>\n",
              "      <td>CHEMBL318804</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>C[C@@]12[C@@H]([C@@H](CC(O1)n3c4ccccc4c5c3c6n2...</td>\n",
              "      <td>86358</td>\n",
              "      <td>CHEMBL162</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...</td>\n",
              "      <td>575087</td>\n",
              "      <td>CHEMBL576683</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8b9346fc-2b67-4466-8b45-89cfcf344e46')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-8b9346fc-2b67-4466-8b45-89cfcf344e46 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-8b9346fc-2b67-4466-8b45-89cfcf344e46');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-fb468182-2709-410b-9261-796c55cdddc2\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-fb468182-2709-410b-9261-796c55cdddc2')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-fb468182-2709-410b-9261-796c55cdddc2 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"active_df[:3]\",\n  \"rows\": 3,\n  \"fields\": [\n    {\n      \"column\": 0,\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCCN5CCOCC5)OC\",\n          \"C[C@@]12[C@@H]([C@@H](CC(O1)n3c4ccccc4c5c3c6n2c7ccccc7c6c8c5C(=O)NC8)NC)OC\",\n          \"Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3)Cl)Nc4cccc5c4OC(O5)(F)F\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 1,\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 261658,\n        \"min\": 86358,\n        \"max\": 575087,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          168691,\n          86358,\n          575087\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 2,\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"CHEMBL318804\",\n          \"CHEMBL162\",\n          \"CHEMBL576683\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 5
        }
      ],
      "source": [
        "active_df = pd.read_csv(\"https://raw.githubusercontent.com/deepchem/DeepLearningLifeSciences/master/Chapter11/mk01/actives_final.ism\",\n",
        "                        header=None,sep=\" \")\n",
        "print(active_df.shape)\n",
        "active_df[:3]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "id": "f81c1b60",
      "metadata": {
        "id": "f81c1b60"
      },
      "outputs": [],
      "source": [
        "active_rows, active_cols = active_df.shape\n",
        "active_df.columns = [\"SMILES\",\"ID\",\"ChEMBL_ID\"]\n",
        "active_df[\"label\"] = [\"Active\"]*active_rows\n",
        "PandasTools.AddMoleculeColumnToFrame(active_df,\"SMILES\",\"Mol\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "id": "49112d7b",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "49112d7b",
        "outputId": "056ea9b8-07d4-47c3-fe23-e8efe15fb22c"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                              SMILES      ID     ChEMBL_ID  \\\n",
              "0  Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCC...  168691  CHEMBL318804   \n",
              "1  C[C@@]12[C@@H]([C@@H](CC(O1)n3c4ccccc4c5c3c6n2...   86358     CHEMBL162   \n",
              "2  Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...  575087  CHEMBL576683   \n",
              "\n",
              "    label                                               Mol  \n",
              "0  Active  <rdkit.Chem.rdchem.Mol object at 0x7f13297f4970>  \n",
              "1  Active  <rdkit.Chem.rdchem.Mol object at 0x7f13297f49e0>  \n",
              "2  Active  <rdkit.Chem.rdchem.Mol object at 0x7f13297f4a50>  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-4c634650-54e9-4519-b2eb-78504e2574e9\" class=\"colab-df-container\">\n",
              "    <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>SMILES</th>\n",
              "      <th>ID</th>\n",
              "      <th>ChEMBL_ID</th>\n",
              "      <th>label</th>\n",
              "      <th>Mol</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCC...</td>\n",
              "      <td>168691</td>\n",
              "      <td>CHEMBL318804</td>\n",
              "      <td>Active</td>\n",
              "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7f13297f4970&gt;</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>C[C@@]12[C@@H]([C@@H](CC(O1)n3c4ccccc4c5c3c6n2...</td>\n",
              "      <td>86358</td>\n",
              "      <td>CHEMBL162</td>\n",
              "      <td>Active</td>\n",
              "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7f13297f49e0&gt;</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...</td>\n",
              "      <td>575087</td>\n",
              "      <td>CHEMBL576683</td>\n",
              "      <td>Active</td>\n",
              "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7f13297f4a50&gt;</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4c634650-54e9-4519-b2eb-78504e2574e9')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-4c634650-54e9-4519-b2eb-78504e2574e9 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-4c634650-54e9-4519-b2eb-78504e2574e9');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-eb5d166b-1212-4417-9e53-44727afd7c62\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-eb5d166b-1212-4417-9e53-44727afd7c62')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-eb5d166b-1212-4417-9e53-44727afd7c62 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"active_df[:3]\",\n  \"rows\": 3,\n  \"fields\": [\n    {\n      \"column\": \"SMILES\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCCN5CCOCC5)OC\",\n          \"C[C@@]12[C@@H]([C@@H](CC(O1)n3c4ccccc4c5c3c6n2c7ccccc7c6c8c5C(=O)NC8)NC)OC\",\n          \"Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3)Cl)Nc4cccc5c4OC(O5)(F)F\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ID\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 261658,\n        \"min\": 86358,\n        \"max\": 575087,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          168691,\n          86358,\n          575087\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ChEMBL_ID\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"CHEMBL318804\",\n          \"CHEMBL162\",\n          \"CHEMBL576683\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"label\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"Active\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Mol\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"<rdkit.Chem.rdchem.Mol object at 0x7f13297f4970>\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 7
        }
      ],
      "source": [
        "active_df[:3]"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "8758aacf",
      "metadata": {
        "id": "8758aacf"
      },
      "source": [
        "## 특성값 추가\n",
        "- 분자량 molecular weight\n",
        "- 분배 계수 partition coefficeint(LogP)\n",
        "- 전하 charge (양인지 음인지)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "id": "caf6c80f",
      "metadata": {
        "id": "caf6c80f"
      },
      "outputs": [],
      "source": [
        "def add_property_columns_to_df(df_in):\n",
        "    df_in[\"mw\"] = [Descriptors.MolWt(mol) for mol in df_in.Mol]\n",
        "    df_in[\"logP\"] = [Descriptors.MolLogP(mol) for mol in df_in.Mol]\n",
        "    df_in[\"charge\"] = [rdmolops.GetFormalCharge(mol) for mol in df_in.Mol]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "id": "a0693035",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "a0693035",
        "outputId": "3ee17107-9eab-4210-e644-b5a06178a0c0"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                              SMILES      ID     ChEMBL_ID  \\\n",
              "0  Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCC...  168691  CHEMBL318804   \n",
              "1  C[C@@]12[C@@H]([C@@H](CC(O1)n3c4ccccc4c5c3c6n2...   86358     CHEMBL162   \n",
              "2  Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...  575087  CHEMBL576683   \n",
              "3  Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...  575065  CHEMBL571484   \n",
              "4  Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...  575047  CHEMBL568937   \n",
              "\n",
              "    label                                               Mol       mw     logP  \\\n",
              "0  Active  <rdkit.Chem.rdchem.Mol object at 0x7f13297f4970>  565.099  5.49788   \n",
              "1  Active  <rdkit.Chem.rdchem.Mol object at 0x7f13297f49e0>  466.541  4.35400   \n",
              "2  Active  <rdkit.Chem.rdchem.Mol object at 0x7f13297f4a50>  527.915  4.96202   \n",
              "3  Active  <rdkit.Chem.rdchem.Mol object at 0x7f13297f4ac0>  491.935  4.36922   \n",
              "4  Active  <rdkit.Chem.rdchem.Mol object at 0x7f13297f4b30>  487.991  5.12922   \n",
              "\n",
              "   charge  \n",
              "0       0  \n",
              "1       0  \n",
              "2       0  \n",
              "3       0  \n",
              "4       0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-fb9ae785-977b-4a75-bd69-1e9ff0f8ca30\" class=\"colab-df-container\">\n",
              "    <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>SMILES</th>\n",
              "      <th>ID</th>\n",
              "      <th>ChEMBL_ID</th>\n",
              "      <th>label</th>\n",
              "      <th>Mol</th>\n",
              "      <th>mw</th>\n",
              "      <th>logP</th>\n",
              "      <th>charge</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCC...</td>\n",
              "      <td>168691</td>\n",
              "      <td>CHEMBL318804</td>\n",
              "      <td>Active</td>\n",
              "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7f13297f4970&gt;</td>\n",
              "      <td>565.099</td>\n",
              "      <td>5.49788</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>C[C@@]12[C@@H]([C@@H](CC(O1)n3c4ccccc4c5c3c6n2...</td>\n",
              "      <td>86358</td>\n",
              "      <td>CHEMBL162</td>\n",
              "      <td>Active</td>\n",
              "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7f13297f49e0&gt;</td>\n",
              "      <td>466.541</td>\n",
              "      <td>4.35400</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...</td>\n",
              "      <td>575087</td>\n",
              "      <td>CHEMBL576683</td>\n",
              "      <td>Active</td>\n",
              "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7f13297f4a50&gt;</td>\n",
              "      <td>527.915</td>\n",
              "      <td>4.96202</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...</td>\n",
              "      <td>575065</td>\n",
              "      <td>CHEMBL571484</td>\n",
              "      <td>Active</td>\n",
              "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7f13297f4ac0&gt;</td>\n",
              "      <td>491.935</td>\n",
              "      <td>4.36922</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...</td>\n",
              "      <td>575047</td>\n",
              "      <td>CHEMBL568937</td>\n",
              "      <td>Active</td>\n",
              "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7f13297f4b30&gt;</td>\n",
              "      <td>487.991</td>\n",
              "      <td>5.12922</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-fb9ae785-977b-4a75-bd69-1e9ff0f8ca30')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-fb9ae785-977b-4a75-bd69-1e9ff0f8ca30 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-fb9ae785-977b-4a75-bd69-1e9ff0f8ca30');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-e34109e7-ae0f-44e4-a355-066ebc2bedff\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-e34109e7-ae0f-44e4-a355-066ebc2bedff')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-e34109e7-ae0f-44e4-a355-066ebc2bedff button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "active_df",
              "summary": "{\n  \"name\": \"active_df\",\n  \"rows\": 79,\n  \"fields\": [\n    {\n      \"column\": \"SMILES\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 79,\n        \"samples\": [\n          \"c1cc(cc(c1)Cl)c2cn[nH]c2c3cc([nH]c3)C(=O)NC(CO)c4cccc(c4)F\",\n          \"Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCCN5CCOCC5)OC\",\n          \"c1c\\\\2c([nH]c1Br)C(=O)NCC/C2=C\\\\3/C(=O)NC(=N3)N\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ID\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 138975,\n        \"min\": 43524,\n        \"max\": 620576,\n        \"num_unique_values\": 79,\n        \"samples\": [\n          370057,\n          168691,\n          250364\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ChEMBL_ID\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 79,\n        \"samples\": [\n          \"CHEMBL219849\",\n          \"CHEMBL318804\",\n          \"CHEMBL359106\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"label\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"Active\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Mol\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 79,\n        \"samples\": [\n          \"<rdkit.Chem.rdchem.Mol object at 0x7f13297f5690>\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"mw\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 66.32895304427547,\n        \"min\": 243.05999999999997,\n        \"max\": 565.0990000000003,\n        \"num_unique_values\": 69,\n        \"samples\": [\n          502.0180000000003\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"logP\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.5156389133024573,\n        \"min\": -0.7704000000000004,\n        \"max\": 5.752080000000006,\n        \"num_unique_values\": 72,\n        \"samples\": [\n          5.129220000000003\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"charge\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 0,\n        \"max\": 0,\n        \"num_unique_values\": 1,\n        \"samples\": [\n          0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 9
        }
      ],
      "source": [
        "add_property_columns_to_df(active_df)\n",
        "active_df.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "796885c6",
      "metadata": {
        "id": "796885c6"
      },
      "source": [
        "## 비활성화 데이터"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "id": "f7d811fd",
      "metadata": {
        "id": "f7d811fd"
      },
      "outputs": [],
      "source": [
        "decoy_df = pd.read_csv(\"https://raw.githubusercontent.com/deepchem/DeepLearningLifeSciences/master/Chapter11/mk01/decoys_final.ism\",\n",
        "                       header=None,sep=\" \")\n",
        "decoy_df.columns = [\"SMILES\",\"ID\"]\n",
        "decoy_rows, decoy_cols = decoy_df.shape\n",
        "decoy_df[\"label\"] = [\"Decoy\"]*decoy_rows\n",
        "PandasTools.AddMoleculeColumnToFrame(decoy_df,\"SMILES\",\"Mol\")\n",
        "add_property_columns_to_df(decoy_df)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "ea38b37d",
      "metadata": {
        "id": "ea38b37d"
      },
      "source": [
        "- 활성 및 비확성 데이터를 합친다"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "id": "edf5384f",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "edf5384f",
        "outputId": "86ddd5c3-fe49-42ce-ab6d-8899cf7caa65"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(4629, 8)"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ],
      "source": [
        "tmp_df = pd.concat([active_df, decoy_df])\n",
        "tmp_df.shape"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "829552c0",
      "metadata": {
        "id": "829552c0"
      },
      "source": [
        "## 데이터 탐색\n",
        "\n",
        "- violin plot"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "id": "4757b0e6",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "4757b0e6",
        "outputId": "e890172c-0e3c-41e1-e4a2-eaf6719cd31f"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 571,
              "height": 432
            }
          }
        }
      ],
      "source": [
        "sns.violinplot(x=tmp_df[\"label\"],y=tmp_df[\"mw\"]);"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "4ec18535",
      "metadata": {
        "id": "4ec18535"
      },
      "source": [
        "- 위 그래프를 보면 분자량 분포는 비슷하다\n",
        "- 400mw 이하의 분자들이 비활성 분자 데이터셋에 좀 더 많이 들어 있다"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "id": "287190d1",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "287190d1",
        "outputId": "4bd63dc2-91df-4cef-9105-7aebf16e8f99"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 565,
              "height": 432
            }
          }
        }
      ],
      "source": [
        "sns.violinplot(x=tmp_df[\"label\"],y=tmp_df[\"logP\"]);"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "id": "4da4aef6",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "4da4aef6",
        "outputId": "175e3d11-4fab-432c-888d-c64ea80ebdb4"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 579,
              "height": 432
            }
          }
        }
      ],
      "source": [
        "sns.violinplot(x=tmp_df[\"label\"],y=tmp_df[\"charge\"]);"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "02fdf707",
      "metadata": {
        "id": "02fdf707"
      },
      "source": [
        "- 활성 분자는 모두 중성이다\n",
        "- 비활성 분자는 일부 +1과 -1이 있다"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "id": "f2e11d9e",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "f2e11d9e",
        "outputId": "5ab12afc-beb0-4d10-8c80-ebe94c6fac59"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.16175824175824177"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ],
      "source": [
        "charged = decoy_df[decoy_df[\"charge\"] != 0]\n",
        "charged.shape[0]/decoy_df.shape[0]"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "fcbb5201",
      "metadata": {
        "id": "fcbb5201"
      },
      "source": [
        "- 비활성 분자의 16%가 전하를 띠고 있다\n",
        "- 전하 차이에 의해 머신러닝 모델이 바이어스가 생기게 되므로, 비활성 분자의 화학 구조를 수정하여 전하를 제거하겠다\n",
        "- 아래는 이를 위한 함수 정의이다"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "id": "a1910a82",
      "metadata": {
        "id": "a1910a82"
      },
      "outputs": [],
      "source": [
        "def _InitialiseNeutralisationReactions():\n",
        "    patts= (\n",
        "        # Imidazoles\n",
        "        ('[n+;H]','n'),\n",
        "        # Amines\n",
        "        ('[N+;!H0]','N'),\n",
        "        # Carboxylic acids and alcohols\n",
        "        ('[$([O-]);!$([O-][#7])]','O'),\n",
        "        # Thiols\n",
        "        ('[S-;X1]','S'),\n",
        "        # Sulfonamides\n",
        "        ('[$([N-;X2]S(=O)=O)]','N'),\n",
        "        # Enamines\n",
        "        ('[$([N-;X2][C,N]=C)]','N'),\n",
        "        # Tetrazoles\n",
        "        ('[n-]','[nH]'),\n",
        "        # Sulfoxides\n",
        "        ('[$([S-]=O)]','S'),\n",
        "        # Amides\n",
        "        ('[$([N-]C=O)]','N'),\n",
        "        )\n",
        "    return [(Chem.MolFromSmarts(x),Chem.MolFromSmiles(y,False)) for x,y in patts]\n",
        "\n",
        "_reactions=None\n",
        "\n",
        "def NeutraliseCharges(smiles, reactions=None):\n",
        "    global _reactions\n",
        "    if reactions is None:\n",
        "        if _reactions is None:\n",
        "            _reactions=_InitialiseNeutralisationReactions()\n",
        "        reactions=_reactions\n",
        "    mol = Chem.MolFromSmiles(smiles)\n",
        "    replaced = False\n",
        "    for i,(reactant, product) in enumerate(reactions):\n",
        "        while mol.HasSubstructMatch(reactant):\n",
        "            replaced = True\n",
        "            rms = AllChem.ReplaceSubstructs(mol, reactant, product)\n",
        "            mol = rms[0]\n",
        "    if replaced:\n",
        "        return (Chem.MolToSmiles(mol,True), True)\n",
        "    else:\n",
        "        return (smiles, False)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "id": "b48a4cb0",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 161
        },
        "id": "b48a4cb0",
        "outputId": "c57fd37b-e871-4a3c-fbc6-27595b751ce8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(4550, 3)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                              SMILES         ID  label\n",
              "0                         C1COCCNC(=S)NCCOCCNC(=S)N1  C03897368  Decoy\n",
              "1            c1cc(ccc1[C@H]2c3ccc(cc3OC(=C2C#N)N)N)O  C00242247  Decoy\n",
              "2  C[C@]12CC[C@H](C[C@@H]1CC[C@@H]3[C@@H]2C(=O)C[...  C04096778  Decoy"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-9bb18b45-df23-43f0-81ea-6ffd9ffc0ba0\" class=\"colab-df-container\">\n",
              "    <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>SMILES</th>\n",
              "      <th>ID</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>C1COCCNC(=S)NCCOCCNC(=S)N1</td>\n",
              "      <td>C03897368</td>\n",
              "      <td>Decoy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>c1cc(ccc1[C@H]2c3ccc(cc3OC(=C2C#N)N)N)O</td>\n",
              "      <td>C00242247</td>\n",
              "      <td>Decoy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>C[C@]12CC[C@H](C[C@@H]1CC[C@@H]3[C@@H]2C(=O)C[...</td>\n",
              "      <td>C04096778</td>\n",
              "      <td>Decoy</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9bb18b45-df23-43f0-81ea-6ffd9ffc0ba0')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-9bb18b45-df23-43f0-81ea-6ffd9ffc0ba0 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-9bb18b45-df23-43f0-81ea-6ffd9ffc0ba0');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-70aad9f3-5c3a-4f61-9138-eeccd70edc8d\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-70aad9f3-5c3a-4f61-9138-eeccd70edc8d')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-70aad9f3-5c3a-4f61-9138-eeccd70edc8d button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"revised_decoy_df[:3]\",\n  \"rows\": 3,\n  \"fields\": [\n    {\n      \"column\": \"SMILES\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"C1COCCNC(=S)NCCOCCNC(=S)N1\",\n          \"c1cc(ccc1[C@H]2c3ccc(cc3OC(=C2C#N)N)N)O\",\n          \"C[C@]12CC[C@H](C[C@@H]1CC[C@@H]3[C@@H]2C(=O)C[C@@]4([C@@H]3CC[C@@]4([C@@H](CO)O)O)C)O\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ID\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"C03897368\",\n          \"C00242247\",\n          \"C04096778\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"label\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"Decoy\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 17
        }
      ],
      "source": [
        "revised_decoy_df = decoy_df[[\"SMILES\",\"ID\",\"label\"]].copy()\n",
        "revised_decoy_df[\"SMILES\"] = [NeutraliseCharges(x)[0] for x in revised_decoy_df[\"SMILES\"]]\n",
        "print(revised_decoy_df.shape)\n",
        "revised_decoy_df[:3]"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "9d5187e4",
      "metadata": {
        "id": "9d5187e4"
      },
      "source": [
        "- 중성화된 SMILES는 첫 번째 컬럼에 들어 있다"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "9266a494",
      "metadata": {
        "id": "9266a494"
      },
      "source": [
        "- 특성을 다시 계산한다\n",
        "- 전하를 띤 샘플의 수가 감소했다"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "id": "41d21038",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 467
        },
        "id": "41d21038",
        "outputId": "58ae7fae-9e61-4cde-8434-ea1ee2fd7d5e"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: xlabel='label', ylabel='charge'>"
            ]
          },
          "metadata": {},
          "execution_count": 18
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 587,
              "height": 432
            }
          }
        }
      ],
      "source": [
        "PandasTools.AddMoleculeColumnToFrame(revised_decoy_df,\"SMILES\",\"Mol\")\n",
        "add_property_columns_to_df(revised_decoy_df)\n",
        "new_tmp_df = pd.concat([active_df, revised_decoy_df])\n",
        "sns.violinplot(x=new_tmp_df[\"label\"],y=new_tmp_df[\"charge\"])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "id": "048b74d5",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "048b74d5",
        "outputId": "ec965590-2696-4c7e-caa1-1ff15d2c92fe"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.0026373626373626374"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ],
      "source": [
        "charged = revised_decoy_df[revised_decoy_df[\"charge\"] != 0]\n",
        "charged.shape[0]/revised_decoy_df.shape[0]"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "3debefec",
      "metadata": {
        "id": "3debefec"
      },
      "source": [
        "- 활성을 1로 비활성을 0으로 인코딩한다"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "id": "327b0f74",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "327b0f74",
        "outputId": "1d930138-ae10-46c0-eb43-fb906020f41a"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                              SMILES      ID  is_active\n",
              "0  Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCC...  168691          1\n",
              "1  C[C@@]12[C@@H]([C@@H](CC(O1)n3c4ccccc4c5c3c6n2...   86358          1\n",
              "2  Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...  575087          1\n",
              "3  Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...  575065          1\n",
              "4  Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...  575047          1"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-6a981211-acac-427c-a050-17c5441df7a4\" class=\"colab-df-container\">\n",
              "    <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>SMILES</th>\n",
              "      <th>ID</th>\n",
              "      <th>is_active</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Cn1ccnc1Sc2ccc(cc2Cl)Nc3c4cc(c(cc4ncc3C#N)OCCC...</td>\n",
              "      <td>168691</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>C[C@@]12[C@@H]([C@@H](CC(O1)n3c4ccccc4c5c3c6n2...</td>\n",
              "      <td>86358</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...</td>\n",
              "      <td>575087</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...</td>\n",
              "      <td>575065</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...</td>\n",
              "      <td>575047</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6a981211-acac-427c-a050-17c5441df7a4')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-6a981211-acac-427c-a050-17c5441df7a4 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-6a981211-acac-427c-a050-17c5441df7a4');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-e610b26e-2711-4904-a96c-7fd86269cf93\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-e610b26e-2711-4904-a96c-7fd86269cf93')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-e610b26e-2711-4904-a96c-7fd86269cf93 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "combined_df",
              "summary": "{\n  \"name\": \"combined_df\",\n  \"rows\": 4629,\n  \"fields\": [\n    {\n      \"column\": \"SMILES\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4629,\n        \"samples\": [\n          \"C[C@H]1CCCC[C@]12C(=O)N(C(=O)N2)CC(=O)N[C@@H](C)c3cc4ccccc4o3\",\n          \"c1cc(c(cc1Cl)Cl)CS(=O)(=O)/C(=C/NC(=S)N)/c2ccc(cc2Cl)Cl\",\n          \"c1cc2c(cc1O)[C@@H]3C=CC[C@H]3[C@@H](N2)c4cc(cc(c4O)Cl)Cl\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ID\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4627,\n        \"samples\": [\n          \"C02520166\",\n          \"C00845575\",\n          \"C14466883\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"is_active\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 0,\n        \"max\": 1,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0,\n          1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 20
        }
      ],
      "source": [
        "active_df[\"is_active\"] = [1] * active_df.shape[0]\n",
        "revised_decoy_df[\"is_active\"] = [0] * revised_decoy_df.shape[0]\n",
        "combined_df = pd.concat([active_df, revised_decoy_df])[[\"SMILES\",\"ID\",\"is_active\"]]\n",
        "combined_df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "id": "dcaba5cf",
      "metadata": {
        "id": "dcaba5cf"
      },
      "outputs": [],
      "source": [
        "# 전처리한 데이터를 파일로 저장해둔다\n",
        "combined_df.to_csv(\"dude_erk1_mk01.csv\")"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "999d254d",
      "metadata": {
        "id": "999d254d"
      },
      "source": [
        "# 분류 모델\n",
        "\n",
        "- GCN 모델을 사용한다"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "6f82357f",
      "metadata": {
        "id": "6f82357f"
      },
      "source": [
        "##  그래프 컨볼류션 모델 생성"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "id": "13ad4121",
      "metadata": {
        "id": "13ad4121"
      },
      "outputs": [],
      "source": [
        "def generate_graph_conv_model():\n",
        "    batch_size = 128\n",
        "    model = GraphConvModel(1, batch_size=batch_size,\n",
        "             mode='classification', model_dir=\"./model_dir\")\n",
        "    return model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "id": "f5b1b662",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "f5b1b662",
        "outputId": "3d6065ec-cf79-40b6-ead7-667ef4a7f269"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<DiskDataset X.shape: (4629,), y.shape: (4629, 1), w.shape: (4629, 1), task_names: ['is_active']>"
            ]
          },
          "metadata": {},
          "execution_count": 23
        }
      ],
      "source": [
        "dataset_file = \"dude_erk1_mk01.csv\"\n",
        "tasks = [\"is_active\"]\n",
        "featurizer = dc.feat.ConvMolFeaturizer()\n",
        "loader = dc.data.CSVLoader(tasks=tasks, feature_field=\"SMILES\", featurizer=featurizer)\n",
        "dataset = loader.create_dataset(dataset_file, shard_size=8192)\n",
        "dataset"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "eccffd7c",
      "metadata": {
        "id": "eccffd7c"
      },
      "source": [
        "- RandomSplitter() 사용 - 데이터를 래덤하게 나눈다\n",
        "- 화학 구조를 기반으로 데이터셋을 나누는 ScaffoleSplit(), 데이터를 군집화 한 후에 데이터셋을 나누는 ButinaSplitter()도 있다\n",
        "- 불균형 데이터에 대한 성능 평가시에는 매튜 상관계수 Matthews Correlarions Coefficients(MCC)를 사용한다\n",
        " - +1이면 완벽한 예측, 0이면 랜덤 예측, -1이면 완전히 반대되는 예측을 의미한다\n",
        "- 모델을 5회 만들어 평균을 구한다 (랜덤 교차검증)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "id": "d116ff56",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 120,
          "referenced_widgets": [
            "1bf89b2ab0c04a5bb748248f4f6776e2",
            "7fa66993224b4d1db04e8490f7ccce5b",
            "fba78101f549436a91ff7bbac7fa2442",
            "a165afbc87ea448ab12365a001621028",
            "266a0218ed8149918108d754703048be",
            "08f443373a714fe4bd38db256f6bf2b2",
            "d951bcfb0d9a49f5ae737587b5cee713",
            "f8acfd5a0d474e3fbab1b492071ec7b4",
            "1e05a6df783049328ccc94c23f4ca319",
            "ac98529749e7459b84d92b3dba11d0b6",
            "97b74e82e83a4bf0a3f82af3799a092c"
          ]
        },
        "id": "d116ff56",
        "outputId": "d18c0271-dc20-45ce-ce4a-01371b5c22b2"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/5 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "1bf89b2ab0c04a5bb748248f4f6776e2"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0.5823908519865434, 0.7275225405977639, 0.8289791101351985, 0.8143774003802622, 0.8108074851854256]\n",
            "[0.7047972264664374, 0.5760992378838319, 0.8433069208152789, 0.6741431187975722, 0.910853532390761]\n",
            "0.7528154776570387\n",
            "0.7418400072707761\n"
          ]
        }
      ],
      "source": [
        "splitter = dc.splits.RandomSplitter()\n",
        "metrics = [dc.metrics.Metric(dc.metrics.matthews_corrcoef, np.mean)]\n",
        "\n",
        "training_score_list = []\n",
        "validation_score_list = []\n",
        "transformers = []\n",
        "cv_folds = 5\n",
        "for i in tqdm(range(0,cv_folds)):\n",
        "    model = generate_graph_conv_model()\n",
        "    train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split(dataset)\n",
        "    model.fit(train_dataset)\n",
        "    train_scores = model.evaluate(train_dataset, metrics, transformers)\n",
        "    training_score_list.append(train_scores[\"mean-matthews_corrcoef\"])\n",
        "    validation_scores = model.evaluate(valid_dataset, metrics, transformers)\n",
        "    validation_score_list.append(validation_scores[\"mean-matthews_corrcoef\"])\n",
        "print(training_score_list)\n",
        "print(validation_score_list)\n",
        "print(np.array(training_score_list).mean())\n",
        "print(np.array(validation_score_list).mean())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "id": "3b71c3f4",
      "metadata": {
        "id": "3b71c3f4",
        "outputId": "d05b1d4f-b78a-4fef-d602-68d89607cf97",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(0.7528154776570387, 0.7418400072707761)"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ],
      "source": [
        "np.array(training_score_list).mean(), np.array(validation_score_list).mean()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "id": "e0095980",
      "metadata": {
        "id": "e0095980",
        "outputId": "15085310-0726-412d-cf91-2b6133d8d933",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        }
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 556,
              "height": 413
            }
          }
        }
      ],
      "source": [
        "sns.boxplot(x=[\"training\"]*cv_folds+[\"validation\"]*cv_folds,y=training_score_list+validation_score_list);"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "be3ebfe1",
      "metadata": {
        "id": "be3ebfe1"
      },
      "source": [
        "## 검증 데이터 분류 예측"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "id": "0b57f7d5",
      "metadata": {
        "id": "0b57f7d5",
        "outputId": "935e64b7-e47d-4427-a5b5-59abe8242ca4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        neg       pos  active  \\\n",
              "0  0.999800  0.000200       0   \n",
              "1  0.999981  0.000019       0   \n",
              "2  0.999964  0.000036       0   \n",
              "3  0.999617  0.000383       0   \n",
              "4  0.999937  0.000062       0   \n",
              "\n",
              "                                              SMILES  \n",
              "0            COc1ccccc1c2nc(cs2)CN3C(=O)c4ccccc4C3=O  \n",
              "1  CCOc1cc(ccc1O)/C=N/NC(=O)CCCSc2nnc(n2N)c3ccccc3Br  \n",
              "2  C[C@@H]1CCCC[C@H]1NC(=O)NC(=O)CSc2nnc(n2N)c3cc...  \n",
              "3  C[C@H]1CCc2c(sc(c2C(=O)N)NC(=O)/C(=C/c3ccccc3O...  \n",
              "4  CC1=C([C@H](NC(=S)N1)c2cccc(c2)OC)C(=O)Nc3ccc(...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c81e7ec2-43ac-4c72-ad3c-1d6cff27d4c1\" class=\"colab-df-container\">\n",
              "    <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>neg</th>\n",
              "      <th>pos</th>\n",
              "      <th>active</th>\n",
              "      <th>SMILES</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.999800</td>\n",
              "      <td>0.000200</td>\n",
              "      <td>0</td>\n",
              "      <td>COc1ccccc1c2nc(cs2)CN3C(=O)c4ccccc4C3=O</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.999981</td>\n",
              "      <td>0.000019</td>\n",
              "      <td>0</td>\n",
              "      <td>CCOc1cc(ccc1O)/C=N/NC(=O)CCCSc2nnc(n2N)c3ccccc3Br</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.999964</td>\n",
              "      <td>0.000036</td>\n",
              "      <td>0</td>\n",
              "      <td>C[C@@H]1CCCC[C@H]1NC(=O)NC(=O)CSc2nnc(n2N)c3cc...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.999617</td>\n",
              "      <td>0.000383</td>\n",
              "      <td>0</td>\n",
              "      <td>C[C@H]1CCc2c(sc(c2C(=O)N)NC(=O)/C(=C/c3ccccc3O...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.999937</td>\n",
              "      <td>0.000062</td>\n",
              "      <td>0</td>\n",
              "      <td>CC1=C([C@H](NC(=S)N1)c2cccc(c2)OC)C(=O)Nc3ccc(...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c81e7ec2-43ac-4c72-ad3c-1d6cff27d4c1')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-c81e7ec2-43ac-4c72-ad3c-1d6cff27d4c1 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-c81e7ec2-43ac-4c72-ad3c-1d6cff27d4c1');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-dd5f40e8-c33f-4b37-aae5-688ece4fee6d\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-dd5f40e8-c33f-4b37-aae5-688ece4fee6d')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-dd5f40e8-c33f-4b37-aae5-688ece4fee6d button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "pred_df",
              "summary": "{\n  \"name\": \"pred_df\",\n  \"rows\": 463,\n  \"fields\": [\n    {\n      \"column\": \"neg\",\n      \"properties\": {\n        \"dtype\": \"float32\",\n        \"num_unique_values\": 444,\n        \"samples\": [\n          0.9994465112686157,\n          0.9802221655845642,\n          0.9998664855957031\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"pos\",\n      \"properties\": {\n        \"dtype\": \"float32\",\n        \"num_unique_values\": 461,\n        \"samples\": [\n          0.0001277863630093634,\n          0.021417411044239998,\n          5.743855126638664e-06\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"active\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 0,\n        \"max\": 1,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          1,\n          0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"SMILES\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 463,\n        \"samples\": [\n          \"Cc1ccccc1C(=O)Nc2ccc(cc2)C(=O)NNC(=S)NC(=O)/C=C/c3ccccc3\",\n          \"C[C@@H](C(=O)N/N=C/c1c2ccccc2ccc1OCc3ccccc3)Nc4c(=O)[nH]c(=O)[nH]n4\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 27
        }
      ],
      "source": [
        "pred = [x.flatten() for x in model.predict(valid_dataset)]\n",
        "pred_df = pd.DataFrame(pred,columns=[\"neg\",\"pos\"])\n",
        "pred_df[\"active\"] = [int(x) for x in valid_dataset.y]\n",
        "pred_df[\"SMILES\"] = valid_dataset.ids\n",
        "pred_df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "id": "73ea15ae",
      "metadata": {
        "id": "73ea15ae",
        "outputId": "fa162945-3c83-4f63-cc08-555e67cf16ea",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          neg       pos  active  \\\n",
              "284  0.040565  0.959435       1   \n",
              "103  0.044354  0.955646       1   \n",
              "338  0.063739  0.936261       1   \n",
              "245  0.093508  0.906492       1   \n",
              "332  0.122533  0.877467       1   \n",
              "\n",
              "                                                SMILES  \n",
              "284  c1ccc(cc1)CNC(=O)c2cc(c[nH]2)c3c(cn[nH]3)c4ccc...  \n",
              "103  c1cc(cc(c1)Cl)c2cn[nH]c2c3cc([nH]c3)C(=O)NC(CO...  \n",
              "338  CN(Cc1ccccc1)C(=O)c2cc(c[nH]2)c3c(cn[nH]3)c4cc...  \n",
              "245    CCN(C)C(=O)c1cc(c[nH]1)c2c(cn[nH]2)c3cccc(c3)Cl  \n",
              "332  Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-be0d755b-6dc6-489b-9f0c-493f65a24d29\" class=\"colab-df-container\">\n",
              "    <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>neg</th>\n",
              "      <th>pos</th>\n",
              "      <th>active</th>\n",
              "      <th>SMILES</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>284</th>\n",
              "      <td>0.040565</td>\n",
              "      <td>0.959435</td>\n",
              "      <td>1</td>\n",
              "      <td>c1ccc(cc1)CNC(=O)c2cc(c[nH]2)c3c(cn[nH]3)c4ccc...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>103</th>\n",
              "      <td>0.044354</td>\n",
              "      <td>0.955646</td>\n",
              "      <td>1</td>\n",
              "      <td>c1cc(cc(c1)Cl)c2cn[nH]c2c3cc([nH]c3)C(=O)NC(CO...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>338</th>\n",
              "      <td>0.063739</td>\n",
              "      <td>0.936261</td>\n",
              "      <td>1</td>\n",
              "      <td>CN(Cc1ccccc1)C(=O)c2cc(c[nH]2)c3c(cn[nH]3)c4cc...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>245</th>\n",
              "      <td>0.093508</td>\n",
              "      <td>0.906492</td>\n",
              "      <td>1</td>\n",
              "      <td>CCN(C)C(=O)c1cc(c[nH]1)c2c(cn[nH]2)c3cccc(c3)Cl</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>332</th>\n",
              "      <td>0.122533</td>\n",
              "      <td>0.877467</td>\n",
              "      <td>1</td>\n",
              "      <td>Cc1cnc(nc1c2cc([nH]c2)C(=O)N[C@H](CO)c3cccc(c3...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-be0d755b-6dc6-489b-9f0c-493f65a24d29')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-be0d755b-6dc6-489b-9f0c-493f65a24d29 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-be0d755b-6dc6-489b-9f0c-493f65a24d29');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-e06cf1db-ff39-4bed-b3f0-91df14fe47b7\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-e06cf1db-ff39-4bed-b3f0-91df14fe47b7')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-e06cf1db-ff39-4bed-b3f0-91df14fe47b7 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"pred_df\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"neg\",\n      \"properties\": {\n        \"dtype\": \"float32\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.04435395449399948,\n          0.12253258377313614,\n          0.06373900920152664\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"pos\",\n      \"properties\": {\n        \"dtype\": \"float32\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.9556459188461304,\n          0.8774673938751221,\n          0.936260998249054\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"active\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 1,\n        \"max\": 1,\n        \"num_unique_values\": 1,\n        \"samples\": [\n          1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"SMILES\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"c1cc(cc(c1)Cl)c2cn[nH]c2c3cc([nH]c3)C(=O)NC(CO)c4ccc(cc4)F\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 28
        }
      ],
      "source": [
        "pred_df.sort_values(\"pos\",ascending=False).head(5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "id": "c5920409",
      "metadata": {
        "id": "c5920409",
        "outputId": "d0124002-62d4-4fbb-e4c2-3b0652e91c5c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 470
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: xlabel='active', ylabel='pos'>"
            ]
          },
          "metadata": {},
          "execution_count": 29
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 567,
              "height": 435
            }
          }
        }
      ],
      "source": [
        "sns.boxplot(x=pred_df.active,y=pred_df.pos)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "61ea9562",
      "metadata": {
        "id": "61ea9562"
      },
      "source": [
        "- 위 그래프를 보면 활성 분자와 비활성 분자를 잘 구분하고 있는 것을 알 수 있다\n",
        "- False Negative 샘플 찾기"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "id": "f84a0733",
      "metadata": {
        "id": "f84a0733"
      },
      "outputs": [],
      "source": [
        "false_negative_df = pred_df.query(\"active == 1 & pos < 0.5\").copy()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "id": "d358e6b4",
      "metadata": {
        "id": "d358e6b4",
        "outputId": "9a4c422f-cbd0-428d-888e-e5c1b5fea325",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 477
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         neg       pos  active  \\\n",
              "25  0.955301  0.044699       1   \n",
              "75  0.990529  0.009471       1   \n",
              "\n",
              "                                               SMILES  \\\n",
              "25  Cn1cc(c2c1nccc2c3ccc(cc3)C(=O)NCCO)/C=C\\4/C(=O...   \n",
              "75  c1ccc2c(c1)c(c[nH]2)c3c(cnc(n3)N)c4ccc(cc4)C(F...   \n",
              "\n",
              "                                                 Mol  \n",
              "25  <rdkit.Chem.rdchem.Mol object at 0x7f131c6e45f0>  \n",
              "75  <rdkit.Chem.rdchem.Mol object at 0x7f131c6e4740>  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-424e8527-27a0-46bb-9ebe-6aeed3fbcb4c\" class=\"colab-df-container\">\n",
              "    <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>neg</th>\n",
              "      <th>pos</th>\n",
              "      <th>active</th>\n",
              "      <th>SMILES</th>\n",
              "      <th>Mol</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>0.955301</td>\n",
              "      <td>0.044699</td>\n",
              "      <td>1</td>\n",
              "      <td>Cn1cc(c2c1nccc2c3ccc(cc3)C(=O)NCCO)/C=C\\4/C(=O...</td>\n",
              "      <td style=\"text-align: center;\"><div style=\"width: 200px; height: 200px\" data-content=\"rdkit/molecule\"><img src=\"data:image/png;base64,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\" alt=\"Mol\"/></div></td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75</th>\n",
              "      <td>0.990529</td>\n",
              "      <td>0.009471</td>\n",
              "      <td>1</td>\n",
              "      <td>c1ccc2c(c1)c(c[nH]2)c3c(cnc(n3)N)c4ccc(cc4)C(F...</td>\n",
              "      <td style=\"text-align: center;\"><div style=\"width: 200px; height: 200px\" data-content=\"rdkit/molecule\"><img src=\"data:image/png;base64,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\" alt=\"Mol\"/></div></td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-424e8527-27a0-46bb-9ebe-6aeed3fbcb4c')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-424e8527-27a0-46bb-9ebe-6aeed3fbcb4c button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-424e8527-27a0-46bb-9ebe-6aeed3fbcb4c');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-d8015cd1-739c-4106-bb5d-5be45f6859d4\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-d8015cd1-739c-4106-bb5d-5be45f6859d4')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-d8015cd1-739c-4106-bb5d-5be45f6859d4 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "  <div id=\"id_82cd6dca-662d-4a2f-91c4-afa4cec8ce7e\">\n",
              "    <style>\n",
              "      .colab-df-generate {\n",
              "        background-color: #E8F0FE;\n",
              "        border: none;\n",
              "        border-radius: 50%;\n",
              "        cursor: pointer;\n",
              "        display: none;\n",
              "        fill: #1967D2;\n",
              "        height: 32px;\n",
              "        padding: 0 0 0 0;\n",
              "        width: 32px;\n",
              "      }\n",
              "\n",
              "      .colab-df-generate:hover {\n",
              "        background-color: #E2EBFA;\n",
              "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "        fill: #174EA6;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate {\n",
              "        background-color: #3B4455;\n",
              "        fill: #D2E3FC;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate:hover {\n",
              "        background-color: #434B5C;\n",
              "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "        fill: #FFFFFF;\n",
              "      }\n",
              "    </style>\n",
              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('false_negative_df')\"\n",
              "            title=\"Generate code using this dataframe.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "    <script>\n",
              "      (() => {\n",
              "      const buttonEl =\n",
              "        document.querySelector('#id_82cd6dca-662d-4a2f-91c4-afa4cec8ce7e button.colab-df-generate');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      buttonEl.onclick = () => {\n",
              "        google.colab.notebook.generateWithVariable('false_negative_df');\n",
              "      }\n",
              "      })();\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "false_negative_df",
              "summary": "{\n  \"name\": \"false_negative_df\",\n  \"rows\": 2,\n  \"fields\": [\n    {\n      \"column\": \"neg\",\n      \"properties\": {\n        \"dtype\": \"float32\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0.9905290007591248,\n          0.9553009271621704\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"pos\",\n      \"properties\": {\n        \"dtype\": \"float32\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0.009471029043197632,\n          0.04469895362854004\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"active\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 1,\n        \"max\": 1,\n        \"num_unique_values\": 1,\n        \"samples\": [\n          1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"SMILES\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"c1ccc2c(c1)c(c[nH]2)c3c(cnc(n3)N)c4ccc(cc4)C(F)(F)F\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Mol\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"<rdkit.Chem.rdchem.Mol object at 0x7f131c6e4740>\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 31
        }
      ],
      "source": [
        "PandasTools.AddMoleculeColumnToFrame(false_negative_df,\"SMILES\",\"Mol\")\n",
        "false_negative_df"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "fa6f7075",
      "metadata": {
        "id": "fa6f7075"
      },
      "source": [
        "- False Positive 샘플 찾기"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "1383ff0b",
      "metadata": {
        "id": "1383ff0b"
      },
      "source": [
        "## 모든 데이터로 학습\n",
        "- 모델의 성능을 향상시키기 위해서 검증데이터를 훈련에 사용"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "id": "265d15cd",
      "metadata": {
        "id": "265d15cd",
        "outputId": "88a1db4d-bdc1-45f2-b07e-88e515113c1c",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.0003516003489494324"
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ],
      "source": [
        "model.fit(dataset)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "20a02dbd",
      "metadata": {
        "id": "20a02dbd"
      },
      "source": [
        "- 앞에서 ERK2 단백질의 활성을 억제하는 분자를 예측하는 모델을 만들었다. 여기에 새로운 데이터셋을 사용하여 분자의 활성을 예측해보겠다\n",
        " - COCONUT 데이터셋을 사용하겠다"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "a5880e8f",
      "metadata": {
        "id": "a5880e8f"
      },
      "source": [
        "## rd_filters 적용\n",
        "\n",
        "- Pat Walters가 만든 스크립트로 잠재적으로 문제가 있는 분자를 걸러낸다\n",
        "- 생물학적 분석이 어려운 분자들을 제거하기 위해서 rd_filters.py를 사용\n",
        "- [rd_filters github](https://github.com/PatWalters/rd_filters)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "id": "3e54b7dc",
      "metadata": {
        "id": "3e54b7dc"
      },
      "outputs": [],
      "source": [
        "model_backup = model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "id": "cc291f7b",
      "metadata": {
        "id": "cc291f7b",
        "scrolled": true,
        "outputId": "767eba30-5236-4cfa-a42a-96211ad1fa0e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting git+https://github.com/PatWalters/rd_filters.git\n",
            "  Cloning https://github.com/PatWalters/rd_filters.git to /tmp/pip-req-build-sgqdpiak\n",
            "  Running command git clone --filter=blob:none --quiet https://github.com/PatWalters/rd_filters.git /tmp/pip-req-build-sgqdpiak\n",
            "  Resolved https://github.com/PatWalters/rd_filters.git to commit 3630f1a59150b4a203c2f06d8376a66d1d4c3f70\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (from rd_filters==0.1) (2.1.4)\n",
            "Requirement already satisfied: docopt in /usr/local/lib/python3.10/dist-packages (from rd_filters==0.1) (0.6.2)\n",
            "Requirement already satisfied: numpy<2,>=1.22.4 in /usr/local/lib/python3.10/dist-packages (from pandas->rd_filters==0.1) (1.26.4)\n",
            "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.10/dist-packages (from pandas->rd_filters==0.1) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas->rd_filters==0.1) (2024.2)\n",
            "Requirement already satisfied: tzdata>=2022.1 in /usr/local/lib/python3.10/dist-packages (from pandas->rd_filters==0.1) (2024.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.2->pandas->rd_filters==0.1) (1.16.0)\n"
          ]
        }
      ],
      "source": [
        "# 현재 폴더에 해당 git 설치\n",
        "!pip install git+https://github.com/PatWalters/rd_filters.git"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "id": "e56aeba7",
      "metadata": {
        "id": "e56aeba7",
        "outputId": "57ac358f-2a23-4aab-ef48-5224ac127730",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Usage:\n",
            "rd_filters filter --in INPUT_FILE --prefix PREFIX [--rules RULES_FILE_NAME] [--alerts ALERT_FILE_NAME][--np NUM_CORES]\n",
            "rd_filters template --out TEMPLATE_FILE [--rules RULES_FILE_NAME]\n",
            "\n",
            "Options:\n",
            "--in INPUT_FILE input file name\n",
            "--prefix PREFIX prefix for output file names\n",
            "--rules RULES_FILE_NAME name of the rules JSON file\n",
            "--alerts ALERTS_FILE_NAME name of the structural alerts file\n",
            "--np NUM_CORES the number of cpu cores to use (default is all)\n",
            "--out TEMPLATE_FILE parameter template file name\n"
          ]
        }
      ],
      "source": [
        "# 사용법 안내\n",
        "!rd_filters -h"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "id": "hj0wafYZqKpw",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hj0wafYZqKpw",
        "outputId": "8b30a59c-8940-4ea8-c0bf-45a6952c6e66"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2024-09-29 14:53:15--  https://coconut.s3.uni-jena.de/prod/downloads/2024-09/coconut-09-2024.csv.zip\n",
            "Resolving coconut.s3.uni-jena.de (coconut.s3.uni-jena.de)... 141.35.104.26, 141.35.104.25, 2001:638:1558:2368::8d23:6819, ...\n",
            "Connecting to coconut.s3.uni-jena.de (coconut.s3.uni-jena.de)|141.35.104.26|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 88935618 (85M) [application/zip]\n",
            "Saving to: ‘coconut-09-2024.csv.zip.1’\n",
            "\n",
            "coconut-09-2024.csv 100%[===================>]  84.82M  22.8MB/s    in 4.1s    \n",
            "\n",
            "2024-09-29 14:53:19 (20.8 MB/s) - ‘coconut-09-2024.csv.zip.1’ saved [88935618/88935618]\n",
            "\n",
            "Archive:  ./coconut-09-2024.csv.zip\n",
            "replace coconut-09-2024.csv? [y]es, [n]o, [A]ll, [N]one, [r]ename: r\n",
            "new name: "
          ]
        }
      ],
      "source": [
        "# coconut database 다운로드 후 파일 압축 해제 및 파일명 확인\n",
        "! wget https://coconut.s3.uni-jena.de/prod/downloads/2024-09/coconut-09-2024.csv.zip\n",
        "!unzip ./coconut-09-2024.csv.zip"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 37,
      "id": "N-LJOLaiqJZG",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "N-LJOLaiqJZG",
        "outputId": "0f37700b-bb16-4a3e-9de3-2f28cd829400"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                           standard_inchi  \\\n",
              "0       InChI=1S/C43H53N9O14S2.Na/c1-5-22(3)35-36(57)4...   \n",
              "1       InChI=1S/C21H32O12/c1-9-14(23)16(25)18(27)21(3...   \n",
              "2       InChI=1S/C36H61N5O7/c1-21(2)18-27-35(47)48-28(...   \n",
              "3       InChI=1S/C22H22O9/c1-28-12-4-2-11(3-5-12)15-9-...   \n",
              "4       InChI=1S/C32H41N5O4/c1-6-18(4)28-32(41)36-12-8...   \n",
              "...                                                   ...   \n",
              "695128  InChI=1S/C19H19N3O4/c1-26-13-8-6-12(7-9-13)20-...   \n",
              "695129  InChI=1S/C30H30N2O10/c1-12-23(34)27(38)28(39)3...   \n",
              "695130  InChI=1S/C21H22O7/c1-11(2)4-5-13-15(23)7-6-14(...   \n",
              "695131  InChI=1S/C20H30O7/c1-17(2)4-3-12(23)18-8-27-20...   \n",
              "695132  InChI=1S/C45H38O18/c46-18-2-4-20-30(12-18)61-4...   \n",
              "\n",
              "                 standard_inchi_key  \\\n",
              "0       DRKUXFLLRIKRHH-QDVYGYDXSA-M   \n",
              "1       OXHVQSRYUNGYOK-NUASCYGXSA-N   \n",
              "2       NEGZFRNAAJQQEG-NOFCQABOSA-N   \n",
              "3       DQIVYFNWBDHNFD-WHCFWRGISA-N   \n",
              "4       HKVSEIVDIONNKB-QWNGKRCASA-N   \n",
              "...                             ...   \n",
              "695128  HHSNDFVMRMIDBG-INIZCTEOSA-N   \n",
              "695129  VVPODVCQSZKNKL-RLOKSPFPSA-N   \n",
              "695130  LMFCHRAKSGPODM-OAQYLSRUSA-N   \n",
              "695131  IJWNAKYUVUUYTE-HMBONYETSA-N   \n",
              "695132  QZNRDKYJASOYMZ-JNGZPCCCSA-N   \n",
              "\n",
              "                                         canonical_smiles    identifier  \n",
              "0       CC[C@@H]1NC(=O)[C@@H](NC(=O)[C@H](C)NC(=O)CNC(...  CNP0437004.1  \n",
              "1       COC1=CC=C(CCO[C@@H]2O[C@H](CO[C@@H]3O[C@@H](C)...  CNP0243002.1  \n",
              "2       C/C1=C\\[C@@H](C(C)(C)C)OC(=O)[C@H](CC(C)C)N(C)...  CNP0458114.1  \n",
              "3       COC1=CC=C(C2=CC(=O)OC3=CC(O[C@@H]4O[C@H](CO)[C...  CNP0252086.2  \n",
              "4       CC[C@H](C)[C@H]1C(=O)N2CCC[C@H]2C(=O)N1C(=O)[C...  CNP0107934.1  \n",
              "...                                                   ...           ...  \n",
              "695128  COC1=CC=C(NC(=O)CC[C@@H]2NC(=O)C3=CC=CC=C3NC2=...  CNP0395779.1  \n",
              "695129  CC(=O)OC1=CC=C2C(=O)C3=C(O)C(CC4=CC=CC(C(N)N)=...  CNP0097600.1  \n",
              "695130  COC1=C([C@]2(O)COC3=CC(O)=CC(O)=C3C2=O)C=CC(O)...  CNP0212403.1  \n",
              "695131  CC1(C)CC[C@H](O)[C@]23COC(O)([C@@H](O)[C@H]12)...  CNP0494455.1  \n",
              "695132  OC1=CC=C2C(=C1)O[C@H](C1=CC(O)=C(O)C(O)=C1)[C@...  CNP0300293.2  \n",
              "\n",
              "[695133 rows x 4 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f7bd7bc7-3ace-4df6-8d41-3e847bff1f87\" class=\"colab-df-container\">\n",
              "    <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>standard_inchi</th>\n",
              "      <th>standard_inchi_key</th>\n",
              "      <th>canonical_smiles</th>\n",
              "      <th>identifier</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>InChI=1S/C43H53N9O14S2.Na/c1-5-22(3)35-36(57)4...</td>\n",
              "      <td>DRKUXFLLRIKRHH-QDVYGYDXSA-M</td>\n",
              "      <td>CC[C@@H]1NC(=O)[C@@H](NC(=O)[C@H](C)NC(=O)CNC(...</td>\n",
              "      <td>CNP0437004.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>InChI=1S/C21H32O12/c1-9-14(23)16(25)18(27)21(3...</td>\n",
              "      <td>OXHVQSRYUNGYOK-NUASCYGXSA-N</td>\n",
              "      <td>COC1=CC=C(CCO[C@@H]2O[C@H](CO[C@@H]3O[C@@H](C)...</td>\n",
              "      <td>CNP0243002.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>InChI=1S/C36H61N5O7/c1-21(2)18-27-35(47)48-28(...</td>\n",
              "      <td>NEGZFRNAAJQQEG-NOFCQABOSA-N</td>\n",
              "      <td>C/C1=C\\[C@@H](C(C)(C)C)OC(=O)[C@H](CC(C)C)N(C)...</td>\n",
              "      <td>CNP0458114.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>InChI=1S/C22H22O9/c1-28-12-4-2-11(3-5-12)15-9-...</td>\n",
              "      <td>DQIVYFNWBDHNFD-WHCFWRGISA-N</td>\n",
              "      <td>COC1=CC=C(C2=CC(=O)OC3=CC(O[C@@H]4O[C@H](CO)[C...</td>\n",
              "      <td>CNP0252086.2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>InChI=1S/C32H41N5O4/c1-6-18(4)28-32(41)36-12-8...</td>\n",
              "      <td>HKVSEIVDIONNKB-QWNGKRCASA-N</td>\n",
              "      <td>CC[C@H](C)[C@H]1C(=O)N2CCC[C@H]2C(=O)N1C(=O)[C...</td>\n",
              "      <td>CNP0107934.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>695128</th>\n",
              "      <td>InChI=1S/C19H19N3O4/c1-26-13-8-6-12(7-9-13)20-...</td>\n",
              "      <td>HHSNDFVMRMIDBG-INIZCTEOSA-N</td>\n",
              "      <td>COC1=CC=C(NC(=O)CC[C@@H]2NC(=O)C3=CC=CC=C3NC2=...</td>\n",
              "      <td>CNP0395779.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>695129</th>\n",
              "      <td>InChI=1S/C30H30N2O10/c1-12-23(34)27(38)28(39)3...</td>\n",
              "      <td>VVPODVCQSZKNKL-RLOKSPFPSA-N</td>\n",
              "      <td>CC(=O)OC1=CC=C2C(=O)C3=C(O)C(CC4=CC=CC(C(N)N)=...</td>\n",
              "      <td>CNP0097600.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>695130</th>\n",
              "      <td>InChI=1S/C21H22O7/c1-11(2)4-5-13-15(23)7-6-14(...</td>\n",
              "      <td>LMFCHRAKSGPODM-OAQYLSRUSA-N</td>\n",
              "      <td>COC1=C([C@]2(O)COC3=CC(O)=CC(O)=C3C2=O)C=CC(O)...</td>\n",
              "      <td>CNP0212403.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>695131</th>\n",
              "      <td>InChI=1S/C20H30O7/c1-17(2)4-3-12(23)18-8-27-20...</td>\n",
              "      <td>IJWNAKYUVUUYTE-HMBONYETSA-N</td>\n",
              "      <td>CC1(C)CC[C@H](O)[C@]23COC(O)([C@@H](O)[C@H]12)...</td>\n",
              "      <td>CNP0494455.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>695132</th>\n",
              "      <td>InChI=1S/C45H38O18/c46-18-2-4-20-30(12-18)61-4...</td>\n",
              "      <td>QZNRDKYJASOYMZ-JNGZPCCCSA-N</td>\n",
              "      <td>OC1=CC=C2C(=C1)O[C@H](C1=CC(O)=C(O)C(O)=C1)[C@...</td>\n",
              "      <td>CNP0300293.2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>695133 rows × 4 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f7bd7bc7-3ace-4df6-8d41-3e847bff1f87')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-f7bd7bc7-3ace-4df6-8d41-3e847bff1f87 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-f7bd7bc7-3ace-4df6-8d41-3e847bff1f87');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-cae610a3-101c-48b9-b980-e353af1207ae\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-cae610a3-101c-48b9-b980-e353af1207ae')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-cae610a3-101c-48b9-b980-e353af1207ae button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "  <div id=\"id_6486e514-1f89-4c11-8b9e-c2f0f11abfb5\">\n",
              "    <style>\n",
              "      .colab-df-generate {\n",
              "        background-color: #E8F0FE;\n",
              "        border: none;\n",
              "        border-radius: 50%;\n",
              "        cursor: pointer;\n",
              "        display: none;\n",
              "        fill: #1967D2;\n",
              "        height: 32px;\n",
              "        padding: 0 0 0 0;\n",
              "        width: 32px;\n",
              "      }\n",
              "\n",
              "      .colab-df-generate:hover {\n",
              "        background-color: #E2EBFA;\n",
              "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "        fill: #174EA6;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate {\n",
              "        background-color: #3B4455;\n",
              "        fill: #D2E3FC;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate:hover {\n",
              "        background-color: #434B5C;\n",
              "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "        fill: #FFFFFF;\n",
              "      }\n",
              "    </style>\n",
              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('coconut')\"\n",
              "            title=\"Generate code using this dataframe.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "    <script>\n",
              "      (() => {\n",
              "      const buttonEl =\n",
              "        document.querySelector('#id_6486e514-1f89-4c11-8b9e-c2f0f11abfb5 button.colab-df-generate');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      buttonEl.onclick = () => {\n",
              "        google.colab.notebook.generateWithVariable('coconut');\n",
              "      }\n",
              "      })();\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "coconut"
            }
          },
          "metadata": {},
          "execution_count": 37
        }
      ],
      "source": [
        "coconut = pd.read_csv('./coconut-09-2024.csv')\n",
        "coconut"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "76d571e9",
      "metadata": {
        "id": "76d571e9"
      },
      "source": [
        "- 필터 수행\n",
        " - 필터 수행 결과 29.5%가 통과하였다\n",
        " - 생물학적 분석이 어려운 화합물을 제거\n",
        " - 필터 결과는 coconut.csv에 저장된다. 필터를 통과한 SMILES 정보는 coconut.smi에 저장된다"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 38,
      "id": "-fU22Eg1HiyO",
      "metadata": {
        "id": "-fU22Eg1HiyO"
      },
      "outputs": [],
      "source": [
        "with open('COCONUT_DB_absoluteSMILES.smi', 'w') as f:\n",
        "        for id, smile in zip(coconut['identifier'], coconut['canonical_smiles']):\n",
        "            f.write(smile + ' ' + id + '\\n')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 39,
      "id": "7e94b92d",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7e94b92d",
        "scrolled": true,
        "outputId": "af8cdc53-eab1-4e1e-cfe4-25bc1ceb2581"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "using 2 cores\n",
            "Using alerts from Inpharmatica\n",
            "[14:54:15] WARNING: not removing hydrogen atom without neighbors\n",
            "[15:02:10] Unusual charge on atom 37 number of radical electrons set to zero\n",
            "[15:14:40] WARNING: not removing hydrogen atom without neighbors\n",
            "[15:15:18] WARNING: not removing hydrogen atom without neighbors\n",
            "[15:15:18] WARNING: not removing hydrogen atom without neighbors\n",
            "[15:16:54] WARNING: not removing hydrogen atom without neighbors\n",
            "Wrote SMILES for molecules passing filters to coconut.smi\n",
            "Wrote detailed data to coconut.csv\n",
            "197794 of 695133 passed filters 28.5%\n",
            "Elapsed time 1438.51 seconds\n"
          ]
        }
      ],
      "source": [
        "!rd_filters filter --in ./COCONUT_DB_absoluteSMILES.smi --prefix coconut"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "08bb1d6f",
      "metadata": {
        "id": "08bb1d6f"
      },
      "source": [
        "## 필터 결과\n",
        "\n",
        "- coconut.csv의 내용\n",
        " - SMILES 문자열,\n",
        " - 분자의 이름(NAME),\n",
        " - 필터를 통과하지 못한 이유(FILTER),\n",
        " - 분자량(MW, 500보다 크면 통과하지 못한다),\n",
        " - 옥탄올/물 분배계수(LogP, 5보다 크면 통과하지 못한다)\n",
        " - 수소결합의 공여체수 (HBD)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "id": "d2a40537",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "d2a40537",
        "outputId": "9e918bdd-512e-4d70-bd93-61afe2f37fbb"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                              SMILES          NAME  \\\n",
              "0  CC[C@@H]1NC(=O)[C@@H](NC(=O)[C@H](C)NC(=O)CNC(...  CNP0437004.1   \n",
              "1  COC1=CC=C(CCO[C@@H]2O[C@H](CO[C@@H]3O[C@@H](C)...  CNP0243002.1   \n",
              "2  C/C1=C\\[C@@H](C(C)(C)C)OC(=O)[C@H](CC(C)C)N(C)...  CNP0458114.1   \n",
              "3  COC1=CC=C(C2=CC(=O)OC3=CC(O[C@@H]4O[C@H](CO)[C...  CNP0252086.2   \n",
              "4  CC[C@H](C)[C@H]1C(=O)N2CCC[C@H]2C(=O)N1C(=O)[C...  CNP0107934.1   \n",
              "\n",
              "                               FILTER        MW    LogP  HBD  HBA    TPSA  Rot  \n",
              "0          Filter41_12_dicarbonyl > 0  1006.062 -4.4544   10   16  360.42   13  \n",
              "1                                  OK   476.475 -2.3884    7   12  187.76    8  \n",
              "2                 Filter87_crowns > 0   675.912  3.2885    2    7  145.43    3  \n",
              "3  Filter14_thio_oxopyrylium_salt > 0   430.409  0.6473    4    9  138.82    5  \n",
              "4                                  OK   559.711  2.9529    2    5  105.82    6  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-742a7bf0-5e11-446a-a2c7-e5c14ab8f263\" class=\"colab-df-container\">\n",
              "    <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>SMILES</th>\n",
              "      <th>NAME</th>\n",
              "      <th>FILTER</th>\n",
              "      <th>MW</th>\n",
              "      <th>LogP</th>\n",
              "      <th>HBD</th>\n",
              "      <th>HBA</th>\n",
              "      <th>TPSA</th>\n",
              "      <th>Rot</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>CC[C@@H]1NC(=O)[C@@H](NC(=O)[C@H](C)NC(=O)CNC(...</td>\n",
              "      <td>CNP0437004.1</td>\n",
              "      <td>Filter41_12_dicarbonyl &gt; 0</td>\n",
              "      <td>1006.062</td>\n",
              "      <td>-4.4544</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "      <td>360.42</td>\n",
              "      <td>13</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>COC1=CC=C(CCO[C@@H]2O[C@H](CO[C@@H]3O[C@@H](C)...</td>\n",
              "      <td>CNP0243002.1</td>\n",
              "      <td>OK</td>\n",
              "      <td>476.475</td>\n",
              "      <td>-2.3884</td>\n",
              "      <td>7</td>\n",
              "      <td>12</td>\n",
              "      <td>187.76</td>\n",
              "      <td>8</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>C/C1=C\\[C@@H](C(C)(C)C)OC(=O)[C@H](CC(C)C)N(C)...</td>\n",
              "      <td>CNP0458114.1</td>\n",
              "      <td>Filter87_crowns &gt; 0</td>\n",
              "      <td>675.912</td>\n",
              "      <td>3.2885</td>\n",
              "      <td>2</td>\n",
              "      <td>7</td>\n",
              "      <td>145.43</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>COC1=CC=C(C2=CC(=O)OC3=CC(O[C@@H]4O[C@H](CO)[C...</td>\n",
              "      <td>CNP0252086.2</td>\n",
              "      <td>Filter14_thio_oxopyrylium_salt &gt; 0</td>\n",
              "      <td>430.409</td>\n",
              "      <td>0.6473</td>\n",
              "      <td>4</td>\n",
              "      <td>9</td>\n",
              "      <td>138.82</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>CC[C@H](C)[C@H]1C(=O)N2CCC[C@H]2C(=O)N1C(=O)[C...</td>\n",
              "      <td>CNP0107934.1</td>\n",
              "      <td>OK</td>\n",
              "      <td>559.711</td>\n",
              "      <td>2.9529</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>105.82</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-742a7bf0-5e11-446a-a2c7-e5c14ab8f263')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-742a7bf0-5e11-446a-a2c7-e5c14ab8f263 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-742a7bf0-5e11-446a-a2c7-e5c14ab8f263');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-8326d0f6-e62d-4945-ac9c-8101958dfea7\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-8326d0f6-e62d-4945-ac9c-8101958dfea7')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-8326d0f6-e62d-4945-ac9c-8101958dfea7 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "df"
            }
          },
          "metadata": {},
          "execution_count": 40
        }
      ],
      "source": [
        "df = pd.read_csv(\"./coconut.csv\")\n",
        "df.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "13fb0222",
      "metadata": {
        "id": "13fb0222"
      },
      "source": [
        "- FILTER의 경우의 수를 파악하기 위해서 collection의 Counter를 사용한다\n",
        " - 통과한 수(OK)는 69148이다\n",
        " - 1,2 디카르보닐 구조를 포하여 통과하지 못한 경우는 19330이다 (이는 아미노산 셀린 또는 시스테인과 공유결합을 형성하는 것으로 알려져 있다)\n",
        " - filter_collection.CSV 파일을 보면 1,2디카르보닐 구조를 의미하는 SMARTS 문자열'*C(=O)C(=O)*'를 확인할 수 있다."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 41,
      "id": "71b1949c",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "71b1949c",
        "outputId": "e0b73312-be41-4df2-8a2f-912a9c1f7156"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                  Rule   Count\n",
              "1                                   OK  355114\n",
              "4       Filter44_michael_acceptor2 > 0  106620\n",
              "3   Filter14_thio_oxopyrylium_salt > 0   65743\n",
              "6               Filter33_c10_alkyl > 0   44924\n",
              "10      Filter40_epoxide_aziridine > 0   24468"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-5ad3db0f-0ce9-416c-97d1-7a4fb5f56b75\" class=\"colab-df-container\">\n",
              "    <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>Rule</th>\n",
              "      <th>Count</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>OK</td>\n",
              "      <td>355114</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Filter44_michael_acceptor2 &gt; 0</td>\n",
              "      <td>106620</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Filter14_thio_oxopyrylium_salt &gt; 0</td>\n",
              "      <td>65743</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Filter33_c10_alkyl &gt; 0</td>\n",
              "      <td>44924</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>Filter40_epoxide_aziridine &gt; 0</td>\n",
              "      <td>24468</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5ad3db0f-0ce9-416c-97d1-7a4fb5f56b75')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-5ad3db0f-0ce9-416c-97d1-7a4fb5f56b75 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-5ad3db0f-0ce9-416c-97d1-7a4fb5f56b75');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-8773145c-face-475f-80cd-46182291b13e\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-8773145c-face-475f-80cd-46182291b13e')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-8773145c-face-475f-80cd-46182291b13e button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "count_df",
              "summary": "{\n  \"name\": \"count_df\",\n  \"rows\": 83,\n  \"fields\": [\n    {\n      \"column\": \"Rule\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 83,\n        \"samples\": [\n          \"Filter6_benzyl_halide > 0\",\n          \"OK\",\n          \"Filter9_metal > 0\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Count\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 41242,\n        \"min\": 1,\n        \"max\": 355114,\n        \"num_unique_values\": 80,\n        \"samples\": [\n          478,\n          355114,\n          1062\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 41
        }
      ],
      "source": [
        "from collections import Counter\n",
        "count_list = list(Counter(df.FILTER).items())\n",
        "count_df = pd.DataFrame(count_list,columns=[\"Rule\",\"Count\"])\n",
        "count_df.sort_values(\"Count\",inplace=True,ascending=False)\n",
        "count_df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 42,
      "id": "358f0ef3",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 167
        },
        "id": "358f0ef3",
        "outputId": "451797e6-3969-4117-e47b-a44c188cde0e"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<rdkit.Chem.rdchem.Mol at 0x7f12e5c40f20>"
            ],
            "image/png": "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\n"
          },
          "metadata": {},
          "execution_count": 42
        }
      ],
      "source": [
        "dicarbonyl = Chem.MolFromSmarts('*C(=O)C(=O)*')\n",
        "dicarbonyl"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 43,
      "id": "b47db4a9",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 417
        },
        "id": "b47db4a9",
        "outputId": "b32b02d5-0c3b-4533-edd6-7a4fbff0683f"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 43
        }
      ],
      "source": [
        "smiles_list = df[df.FILTER == \"Filter41_12_dicarbonyl > 0\"].SMILES[:10]\n",
        "mol_list = [Chem.MolFromSmiles(x) for x in smiles_list]\n",
        "match_list = [mol.GetSubstructMatch(dicarbonyl) for mol in mol_list]\n",
        "Draw.MolsToGridImage(mol_list,highlightAtomLists=match_list,molsPerRow=5)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "621e48b7",
      "metadata": {
        "id": "621e48b7"
      },
      "source": [
        "# 모델 사용하기\n",
        "- GCN 모델을 불러오고, 피처화기를 만든후, 분자 데이터를 피처화 한다\n",
        "- 예측값을 확인하고 가장 높은 점수를 얻은 화학구조를 확인한다\n",
        "- 분자 데이터를 CSV로 저장한다"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "id": "c78b1e1d",
      "metadata": {
        "id": "c78b1e1d"
      },
      "outputs": [],
      "source": [
        "# 저장된 모델 불러오기\n",
        "# model = dc.models.GraphConvModel(1, batch_size=128,\n",
        "#           mode='classification',model_dir=\"model_dir\")\n",
        "# model.restore()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 45,
      "id": "16ae7bd5",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "16ae7bd5",
        "outputId": "97c209c8-f31a-422c-b34d-0ffbfaf59e7b"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                              SMILES          Name  Val\n",
              "0  CC[C@@H]1NC(=O)[C@@H](NC(=O)[C@H](C)NC(=O)CNC(...  CNP0437004.1    0\n",
              "1  COC1=CC=C(CCO[C@@H]2O[C@H](CO[C@@H]3O[C@@H](C)...  CNP0243002.1    0\n",
              "2  C/C1=C\\[C@@H](C(C)(C)C)OC(=O)[C@H](CC(C)C)N(C)...  CNP0458114.1    0\n",
              "3  COC1=CC=C(C2=CC(=O)OC3=CC(O[C@@H]4O[C@H](CO)[C...  CNP0252086.2    0\n",
              "4  CC[C@H](C)[C@H]1C(=O)N2CCC[C@H]2C(=O)N1C(=O)[C...  CNP0107934.1    0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-56195d3f-a9c5-4a80-bff5-b8e2d7477f0c\" class=\"colab-df-container\">\n",
              "    <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>SMILES</th>\n",
              "      <th>Name</th>\n",
              "      <th>Val</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>CC[C@@H]1NC(=O)[C@@H](NC(=O)[C@H](C)NC(=O)CNC(...</td>\n",
              "      <td>CNP0437004.1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>COC1=CC=C(CCO[C@@H]2O[C@H](CO[C@@H]3O[C@@H](C)...</td>\n",
              "      <td>CNP0243002.1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>C/C1=C\\[C@@H](C(C)(C)C)OC(=O)[C@H](CC(C)C)N(C)...</td>\n",
              "      <td>CNP0458114.1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>COC1=CC=C(C2=CC(=O)OC3=CC(O[C@@H]4O[C@H](CO)[C...</td>\n",
              "      <td>CNP0252086.2</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>CC[C@H](C)[C@H]1C(=O)N2CCC[C@H]2C(=O)N1C(=O)[C...</td>\n",
              "      <td>CNP0107934.1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-56195d3f-a9c5-4a80-bff5-b8e2d7477f0c')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-56195d3f-a9c5-4a80-bff5-b8e2d7477f0c button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-56195d3f-a9c5-4a80-bff5-b8e2d7477f0c');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-2239f9fa-aa36-401c-9260-705a812aa9bd\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-2239f9fa-aa36-401c-9260-705a812aa9bd')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-2239f9fa-aa36-401c-9260-705a812aa9bd button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "df"
            }
          },
          "metadata": {},
          "execution_count": 45
        }
      ],
      "source": [
        "df = pd.read_csv(\"./COCONUT_DB_absoluteSMILES.smi\",sep=\" \",header=None)\n",
        "df.columns=[\"SMILES\",\"Name\"]\n",
        "rows,cols = df.shape\n",
        "df[\"Val\"] = [0] * rows # 피처화를 위한 빈 자리\n",
        "df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 46,
      "id": "743666b8",
      "metadata": {
        "id": "743666b8"
      },
      "outputs": [],
      "source": [
        "infile_name = \"coconut_filtered.csv\"\n",
        "df.to_csv(infile_name,index=False)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "d0ad6f4f",
      "metadata": {
        "id": "d0ad6f4f"
      },
      "source": [
        "- CSV 파일을 읽어온다\n",
        "- dataset을 만든후 모델을 사용하기 위해서 피처화를 수행한다\n",
        " - ConvMolFeaturizer 사용\n",
        "- 시간이 오래걸린다 (7분)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "51e96a6f",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "51e96a6f",
        "outputId": "0367a2ac-c295-4672-c17f-dab879c7195f"
      },
      "outputs": [
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "[15:18:20] WARNING: not removing hydrogen atom without neighbors\n",
            "[15:36:34] Unusual charge on atom 37 number of radical electrons set to zero\n",
            "[15:52:52] WARNING: not removing hydrogen atom without neighbors\n",
            "[15:54:50] WARNING: not removing hydrogen atom without neighbors\n"
          ]
        }
      ],
      "source": [
        "loader = dc.data.CSVLoader(tasks=['Val'], feature_field=\"SMILES\", featurizer=featurizer)\n",
        "dataset = loader.create_dataset(infile_name, shard_size=8192)\n",
        "\n",
        "pred = model.predict(dataset)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "a9095529",
      "metadata": {
        "id": "a9095529"
      },
      "source": [
        "- 활성화와 비활성화를 구분하기 위해서 점수를 시각화 한다\n",
        "- SMILES와 예측값이 있는 데이터프레임을 합친다\n",
        "- 예측값이 높은 분자를 살펴본다\n",
        " - 아래 그림에서 0.3 이상을 선택하겠다"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "bd973049",
      "metadata": {
        "id": "bd973049"
      },
      "outputs": [],
      "source": [
        "pred_df = pd.DataFrame([x.flatten() for x in pred],columns=[\"Neg\",\"Pos\"])\n",
        "sns.displot(x=pred_df.Pos,kind=\"kde\",rug=True)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "c3d55c1b",
      "metadata": {
        "id": "c3d55c1b"
      },
      "outputs": [],
      "source": [
        "combo_df = df.join(pred_df,how=\"outer\")\n",
        "combo_df.sort_values(\"Pos\",inplace=True,ascending=False)\n",
        "PandasTools.AddMoleculeColumnToFrame(combo_df,\"SMILES\",\"Mol\")\n",
        "combo_df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "85fb3812",
      "metadata": {
        "id": "85fb3812"
      },
      "outputs": [],
      "source": [
        "Draw.MolsToGridImage(combo_df.Mol[:10],molsPerRow=5,legends=[\"%.2f\" % x for x in combo_df.Pos[:10]])\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "2332d14d",
      "metadata": {
        "id": "2332d14d"
      },
      "source": [
        "- 예측값이 높는 분자들의 모양이 비슷한 것을 알 수 있다\n",
        "- 중복되는 것을 피하기 위해서 군집화를 수행한다\n",
        " - 화학적 유사성을 사용하는 Butina 군집화를 사용하겠다\n",
        " - 군집화를 하려면 기준값을 주어야 한다 (아래에서 Tanimoto 유사도 > 0.35)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "ab4fe246",
      "metadata": {
        "id": "ab4fe246"
      },
      "outputs": [],
      "source": [
        "def butina_cluster(mol_list,cutoff=0.35):\n",
        "    fp_list = [rdmd.GetMorganFingerprintAsBitVect(m, 3, nBits=2048) for m in mol_list]\n",
        "    dists = []\n",
        "    nfps = len(fp_list)\n",
        "    for i in range(1,nfps):\n",
        "        sims = DataStructs.BulkTanimotoSimilarity(fp_list[i],fp_list[:i])\n",
        "        dists.extend([1-x for x in sims])\n",
        "    mol_clusters = Butina.ClusterData(dists,nfps,cutoff,isDistData=True)\n",
        "    cluster_id_list = [0]*nfps\n",
        "    for idx,cluster in enumerate(mol_clusters,1):\n",
        "        for member in cluster:\n",
        "            cluster_id_list[member] = idx\n",
        "    return cluster_id_list"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "d046cc38",
      "metadata": {
        "id": "d046cc38"
      },
      "outputs": [],
      "source": [
        "best_100_df = combo_df.head(100).copy()\n",
        "best_100_df[\"Cluster\"] = butina_cluster(best_100_df.Mol)\n",
        "best_100_df[:5]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "fa26c6c5",
      "metadata": {
        "id": "fa26c6c5"
      },
      "outputs": [],
      "source": [
        "best_cluster_rep_df = best_100_df.drop_duplicates(\"Cluster\")\n",
        "print(best_cluster_rep_df.shape)\n",
        "best_cluster_rep_df"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "4c31f08a",
      "metadata": {
        "id": "4c31f08a"
      },
      "source": [
        "- 결과를 파일로 저장하기"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "b1b14625",
      "metadata": {
        "id": "b1b14625"
      },
      "outputs": [],
      "source": [
        "best_cluster_rep_df.to_csv(\"best_cluster_represenatives.csv\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "c1420173",
      "metadata": {
        "id": "c1420173"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "AI_study",
      "language": "python",
      "name": "ai_study"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.10.12"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "1bf89b2ab0c04a5bb748248f4f6776e2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_7fa66993224b4d1db04e8490f7ccce5b",
              "IPY_MODEL_fba78101f549436a91ff7bbac7fa2442",
              "IPY_MODEL_a165afbc87ea448ab12365a001621028"
            ],
            "layout": "IPY_MODEL_266a0218ed8149918108d754703048be"
          }
        },
        "7fa66993224b4d1db04e8490f7ccce5b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_08f443373a714fe4bd38db256f6bf2b2",
            "placeholder": "​",
            "style": "IPY_MODEL_d951bcfb0d9a49f5ae737587b5cee713",
            "value": "100%"
          }
        },
        "fba78101f549436a91ff7bbac7fa2442": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f8acfd5a0d474e3fbab1b492071ec7b4",
            "max": 5,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_1e05a6df783049328ccc94c23f4ca319",
            "value": 5
          }
        },
        "a165afbc87ea448ab12365a001621028": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ac98529749e7459b84d92b3dba11d0b6",
            "placeholder": "​",
            "style": "IPY_MODEL_97b74e82e83a4bf0a3f82af3799a092c",
            "value": " 5/5 [03:31&lt;00:00, 43.54s/it]"
          }
        },
        "266a0218ed8149918108d754703048be": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "08f443373a714fe4bd38db256f6bf2b2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "d951bcfb0d9a49f5ae737587b5cee713": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "f8acfd5a0d474e3fbab1b492071ec7b4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "1e05a6df783049328ccc94c23f4ca319": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "ac98529749e7459b84d92b3dba11d0b6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "97b74e82e83a4bf0a3f82af3799a092c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}