train_ml_models/train_linear_ML_models.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "train_linear_ML_models.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-T9havD0hte-",
        "colab_type": "text"
      },
      "source": [
        "# Code to train linear ML model \n",
        "This notebook was made on Google Colab ([link to notebook](https://colab.research.google.com/drive/1v9TvoMjIgbUIj1Mna2setqvZNuOrSbXv?usp=sharing)). If you are running on your local machine remove the cells related to data retrieval and RDKit installation, and change the paths to the files as needed.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RFB2J63uGWKn",
        "colab_type": "text"
      },
      "source": [
        "# Software installation and data retrieval, etc.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ezQT0icZlP4w",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "%%capture\n",
        "# Install RDKit\n",
        "# takes 2-3 minutes\n",
        "!wget -c https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh\n",
        "!chmod +x Miniconda3-latest-Linux-x86_64.sh\n",
        "!time bash ./Miniconda3-latest-Linux-x86_64.sh -b -f -p /usr/local\n",
        "!time conda install -q -y -c conda-forge rdkit"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y3TqLnDnGcKy",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 399
        },
        "outputId": "460640c3-f63b-4912-a67d-035fd5705c40"
      },
      "source": [
        "# Download training data\n",
        "! rm -f linear_ml_density_train_data.csv linear_ml_barrier_train_data.csv\n",
        "! wget https://sid.erda.dk/share_redirect/EwaEr2JMrb/all_230_billion/ml_data/linear_ml_density_train_data.csv\n",
        "! wget https://sid.erda.dk/share_redirect/EwaEr2JMrb/all_230_billion/ml_data/linear_ml_barrier_train_data.csv"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-07-03 09:00:59--  https://sid.erda.dk/share_redirect/EwaEr2JMrb/all_230_billion/ml_data/linear_ml_density_train_data.csv\n",
            "Resolving sid.erda.dk (sid.erda.dk)... 130.225.104.13\n",
            "Connecting to sid.erda.dk (sid.erda.dk)|130.225.104.13|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 10918996 (10M) [text/csv]\n",
            "Saving to: ‘linear_ml_density_train_data.csv’\n",
            "\n",
            "linear_ml_density_t 100%[===================>]  10.41M  7.32MB/s    in 1.4s    \n",
            "\n",
            "2020-07-03 09:01:01 (7.32 MB/s) - ‘linear_ml_density_train_data.csv’ saved [10918996/10918996]\n",
            "\n",
            "--2020-07-03 09:01:02--  https://sid.erda.dk/share_redirect/EwaEr2JMrb/all_230_billion/ml_data/linear_ml_barrier_train_data.csv\n",
            "Resolving sid.erda.dk (sid.erda.dk)... 130.225.104.13\n",
            "Connecting to sid.erda.dk (sid.erda.dk)|130.225.104.13|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 2565910 (2.4M) [text/csv]\n",
            "Saving to: ‘linear_ml_barrier_train_data.csv’\n",
            "\n",
            "linear_ml_barrier_t 100%[===================>]   2.45M  2.68MB/s    in 0.9s    \n",
            "\n",
            "2020-07-03 09:01:03 (2.68 MB/s) - ‘linear_ml_barrier_train_data.csv’ saved [2565910/2565910]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bdN9Rw8zhKA3",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "705859b0-6342-423b-d306-636d4c89975e"
      },
      "source": [
        "import sys\n",
        "import re\n",
        "import pandas as pd \n",
        "import numpy as np\n",
        "\n",
        "from collections import OrderedDict\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "# RDKit\n",
        "sys.path.append('/usr/local/lib/python3.7/site-packages/')\n",
        "from rdkit import Chem\n",
        "from rdkit.Chem import Descriptors\n",
        "\n",
        "# ML\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.linear_model import LinearRegression"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jLvU76S-hm2O",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "plt.rcParams['xtick.labelsize'] = 16 \n",
        "plt.rcParams['ytick.labelsize'] = 16 \n",
        "plt.rcParams['axes.labelsize'] = 16"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9eU9aUCElCOQ",
        "colab_type": "text"
      },
      "source": [
        "# Helper Functions\n",
        "Functions to make RDKit object and the find number of substituents from gene names."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tff3gOxdlA5t",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "ligands = ['H', 'F', 'Cl', 'Br', 'C(F)(F)(F)', 'C#N',\n",
        "           '[N+](=O)([O-])', 'C(=O)[H]', 'C(=O)O',\n",
        "           'C(=O)C', 'C(=O)N', 'C#C', 'S(=O)(=O)(C)',\n",
        "           'C=N', 'O', 'OC', 'N', 'N(C)(C)', 'C',\n",
        "           'N(C(=O)(C))', 'SC', 'C3=CC=C(F)C=C3',\n",
        "           'C3=CC=C(Cl)C=C3', 'C3=CC=C(Br)C=C3',\n",
        "           'C3=CC=C(C(F)(F)(F))C=C3',\n",
        "           'C3=CC=C(C#N)C=C3', 'C3=CC=C([N+](=O)([O-]))C=C3',\n",
        "           'C3=CC=C(C(=O)[H])C=C3', 'C3=CC=C(C(=O)O)C=C3',\n",
        "           'C3=CC=C(C(=O)C)C=C3', 'C3=CC=C(C(=O)N)C=C3',\n",
        "           'C3=CC=C(C#C)C=C3', 'C3=CC=C(S(=O)(=O)(C))C=C3',\n",
        "           'C3=CC=C(C=N)C=C3', 'C3=CC=C(O)C=C3',\n",
        "           'C3=CC=C(OC)C=C3', 'C3=CC=C(N)C=C3',\n",
        "           'C3=CC=C(N(C)(C))C=C3', 'C3=CC=C(C)C=C3',\n",
        "           'C3=CC=C(N(C(=O)(C)))C=C3', 'C3=CC=C(SC)C=C3',\n",
        "           'C3=CC=CC=C3']\n",
        "\n",
        "\n",
        "ligand_num2smiles = OrderedDict(zip(list(range(len(ligands))),ligands))\n",
        "\n",
        "def get_mol(gene):\n",
        "    global ligands\n",
        "\n",
        "    gene = gene.split('-')\n",
        "    \n",
        "    core = 'N#CC1(C#N)C(1)=C(2)C2=C(3)C(4)=C(5)C(6)=C(7)C21'\n",
        "    \n",
        "    gene_smiles = core\n",
        "    for i, ligand_idx in enumerate(gene):\n",
        "        substr = ligands[int(ligand_idx)]\n",
        "        if substr == 'H':\n",
        "            continue\n",
        "        gene_smiles = gene_smiles.replace('('+str(i+1)+')', f'({substr})')\n",
        "    \n",
        "    mol = Chem.MolFromSmiles(re.sub('\\([0-9]\\)', '', gene_smiles))\n",
        "\n",
        "    return mol\n",
        "\n",
        "def num_subs(gene):\n",
        "    gene = [int(x) for x in gene.split('-')]\n",
        "    return 7 - gene.count(0)\n",
        "\n",
        "def get_smiles(rdkit_mol):\n",
        "    return Chem.MolToSmiles(Chem.RemoveHs(rdkit_mol))"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yMGA1iQWn00u",
        "colab_type": "text"
      },
      "source": [
        "# Train Linear Regression Model - Density"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "H7ljUVs1h0Ug",
        "colab_type": "text"
      },
      "source": [
        "## Prepare Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tKPQf22PliFd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_density = pd.read_csv('linear_ml_density_train_data.csv')"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-OkVLIcZlyGB",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 344
        },
        "outputId": "5635c297-84ce-4fcf-d064-171bbbdf9f89"
      },
      "source": [
        "fig, ax = plt.subplots(ncols=1, nrows=1, figsize=(7,5))\n",
        "ax.hist(df_density['xtb_density'], bins=60, range=(0,1), label='All Density Data')\n",
        "\n",
        "ax.set(xlabel='Storage Density [kJ/g]', ylabel='Counts');"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 504x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OkAsOCSpoHcP",
        "colab_type": "text"
      },
      "source": [
        "## Split into two models\n",
        "Train two different models. One that predicts on 1-4 subs, and another for 5-7 subs."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EflLrqy6oF_x",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df1 = df_density.loc[df_density['num_subs'].isin([1])].copy()\n",
        "\n",
        "df24 = df_density.loc[df_density['num_subs'].isin([2,3,4])].copy()\n",
        "df57 = df_density.loc[df_density['num_subs'].isin([5,6,7])].copy()"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ULIf8uz0oh1e",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "0bb4f307-43aa-44cc-eda5-1080778e002f"
      },
      "source": [
        "df_train57, df_test57 =  train_test_split(df57, test_size=0.5, random_state=80)\n",
        "df_test57 = df_test57.copy()\n",
        "\n",
        "# Add all ones to 2-4 subs\n",
        "df_train24, df_test24 =  train_test_split(df24, test_size=0.5, random_state=80)\n",
        "df_train24 = pd.concat([df_train24, df1])\n",
        "df_train24 = df_train24.copy()\n",
        "\n",
        "print(f\"train 5-7: {df_train57.shape[0]}, test2-4: {df_test57.shape[0]}\")\n",
        "print(f\"train 2-4: {df_train24.shape[0]}, test5-7: {df_test24.shape[0]}\")"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "train 5-7: 25940, test2-4: 25940\n",
            "train 2-4: 22158, test5-7: 21884\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8lu7z4U3pE4l",
        "colab_type": "text"
      },
      "source": [
        "## Train Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pJhF00D-nkLD",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def one_hot(gene):\n",
        "    \"\"\"Creates one-hot representation\"\"\"\n",
        "    one_hot = np.zeros((7,41), dtype=np.float32)\n",
        "    for i, base in enumerate([int(x) for x in gene.split('-')]):\n",
        "        if base == 0:\n",
        "            continue\n",
        "        one_hot[i,base - 1] = 1.\n",
        "    return one_hot\n",
        "\n",
        "def training_data_one_hot(df, prop='xtb_density'):\n",
        "    \"\"\" Create Matrices and Vectors with representation and targets repectively \"\"\"\n",
        "    Xs = np.empty((df.shape[0], 287))\n",
        "    ys = np.empty(df.shape[0])\n",
        "    \n",
        "    for i, row in enumerate(df.itertuples()):\n",
        "        Xs[i] = one_hot(row.gene).flatten()\n",
        "        ys[i] = getattr(row, prop)\n",
        "    \n",
        "    return Xs, ys"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LonvpRfOo1jB",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 121
        },
        "outputId": "bb35175e-48d9-4c82-f9f5-00a7a965de5f"
      },
      "source": [
        "# Predict 2-4\n",
        "Xtrain24_s, ytrain24_s = training_data_one_hot(df_train24, prop='xtb_density')\n",
        "reg_s24 = LinearRegression().fit(X=Xtrain24_s, y=ytrain24_s)\n",
        "\n",
        "Xtest24_s, ytest24_s = training_data_one_hot(df_test24, prop='xtb_density')\n",
        "ypred_24 = reg_s24.predict(Xtest24_s)\n",
        "df_test24['pred_density'] = ypred_24 "
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:7: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  import sys\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2BOzb7lbpXvl",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Predict 5-7\n",
        "Xtrain57_s, ytrain57_s = training_data_one_hot(df_train57, prop='xtb_density')\n",
        "reg_s57 = LinearRegression().fit(X=Xtrain57_s, y=ytrain57_s)\n",
        "\n",
        "Xtest57_s, ytest57_s = training_data_one_hot(df_test57, prop='xtb_density')\n",
        "ypred_57 = reg_s57.predict(Xtest57_s)\n",
        "df_test57['pred_density'] = ypred_57"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5FDUSTMzq_AI",
        "colab_type": "text"
      },
      "source": [
        "## Make Figure S4"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SP-SPSmbrAW8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 369
        },
        "outputId": "df721a72-c83e-4052-9b96-4875d65085fd"
      },
      "source": [
        "fig, ax = plt.subplots(ncols=2, nrows=1, figsize=(9,5), sharey=True)\n",
        "\n",
        "num_samples = 8000\n",
        "\n",
        "# Plot 2-4 subs\n",
        "sns.scatterplot(x='xtb_density', y='pred_density', data=df_test24.sample(num_samples, random_state=31), ax=ax[0])\n",
        "\n",
        "error24 = df_test24.pred_density - df_test24.xtb_density\n",
        "mae = abs(error24).mean()\n",
        "rms = (error24).std()\n",
        "\n",
        "ax[0].plot([0.0,0.7], [0.0,0.7], '--k')\n",
        "ax[0].set_title('2-4 Substituents', fontsize=16)\n",
        "ax[0].set(xlabel='Storage Density [kJ/g]', ylabel='Linear Storage Density [kJ/g]')\n",
        "\n",
        "s24 = f'RMS: {rms:.4f} \\nMAE: {mae:.4f}'\n",
        "ax[0].text(0.4,0.05, s24, {'fontsize': 16});\n",
        "\n",
        "# Plot 5-7 subs\n",
        "sns.scatterplot(x='xtb_density', y='pred_density', data=df_test57.sample(num_samples, random_state=31), ax=ax[1])\n",
        "\n",
        "error57 = df_test57.pred_density - df_test57.xtb_density\n",
        "mae = abs(error57).mean()\n",
        "rms = (error57).std()\n",
        "\n",
        "ax[1].plot([0.0,0.7], [0.0,0.7], '--k')\n",
        "ax[1].set_title('5-7 Substituents', fontsize=16)\n",
        "ax[1].set(xlabel='Storage Density [kJ/g]', ylabel='Linear Storage Density [kJ/g]')\n",
        "\n",
        "s57 = f'RMS: {rms:.4f} \\nMAE: {mae:.4f}'\n",
        "ax[1].text(0.4,0.05, s57, {'fontsize': 16});\n",
        "\n",
        "fig.tight_layout()"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 648x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zf4IAjpZx0Ks",
        "colab_type": "text"
      },
      "source": [
        "# Train Linear Regression Model - Barrier \n",
        "Only for testing GA."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fXN9ddnJrUpx",
        "colab_type": "text"
      },
      "source": [
        "## Prepare Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SaBofB24lxag",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "b408f7a7-bc16-4807-b416-4f146eecbcbe"
      },
      "source": [
        "df_barrier = pd.read_csv('linear_ml_barrier_train_data.csv')\n",
        "df_barrier.head()"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>gene</th>\n",
              "      <th>num_subs</th>\n",
              "      <th>smiles</th>\n",
              "      <th>pm3_barrier</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0-0-6-0-0-0-0</td>\n",
              "      <td>1</td>\n",
              "      <td>N#CC1(C#N)C=CC2=C([N+](=O)[O-])C=CC=CC21</td>\n",
              "      <td>195.204558</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0-0-0-0-25-0-0</td>\n",
              "      <td>1</td>\n",
              "      <td>N#Cc1ccc(C2=CC=C3C=CC(C#N)(C#N)C3C=C2)cc1</td>\n",
              "      <td>186.308338</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0-0-0-0-0-0-20</td>\n",
              "      <td>1</td>\n",
              "      <td>CSC1=CC=CC=C2C=CC(C#N)(C#N)C21</td>\n",
              "      <td>195.950349</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0-36-0-0-0-0-0</td>\n",
              "      <td>1</td>\n",
              "      <td>N#CC1(C#N)C=C(c2ccc(N)cc2)C2=CC=CC=CC21</td>\n",
              "      <td>163.975406</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0-0-0-0-33-0-0</td>\n",
              "      <td>1</td>\n",
              "      <td>N#CC1(C#N)C=CC2=CC=C(c3ccc(C=N)cc3)C=CC21</td>\n",
              "      <td>183.751149</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             gene  ...  pm3_barrier\n",
              "0   0-0-6-0-0-0-0  ...   195.204558\n",
              "1  0-0-0-0-25-0-0  ...   186.308338\n",
              "2  0-0-0-0-0-0-20  ...   195.950349\n",
              "3  0-36-0-0-0-0-0  ...   163.975406\n",
              "4  0-0-0-0-33-0-0  ...   183.751149\n",
              "\n",
              "[5 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KyudggBAsYkP",
        "colab_type": "text"
      },
      "source": [
        "How many samples for each number of subs?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OmRiygU9r_g0",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "outputId": "a2be26d1-5e8e-4b6b-efb1-c996f863068d"
      },
      "source": [
        "df_barrier[['gene', 'num_subs']].groupby(\"num_subs\").count()"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>gene</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>num_subs</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>262</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>927</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4497</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4316</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>2294</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>5861</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>3748</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          gene\n",
              "num_subs      \n",
              "1          262\n",
              "2          927\n",
              "3         4497\n",
              "4         4316\n",
              "5         2294\n",
              "6         5861\n",
              "7         3748"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0Qwc3qs7sdUZ",
        "colab_type": "text"
      },
      "source": [
        "## Train Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BWZKQTbUsE2j",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "f5b4a385-08da-4528-b63b-d5bc3a7091b6"
      },
      "source": [
        "# Split Data\n",
        "df_train_barrier, df_test_barrier = train_test_split(df_barrier, test_size=0.2, random_state=42)\n",
        "\n",
        "df_train_barrier = pd.concat([df_train_barrier, df_test_barrier.loc[df_test_barrier[\"num_subs\"] == 1]])\n",
        "df_test_barrier =  df_test_barrier.loc[df_test_barrier[\"num_subs\"] != 1]\n",
        "\n",
        "print(f\"train: {df_train_barrier.shape[0]}, test: {df_test_barrier.shape[0]}\")"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "train: 17567, test: 4338\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IY1MES8qtUoV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Train Model\n",
        "Xtrain_b, ytrain_b = training_data_one_hot(df_train_barrier, prop='pm3_barrier')\n",
        "reg_barrier = LinearRegression().fit(X=Xtrain_b, y=ytrain_b)\n",
        "\n",
        "# Predict on test\n",
        "Xtest_b, ytest_b = training_data_one_hot(df_test_barrier, prop='pm3_barrier')\n",
        "ypred_b = reg_barrier.predict(Xtest_b)\n",
        "\n",
        "df_test_barrier[\"pred_barrier\"] = ypred_b"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0SKoNd75tvNi",
        "colab_type": "text"
      },
      "source": [
        "## Make Figure S6"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "azq1W1QttmsC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Compute error\n",
        "error = df_test_barrier[\"pred_barrier\"] - df_test_barrier['pm3_barrier']\n",
        "\n",
        "mae = abs(error).mean()\n",
        "rms = error.std()"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y9OrBUqnt93f",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 369
        },
        "outputId": "d66ea8b5-3489-4a68-8c84-7918e3dd51ed"
      },
      "source": [
        "fig2, ax2 = plt.subplots(ncols=1, nrows=1, figsize=(5,5))\n",
        "\n",
        "sns.scatterplot(x=\"pm3_barrier\", y=\"pred_barrier\", data=df_test_barrier)\n",
        "ax2.plot([100,300], [100,300], '--k')\n",
        "\n",
        "ax2.set_title('2-7 Subtituents', fontsize=16)\n",
        "ax2.set(xlabel=r'PM3 $\\Delta \\Delta H_f^{\\ddag}$ [kJ/mol]', ylabel='Pred. $\\Delta \\Delta H_f^{\\ddag}$ [kJ/mol]')\n",
        "\n",
        "s_barrier = f'RMS: {rms:.2f} \\nMAE: {mae:.2f}'\n",
        "ax2.text(220,120, s_barrier, {'fontsize': 16});\n",
        "\n",
        "fig2.tight_layout()"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}