From c3ba93c5c2edcdd80985c5e219398f71777aa08a Mon Sep 17 00:00:00 2001
From: Chanin Nantasenamat <51851491+dataprofessor@users.noreply.github.com>
Date: Sun, 9 Aug 2020 22:43:08 +0700
Subject: [PATCH] Add files via upload

---
 ...le_linear_regression_model_in_python.ipynb | 1349 +++++++++++++++++
 1 file changed, 1349 insertions(+)
 create mode 100644 python/How_to_build_a_simple_linear_regression_model_in_python.ipynb

diff --git a/python/How_to_build_a_simple_linear_regression_model_in_python.ipynb b/python/How_to_build_a_simple_linear_regression_model_in_python.ipynb
new file mode 100644
index 0000000..4049574
--- /dev/null
+++ b/python/How_to_build_a_simple_linear_regression_model_in_python.ipynb
@@ -0,0 +1,1349 @@
+{
+  "nbformat": 4,
+  "nbformat_minor": 0,
+  "metadata": {
+    "colab": {
+      "name": "How_to_build_a_simple_linear_regression_model_in_python.ipynb",
+      "provenance": [],
+      "collapsed_sections": []
+    },
+    "kernelspec": {
+      "name": "python3",
+      "display_name": "Python 3"
+    }
+  },
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "OQi3X7TNUl5Y",
+        "colab_type": "text"
+      },
+      "source": [
+        "# **How to Build a Simple Linear Regression Model in Python** \n",
+        "\n",
+        "Chanin Nantasenamat\n",
+        "\n",
+        "<i>[Data Professor YouTube channel](http://youtube.com/dataprofessor), http://youtube.com/dataprofessor </i>\n",
+        "\n",
+        "In this Jupyter notebook, we will building a simple linear regression model using the **Delaney Molecular Solubility** dataset."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "H661uGwCNFMC",
+        "colab_type": "text"
+      },
+      "source": [
+        "## **1. Retrieving the Dataset**"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "ZkglmVcwpoXG",
+        "colab_type": "text"
+      },
+      "source": [
+        "### **1.1. Original dataset**\n",
+        "\n",
+        "The original [Delaney's dataset](https://pubs.acs.org/doi/10.1021/ci034243x) available as a [Supplementary file](https://pubs.acs.org/doi/10.1021/ci034243x)$^4$. The full paper is entitled [ESOL:  Estimating Aqueous Solubility Directly from Molecular Structure](https://pubs.acs.org/doi/10.1021/ci034243x).$^1$"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "mupO58ZfpiqE",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "import pandas as pd"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "5NSgd-6Mol_T",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 419
+        },
+        "outputId": "26d12350-2f85-4955-fdcf-c7ca8366f2b4"
+      },
+      "source": [
+        "delaney_url = 'https://raw.githubusercontent.com/dataprofessor/data/master/delaney.csv'\n",
+        "delaney_df = pd.read_csv(delaney_url)\n",
+        "delaney_df"
+      ],
+      "execution_count": null,
+      "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>Compound ID</th>\n",
+              "      <th>measured log(solubility:mol/L)</th>\n",
+              "      <th>ESOL predicted log(solubility:mol/L)</th>\n",
+              "      <th>SMILES</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>1,1,1,2-Tetrachloroethane</td>\n",
+              "      <td>-2.180</td>\n",
+              "      <td>-2.794</td>\n",
+              "      <td>ClCC(Cl)(Cl)Cl</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>1,1,1-Trichloroethane</td>\n",
+              "      <td>-2.000</td>\n",
+              "      <td>-2.232</td>\n",
+              "      <td>CC(Cl)(Cl)Cl</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>1,1,2,2-Tetrachloroethane</td>\n",
+              "      <td>-1.740</td>\n",
+              "      <td>-2.549</td>\n",
+              "      <td>ClC(Cl)C(Cl)Cl</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>1,1,2-Trichloroethane</td>\n",
+              "      <td>-1.480</td>\n",
+              "      <td>-1.961</td>\n",
+              "      <td>ClCC(Cl)Cl</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>1,1,2-Trichlorotrifluoroethane</td>\n",
+              "      <td>-3.040</td>\n",
+              "      <td>-3.077</td>\n",
+              "      <td>FC(F)(Cl)C(F)(Cl)Cl</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>1139</th>\n",
+              "      <td>vamidothion</td>\n",
+              "      <td>1.144</td>\n",
+              "      <td>-1.446</td>\n",
+              "      <td>CNC(=O)C(C)SCCSP(=O)(OC)(OC)</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1140</th>\n",
+              "      <td>Vinclozolin</td>\n",
+              "      <td>-4.925</td>\n",
+              "      <td>-4.377</td>\n",
+              "      <td>CC1(OC(=O)N(C1=O)c2cc(Cl)cc(Cl)c2)C=C</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1141</th>\n",
+              "      <td>Warfarin</td>\n",
+              "      <td>-3.893</td>\n",
+              "      <td>-3.913</td>\n",
+              "      <td>CC(=O)CC(c1ccccc1)c3c(O)c2ccccc2oc3=O</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1142</th>\n",
+              "      <td>Xipamide</td>\n",
+              "      <td>-3.790</td>\n",
+              "      <td>-3.642</td>\n",
+              "      <td>Cc1cccc(C)c1NC(=O)c2cc(c(Cl)cc2O)S(N)(=O)=O</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1143</th>\n",
+              "      <td>XMC</td>\n",
+              "      <td>-2.581</td>\n",
+              "      <td>-2.688</td>\n",
+              "      <td>CNC(=O)Oc1cc(C)cc(C)c1</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "<p>1144 rows × 4 columns</p>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "                         Compound ID  ...                                       SMILES\n",
+              "0          1,1,1,2-Tetrachloroethane  ...                               ClCC(Cl)(Cl)Cl\n",
+              "1              1,1,1-Trichloroethane  ...                                 CC(Cl)(Cl)Cl\n",
+              "2          1,1,2,2-Tetrachloroethane  ...                               ClC(Cl)C(Cl)Cl\n",
+              "3              1,1,2-Trichloroethane  ...                                   ClCC(Cl)Cl\n",
+              "4     1,1,2-Trichlorotrifluoroethane  ...                          FC(F)(Cl)C(F)(Cl)Cl\n",
+              "...                              ...  ...                                          ...\n",
+              "1139                     vamidothion  ...                 CNC(=O)C(C)SCCSP(=O)(OC)(OC)\n",
+              "1140                     Vinclozolin  ...        CC1(OC(=O)N(C1=O)c2cc(Cl)cc(Cl)c2)C=C\n",
+              "1141                        Warfarin  ...       CC(=O)CC(c1ccccc1)c3c(O)c2ccccc2oc3=O \n",
+              "1142                        Xipamide  ...  Cc1cccc(C)c1NC(=O)c2cc(c(Cl)cc2O)S(N)(=O)=O\n",
+              "1143                             XMC  ...                       CNC(=O)Oc1cc(C)cc(C)c1\n",
+              "\n",
+              "[1144 rows x 4 columns]"
+            ]
+          },
+          "metadata": {
+            "tags": []
+          },
+          "execution_count": 14
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "biqQ78_hqdb6",
+        "colab_type": "text"
+      },
+      "source": [
+        "### **1.2. Delaney dataset with computed molecular descriptors**\n",
+        "\n",
+        "As demonstrated in a previous YouTube video [Data Science for Computational Drug Discovery using Python](https://www.youtube.com/watch?v=VXFFHHoE1wk) on the Data Professor YouTube channel, SMILES notation from the Delaney dataset was used as *input* for molecular descriptor calculation using the **rdkit** Python library. This produced the 4 molecular descriptors as used by the authors in their published research article."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "olyPX1TjQMvr",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### **Definition of variables**\n",
+        "\n",
+        "The **Y** variable (response variable) is **LogS** (log of the aqueous solubility).\n",
+        "\n",
+        "The **X** variables are comprised of 4 molecular descriptors:\n",
+        "1. **cLogP** *(Octanol-water partition coefficient)*\n",
+        "2. **MW** *(Molecular weight)*\n",
+        "3. **RB** *(Number of rotatable bonds)*\n",
+        "4. **AP** *(Aromatic proportion = number of aromatic atoms / total number of heavy atoms)*"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "q9kna0SWkamZ",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 419
+        },
+        "outputId": "3d948275-2329-4cc0-b465-0198efafb183"
+      },
+      "source": [
+        "delaney_descriptors_url = 'https://raw.githubusercontent.com/dataprofessor/data/master/delaney_solubility_with_descriptors.csv'\n",
+        "delaney_descriptors_df = pd.read_csv(delaney_descriptors_url)\n",
+        "delaney_descriptors_df"
+      ],
+      "execution_count": null,
+      "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>MolLogP</th>\n",
+              "      <th>MolWt</th>\n",
+              "      <th>NumRotatableBonds</th>\n",
+              "      <th>AromaticProportion</th>\n",
+              "      <th>logS</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>2.59540</td>\n",
+              "      <td>167.850</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "      <td>-2.180</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>2.37650</td>\n",
+              "      <td>133.405</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "      <td>-2.000</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>2.59380</td>\n",
+              "      <td>167.850</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "      <td>-1.740</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>2.02890</td>\n",
+              "      <td>133.405</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "      <td>-1.480</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>2.91890</td>\n",
+              "      <td>187.375</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "      <td>-3.040</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>...</th>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1139</th>\n",
+              "      <td>1.98820</td>\n",
+              "      <td>287.343</td>\n",
+              "      <td>8.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "      <td>1.144</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1140</th>\n",
+              "      <td>3.42130</td>\n",
+              "      <td>286.114</td>\n",
+              "      <td>2.0</td>\n",
+              "      <td>0.333333</td>\n",
+              "      <td>-4.925</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1141</th>\n",
+              "      <td>3.60960</td>\n",
+              "      <td>308.333</td>\n",
+              "      <td>4.0</td>\n",
+              "      <td>0.695652</td>\n",
+              "      <td>-3.893</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1142</th>\n",
+              "      <td>2.56214</td>\n",
+              "      <td>354.815</td>\n",
+              "      <td>3.0</td>\n",
+              "      <td>0.521739</td>\n",
+              "      <td>-3.790</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1143</th>\n",
+              "      <td>2.02164</td>\n",
+              "      <td>179.219</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.461538</td>\n",
+              "      <td>-2.581</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "<p>1144 rows × 5 columns</p>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "      MolLogP    MolWt  NumRotatableBonds  AromaticProportion   logS\n",
+              "0     2.59540  167.850                0.0            0.000000 -2.180\n",
+              "1     2.37650  133.405                0.0            0.000000 -2.000\n",
+              "2     2.59380  167.850                1.0            0.000000 -1.740\n",
+              "3     2.02890  133.405                1.0            0.000000 -1.480\n",
+              "4     2.91890  187.375                1.0            0.000000 -3.040\n",
+              "...       ...      ...                ...                 ...    ...\n",
+              "1139  1.98820  287.343                8.0            0.000000  1.144\n",
+              "1140  3.42130  286.114                2.0            0.333333 -4.925\n",
+              "1141  3.60960  308.333                4.0            0.695652 -3.893\n",
+              "1142  2.56214  354.815                3.0            0.521739 -3.790\n",
+              "1143  2.02164  179.219                1.0            0.461538 -2.581\n",
+              "\n",
+              "[1144 rows x 5 columns]"
+            ]
+          },
+          "metadata": {
+            "tags": []
+          },
+          "execution_count": 15
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "qQYE-jCRSmCn",
+        "colab_type": "text"
+      },
+      "source": [
+        "---"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "WRXEmm941h13",
+        "colab_type": "text"
+      },
+      "source": [
+        "## **2. Create X and Y variables**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "JeSbdwYA1l3K",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 419
+        },
+        "outputId": "5e139655-ff97-4f95-947a-c6ecbc9c8861"
+      },
+      "source": [
+        "X = delaney_descriptors_df.drop('logS', axis=1)\n",
+        "X"
+      ],
+      "execution_count": 19,
+      "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>MolLogP</th>\n",
+              "      <th>MolWt</th>\n",
+              "      <th>NumRotatableBonds</th>\n",
+              "      <th>AromaticProportion</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>2.59540</td>\n",
+              "      <td>167.850</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>2.37650</td>\n",
+              "      <td>133.405</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>2.59380</td>\n",
+              "      <td>167.850</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>2.02890</td>\n",
+              "      <td>133.405</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>2.91890</td>\n",
+              "      <td>187.375</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.000000</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>1139</th>\n",
+              "      <td>1.98820</td>\n",
+              "      <td>287.343</td>\n",
+              "      <td>8.0</td>\n",
+              "      <td>0.000000</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1140</th>\n",
+              "      <td>3.42130</td>\n",
+              "      <td>286.114</td>\n",
+              "      <td>2.0</td>\n",
+              "      <td>0.333333</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1141</th>\n",
+              "      <td>3.60960</td>\n",
+              "      <td>308.333</td>\n",
+              "      <td>4.0</td>\n",
+              "      <td>0.695652</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1142</th>\n",
+              "      <td>2.56214</td>\n",
+              "      <td>354.815</td>\n",
+              "      <td>3.0</td>\n",
+              "      <td>0.521739</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1143</th>\n",
+              "      <td>2.02164</td>\n",
+              "      <td>179.219</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.461538</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "<p>1144 rows × 4 columns</p>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "      MolLogP    MolWt  NumRotatableBonds  AromaticProportion\n",
+              "0     2.59540  167.850                0.0            0.000000\n",
+              "1     2.37650  133.405                0.0            0.000000\n",
+              "2     2.59380  167.850                1.0            0.000000\n",
+              "3     2.02890  133.405                1.0            0.000000\n",
+              "4     2.91890  187.375                1.0            0.000000\n",
+              "...       ...      ...                ...                 ...\n",
+              "1139  1.98820  287.343                8.0            0.000000\n",
+              "1140  3.42130  286.114                2.0            0.333333\n",
+              "1141  3.60960  308.333                4.0            0.695652\n",
+              "1142  2.56214  354.815                3.0            0.521739\n",
+              "1143  2.02164  179.219                1.0            0.461538\n",
+              "\n",
+              "[1144 rows x 4 columns]"
+            ]
+          },
+          "metadata": {
+            "tags": []
+          },
+          "execution_count": 19
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "VGwuoNKs2ReN",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 221
+        },
+        "outputId": "27773f04-d07f-4d6f-80b8-323944ddac7a"
+      },
+      "source": [
+        "Y = delaney_descriptors_df.logS\n",
+        "Y"
+      ],
+      "execution_count": 21,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "0      -2.180\n",
+              "1      -2.000\n",
+              "2      -1.740\n",
+              "3      -1.480\n",
+              "4      -3.040\n",
+              "        ...  \n",
+              "1139    1.144\n",
+              "1140   -4.925\n",
+              "1141   -3.893\n",
+              "1142   -3.790\n",
+              "1143   -2.581\n",
+              "Name: logS, Length: 1144, dtype: float64"
+            ]
+          },
+          "metadata": {
+            "tags": []
+          },
+          "execution_count": 21
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "SzrfuUZNFg_X",
+        "colab_type": "text"
+      },
+      "source": [
+        "## **3. Data split**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "dMRn8EVjFlrT",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "from sklearn.model_selection import train_test_split"
+      ],
+      "execution_count": 22,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "aOIAljc1FmXb",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.2)"
+      ],
+      "execution_count": 23,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "39nTAc3UFUMW",
+        "colab_type": "text"
+      },
+      "source": [
+        "## **4. Linear Regression Model**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "K0MokzGBCimk",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "from sklearn import linear_model\n",
+        "from sklearn.metrics import mean_squared_error, r2_score"
+      ],
+      "execution_count": 24,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "vkR1siPuFZ6X",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 34
+        },
+        "outputId": "e20472e7-0179-419f-8dc8-8710cfd008cc"
+      },
+      "source": [
+        "model = linear_model.LinearRegression()\n",
+        "model.fit(X_train, Y_train)"
+      ],
+      "execution_count": 25,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False)"
+            ]
+          },
+          "metadata": {
+            "tags": []
+          },
+          "execution_count": 25
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "aG4DMzc5Rks9",
+        "colab_type": "text"
+      },
+      "source": [
+        "### **Predicts the X_train**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "tZr9CBGvRp1F",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "Y_pred_train = model.predict(X_train)"
+      ],
+      "execution_count": 26,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "0x3saPCyRtJP",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 85
+        },
+        "outputId": "68235d8c-e66a-4acb-e4d3-83ef26e92629"
+      },
+      "source": [
+        "print('Coefficients:', model.coef_)\n",
+        "print('Intercept:', model.intercept_)\n",
+        "print('Mean squared error (MSE): %.2f'\n",
+        "      % mean_squared_error(Y_train, Y_pred_train))\n",
+        "print('Coefficient of determination (R^2): %.2f'\n",
+        "      % r2_score(Y_train, Y_pred_train))"
+      ],
+      "execution_count": 27,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "Coefficients: [-0.76779153 -0.00668131  0.00654032 -0.36959403]\n",
+            "Intercept: 0.3108998121270652\n",
+            "Mean squared error (MSE): 1.01\n",
+            "Coefficient of determination (R^2): 0.77\n"
+          ],
+          "name": "stdout"
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "M6evZTPNRecd",
+        "colab_type": "text"
+      },
+      "source": [
+        "### **Predicts the X_test**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "I_eFbrlaHhPU",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "Y_pred_test = model.predict(X_test)"
+      ],
+      "execution_count": 28,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "TQnDfyl5HkUr",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 85
+        },
+        "outputId": "32b22212-424a-458f-8f0f-57095f5903ec"
+      },
+      "source": [
+        "print('Coefficients:', model.coef_)\n",
+        "print('Intercept:', model.intercept_)\n",
+        "print('Mean squared error (MSE): %.2f'\n",
+        "      % mean_squared_error(Y_test, Y_pred_test))\n",
+        "print('Coefficient of determination (R^2): %.2f'\n",
+        "      % r2_score(Y_test, Y_pred_test))"
+      ],
+      "execution_count": 29,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "Coefficients: [-0.76779153 -0.00668131  0.00654032 -0.36959403]\n",
+            "Intercept: 0.3108998121270652\n",
+            "Mean squared error (MSE): 1.00\n",
+            "Coefficient of determination (R^2): 0.74\n"
+          ],
+          "name": "stdout"
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "nERFfdQBRFF5",
+        "colab_type": "text"
+      },
+      "source": [
+        "### **Linear Regression Equation**"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "j3xLiGWHFiY1",
+        "colab_type": "text"
+      },
+      "source": [
+        "The work of Delaney$^1$ provided the following linear regression equation:\n",
+        "\n",
+        "> LogS = 0.16 -  0.63 cLogP - 0.0062 MW + 0.066 RB - 0.74 AP\n",
+        "\n",
+        "The reproduction by Pat Walters$^2$ provided the following:\n",
+        "\n",
+        "> LogS = 0.26 -  0.74 LogP - 0.0066 MW + 0.0034 RB - 0.42 AP\n",
+        "\n",
+        "This notebook's reproduction gave the following equation:\n",
+        "\n",
+        "* Based on the Train set\n",
+        "> LogS = 0.30 -0.75 LogP - .0066 MW -0.0041 RB - 0.36 AP\n",
+        "\n",
+        "* Based on the Full dataset\n",
+        "> LogS =  0.26 -0.74 LogP - 0.0066 + MW 0.0032 RB - 0.42 AP"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "FaWyYnMbWtYu",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### **Our linear regression equation**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "0TH6J9evHIIE",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 34
+        },
+        "outputId": "9c960b7b-d2cf-4309-d187-60fa783a9529"
+      },
+      "source": [
+        "print('LogS = %.2f %.2f LogP %.4f MW + %.4f RB %.2f AP' % (model.intercept_, model.coef_[0], model.coef_[1], model.coef_[2], model.coef_[3] ) )"
+      ],
+      "execution_count": 33,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "LogS = 0.31 -0.77 LogP -0.0067 MW + 0.0065 RB -0.37 AP\n"
+          ],
+          "name": "stdout"
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "VcJyUzsLSz2A",
+        "colab_type": "text"
+      },
+      "source": [
+        "The same equation can also be produced with the following code (which breaks up the previous one-line code into several comprehensible lines."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "byUbJ9QqK5gA",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "yintercept = '%.2f' % model.intercept_\n",
+        "LogP = '%.2f LogP' % model.coef_[0]\n",
+        "MW = '%.4f MW' % model.coef_[1]\n",
+        "RB = '%.4f RB' % model.coef_[2]\n",
+        "AP = '%.2f AP' % model.coef_[3]"
+      ],
+      "execution_count": 31,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "QY-9rh--S-6g",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 34
+        },
+        "outputId": "095cefae-a92c-466d-8fb2-67abe856494f"
+      },
+      "source": [
+        "print('LogS = ' + \n",
+        "      ' ' + \n",
+        "      yintercept + \n",
+        "      ' ' + \n",
+        "      LogP + \n",
+        "      ' ' + \n",
+        "      MW + \n",
+        "      ' + ' + \n",
+        "      RB + \n",
+        "      ' ' + \n",
+        "      AP)"
+      ],
+      "execution_count": 34,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "LogS =  0.31 -0.77 LogP -0.0067 MW + 0.0065 RB -0.37 AP\n"
+          ],
+          "name": "stdout"
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "R3lRkSOJRm1q",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### **Use entire dataset for model training (For Comparison)**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "QUye6SsIRl9T",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 34
+        },
+        "outputId": "d93a60d1-cfb8-4ddd-843c-b3bd42be4161"
+      },
+      "source": [
+        "full = linear_model.LinearRegression()\n",
+        "full.fit(X, Y)"
+      ],
+      "execution_count": 35,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False)"
+            ]
+          },
+          "metadata": {
+            "tags": []
+          },
+          "execution_count": 35
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "6tMI8n0oR1b5",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "full_pred = model.predict(X)"
+      ],
+      "execution_count": 36,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "7ZVD8Fg1R6zt",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 85
+        },
+        "outputId": "323a7815-5a7d-49c6-cb13-ad60c7f57e69"
+      },
+      "source": [
+        "print('Coefficients:', full.coef_)\n",
+        "print('Intercept:', full.intercept_)\n",
+        "print('Mean squared error (MSE): %.2f'\n",
+        "      % mean_squared_error(Y, full_pred))\n",
+        "print('Coefficient of determination (R^2): %.2f'\n",
+        "      % r2_score(Y, full_pred))"
+      ],
+      "execution_count": 37,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "Coefficients: [-0.74173609 -0.00659927  0.00320051 -0.42316387]\n",
+            "Intercept: 0.2565006830997194\n",
+            "Mean squared error (MSE): 1.01\n",
+            "Coefficient of determination (R^2): 0.77\n"
+          ],
+          "name": "stdout"
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "AFYYzcc1VqIo",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "full_yintercept = '%.2f' % full.intercept_\n",
+        "full_LogP = '%.2f LogP' % full.coef_[0]\n",
+        "full_MW = '%.4f MW' % full.coef_[1]\n",
+        "full_RB = '+ %.4f RB' % full.coef_[2]\n",
+        "full_AP = '%.2f AP' % full.coef_[3]"
+      ],
+      "execution_count": 38,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "zwU4QJhhVsKb",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 34
+        },
+        "outputId": "7e3042f6-6447-497b-e6a8-6b58a57599e7"
+      },
+      "source": [
+        "print('LogS = ' + \n",
+        "      ' ' + \n",
+        "      full_yintercept + \n",
+        "      ' ' + \n",
+        "      full_LogP + \n",
+        "      ' ' + \n",
+        "      full_MW + \n",
+        "      ' ' + \n",
+        "      full_RB + \n",
+        "      ' ' + \n",
+        "      full_AP)"
+      ],
+      "execution_count": 39,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "LogS =  0.26 -0.74 LogP -0.0066 MW + 0.0032 RB -0.42 AP\n"
+          ],
+          "name": "stdout"
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "qp-hjUv4IWe-",
+        "colab_type": "text"
+      },
+      "source": [
+        "## **Scatter plot of experimental vs. predicted LogS**"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Q6bP41fKEY9O",
+        "colab_type": "text"
+      },
+      "source": [
+        "### **Quick check of the variable dimensions of Train and Test sets**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "LA5dH5oiEUnP",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 34
+        },
+        "outputId": "11d5f348-d3da-4bc3-ae52-96c6d607e14f"
+      },
+      "source": [
+        "Y_train.shape, Y_pred_train.shape"
+      ],
+      "execution_count": 41,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "((915,), (915,))"
+            ]
+          },
+          "metadata": {
+            "tags": []
+          },
+          "execution_count": 41
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "HIu7YbbFP-7o",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 34
+        },
+        "outputId": "6460df10-4f85-4b2b-bf60-d555afb7dc41"
+      },
+      "source": [
+        "Y_test.shape, Y_pred_test.shape"
+      ],
+      "execution_count": 42,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "((229,), (229,))"
+            ]
+          },
+          "metadata": {
+            "tags": []
+          },
+          "execution_count": 42
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "OHqv3TlYa5qF",
+        "colab_type": "text"
+      },
+      "source": [
+        "### **Vertical plot**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "shQPfrHIOmRD",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 660
+        },
+        "outputId": "1ddf9dac-2be1-482f-8451-3e4a978fffd0"
+      },
+      "source": [
+        "import matplotlib.pyplot as plt\n",
+        "import numpy as np\n",
+        "\n",
+        "plt.figure(figsize=(5,11))\n",
+        "\n",
+        "# 2 row, 1 column, plot 1\n",
+        "plt.subplot(2, 1, 1)\n",
+        "plt.scatter(x=Y_train, y=Y_pred_train, c=\"#7CAE00\", alpha=0.3)\n",
+        "\n",
+        "# Add trendline\n",
+        "# https://stackoverflow.com/questions/26447191/how-to-add-trendline-in-python-matplotlib-dot-scatter-graphs\n",
+        "z = np.polyfit(Y_train, Y_pred_train, 1)\n",
+        "p = np.poly1d(z)\n",
+        "plt.plot(Y_test,p(Y_test),\"#F8766D\")\n",
+        "\n",
+        "plt.ylabel('Predicted LogS')\n",
+        "\n",
+        "\n",
+        "# 2 row, 1 column, plot 2\n",
+        "plt.subplot(2, 1, 2)\n",
+        "plt.scatter(x=Y_test, y=Y_pred_test, c=\"#619CFF\", alpha=0.3)\n",
+        "\n",
+        "z = np.polyfit(Y_test, Y_pred_test, 1)\n",
+        "p = np.poly1d(z)\n",
+        "plt.plot(Y_test,p(Y_test),\"#F8766D\")\n",
+        "\n",
+        "plt.ylabel('Predicted LogS')\n",
+        "plt.xlabel('Experimental LogS')\n",
+        "\n",
+        "plt.savefig('plot_vertical_logS.png')\n",
+        "plt.savefig('plot_vertical_logS.pdf')\n",
+        "plt.show()"
+      ],
+      "execution_count": 44,
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 360x792 with 2 Axes>"
+            ]
+          },
+          "metadata": {
+            "tags": [],
+            "needs_background": "light"
+          }
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "PswCQ7Yra_CW",
+        "colab_type": "text"
+      },
+      "source": [
+        "### **Horizontal plot**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "xG7NWEscT8QO",
+        "colab_type": "code",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 334
+        },
+        "outputId": "c22932ab-ba03-46d6-a83f-ae0ebe0d3dca"
+      },
+      "source": [
+        "plt.figure(figsize=(11,5))\n",
+        "\n",
+        "# 1 row, 2 column, plot 1\n",
+        "plt.subplot(1, 2, 1)\n",
+        "plt.scatter(x=Y_train, y=Y_pred_train, c=\"#7CAE00\", alpha=0.3)\n",
+        "\n",
+        "z = np.polyfit(Y_train, Y_pred_train, 1)\n",
+        "p = np.poly1d(z)\n",
+        "plt.plot(Y_test,p(Y_test),\"#F8766D\")\n",
+        "\n",
+        "plt.ylabel('Predicted LogS')\n",
+        "plt.xlabel('Experimental LogS')\n",
+        "\n",
+        "# 1 row, 2 column, plot 2\n",
+        "plt.subplot(1, 2, 2)\n",
+        "plt.scatter(x=Y_test, y=Y_pred_test, c=\"#619CFF\", alpha=0.3)\n",
+        "\n",
+        "z = np.polyfit(Y_test, Y_pred_test, 1)\n",
+        "p = np.poly1d(z)\n",
+        "plt.plot(Y_test,p(Y_test),\"#F8766D\")\n",
+        "\n",
+        "plt.xlabel('Experimental LogS')\n",
+        "\n",
+        "plt.savefig('plot_horizontal_logS.png')\n",
+        "plt.savefig('plot_horizontal_logS.pdf')\n",
+        "plt.show()"
+      ],
+      "execution_count": 45,
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 792x360 with 2 Axes>"
+            ]
+          },
+          "metadata": {
+            "tags": [],
+            "needs_background": "light"
+          }
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "ARiv3f1iC565",
+        "colab_type": "text"
+      },
+      "source": [
+        "---"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "jwM1QHeLbxJl",
+        "colab_type": "text"
+      },
+      "source": [
+        "## **Reference**\n",
+        "\n",
+        "1. John S. Delaney. [ESOL:  Estimating Aqueous Solubility Directly from Molecular Structure](https://pubs.acs.org/doi/10.1021/ci034243x). ***J. Chem. Inf. Comput. Sci.*** 2004, 44, 3, 1000-1005.\n",
+        "\n",
+        "2. Pat Walters. [Predicting Aqueous Solubility - It's Harder Than It Looks](http://practicalcheminformatics.blogspot.com/2018/09/predicting-aqueous-solubility-its.html). ***Practical Cheminformatics Blog***\n",
+        "\n",
+        "3. Bharath Ramsundar, Peter Eastman, Patrick Walters, and Vijay Pande. [Deep Learning for the Life Sciences: Applying Deep Learning to Genomics, Microscopy, Drug Discovery, and More](https://learning.oreilly.com/library/view/deep-learning-for/9781492039822/), O'Reilly, 2019.\n",
+        "\n",
+        "4. [Supplementary file](https://pubs.acs.org/doi/10.1021/ci034243x) from Delaney's ESOL:  Estimating Aqueous Solubility Directly from Molecular Structure."
+      ]
+    }
+  ]
+}
\ No newline at end of file