6 QSAR Model Building of Beta-Secretase 1 Inhibitors.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "4bb766bb",
   "metadata": {},
   "source": [
    "## QSAR Model Building of Beta-Secretase 1 Inhibitors"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8cac5e1d",
   "metadata": {},
   "source": [
    "### Read in data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "b944730f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3274669c",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PubchemFP0</th>\n",
       "      <th>PubchemFP1</th>\n",
       "      <th>PubchemFP2</th>\n",
       "      <th>PubchemFP3</th>\n",
       "      <th>PubchemFP4</th>\n",
       "      <th>PubchemFP5</th>\n",
       "      <th>PubchemFP6</th>\n",
       "      <th>PubchemFP7</th>\n",
       "      <th>PubchemFP8</th>\n",
       "      <th>PubchemFP9</th>\n",
       "      <th>...</th>\n",
       "      <th>PubchemFP872</th>\n",
       "      <th>PubchemFP873</th>\n",
       "      <th>PubchemFP874</th>\n",
       "      <th>PubchemFP875</th>\n",
       "      <th>PubchemFP876</th>\n",
       "      <th>PubchemFP877</th>\n",
       "      <th>PubchemFP878</th>\n",
       "      <th>PubchemFP879</th>\n",
       "      <th>PubchemFP880</th>\n",
       "      <th>pIC50</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.384050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.698970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.337242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.045757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.251812</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7229</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.356547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7230</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.036212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7231</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7232</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.966576</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7233</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.567031</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7234 rows × 882 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      PubchemFP0  PubchemFP1  PubchemFP2  PubchemFP3  PubchemFP4  PubchemFP5  \\\n",
       "0              1           1           1           1           0           0   \n",
       "1              1           1           1           1           0           0   \n",
       "2              1           1           1           1           0           0   \n",
       "3              1           1           1           1           0           0   \n",
       "4              1           1           1           1           0           0   \n",
       "...          ...         ...         ...         ...         ...         ...   \n",
       "7229           1           1           1           1           0           0   \n",
       "7230           1           1           1           1           0           0   \n",
       "7231           1           1           1           1           0           0   \n",
       "7232           1           1           1           1           0           0   \n",
       "7233           1           1           1           1           0           0   \n",
       "\n",
       "      PubchemFP6  PubchemFP7  PubchemFP8  PubchemFP9  ...  PubchemFP872  \\\n",
       "0              0           0           0           1  ...             0   \n",
       "1              0           0           0           1  ...             0   \n",
       "2              0           0           0           1  ...             0   \n",
       "3              0           0           0           1  ...             0   \n",
       "4              0           0           0           1  ...             0   \n",
       "...          ...         ...         ...         ...  ...           ...   \n",
       "7229           0           0           0           1  ...             0   \n",
       "7230           0           0           0           1  ...             0   \n",
       "7231           0           0           0           1  ...             0   \n",
       "7232           0           0           0           1  ...             0   \n",
       "7233           0           0           0           1  ...             0   \n",
       "\n",
       "      PubchemFP873  PubchemFP874  PubchemFP875  PubchemFP876  PubchemFP877  \\\n",
       "0                0             0             0             0             0   \n",
       "1                0             0             0             0             0   \n",
       "2                0             0             0             0             0   \n",
       "3                0             0             0             0             0   \n",
       "4                0             0             0             0             0   \n",
       "...            ...           ...           ...           ...           ...   \n",
       "7229             0             0             0             0             0   \n",
       "7230             0             0             0             0             0   \n",
       "7231             0             0             0             0             0   \n",
       "7232             0             0             0             0             0   \n",
       "7233             0             0             0             0             0   \n",
       "\n",
       "      PubchemFP878  PubchemFP879  PubchemFP880     pIC50  \n",
       "0                0             0             0  6.384050  \n",
       "1                0             0             0  8.698970  \n",
       "2                0             0             0  6.337242  \n",
       "3                0             0             0  5.045757  \n",
       "4                0             0             0  5.251812  \n",
       "...            ...           ...           ...       ...  \n",
       "7229             0             0             0  5.356547  \n",
       "7230             0             0             0  5.036212  \n",
       "7231             0             0             0  5.000000  \n",
       "7232             0             0             0  4.966576  \n",
       "7233             0             0             0  5.567031  \n",
       "\n",
       "[7234 rows x 882 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# using calculated pubchem fingerprints\n",
    "dataset_path = 'beta_secretase_bioactivity_data_3class_pIC50_pubchem_fp.csv'\n",
    "dataset = pd.read_csv(dataset_path)\n",
    "dataset\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "4e6b3894",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PubchemFP0</th>\n",
       "      <th>PubchemFP1</th>\n",
       "      <th>PubchemFP2</th>\n",
       "      <th>PubchemFP3</th>\n",
       "      <th>PubchemFP4</th>\n",
       "      <th>PubchemFP5</th>\n",
       "      <th>PubchemFP6</th>\n",
       "      <th>PubchemFP7</th>\n",
       "      <th>PubchemFP8</th>\n",
       "      <th>PubchemFP9</th>\n",
       "      <th>...</th>\n",
       "      <th>PubchemFP871</th>\n",
       "      <th>PubchemFP872</th>\n",
       "      <th>PubchemFP873</th>\n",
       "      <th>PubchemFP874</th>\n",
       "      <th>PubchemFP875</th>\n",
       "      <th>PubchemFP876</th>\n",
       "      <th>PubchemFP877</th>\n",
       "      <th>PubchemFP878</th>\n",
       "      <th>PubchemFP879</th>\n",
       "      <th>PubchemFP880</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7229</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7230</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7231</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7232</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7233</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7234 rows × 881 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      PubchemFP0  PubchemFP1  PubchemFP2  PubchemFP3  PubchemFP4  PubchemFP5  \\\n",
       "0              1           1           1           1           0           0   \n",
       "1              1           1           1           1           0           0   \n",
       "2              1           1           1           1           0           0   \n",
       "3              1           1           1           1           0           0   \n",
       "4              1           1           1           1           0           0   \n",
       "...          ...         ...         ...         ...         ...         ...   \n",
       "7229           1           1           1           1           0           0   \n",
       "7230           1           1           1           1           0           0   \n",
       "7231           1           1           1           1           0           0   \n",
       "7232           1           1           1           1           0           0   \n",
       "7233           1           1           1           1           0           0   \n",
       "\n",
       "      PubchemFP6  PubchemFP7  PubchemFP8  PubchemFP9  ...  PubchemFP871  \\\n",
       "0              0           0           0           1  ...             0   \n",
       "1              0           0           0           1  ...             0   \n",
       "2              0           0           0           1  ...             0   \n",
       "3              0           0           0           1  ...             0   \n",
       "4              0           0           0           1  ...             0   \n",
       "...          ...         ...         ...         ...  ...           ...   \n",
       "7229           0           0           0           1  ...             0   \n",
       "7230           0           0           0           1  ...             0   \n",
       "7231           0           0           0           1  ...             0   \n",
       "7232           0           0           0           1  ...             0   \n",
       "7233           0           0           0           1  ...             0   \n",
       "\n",
       "      PubchemFP872  PubchemFP873  PubchemFP874  PubchemFP875  PubchemFP876  \\\n",
       "0                0             0             0             0             0   \n",
       "1                0             0             0             0             0   \n",
       "2                0             0             0             0             0   \n",
       "3                0             0             0             0             0   \n",
       "4                0             0             0             0             0   \n",
       "...            ...           ...           ...           ...           ...   \n",
       "7229             0             0             0             0             0   \n",
       "7230             0             0             0             0             0   \n",
       "7231             0             0             0             0             0   \n",
       "7232             0             0             0             0             0   \n",
       "7233             0             0             0             0             0   \n",
       "\n",
       "      PubchemFP877  PubchemFP878  PubchemFP879  PubchemFP880  \n",
       "0                0             0             0             0  \n",
       "1                0             0             0             0  \n",
       "2                0             0             0             0  \n",
       "3                0             0             0             0  \n",
       "4                0             0             0             0  \n",
       "...            ...           ...           ...           ...  \n",
       "7229             0             0             0             0  \n",
       "7230             0             0             0             0  \n",
       "7231             0             0             0             0  \n",
       "7232             0             0             0             0  \n",
       "7233             0             0             0             0  \n",
       "\n",
       "[7234 rows x 881 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = dataset.drop(['pIC50'], axis = 1)\n",
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "222f9347",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       6.384050\n",
       "1       8.698970\n",
       "2       6.337242\n",
       "3       5.045757\n",
       "4       5.251812\n",
       "          ...   \n",
       "7229    5.356547\n",
       "7230    5.036212\n",
       "7231    5.000000\n",
       "7232    4.966576\n",
       "7233    5.567031\n",
       "Name: pIC50, Length: 7234, dtype: float64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y = dataset.iloc[:,-1]\n",
    "Y"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2aaf08b9",
   "metadata": {},
   "source": [
    "### Remove low variance features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "5fd0c62e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PubchemFP3</th>\n",
       "      <th>PubchemFP13</th>\n",
       "      <th>PubchemFP16</th>\n",
       "      <th>PubchemFP19</th>\n",
       "      <th>PubchemFP20</th>\n",
       "      <th>PubchemFP23</th>\n",
       "      <th>PubchemFP24</th>\n",
       "      <th>PubchemFP33</th>\n",
       "      <th>PubchemFP37</th>\n",
       "      <th>PubchemFP115</th>\n",
       "      <th>...</th>\n",
       "      <th>PubchemFP755</th>\n",
       "      <th>PubchemFP756</th>\n",
       "      <th>PubchemFP776</th>\n",
       "      <th>PubchemFP777</th>\n",
       "      <th>PubchemFP797</th>\n",
       "      <th>PubchemFP798</th>\n",
       "      <th>PubchemFP800</th>\n",
       "      <th>PubchemFP818</th>\n",
       "      <th>PubchemFP819</th>\n",
       "      <th>PubchemFP860</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7229</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7230</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7231</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7232</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7233</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7234 rows × 217 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      PubchemFP3  PubchemFP13  PubchemFP16  PubchemFP19  PubchemFP20  \\\n",
       "0              1            1            1            1            1   \n",
       "1              1            1            1            1            1   \n",
       "2              1            1            1            1            1   \n",
       "3              1            1            1            1            1   \n",
       "4              1            1            1            1            1   \n",
       "...          ...          ...          ...          ...          ...   \n",
       "7229           1            1            1            1            1   \n",
       "7230           1            1            1            1            1   \n",
       "7231           1            1            1            1            1   \n",
       "7232           1            0            1            1            0   \n",
       "7233           1            1            0            1            0   \n",
       "\n",
       "      PubchemFP23  PubchemFP24  PubchemFP33  PubchemFP37  PubchemFP115  ...  \\\n",
       "0               0            0            1            0             0  ...   \n",
       "1               0            0            1            0             0  ...   \n",
       "2               0            0            0            0             0  ...   \n",
       "3               0            0            1            0             0  ...   \n",
       "4               0            0            0            0             0  ...   \n",
       "...           ...          ...          ...          ...           ...  ...   \n",
       "7229            0            0            1            0             0  ...   \n",
       "7230            0            0            1            0             0  ...   \n",
       "7231            0            0            1            0             0  ...   \n",
       "7232            0            0            1            1             0  ...   \n",
       "7233            0            0            0            1             0  ...   \n",
       "\n",
       "      PubchemFP755  PubchemFP756  PubchemFP776  PubchemFP777  PubchemFP797  \\\n",
       "0                0             0             1             0             0   \n",
       "1                0             0             1             0             0   \n",
       "2                0             0             0             0             0   \n",
       "3                0             0             0             0             0   \n",
       "4                0             0             1             0             0   \n",
       "...            ...           ...           ...           ...           ...   \n",
       "7229             0             0             0             0             0   \n",
       "7230             0             0             0             0             0   \n",
       "7231             1             0             0             0             0   \n",
       "7232             0             0             1             0             1   \n",
       "7233             0             0             1             0             1   \n",
       "\n",
       "      PubchemFP798  PubchemFP800  PubchemFP818  PubchemFP819  PubchemFP860  \n",
       "0                0             0             0             0             0  \n",
       "1                0             0             0             0             0  \n",
       "2                0             0             0             0             0  \n",
       "3                0             0             0             0             0  \n",
       "4                0             0             0             0             0  \n",
       "...            ...           ...           ...           ...           ...  \n",
       "7229             0             0             0             0             0  \n",
       "7230             0             0             0             0             0  \n",
       "7231             0             1             1             0             0  \n",
       "7232             0             1             1             0             0  \n",
       "7233             1             1             1             0             0  \n",
       "\n",
       "[7234 rows x 217 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# to remove excessive, redundant features\n",
    "from sklearn.feature_selection import VarianceThreshold\n",
    "\n",
    "def remove_low_variance(input_data, threshold = 0.1):\n",
    "    selection = VarianceThreshold(threshold)\n",
    "    selection.fit(input_data)\n",
    "    return input_data[input_data.columns[selection.get_support(indices = True)]]\n",
    "X = remove_low_variance(X, threshold = 0.1)\n",
    "X    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "0c18a279",
   "metadata": {},
   "outputs": [],
   "source": [
    "X.to_csv('descriptor_list.csv', index = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a672c397",
   "metadata": {},
   "source": [
    "### Random Forest Regression Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "ed909ce3",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.metrics import mean_squared_error, r2_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "6ed06b41",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8304362381476093"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = RandomForestRegressor(n_estimators = 500, random_state = 42)\n",
    "model.fit(X, Y)\n",
    "r2 = model.score(X, Y)\n",
    "r2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3116de74",
   "metadata": {},
   "source": [
    "### Model Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "b2c36938",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([7.20207561, 7.20207561, 6.09079988, ..., 5.2011968 , 5.28940958,\n",
       "       5.88575816])"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_pred = model.predict(X)\n",
    "Y_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8caaed28",
   "metadata": {},
   "source": [
    "### Model Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "c6f1a0d6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean squared error (MSE): 0.33\n",
      "Coefficient of determination (R^2): 0.83\n"
     ]
    }
   ],
   "source": [
    "print('Mean squared error (MSE): %.2f'\n",
    "      % mean_squared_error(Y, Y_pred))\n",
    "print('Coefficient of determination (R^2): %.2f'\n",
    "      % r2_score(Y, Y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42821081",
   "metadata": {},
   "source": [
    "### Data Visualization (Experimental vs Predicted pIC50 for Training Data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "6812cb00",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "2ed3d1bb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Experimental pIC50')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (5, 5))\n",
    "plt.scatter(x = Y, y = Y_pred, c = '#9EBC9F', alpha = 0.3)\n",
    "\n",
    "z = np.polyfit(Y, Y_pred, 1)\n",
    "p = np.poly1d(z)\n",
    "\n",
    "plt.plot(Y, p(Y),'#953019')\n",
    "plt.ylabel('Predicted pIC50')\n",
    "plt.xlabel('Experimental pIC50')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1eb06099",
   "metadata": {},
   "source": [
    "### Save Model as Pickle Object "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "314d51ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "df2e2938",
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(model, open('beta_secretase_model.pkl', 'wb'))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:rdkit-env-test]",
   "language": "python",
   "name": "conda-env-rdkit-env-test-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}