2_Encodings.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6db51034-1c6d-421f-864e-b6bfc7d8a3a6",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Encodings for Categorical Variables\n",
    "\n",
    "How to use chemical and custom encodigns for your categorical variables.\n",
    "\n",
    "Find more information here:\n",
    "- [Examples in Different Search Spaces](https://emdgroup.github.io/baybe/examples/Searchspaces/Searchspaces.html)\n",
    "- [Backtesting Examples](https://emdgroup.github.io/baybe/examples/Backtesting/Backtesting.html)\n",
    "- [BayBE GitHub](https://github.com/emdgroup/baybe)\n",
    "- [BayBE Userguide](https://emdgroup.github.io/baybe/userguide/userguide.html)\n",
    "\n",
    "<a target=\"_blank\" href=\"https://colab.research.google.com/github/emdgroup/baybe-ac24-workshop/blob/main/2_Encodings.ipynb\">\n",
    "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
    "</a>\n",
    "<a target=\"_blank\" href=\"https://mybinder.org/v2/gh/emdgroup/baybe-ac24-workshop/HEAD?labpath=2_Encodings.ipynb\">\n",
    "  <img src=\"https://mybinder.org/badge_logo.svg\" alt=\"Launch Binder\"/>\n",
    "</a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08ad7e38-ef54-467d-91d4-ea2a657cb1b9",
   "metadata": {},
   "source": [
    "# Install and Import"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "463b897f-fb8f-4fd2-9179-f343b8c76f32",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install 'baybe[chem,simulation]==0.10.0' seaborn matplotlib pandas[excel]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "907fbf7f-dd1c-4b0a-80d9-3e0195a72453",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "from botorch.exceptions import InputDataWarning\n",
    "warnings.filterwarnings('ignore', category = InputDataWarning)\n",
    "warnings.filterwarnings('ignore', message = '.*BayBE.*')\n",
    "\n",
    "import baybe\n",
    "baybe.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e330bf48-8266-4a11-9ec7-93fe0fe0f697",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "os.environ['BAYBE_CACHE_DIR']='' # turn descriptor caching off\n",
    "from baybe.utils.random import set_random_seed\n",
    "set_random_seed(1337)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f9db81d0-c711-46ae-b55e-83bcedb167cc",
   "metadata": {},
   "source": [
    "# Create the Lookup\n",
    "The problem we study here, corresponds to the chemical reaciton yield optimization from [Shields, B.J., Stevens et al. Nature 590, 89–96 (2021)](https://doi.org/10.1038/s41586-021-03213-y).\n",
    "![reaction](https://raw.githubusercontent.com/emdgroup/baybe-ac24-workshop/main/files/reaction.png)\n",
    "\n",
    "The parameters screened are:\n",
    "- Solvent\n",
    "- Base\n",
    "- Ligand\n",
    "- Concentration (of the solvent)\n",
    "- Temperature\n",
    "\n",
    "There is one target that should be maximized:\n",
    "- yield"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "5943f9e7-52e0-4eac-8f72-cae47e2be66e",
   "metadata": {
    "tags": []
   },
   "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>Base_SMILES</th>\n",
       "      <th>Ligand_SMILES</th>\n",
       "      <th>Solvent_SMILES</th>\n",
       "      <th>Concentration</th>\n",
       "      <th>Temp_C</th>\n",
       "      <th>yield</th>\n",
       "      <th>Base</th>\n",
       "      <th>Ligand</th>\n",
       "      <th>Solvent</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entry</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>O=C([O-])C.[K+]</td>\n",
       "      <td>CC(C)C1=CC(C(C)C)=C(C(C(C)C)=C1)C2=C(P(C3CCCCC...</td>\n",
       "      <td>CC(N(C)C)=O</td>\n",
       "      <td>0.1</td>\n",
       "      <td>105</td>\n",
       "      <td>5.47</td>\n",
       "      <td>Potassium acetate</td>\n",
       "      <td>BrettPhos</td>\n",
       "      <td>DMAc</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>O=C([O-])C.[K+]</td>\n",
       "      <td>CC(C)(C)P(C1=CC=CC=C1)C(C)(C)C</td>\n",
       "      <td>CC(N(C)C)=O</td>\n",
       "      <td>0.1</td>\n",
       "      <td>105</td>\n",
       "      <td>0.00</td>\n",
       "      <td>Potassium acetate</td>\n",
       "      <td>Di-tert-butylphenylphosphine</td>\n",
       "      <td>DMAc</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>O=C([O-])C.[K+]</td>\n",
       "      <td>CN(C)C1=CC=CC(N(C)C)=C1C2=CC=CC=C2P(C(C)(C)C)C...</td>\n",
       "      <td>CC(N(C)C)=O</td>\n",
       "      <td>0.1</td>\n",
       "      <td>105</td>\n",
       "      <td>78.95</td>\n",
       "      <td>Potassium acetate</td>\n",
       "      <td>(t-Bu)PhCPhos</td>\n",
       "      <td>DMAc</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>O=C([O-])C.[K+]</td>\n",
       "      <td>P(C1CCCCC1)(C2CCCCC2)C3CCCCC3</td>\n",
       "      <td>CC(N(C)C)=O</td>\n",
       "      <td>0.1</td>\n",
       "      <td>105</td>\n",
       "      <td>7.26</td>\n",
       "      <td>Potassium acetate</td>\n",
       "      <td>Tricyclohexylphosphine</td>\n",
       "      <td>DMAc</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>O=C([O-])C.[K+]</td>\n",
       "      <td>P(C1=CC=CC=C1)(C2=CC=CC=C2)C3=CC=CC=C3</td>\n",
       "      <td>CC(N(C)C)=O</td>\n",
       "      <td>0.1</td>\n",
       "      <td>105</td>\n",
       "      <td>28.15</td>\n",
       "      <td>Potassium acetate</td>\n",
       "      <td>PPh3</td>\n",
       "      <td>DMAc</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Base_SMILES                                      Ligand_SMILES  \\\n",
       "entry                                                                       \n",
       "0      O=C([O-])C.[K+]  CC(C)C1=CC(C(C)C)=C(C(C(C)C)=C1)C2=C(P(C3CCCCC...   \n",
       "1      O=C([O-])C.[K+]                     CC(C)(C)P(C1=CC=CC=C1)C(C)(C)C   \n",
       "2      O=C([O-])C.[K+]  CN(C)C1=CC=CC(N(C)C)=C1C2=CC=CC=C2P(C(C)(C)C)C...   \n",
       "3      O=C([O-])C.[K+]                      P(C1CCCCC1)(C2CCCCC2)C3CCCCC3   \n",
       "4      O=C([O-])C.[K+]             P(C1=CC=CC=C1)(C2=CC=CC=C2)C3=CC=CC=C3   \n",
       "\n",
       "      Solvent_SMILES  Concentration  Temp_C  yield               Base  \\\n",
       "entry                                                                   \n",
       "0        CC(N(C)C)=O            0.1     105   5.47  Potassium acetate   \n",
       "1        CC(N(C)C)=O            0.1     105   0.00  Potassium acetate   \n",
       "2        CC(N(C)C)=O            0.1     105  78.95  Potassium acetate   \n",
       "3        CC(N(C)C)=O            0.1     105   7.26  Potassium acetate   \n",
       "4        CC(N(C)C)=O            0.1     105  28.15  Potassium acetate   \n",
       "\n",
       "                             Ligand Solvent  \n",
       "entry                                        \n",
       "0                         BrettPhos    DMAc  \n",
       "1      Di-tert-butylphenylphosphine    DMAc  \n",
       "2                     (t-Bu)PhCPhos    DMAc  \n",
       "3            Tricyclohexylphosphine    DMAc  \n",
       "4                              PPh3    DMAc  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url = 'https://raw.githubusercontent.com/emdgroup/baybe/main/examples/Backtesting/lookup.xlsx'\n",
    "lookup = pd.read_excel(url, index_col=0)\n",
    "F_BEST = lookup['yield'].max()\n",
    "lookup.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e8ff6c6e-1bca-4e1e-b7d0-399ad03396e6",
   "metadata": {},
   "source": [
    "### SMILES\n",
    "The structure of the molecules is represented by a SMILES, a text based representation of molecules.\n",
    "In particular for small molecuels it is a convenient way of presenting a molecule in the computer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b0926d9b-2d74-44f5-81c6-6286a1a56508",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/jpeg": "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",
      "image/png": "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",
      "text/plain": [
       "<PIL.PngImagePlugin.PngImageFile image mode=RGB size=300x300>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from rdkit import Chem\n",
    "from rdkit.Chem import Draw\n",
    "\n",
    "mol = Chem.MolFromSmiles('P(C1=CC=CC=C1)(C2=CC=CC=C2)C3=CC=CC=C3') #PPh3\n",
    "Draw.MolToImage(mol)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b66c30b-e619-4525-9477-e66fc3918228",
   "metadata": {},
   "source": [
    "### Target Distribution\n",
    "This is a needle-in-a-haystack problem."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "74bd7020-356b-4055-a807-bf3e60c7932a",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.histplot(data = lookup, x = 'yield', edgecolor = 'black');\n",
    "plt.title('Reaction Yield Histogram');"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d7c3c511-e709-40e8-bda9-2bcdf2f7fc24",
   "metadata": {},
   "source": [
    "### Assemble Label-Structure Pairs Needed for the Parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6476d3b9-7f0a-4330-8a4e-f86a101ccf50",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Butyl Ester': 'CCCCOC(C)=O',\n",
       " 'Butyornitrile': 'CCCC#N',\n",
       " 'DMAc': 'CC(N(C)C)=O',\n",
       " 'p-Xylene': 'CC1=CC=C(C)C=C1'}"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "solvent_data = dict(sorted(set(zip(lookup.Solvent, lookup.Solvent_SMILES))))\n",
    "solvent_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "3dd023f4-ff04-4f94-ac6c-5c7ad4cfbfda",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Cesium acetate': 'O=C([O-])C.[Cs+]',\n",
       " 'Cesium pivalate': 'O=C([O-])C(C)(C)C.[Cs+]',\n",
       " 'Potassium acetate': 'O=C([O-])C.[K+]',\n",
       " 'Potassium pivalate': 'O=C([O-])C(C)(C)C.[K+]'}"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "base_data = dict(sorted(set(zip(lookup.Base, lookup.Base_SMILES))))\n",
    "base_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ffa0d324-ef4b-4ac3-bb56-a255d0456037",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'(t-Bu)PhCPhos': 'CN(C)C1=CC=CC(N(C)C)=C1C2=CC=CC=C2P(C(C)(C)C)C3=CC=CC=C3',\n",
       " '1268824-69-6': 'CC(OC1=C(P(C2CCCCC2)C3CCCCC3)C(OC(C)C)=CC=C1)C',\n",
       " 'BrettPhos': 'CC(C)C1=CC(C(C)C)=C(C(C(C)C)=C1)C2=C(P(C3CCCCC3)C4CCCCC4)C(OC)=CC=C2OC',\n",
       " 'Di-tert-butylphenylphosphine': 'CC(C)(C)P(C1=CC=CC=C1)C(C)(C)C',\n",
       " 'JackiePhos': 'FC(F)(F)C1=CC(P(C2=C(C3=C(C(C)C)C=C(C(C)C)C=C3C(C)C)C(OC)=CC=C2OC)C4=CC(C(F)(F)F)=CC(C(F)(F)F)=C4)=CC(C(F)(F)F)=C1',\n",
       " 'Me2PPh': 'CP(C)C1=CC=CC=C1',\n",
       " 'Methyldiphenylphosphine': 'CP(C1=CC=CC=C1)C2=CC=CC=C2',\n",
       " 'P(2-furyl)3': 'P(C1=CC=CO1)(C2=CC=CO2)C3=CC=CO3',\n",
       " 'PPh3': 'P(C1=CC=CC=C1)(C2=CC=CC=C2)C3=CC=CC=C3',\n",
       " 'SCHEMBL15068049': 'C[C@]1(O2)O[C@](C[C@]2(C)P3C4=CC=CC=C4)(C)O[C@]3(C)C1',\n",
       " 'Tricyclohexylphosphine': 'P(C1CCCCC1)(C2CCCCC2)C3CCCCC3',\n",
       " 'XPhos': 'CC(C1=C(C2=CC=CC=C2P(C3CCCCC3)C4CCCCC4)C(C(C)C)=CC(C(C)C)=C1)C'}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ligand_data = dict(sorted(set(zip( lookup.Ligand, lookup.Ligand_SMILES))))\n",
    "ligand_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "83ca5a38-d577-40f5-aab1-e43044ca6be8",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{90, 105, 120}"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temperature_values = set(lookup.Temp_C)\n",
    "temperature_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "0f8ab10c-780b-4033-acbc-b6763b737b52",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0.057, 0.1, 0.153}"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "concentration_values = set(lookup.Concentration)\n",
    "concentration_values"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c0a248d6-2bab-4df6-a5ee-93adba8a2d7f",
   "metadata": {},
   "source": [
    "# Create Different Parametrizations"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4556764-4c8a-447c-a988-7f1b8d21c856",
   "metadata": {},
   "source": [
    "## Set up Parametrization for Solvent, Base, Temp and Concentration\n",
    "\n",
    "For four out of the five parameters we will keep the parameteriyation the same.\n",
    "\n",
    "We will encode the base and solvent with the [`SubstanceParameter`](https://emdgroup.github.io/baybe/userguide/parameters.html#substanceparameter) which requires `data` in the form of the above created dictionary mappings between substance labels and their SMILES.\n",
    "\n",
    "For the temperature and concentration we use the\n",
    "[`NumericalDiscreteParameter`](https://emdgroup.github.io/baybe/userguide/parameters.html#numericaldiscreteparameter), which simply needs a sequence of `values`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "86acc617-0c0a-4233-a5c9-677300b98227",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from baybe.parameters import (SubstanceParameter, \n",
    "                              CustomDiscreteParameter, \n",
    "                              CategoricalParameter, \n",
    "                              NumericalDiscreteParameter)\n",
    "\n",
    "# Substance parameters including chemical descriptors\n",
    "p_solvent = SubstanceParameter(name = \"Solvent\", data = solvent_data)\n",
    "p_base = SubstanceParameter(name = \"Base\", data = base_data)\n",
    "\n",
    "# Discrete numerical parameters\n",
    "p_temp = NumericalDiscreteParameter(name = \"Temp_C\", values = temperature_values)\n",
    "p_concentration = NumericalDiscreteParameter(name = \"Concentration\", values = concentration_values)\n",
    "\n",
    "shared_parameters = [\n",
    "    p_solvent,\n",
    "    p_base, \n",
    "    p_concentration, \n",
    "    p_temp\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e871045-6ee1-4e7e-abed-291b832f7b92",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Create Different Ligand Parameterizations\n",
    "\n",
    "To illustrate the role of featurization for categorical parameters we will create three differen variants for the ligand. In this type of reaction, the ligand is known to play a crucial role."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "28685748-67c9-4e13-a507-cd5e4bddb8de",
   "metadata": {},
   "source": [
    "### MORDRED Encoding\n",
    "\n",
    "[MORDRED](https://github.com/JacksonBurns/mordred-community) and [RDKIT](https://github.com/rdkit/rdkit) are well known cheminformatics packages which allow manipulation of small molecules, but also calculation of descriptors for them. There exist many decades of work and over the years thousands of descriptors have been developed. In general, these descriptors are quickly calculated, but include several approximations to enable this.\n",
    "\n",
    "Since surrogate models generally scale with data dimensionality, it is a bad idea to use all of these descriptors. For this, BayBE provides the `decorrelate` keyword. With this you can specify a threshold of pairwise column correlation above which one of the columns will be dropped.\n",
    "\n",
    "Below we explicitly turn decorrelation off to see the resulting (bad) featurization."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "f03e790a-206b-499b-a378-049b100917b3",
   "metadata": {
    "tags": []
   },
   "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>Ligand_MORDRED_ABC</th>\n",
       "      <th>Ligand_MORDRED_ABCGG</th>\n",
       "      <th>...</th>\n",
       "      <th>Ligand_MORDRED_mZagreb1</th>\n",
       "      <th>Ligand_MORDRED_mZagreb2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>(t-Bu)PhCPhos</th>\n",
       "      <td>22.449495</td>\n",
       "      <td>19.292201</td>\n",
       "      <td>...</td>\n",
       "      <td>11.062500</td>\n",
       "      <td>6.305556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1268824-69-6</th>\n",
       "      <td>20.741455</td>\n",
       "      <td>17.344789</td>\n",
       "      <td>...</td>\n",
       "      <td>8.638889</td>\n",
       "      <td>6.055556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BrettPhos</th>\n",
       "      <td>29.162435</td>\n",
       "      <td>24.462781</td>\n",
       "      <td>...</td>\n",
       "      <td>13.555556</td>\n",
       "      <td>8.583333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Di-tert-butylphenylphosphine</th>\n",
       "      <td>11.396454</td>\n",
       "      <td>10.965064</td>\n",
       "      <td>...</td>\n",
       "      <td>7.597222</td>\n",
       "      <td>3.111111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JackiePhos</th>\n",
       "      <td>42.136729</td>\n",
       "      <td>35.374348</td>\n",
       "      <td>...</td>\n",
       "      <td>25.250000</td>\n",
       "      <td>11.250000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 1107 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                              Ligand_MORDRED_ABC  Ligand_MORDRED_ABCGG  ...  \\\n",
       "(t-Bu)PhCPhos                          22.449495             19.292201  ...   \n",
       "1268824-69-6                           20.741455             17.344789  ...   \n",
       "BrettPhos                              29.162435             24.462781  ...   \n",
       "Di-tert-butylphenylphosphine           11.396454             10.965064  ...   \n",
       "JackiePhos                             42.136729             35.374348  ...   \n",
       "\n",
       "                              Ligand_MORDRED_mZagreb1  Ligand_MORDRED_mZagreb2  \n",
       "(t-Bu)PhCPhos                               11.062500                 6.305556  \n",
       "1268824-69-6                                 8.638889                 6.055556  \n",
       "BrettPhos                                   13.555556                 8.583333  \n",
       "Di-tert-butylphenylphosphine                 7.597222                 3.111111  \n",
       "JackiePhos                                  25.250000                11.250000  \n",
       "\n",
       "[5 rows x 1107 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p_ligand_mordred_bad = SubstanceParameter(name = \"Ligand\",\n",
    "                                          data = ligand_data,\n",
    "                                          encoding = 'MORDRED', # MORDRED or RDKIT or MORGAN_FP\n",
    "                                          decorrelate = False) # Max allowed pairwise column correlation\n",
    "\n",
    "# The comp_df property of the parameter contains the descriptors as data frame.\n",
    "# This is essentially a tabular mapping from labels (indices) to numbers (columns).\n",
    "with pd.option_context('display.max_columns', 5):\n",
    "    display(p_ligand_mordred_bad.comp_df.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e1119a9-e4ae-4f2a-b023-c472f629e906",
   "metadata": {},
   "source": [
    "Now we set the decorrelation to its default value and observe a dramatic reduction in descriptors necessary."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "fb0de00f-a210-4587-b6cd-a0ef07beb50a",
   "metadata": {
    "tags": []
   },
   "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>Ligand_MORDRED_ABC</th>\n",
       "      <th>Ligand_MORDRED_SpMax_A</th>\n",
       "      <th>...</th>\n",
       "      <th>Ligand_MORDRED_JGI7</th>\n",
       "      <th>Ligand_MORDRED_TopoShapeIndex</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>(t-Bu)PhCPhos</th>\n",
       "      <td>22.449495</td>\n",
       "      <td>2.507279</td>\n",
       "      <td>...</td>\n",
       "      <td>0.020557</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1268824-69-6</th>\n",
       "      <td>20.741455</td>\n",
       "      <td>2.468664</td>\n",
       "      <td>...</td>\n",
       "      <td>0.015783</td>\n",
       "      <td>0.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BrettPhos</th>\n",
       "      <td>29.162435</td>\n",
       "      <td>2.555080</td>\n",
       "      <td>...</td>\n",
       "      <td>0.014660</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Di-tert-butylphenylphosphine</th>\n",
       "      <td>11.396454</td>\n",
       "      <td>2.429653</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JackiePhos</th>\n",
       "      <td>42.136729</td>\n",
       "      <td>2.572820</td>\n",
       "      <td>...</td>\n",
       "      <td>0.017747</td>\n",
       "      <td>0.857143</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 48 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                              Ligand_MORDRED_ABC  Ligand_MORDRED_SpMax_A  ...  \\\n",
       "(t-Bu)PhCPhos                          22.449495                2.507279  ...   \n",
       "1268824-69-6                           20.741455                2.468664  ...   \n",
       "BrettPhos                              29.162435                2.555080  ...   \n",
       "Di-tert-butylphenylphosphine           11.396454                2.429653  ...   \n",
       "JackiePhos                             42.136729                2.572820  ...   \n",
       "\n",
       "                              Ligand_MORDRED_JGI7  \\\n",
       "(t-Bu)PhCPhos                            0.020557   \n",
       "1268824-69-6                             0.015783   \n",
       "BrettPhos                                0.014660   \n",
       "Di-tert-butylphenylphosphine             0.000000   \n",
       "JackiePhos                               0.017747   \n",
       "\n",
       "                              Ligand_MORDRED_TopoShapeIndex  \n",
       "(t-Bu)PhCPhos                                      1.000000  \n",
       "1268824-69-6                                       0.800000  \n",
       "BrettPhos                                          1.000000  \n",
       "Di-tert-butylphenylphosphine                       1.000000  \n",
       "JackiePhos                                         0.857143  \n",
       "\n",
       "[5 rows x 48 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p_ligand_mordred = SubstanceParameter(name = \"Ligand\",\n",
    "                                      data = ligand_data,\n",
    "                                      encoding = 'MORDRED', # MORDRED or RDKIT or MORGAN_FP\n",
    "                                      decorrelate = 0.7) # This is the default\n",
    "\n",
    "with pd.option_context('display.max_columns', 5):\n",
    "    display(p_ligand_mordred.comp_df.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7357918a-1856-40b7-bdc3-e7abfe515a95",
   "metadata": {},
   "source": [
    "### Ohe-Hot Encoding\n",
    "\n",
    "For this we can use a traditional [`CategoricalParameter`](https://emdgroup.github.io/baybe/userguide/parameters.html#categoricalparameter)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "bea5159d-5876-4dde-89c7-0303ae08357e",
   "metadata": {
    "tags": []
   },
   "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>Ligand_(t-Bu)PhCPhos</th>\n",
       "      <th>Ligand_1268824-69-6</th>\n",
       "      <th>...</th>\n",
       "      <th>Ligand_Tricyclohexylphosphine</th>\n",
       "      <th>Ligand_XPhos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>(t-Bu)PhCPhos</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1268824-69-6</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BrettPhos</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Di-tert-butylphenylphosphine</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JackiePhos</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                              Ligand_(t-Bu)PhCPhos  Ligand_1268824-69-6  ...  \\\n",
       "(t-Bu)PhCPhos                                  1.0                  0.0  ...   \n",
       "1268824-69-6                                   0.0                  1.0  ...   \n",
       "BrettPhos                                      0.0                  0.0  ...   \n",
       "Di-tert-butylphenylphosphine                   0.0                  0.0  ...   \n",
       "JackiePhos                                     0.0                  0.0  ...   \n",
       "\n",
       "                              Ligand_Tricyclohexylphosphine  Ligand_XPhos  \n",
       "(t-Bu)PhCPhos                                           0.0           0.0  \n",
       "1268824-69-6                                            0.0           0.0  \n",
       "BrettPhos                                               0.0           0.0  \n",
       "Di-tert-butylphenylphosphine                            0.0           0.0  \n",
       "JackiePhos                                              0.0           0.0  \n",
       "\n",
       "[5 rows x 12 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p_ligand_ohe = CategoricalParameter(name = \"Ligand\",\n",
    "                                    values = ligand_data.keys(),\n",
    "                                    encoding = 'OHE') # INT or OHE\n",
    "\n",
    "with pd.option_context('display.max_columns', 5):\n",
    "    display(p_ligand_ohe.comp_df.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "804f93cc-0c7e-4af6-9014-46310c82cb0a",
   "metadata": {},
   "source": [
    "### CUSTOM encoding\n",
    "\n",
    "BayBE also has the [`CustomDiscreteParameter`](https://emdgroup.github.io/baybe/userguide/parameters.html#customdiscreteparameter) which takes your very own descriptor mapping as data frame as `data`.\n",
    "\n",
    "To illsutrate its usage we create a PCA-based reduction based on the large MORDRED descriptor map caluclated previously."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "38e1dc95-bf8d-489c-802e-901c85c1de28",
   "metadata": {
    "tags": []
   },
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>(t-Bu)PhCPhos</th>\n",
       "      <td>8.435875</td>\n",
       "      <td>-3.408601</td>\n",
       "      <td>0.113716</td>\n",
       "      <td>5.114928</td>\n",
       "      <td>-15.921587</td>\n",
       "      <td>7.060074</td>\n",
       "      <td>1.650195</td>\n",
       "      <td>5.221354</td>\n",
       "      <td>6.385881</td>\n",
       "      <td>0.082431</td>\n",
       "      <td>-0.184933</td>\n",
       "      <td>-8.141916e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1268824-69-6</th>\n",
       "      <td>6.755321</td>\n",
       "      <td>-12.111389</td>\n",
       "      <td>-2.594902</td>\n",
       "      <td>6.280692</td>\n",
       "      <td>5.512184</td>\n",
       "      <td>0.937279</td>\n",
       "      <td>5.448890</td>\n",
       "      <td>-8.351234</td>\n",
       "      <td>4.511744</td>\n",
       "      <td>-3.311887</td>\n",
       "      <td>-0.995624</td>\n",
       "      <td>-8.141916e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BrettPhos</th>\n",
       "      <td>24.538098</td>\n",
       "      <td>-17.319861</td>\n",
       "      <td>-3.928779</td>\n",
       "      <td>2.821141</td>\n",
       "      <td>1.053323</td>\n",
       "      <td>5.245916</td>\n",
       "      <td>4.073336</td>\n",
       "      <td>-2.586212</td>\n",
       "      <td>-3.568789</td>\n",
       "      <td>4.924581</td>\n",
       "      <td>1.462626</td>\n",
       "      <td>-8.141916e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Di-tert-butylphenylphosphine</th>\n",
       "      <td>-18.056464</td>\n",
       "      <td>-0.054137</td>\n",
       "      <td>20.290740</td>\n",
       "      <td>-0.085171</td>\n",
       "      <td>-11.794890</td>\n",
       "      <td>-10.692620</td>\n",
       "      <td>2.950254</td>\n",
       "      <td>-3.649690</td>\n",
       "      <td>-3.127553</td>\n",
       "      <td>0.384644</td>\n",
       "      <td>-0.364512</td>\n",
       "      <td>-8.141916e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JackiePhos</th>\n",
       "      <td>56.639889</td>\n",
       "      <td>21.698687</td>\n",
       "      <td>3.587066</td>\n",
       "      <td>-13.057508</td>\n",
       "      <td>2.007929</td>\n",
       "      <td>-4.200061</td>\n",
       "      <td>-0.852781</td>\n",
       "      <td>-0.312617</td>\n",
       "      <td>1.061784</td>\n",
       "      <td>-0.151094</td>\n",
       "      <td>-0.100496</td>\n",
       "      <td>-8.141916e-15</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     0          1          2          3   \\\n",
       "(t-Bu)PhCPhos                  8.435875  -3.408601   0.113716   5.114928   \n",
       "1268824-69-6                   6.755321 -12.111389  -2.594902   6.280692   \n",
       "BrettPhos                     24.538098 -17.319861  -3.928779   2.821141   \n",
       "Di-tert-butylphenylphosphine -18.056464  -0.054137  20.290740  -0.085171   \n",
       "JackiePhos                    56.639889  21.698687   3.587066 -13.057508   \n",
       "\n",
       "                                     4          5         6         7   \\\n",
       "(t-Bu)PhCPhos                -15.921587   7.060074  1.650195  5.221354   \n",
       "1268824-69-6                   5.512184   0.937279  5.448890 -8.351234   \n",
       "BrettPhos                      1.053323   5.245916  4.073336 -2.586212   \n",
       "Di-tert-butylphenylphosphine -11.794890 -10.692620  2.950254 -3.649690   \n",
       "JackiePhos                     2.007929  -4.200061 -0.852781 -0.312617   \n",
       "\n",
       "                                    8         9         10            11  \n",
       "(t-Bu)PhCPhos                 6.385881  0.082431 -0.184933 -8.141916e-15  \n",
       "1268824-69-6                  4.511744 -3.311887 -0.995624 -8.141916e-15  \n",
       "BrettPhos                    -3.568789  4.924581  1.462626 -8.141916e-15  \n",
       "Di-tert-butylphenylphosphine -3.127553  0.384644 -0.364512 -8.141916e-15  \n",
       "JackiePhos                    1.061784 -0.151094 -0.100496 -8.141916e-15  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# Consider the large table of MORDRED features\n",
    "features_mordred = p_ligand_mordred_bad.comp_df.values\n",
    "\n",
    "# Create a table of PC's from these features\n",
    "pca = PCA(n_components = 12)\n",
    "features_pca = pca.fit_transform(StandardScaler().fit_transform(features_mordred))\n",
    "custom_features = pd.DataFrame(features_pca,\n",
    "                               index = p_ligand_mordred.comp_df.index)\n",
    "custom_features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "21ccebc1",
   "metadata": {
    "tags": []
   },
   "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>Ligand_0</th>\n",
       "      <th>Ligand_1</th>\n",
       "      <th>Ligand_2</th>\n",
       "      <th>Ligand_3</th>\n",
       "      <th>Ligand_4</th>\n",
       "      <th>Ligand_5</th>\n",
       "      <th>Ligand_6</th>\n",
       "      <th>Ligand_7</th>\n",
       "      <th>Ligand_8</th>\n",
       "      <th>Ligand_9</th>\n",
       "      <th>Ligand_10</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>(t-Bu)PhCPhos</th>\n",
       "      <td>8.435875</td>\n",
       "      <td>-3.408601</td>\n",
       "      <td>0.113716</td>\n",
       "      <td>5.114928</td>\n",
       "      <td>-15.921587</td>\n",
       "      <td>7.060074</td>\n",
       "      <td>1.650195</td>\n",
       "      <td>5.221354</td>\n",
       "      <td>6.385881</td>\n",
       "      <td>0.082431</td>\n",
       "      <td>-0.184933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1268824-69-6</th>\n",
       "      <td>6.755321</td>\n",
       "      <td>-12.111389</td>\n",
       "      <td>-2.594902</td>\n",
       "      <td>6.280692</td>\n",
       "      <td>5.512184</td>\n",
       "      <td>0.937279</td>\n",
       "      <td>5.448890</td>\n",
       "      <td>-8.351234</td>\n",
       "      <td>4.511744</td>\n",
       "      <td>-3.311887</td>\n",
       "      <td>-0.995624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BrettPhos</th>\n",
       "      <td>24.538098</td>\n",
       "      <td>-17.319861</td>\n",
       "      <td>-3.928779</td>\n",
       "      <td>2.821141</td>\n",
       "      <td>1.053323</td>\n",
       "      <td>5.245916</td>\n",
       "      <td>4.073336</td>\n",
       "      <td>-2.586212</td>\n",
       "      <td>-3.568789</td>\n",
       "      <td>4.924581</td>\n",
       "      <td>1.462626</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Di-tert-butylphenylphosphine</th>\n",
       "      <td>-18.056464</td>\n",
       "      <td>-0.054137</td>\n",
       "      <td>20.290740</td>\n",
       "      <td>-0.085171</td>\n",
       "      <td>-11.794890</td>\n",
       "      <td>-10.692620</td>\n",
       "      <td>2.950254</td>\n",
       "      <td>-3.649690</td>\n",
       "      <td>-3.127553</td>\n",
       "      <td>0.384644</td>\n",
       "      <td>-0.364512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JackiePhos</th>\n",
       "      <td>56.639889</td>\n",
       "      <td>21.698687</td>\n",
       "      <td>3.587066</td>\n",
       "      <td>-13.057508</td>\n",
       "      <td>2.007929</td>\n",
       "      <td>-4.200061</td>\n",
       "      <td>-0.852781</td>\n",
       "      <td>-0.312617</td>\n",
       "      <td>1.061784</td>\n",
       "      <td>-0.151094</td>\n",
       "      <td>-0.100496</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Ligand_0   Ligand_1   Ligand_2   Ligand_3  \\\n",
       "(t-Bu)PhCPhos                  8.435875  -3.408601   0.113716   5.114928   \n",
       "1268824-69-6                   6.755321 -12.111389  -2.594902   6.280692   \n",
       "BrettPhos                     24.538098 -17.319861  -3.928779   2.821141   \n",
       "Di-tert-butylphenylphosphine -18.056464  -0.054137  20.290740  -0.085171   \n",
       "JackiePhos                    56.639889  21.698687   3.587066 -13.057508   \n",
       "\n",
       "                               Ligand_4   Ligand_5  Ligand_6  Ligand_7  \\\n",
       "(t-Bu)PhCPhos                -15.921587   7.060074  1.650195  5.221354   \n",
       "1268824-69-6                   5.512184   0.937279  5.448890 -8.351234   \n",
       "BrettPhos                      1.053323   5.245916  4.073336 -2.586212   \n",
       "Di-tert-butylphenylphosphine -11.794890 -10.692620  2.950254 -3.649690   \n",
       "JackiePhos                     2.007929  -4.200061 -0.852781 -0.312617   \n",
       "\n",
       "                              Ligand_8  Ligand_9  Ligand_10  \n",
       "(t-Bu)PhCPhos                 6.385881  0.082431  -0.184933  \n",
       "1268824-69-6                  4.511744 -3.311887  -0.995624  \n",
       "BrettPhos                    -3.568789  4.924581   1.462626  \n",
       "Di-tert-butylphenylphosphine -3.127553  0.384644  -0.364512  \n",
       "JackiePhos                    1.061784 -0.151094  -0.100496  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p_ligand_custom = CustomDiscreteParameter(name = \"Ligand\",\n",
    "                                          data = custom_features)\n",
    "p_ligand_custom.comp_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "910c8df9-f093-4efb-82a4-26661248076c",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Create the Scenarios"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "657ea218-bdd3-4089-98f5-025999c6e173",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from baybe.objectives import SingleTargetObjective\n",
    "from baybe.targets import NumericalTarget\n",
    "\n",
    "yield_target = NumericalTarget(name = 'yield', mode = 'MAX')\n",
    "objective = SingleTargetObjective(target = yield_target)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "e1632776-e6e0-449e-9718-57fb252c15bc",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from baybe.searchspace import SearchSpace\n",
    "\n",
    "searchspace_mordred = SearchSpace.from_product(parameters=[*shared_parameters, p_ligand_mordred])\n",
    "searchspace_ohe = SearchSpace.from_product(parameters=[*shared_parameters, p_ligand_ohe])\n",
    "searchspace_custom = SearchSpace.from_product(parameters=[*shared_parameters, p_ligand_custom])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38d5fa89-1e2b-438f-aac6-27c10a68f189",
   "metadata": {
    "tags": []
   },
   "source": [
    "We will create one scenario for each of the ligand descriptors created before plus an additional random baseline campaign."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "2f958eaf-37fd-4946-a0aa-91aacfc5a029",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from baybe.campaign import Campaign\n",
    "from baybe.recommenders import RandomRecommender\n",
    "\n",
    "# All campaigns share the same objective.\n",
    "# All campaigns except the random baseline use the default recommender.\n",
    "scenarios = {\n",
    "    'MORDRED': Campaign(objective = objective,\n",
    "                        searchspace = searchspace_mordred),\n",
    "    'OHE': Campaign(objective = objective,\n",
    "                    searchspace = searchspace_ohe),\n",
    "    'MORDRED+PCA': Campaign(objective = objective,\n",
    "                            searchspace = searchspace_custom),\n",
    "    'Random': Campaign(objective = objective,\n",
    "                       searchspace = searchspace_mordred,\n",
    "                       recommender = RandomRecommender())\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d477eb21-d6d0-4719-85ae-303e88920c64",
   "metadata": {},
   "source": [
    "# Perform the Backtest Simulations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "840da1bb-d067-4b63-9c03-001826e08949",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "I0000 00:00:1722731537.691507   90153 work_stealing_thread_pool.cc:320] WorkStealingThreadPoolImpl::PrepareFork\n",
      "100%|###########################################################################################################################################################################| 200/200 [04:34<00:00,  1.37s/it]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ligand Encoding</th>\n",
       "      <th>Monte_Carlo_Run</th>\n",
       "      <th>Iteration</th>\n",
       "      <th>Num_Experiments</th>\n",
       "      <th>yield_Measurements</th>\n",
       "      <th>yield_IterBest</th>\n",
       "      <th>yield_CumBest</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MORDRED</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>[0.0, 13.12]</td>\n",
       "      <td>13.12</td>\n",
       "      <td>13.12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>MORDRED</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>[15.92, 6.48]</td>\n",
       "      <td>15.92</td>\n",
       "      <td>15.92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>MORDRED</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>[25.45, 37.3]</td>\n",
       "      <td>37.30</td>\n",
       "      <td>37.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>MORDRED</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>8</td>\n",
       "      <td>[9.88, 52.2]</td>\n",
       "      <td>52.20</td>\n",
       "      <td>52.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>MORDRED</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>[44.4, 0.85]</td>\n",
       "      <td>44.40</td>\n",
       "      <td>52.20</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Ligand Encoding  Monte_Carlo_Run  Iteration  Num_Experiments  \\\n",
       "0         MORDRED                0          0                2   \n",
       "1         MORDRED                0          1                4   \n",
       "2         MORDRED                0          2                6   \n",
       "3         MORDRED                0          3                8   \n",
       "4         MORDRED                0          4               10   \n",
       "\n",
       "  yield_Measurements  yield_IterBest  yield_CumBest  \n",
       "0       [0.0, 13.12]           13.12          13.12  \n",
       "1      [15.92, 6.48]           15.92          15.92  \n",
       "2      [25.45, 37.3]           37.30          37.30  \n",
       "3       [9.88, 52.2]           52.20          52.20  \n",
       "4       [44.4, 0.85]           44.40          52.20  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from baybe.simulation import simulate_scenarios\n",
    "\n",
    "# Backtest settings\n",
    "MC_RUNS = 50 # this will take fairly long on typical free-of-charge cloud compute\n",
    "DOE_ITERATIONS = 16\n",
    "BATCH_SIZE = 2\n",
    "\n",
    "# Run the utility for backtesting\n",
    "results = simulate_scenarios(\n",
    "    scenarios,\n",
    "    lookup, # the initial dataframe with the experimental results is our lookup\n",
    "    batch_size = BATCH_SIZE,\n",
    "    n_doe_iterations = DOE_ITERATIONS,\n",
    "    n_mc_iterations = MC_RUNS,\n",
    ")\n",
    "results.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "da9c5aa1-23ce-4657-a96c-69af0747e445",
   "metadata": {},
   "source": [
    "# Plot the Results\n",
    "\n",
    "We find a strong influence of the encoding on the overall performance. While the MORDRED encoding (with `decorrelate` set to its default 0.7) performs best, our custom MORDRED+PCA version also does well. The OHE performs worst, almost as bad as random. OHE only becomes better than random towards the late stage of the optimization (not shown here).\n",
    "\n",
    "Keep in mind that there are other factors that could influence the performance, such as the priors, the fitting procedure settings and so on."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "5429e144-f52e-46b9-a6f9-c7483590ebd1",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "PLOTARGS = {\n",
    "    'linestyle': 'solid',\n",
    "    'marker': 'o',\n",
    "    'markersize': 6, \n",
    "    'markeredgecolor': 'none'\n",
    "}\n",
    "FIGSIZE = (11,6)\n",
    "\n",
    "results.rename(columns = {\"Scenario\": \"Ligand Encoding\"}, inplace = True)\n",
    "sns.lineplot(data = results, \n",
    "             x = \"Num_Experiments\", \n",
    "             y = \"yield_CumBest\", \n",
    "             hue = \"Ligand Encoding\",\n",
    "             **PLOTARGS)\n",
    "\n",
    "plt.axhline(y = F_BEST, color = 'red', linestyle = '--', label = 'Best Possible')\n",
    "plt.gcf().set_size_inches(FIGSIZE)\n",
    "plt.gca().set_ylim(plt.gca().get_ylim()[0], F_BEST+5)\n",
    "plt.title('Influence of the Ligand Substance Encoding');"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "BayBE-dev",
   "language": "python",
   "name": "baybe-dev"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}