{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Interpretability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from molgraph.chemistry import MolecularGraphEncoder\n",
    "from molgraph.chemistry import Featurizer\n",
    "from molgraph.chemistry import features\n",
    "\n",
    "from molgraph.models import GradientActivationMapping\n",
    "from molgraph.models import IntegratedSaliencyMapping\n",
    "\n",
    "import tensorflow as tf\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "np.set_printoptions(suppress=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Construct a `MolecularGraphEncoder`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MolecularGraphEncoder(atom_encoder=Featurizer(features=[Symbol(allowable_set=['[OOV:0]', 'C', 'N', 'O'], ordinal=False, oov_size=1), Hybridization(allowable_set=['[OOV:0]', 'SP', 'SP2', 'SP3'], ordinal=False, oov_size=1), HydrogenDonor(), HydrogenAcceptor(), Hetero()]), bond_encoder=Featurizer(features=[BondType(allowable_set=['AROMATIC', 'DOUBLE', 'SINGLE', 'TRIPLE'], ordinal=False, oov_size=0), Rotatable()]), positional_encoding_dim=None, self_loops=False)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "atom_encoder = Featurizer([\n",
    "    features.Symbol({'C', 'N', 'O'}, oov_size=1),\n",
    "    features.Hybridization({'SP', 'SP2', 'SP3'}, oov_size=1),\n",
    "    features.HydrogenDonor(),\n",
    "    features.HydrogenAcceptor(),\n",
    "    features.Hetero()\n",
    "])\n",
    "\n",
    "bond_encoder = Featurizer([\n",
    "    features.BondType({'SINGLE', 'DOUBLE', 'TRIPLE', 'AROMATIC'}),\n",
    "    features.Rotatable()\n",
    "])\n",
    "\n",
    "mol_encoder = MolecularGraphEncoder(atom_encoder, bond_encoder, positional_encoding_dim=None)\n",
    "mol_encoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Obtain dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Compound ID</th>\n",
       "      <th>ESOL predicted log solubility in mols per litre</th>\n",
       "      <th>Minimum Degree</th>\n",
       "      <th>Molecular Weight</th>\n",
       "      <th>Number of H-Bond Donors</th>\n",
       "      <th>Number of Rings</th>\n",
       "      <th>Number of Rotatable Bonds</th>\n",
       "      <th>Polar Surface Area</th>\n",
       "      <th>measured log solubility in mols per litre</th>\n",
       "      <th>smiles</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Amigdalin</td>\n",
       "      <td>-0.974</td>\n",
       "      <td>1</td>\n",
       "      <td>457.432</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>202.32</td>\n",
       "      <td>-0.77</td>\n",
       "      <td>OCC3OC(OCC2OC(OC(C#N)c1ccccc1)C(O)C(O)C2O)C(O)...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Fenfuram</td>\n",
       "      <td>-2.885</td>\n",
       "      <td>1</td>\n",
       "      <td>201.225</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>42.24</td>\n",
       "      <td>-3.30</td>\n",
       "      <td>Cc1occc1C(=O)Nc2ccccc2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>citral</td>\n",
       "      <td>-2.579</td>\n",
       "      <td>1</td>\n",
       "      <td>152.237</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>17.07</td>\n",
       "      <td>-2.06</td>\n",
       "      <td>CC(C)=CCCC(C)=CC(=O)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Compound ID  ESOL predicted log solubility in mols per litre  \\\n",
       "0   Amigdalin                                           -0.974   \n",
       "1    Fenfuram                                           -2.885   \n",
       "2      citral                                           -2.579   \n",
       "\n",
       "   Minimum Degree  Molecular Weight  Number of H-Bond Donors  Number of Rings  \\\n",
       "0               1           457.432                        7                3   \n",
       "1               1           201.225                        1                2   \n",
       "2               1           152.237                        0                0   \n",
       "\n",
       "   Number of Rotatable Bonds  Polar Surface Area  \\\n",
       "0                          7              202.32   \n",
       "1                          2               42.24   \n",
       "2                          4               17.07   \n",
       "\n",
       "   measured log solubility in mols per litre  \\\n",
       "0                                      -0.77   \n",
       "1                                      -3.30   \n",
       "2                                      -2.06   \n",
       "\n",
       "                                              smiles  \n",
       "0  OCC3OC(OCC2OC(OC(C#N)c1ccccc1)C(O)C(O)C2O)C(O)...  \n",
       "1                             Cc1occc1C(=O)Nc2ccccc2  \n",
       "2                               CC(C)=CCCC(C)=CC(=O)  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = tf.keras.utils.get_file(\n",
    "    fname='ESOL.csv',\n",
    "    origin='http://deepchem.io.s3-website-us-west-1.amazonaws.com/datasets/ESOL.csv',\n",
    ")\n",
    "df = pd.read_csv(path)\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Obtain SMILES `x`and associated labels `y`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "x, y = df['smiles'].values, df['measured log solubility in mols per litre'].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Obtain `GraphTensor` from `x`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GraphTensor(\n",
      "  sizes=<tf.Tensor: shape=(1128,), dtype=int64>,\n",
      "  node_feature=<tf.Tensor: shape=(14991, 11), dtype=float32>,\n",
      "  edge_src=<tf.Tensor: shape=(30856,), dtype=int32>,\n",
      "  edge_dst=<tf.Tensor: shape=(30856,), dtype=int32>,\n",
      "  edge_feature=<tf.Tensor: shape=(30856, 5), dtype=float32>)\n",
      "\n",
      "node_feature shape: (14991, 11)\n",
      "edge_dst shape:     (30856,)\n",
      "edge_src shape:     (30856,)\n",
      "edge_feature shape: (30856, 5)\n"
     ]
    }
   ],
   "source": [
    "x_data = mol_encoder(x)\n",
    "\n",
    "print(x_data, end='\\n\\n')\n",
    "print('node_feature shape:', x_data.node_feature.shape)\n",
    "print('edge_dst shape:    ', x_data.edge_dst.shape)\n",
    "print('edge_src shape:    ', x_data.edge_src.shape)\n",
    "print('edge_feature shape:', x_data.edge_feature.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Build Keras GNN **model** with GNN **layers**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_6\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv_1 (GCNConv)            (None, None, 128)         4608      \n",
      "                                                                 \n",
      " conv_2 (GCNConv)            (None, None, 128)         33152     \n",
      "                                                                 \n",
      " conv_3 (GCNConv)            (None, None, 128)         33152     \n",
      "                                                                 \n",
      " segment_pooling_readout_6   (None, 128)               0         \n",
      " (SegmentPoolingReadout)                                         \n",
      "                                                                 \n",
      " dense_12 (Dense)            (None, 512)               66048     \n",
      "                                                                 \n",
      " dense_13 (Dense)            (None, 1)                 513       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 137473 (537.00 KB)\n",
      "Trainable params: 137473 (537.00 KB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from molgraph.layers import GCNConv, Readout\n",
    "\n",
    "sequential_model = tf.keras.Sequential([\n",
    "    tf.keras.layers.Input(type_spec=x_data.spec),\n",
    "    GCNConv(128, name='conv_1'),\n",
    "    GCNConv(128, name='conv_2'),\n",
    "    GCNConv(128, name='conv_3'),\n",
    "    GCNConv(128, name='conv_4'),\n",
    "    Readout(),\n",
    "    tf.keras.layers.Dense(512, activation='relu'),\n",
    "    tf.keras.layers.Dense(1)\n",
    "])\n",
    "\n",
    "sequential_model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Compile and fit GNN **model**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "36/36 - 3s - loss: 4.2682 - mae: 1.6230 - 3s/epoch - 84ms/step\n",
      "Epoch 2/50\n",
      "36/36 - 0s - loss: 3.3620 - mae: 1.4817 - 179ms/epoch - 5ms/step\n",
      "Epoch 3/50\n",
      "36/36 - 0s - loss: 3.3828 - mae: 1.4810 - 188ms/epoch - 5ms/step\n",
      "Epoch 4/50\n",
      "36/36 - 0s - loss: 3.1614 - mae: 1.4371 - 186ms/epoch - 5ms/step\n",
      "Epoch 5/50\n",
      "36/36 - 0s - loss: 2.9661 - mae: 1.3852 - 183ms/epoch - 5ms/step\n",
      "Epoch 6/50\n",
      "36/36 - 0s - loss: 2.8416 - mae: 1.3507 - 188ms/epoch - 5ms/step\n",
      "Epoch 7/50\n",
      "36/36 - 0s - loss: 2.7258 - mae: 1.3335 - 176ms/epoch - 5ms/step\n",
      "Epoch 8/50\n",
      "36/36 - 0s - loss: 2.5141 - mae: 1.2632 - 186ms/epoch - 5ms/step\n",
      "Epoch 9/50\n",
      "36/36 - 0s - loss: 2.6962 - mae: 1.2898 - 186ms/epoch - 5ms/step\n",
      "Epoch 10/50\n",
      "36/36 - 0s - loss: 2.5491 - mae: 1.2679 - 180ms/epoch - 5ms/step\n",
      "Epoch 11/50\n",
      "36/36 - 0s - loss: 2.3768 - mae: 1.2089 - 180ms/epoch - 5ms/step\n",
      "Epoch 12/50\n",
      "36/36 - 0s - loss: 2.3904 - mae: 1.2302 - 177ms/epoch - 5ms/step\n",
      "Epoch 13/50\n",
      "36/36 - 0s - loss: 2.1581 - mae: 1.1444 - 185ms/epoch - 5ms/step\n",
      "Epoch 14/50\n",
      "36/36 - 0s - loss: 2.4210 - mae: 1.2226 - 187ms/epoch - 5ms/step\n",
      "Epoch 15/50\n",
      "36/36 - 0s - loss: 2.2568 - mae: 1.1956 - 197ms/epoch - 5ms/step\n",
      "Epoch 16/50\n",
      "36/36 - 0s - loss: 2.1405 - mae: 1.1333 - 201ms/epoch - 6ms/step\n",
      "Epoch 17/50\n",
      "36/36 - 0s - loss: 1.9765 - mae: 1.0873 - 203ms/epoch - 6ms/step\n",
      "Epoch 18/50\n",
      "36/36 - 0s - loss: 1.9263 - mae: 1.0707 - 204ms/epoch - 6ms/step\n",
      "Epoch 19/50\n",
      "36/36 - 0s - loss: 1.9688 - mae: 1.0793 - 186ms/epoch - 5ms/step\n",
      "Epoch 20/50\n",
      "36/36 - 0s - loss: 1.8725 - mae: 1.0718 - 170ms/epoch - 5ms/step\n",
      "Epoch 21/50\n",
      "36/36 - 0s - loss: 1.9170 - mae: 1.0576 - 176ms/epoch - 5ms/step\n",
      "Epoch 22/50\n",
      "36/36 - 0s - loss: 1.9656 - mae: 1.0912 - 181ms/epoch - 5ms/step\n",
      "Epoch 23/50\n",
      "36/36 - 0s - loss: 1.7675 - mae: 1.0313 - 185ms/epoch - 5ms/step\n",
      "Epoch 24/50\n",
      "36/36 - 0s - loss: 1.7883 - mae: 1.0274 - 172ms/epoch - 5ms/step\n",
      "Epoch 25/50\n",
      "36/36 - 0s - loss: 1.9393 - mae: 1.0915 - 182ms/epoch - 5ms/step\n",
      "Epoch 26/50\n",
      "36/36 - 0s - loss: 1.7453 - mae: 1.0155 - 179ms/epoch - 5ms/step\n",
      "Epoch 27/50\n",
      "36/36 - 0s - loss: 1.7054 - mae: 1.0131 - 182ms/epoch - 5ms/step\n",
      "Epoch 28/50\n",
      "36/36 - 0s - loss: 1.7013 - mae: 1.0174 - 179ms/epoch - 5ms/step\n",
      "Epoch 29/50\n",
      "36/36 - 0s - loss: 1.7120 - mae: 1.0070 - 179ms/epoch - 5ms/step\n",
      "Epoch 30/50\n",
      "36/36 - 0s - loss: 1.7977 - mae: 1.0334 - 180ms/epoch - 5ms/step\n",
      "Epoch 31/50\n",
      "36/36 - 0s - loss: 1.6766 - mae: 0.9960 - 186ms/epoch - 5ms/step\n",
      "Epoch 32/50\n",
      "36/36 - 0s - loss: 1.6374 - mae: 0.9735 - 180ms/epoch - 5ms/step\n",
      "Epoch 33/50\n",
      "36/36 - 0s - loss: 1.6437 - mae: 0.9936 - 180ms/epoch - 5ms/step\n",
      "Epoch 34/50\n",
      "36/36 - 0s - loss: 1.5678 - mae: 0.9722 - 184ms/epoch - 5ms/step\n",
      "Epoch 35/50\n",
      "36/36 - 0s - loss: 1.4954 - mae: 0.9241 - 179ms/epoch - 5ms/step\n",
      "Epoch 36/50\n",
      "36/36 - 0s - loss: 1.5489 - mae: 0.9540 - 175ms/epoch - 5ms/step\n",
      "Epoch 37/50\n",
      "36/36 - 0s - loss: 1.4601 - mae: 0.9223 - 183ms/epoch - 5ms/step\n",
      "Epoch 38/50\n",
      "36/36 - 0s - loss: 1.5156 - mae: 0.9418 - 176ms/epoch - 5ms/step\n",
      "Epoch 39/50\n",
      "36/36 - 0s - loss: 1.5280 - mae: 0.9391 - 178ms/epoch - 5ms/step\n",
      "Epoch 40/50\n",
      "36/36 - 0s - loss: 1.5135 - mae: 0.9387 - 179ms/epoch - 5ms/step\n",
      "Epoch 41/50\n",
      "36/36 - 0s - loss: 1.4961 - mae: 0.9250 - 181ms/epoch - 5ms/step\n",
      "Epoch 42/50\n",
      "36/36 - 0s - loss: 1.4935 - mae: 0.9296 - 176ms/epoch - 5ms/step\n",
      "Epoch 43/50\n",
      "36/36 - 0s - loss: 1.4748 - mae: 0.9282 - 179ms/epoch - 5ms/step\n",
      "Epoch 44/50\n",
      "36/36 - 0s - loss: 1.4566 - mae: 0.9082 - 170ms/epoch - 5ms/step\n",
      "Epoch 45/50\n",
      "36/36 - 0s - loss: 1.3972 - mae: 0.9027 - 186ms/epoch - 5ms/step\n",
      "Epoch 46/50\n",
      "36/36 - 0s - loss: 1.4849 - mae: 0.9236 - 176ms/epoch - 5ms/step\n",
      "Epoch 47/50\n",
      "36/36 - 0s - loss: 1.3397 - mae: 0.8696 - 183ms/epoch - 5ms/step\n",
      "Epoch 48/50\n",
      "36/36 - 0s - loss: 1.4117 - mae: 0.8968 - 172ms/epoch - 5ms/step\n",
      "Epoch 49/50\n",
      "36/36 - 0s - loss: 1.2797 - mae: 0.8465 - 177ms/epoch - 5ms/step\n",
      "Epoch 50/50\n",
      "36/36 - 0s - loss: 1.2962 - mae: 0.8538 - 179ms/epoch - 5ms/step\n"
     ]
    }
   ],
   "source": [
    "sequential_model.compile('adam', 'mse', ['mae'])\n",
    "sequential_model.fit(x_data, y, epochs=50, verbose=2);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Pass GNN **model** to **GradientActivationMapping**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "gam_model = GradientActivationMapping(\n",
    "    sequential_model, \n",
    "    ['conv_1', 'conv_2', 'conv_3', 'conv_4'], \n",
    "    output_activation=None,\n",
    "    discard_negative_values=False,\n",
    ")    \n",
    "\n",
    "gam = gam_model(x_data.separate())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Visualize **maps** on molecule"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "tags": [
     "nbsphinx-thumbnail"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<PIL.PngImagePlugin.PngImageFile image mode=RGB size=360x360 at 0x7FDEB5B65610>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from molgraph.chemistry import vis\n",
    "\n",
    "vis.visualize_maps(molecule=x[42], maps=gam[42])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}