surrogate_model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PROTAC Activity Surrogate Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "InteractiveShell.ast_node_interactivity = \"all\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'3.2.1'"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "numpy 1.21.2\n",
      "pandas 1.4.1\n"
     ]
    }
   ],
   "source": [
    "from IPython.display import display_html\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "pd.set_option(\"display.max_columns\",100, \"display.width\",200, \"display.max_colwidth\",40)\n",
    "import pickle\n",
    "from rdkit import Chem\n",
    "from rdkit.Chem import AllChem, DataStructs\n",
    "import re\n",
    "from datetime import date\n",
    "pd.set_option(\"display.max_columns\",100, \"display.width\",2000, \"display.max_colwidth\",100)\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.model_selection import train_test_split\n",
    "import sklearn.metrics\n",
    "import optuna\n",
    "optuna.logging.set_verbosity(optuna.logging.WARN) #INFO, WARN\n",
    "import lightgbm as lgb; lgb.__version__\n",
    "import pickle\n",
    "import requests as r\n",
    "from scipy.sparse import csr_matrix, vstack\n",
    "from sklearn.metrics import classification_report, confusion_matrix, f1_score, roc_auc_score\n",
    "from sklearn.utils import class_weight\n",
    "import collections\n",
    "from binary_label_metrics import BinaryLabelMetrics\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "for x in [(\"numpy\",\"np\"),(\"pandas\",\"pd\")]:\n",
    "  print(f\"{x[0]} {eval(f'{x[1]}.__version__')}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Parameters\n",
      "NTRIALS: 20\n",
      " NBOOST: 50\n",
      " NCLASS: 2\n",
      "FP_BITS: 1024\n"
     ]
    }
   ],
   "source": [
    "prm = dict(\n",
    "  #number of Optuna trials\n",
    "  NTRIALS = 20,\n",
    "  \n",
    "  #number of boosted trees to be created\n",
    "  NBOOST = 50,\n",
    "\n",
    "  #number of classes in response variable\n",
    "  NCLASS = 2,\n",
    "\n",
    "  #morgan fingerpring bit length\n",
    "  FP_BITS = 1024\n",
    ")\n",
    "\n",
    "offs = max(map(len,prm.keys()))\n",
    "print(\"Parameters\")\n",
    "for k,v in prm.items():\n",
    "  print(f\"{k:>{offs}}: {v}\")\n",
    "\n",
    "del k, v, offs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Reading Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input_df: 5,388 x 83 (input df) \n",
      "\n",
      "Counts of distinct values in columns\n",
      "col:Target  (#unique:323)\n",
      "         0     1    2     3    4     5    6        7     8     9       10    11    12     13                14    15    16         17    18    19\n",
      "Target   AR  BRD4   ER  CDK4  BTK  CDK6  ALK  BCR-ABL  BRD2  BRD3  BCL-xL  CDK9  MDM2  HDAC6  EGFR L858R/T790M  MEK1  CDK2  EGFR e19d  AKT3  AKT2\n",
      "size    248   222  183   147  146   143  120      118   106    98      96    92    89     78                76    76    74         65    64    64 \n",
      "\n",
      "col:E3ligase  (#unique:13)\n",
      "            0     1      2     3     4    5    6      7       8       9       10      11    12\n",
      "E3ligase  CRBN   VHL  cIAP1  MDM2  XIAP  IAP  AhR  FEM1B  DCAF15  DCAF16  RNF114  DCAF11  RNF4\n",
      "size      3530  1578     95    56    49   46    9      9       4       4       4       3     1 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "input_df = pd.read_csv(\"data/protac/data.csv\").reset_index(drop=True)\n",
    "input_df = input_df.rename(columns={\"E3 ligase\":\"E3ligase\"})\n",
    "print(\"input_df: {:,} x {:,}\".format(*input_df.shape), \"(input df) \\n\")\n",
    "\n",
    "print(\"Counts of distinct values in columns\")\n",
    "for x in [\"Target\",\"E3ligase\"]:\n",
    "  tmp = input_df.groupby(x, as_index=False).size().sort_values(\"size\", ascending=False, ignore_index=True)\n",
    "  print(f\"col:{x}  (#unique:{tmp.shape[0]:,})\")\n",
    "  print(tmp.head(20).T, \"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Creating E3 Ligase Classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "col:E3ligase  (#unique:8)\n",
      "             0     1    2     3     4    5      6    7\n",
      "E3ligase  CRBN   VHL  IAP  MDM2  DCAF  AhR  FEM1B  RNF\n",
      "size      3530  1578  190    56    11    9      9    5 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(input_df[\"E3ligase\"])):\n",
    "    if 'IAP' in str(input_df.loc[i].at[\"E3ligase\"]):\n",
    "        input_df.at[i,'E3ligase']='IAP'\n",
    "    elif 'DCAF' in str(input_df.loc[i].at[\"E3ligase\"]):\n",
    "        input_df.at[i,'E3ligase']='DCAF'\n",
    "    elif 'RNF' in str(input_df.loc[i].at[\"E3ligase\"]):\n",
    "        input_df.at[i,'E3ligase']='RNF'\n",
    "\n",
    "tmp = input_df.groupby('E3ligase', as_index=False).size().sort_values(\"size\", ascending=False, ignore_index=True)\n",
    "print(f\"col:{'E3ligase'}  (#unique:{tmp.shape[0]:,})\")\n",
    "print(tmp.head(20).T, \"\\n\")\n",
    "del tmp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dropping Rows with no DC50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of Rows with explicit DC50: 905\n",
      "input_df: 833 x 83 (input df) \n",
      "\n"
     ]
    }
   ],
   "source": [
    "print('Number of Rows with explicit DC50: ' + str((input_df.shape[0] - input_df['DC50 (nM)'].isna().sum())))\n",
    "input_df = input_df.dropna(subset=['DC50 (nM)'])\n",
    "input_df = input_df.dropna(subset=['Uniprot'])\n",
    "print(\"input_df: {:,} x {:,}\".format(*input_df.shape), \"(input df) \\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Preparing Features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "E3 Ligase"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model_df: 833 x 6 (E3ligase OHE features)\n"
     ]
    }
   ],
   "source": [
    "model_df = pd.concat([\n",
    "    pd.get_dummies(input_df[\"E3ligase\"], prefix=\"e3\"),\n",
    "    input_df[\"Smiles\"]\n",
    "  ], axis=1).reset_index(drop=True)\n",
    "\n",
    "print(\"model_df: {:,} x {:,}\".format(*model_df.shape), \"(E3ligase OHE features)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Receptor Sequence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "cID= list(set(input_df['Uniprot'].tolist()))\n",
    "seqs = {}\n",
    "\n",
    "for c in cID:\n",
    "    try:\n",
    "        baseUrl=\"http://www.uniprot.org/uniprot/\"\n",
    "        currentUrl=baseUrl+c+\".fasta\"\n",
    "        response = r.post(currentUrl)\n",
    "        cData=''.join(response.text)\n",
    "        i = cData.index('\\n')+1\n",
    "        seq = cData[i:].strip()\n",
    "        seqs[c] = seq\n",
    "    except:pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model_df: 833 x 7 (rec seq feature)\n"
     ]
    }
   ],
   "source": [
    "input_df[\"rec_seq\"] = input_df.apply(lambda row: seqs[row[\"Uniprot\"]], axis=1)\n",
    "model_df = model_df.assign(rec_seq = input_df[\"rec_seq\"].tolist())\n",
    "print(\"model_df: {:,} x {:,}\".format(*model_df.shape), \"(rec seq feature)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Molecular Fingerprint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model_df: 833 x 7\n",
      "fp_df: 833 x 1,024\n",
      "model_df: 833 x 1,031 (added fingerprint features)\n"
     ]
    }
   ],
   "source": [
    "def get_morgan_fp(smiles:str, nbits:int=prm[\"FP_BITS\"]):\n",
    "  m = Chem.MolFromSmiles(smiles)\n",
    "  fingerprint = AllChem.GetMorganFingerprintAsBitVect(m, 2, nBits=nbits)\n",
    "  array = np.zeros((0,), dtype=np.int8)\n",
    "  DataStructs.ConvertToNumpyArray(fingerprint, array)\n",
    "  return array\n",
    "\n",
    "fingerprints = [get_morgan_fp(x).tolist() for x in input_df[\"Smiles\"]]\n",
    "fp_df = pd.DataFrame(fingerprints, columns=[\"sm\"+str(n) for n in range(1,prm[\"FP_BITS\"]+1)])\n",
    "print(\"model_df: {:,} x {:,}\".format(*model_df.shape))\n",
    "\n",
    "#assert model_df.columns == smiles_df.columns\n",
    "\n",
    "model_df = pd.concat([model_df,fp_df],axis=1)\n",
    "print(\"fp_df: {:,} x {:,}\".format(*fp_df.shape))\n",
    "print(\"model_df: {:,} x {:,}\".format(*model_df.shape), \"(added fingerprint features)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cell Type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Most Common Cell Type: SRD15\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "148"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model_df: 833 x 1,032\n",
      "ct_df: 833 x 148\n",
      "model_df: 833 x 1,180 (added cell type OHE features)\n"
     ]
    }
   ],
   "source": [
    "celltypes = []\n",
    "for i in range(input_df.shape[0]):\n",
    "    row = str(input_df.iloc[i].to_list())\n",
    "    cell = re.findall('[\\w\\.;/-]+ cells', row)\n",
    "    if len(cell)==0:\n",
    "        celltypes.append('Unknown')\n",
    "    else:\n",
    "        celltypes.append(cell[0][:-6])\n",
    "\n",
    "def most_frequent(List):\n",
    "    return max(set(List), key = List.count)\n",
    "\n",
    "print(\"Most Common Cell Type: \" + most_frequent(celltypes))\n",
    "\n",
    "len(set(celltypes))\n",
    "model_df['ct'] = celltypes\n",
    "ct_df = pd.get_dummies(model_df['ct'], prefix='ct')\n",
    "print(\"model_df: {:,} x {:,}\".format(*model_df.shape))\n",
    "model_df = pd.concat([model_df,ct_df],axis=1)\n",
    "print(\"ct_df: {:,} x {:,}\".format(*ct_df.shape))\n",
    "print(\"model_df: {:,} x {:,}\".format(*model_df.shape), \"(added cell type OHE features)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Response Variable (DC50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/vs/8ts0hp5d0wzf8zdrzyqly4840000gn/T/ipykernel_71585/1778872113.py:4: FutureWarning: The default value of regex will change from True to False in a future version.\n",
      "  model_df['resp'] = model_df[\"resp\"].str.replace(\"[<=>]\",\"\")\n",
      "/var/folders/vs/8ts0hp5d0wzf8zdrzyqly4840000gn/T/ipykernel_71585/1778872113.py:5: FutureWarning: The default value of regex will change from True to False in a future version.\n",
      "  model_df['resp'] = model_df[\"resp\"].str.replace(\"\\(n/a\\)\",\"0\")\n",
      "/var/folders/vs/8ts0hp5d0wzf8zdrzyqly4840000gn/T/ipykernel_71585/1778872113.py:6: FutureWarning: The default value of regex will change from True to False in a future version.\n",
      "  model_df['resp'] = model_df[\"resp\"].str.replace('[\\d]+[-~]',\"\")\n"
     ]
    }
   ],
   "source": [
    "#add and process DC50 column\n",
    "\n",
    "model_df['resp'] = list(input_df['DC50 (nM)'])\n",
    "model_df['resp'] = model_df[\"resp\"].str.replace(\"[<=>]\",\"\")\n",
    "model_df['resp'] = model_df[\"resp\"].str.replace(\"\\(n/a\\)\",\"0\")\n",
    "model_df['resp'] = model_df[\"resp\"].str.replace('[\\d]+[-~]',\"\")\n",
    "model_df = model_df.loc[~model_df['resp'].str.contains('[a-zA-Z]')]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Split DC50 by Cell Type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [],
   "source": [
    "out_nochange = []\n",
    "out_v1 = []\n",
    "out_v2 = []\n",
    "for row in model_df.to_dict(orient=\"records\"):\n",
    "  #no slashes\n",
    "  if \"/\" not in row[\"resp\"]:\n",
    "    out_nochange.append(row)\n",
    "    continue\n",
    "  \n",
    "  #slashes in resp matched by a column with ct which has same number of slashes\n",
    "  lst = [k for k,v in row.items() if k.startswith(\"ct\") and \"/\" in k and v==1]\n",
    "  assert len(lst) in [0,1]\n",
    "  if len(lst) == 1:\n",
    "    #test.append((lst[0],row[\"resp\"]))\n",
    "    cols = lst[0].split(\"/\")\n",
    "    vals = row[\"resp\"].split(\"/\")\n",
    "    # assert len(cols)==len(vals)\n",
    "    # assert len(cols) in [2,3]\n",
    "    \n",
    "    #adding 2 or 3 rows\n",
    "    for c,v in zip(cols,vals):\n",
    "      dct = row.copy()\n",
    "      k = c if c.startswith(\"ct_\") else \"ct_\"+c\n",
    "      dct[k] = 1\n",
    "      dct[\"resp\"] = v\n",
    "      del dct[lst[0]]\n",
    "      out_v1.append(dct)\n",
    "  else:\n",
    "    vals = row[\"resp\"].split(\"/\")\n",
    "    dct = row.copy()\n",
    "    dct[\"resp\"] = vals[0]\n",
    "    out_v2.append(dct)\n",
    "    \n",
    "model_df = pd.DataFrame(out_nochange+out_v1+out_v2)\n",
    "model_df = model_df.fillna(0)\n",
    "model_df['resp'] = pd.to_numeric(model_df['resp'])\n",
    "model_df = model_df.drop(['ct'],axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "DC50 Distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([306.,  69.,  59.,  53.,  53.,  17.,  13.,  23.,  11., 384.]),\n",
       " array([1.0000e-02, 1.0109e+01, 2.0208e+01, 3.0307e+01, 4.0406e+01,\n",
       "        5.0505e+01, 6.0604e+01, 7.0703e+01, 8.0802e+01, 9.0901e+01,\n",
       "        1.0100e+02]),\n",
       " <BarContainer object of 10 artists>)"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Count')"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'DC50 (nM)')"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#dropping rows with outlier DC50 values\n",
    "model_df.drop(model_df.index[model_df['resp'] <= 0], inplace=True)\n",
    "model_df.loc[model_df.resp>100, 'resp'] = 101\n",
    "\n",
    "plt.hist(model_df['resp'].astype(float), color = '#00A4CCFF')\n",
    "plt.grid()\n",
    "plt.ylabel('Count')\n",
    "plt.xlabel('DC50 (nM)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model_df: 987 x 1,240\n"
     ]
    }
   ],
   "source": [
    "model_df.loc[model_df.resp<=100.0, 'resp_categorical'] = '1'\n",
    "model_df['resp_categorical'] = model_df['resp_categorical'].fillna('0')\n",
    "model_df.drop(model_df.tail(1).index,inplace=True) # drop last row of nans\n",
    "print(\"model_df: {:,} x {:,}\".format(*model_df.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Receptor Seq to Ngrams"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['a a', 'a a a', 'a a c', ..., 'y y v', 'y y w', 'y y y'],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model_df: 987 x 1,240\n",
      "rec_n_grams_df: 987 x 7,841\n",
      "model_df: 989 x 9,081 (added n-gram features)\n"
     ]
    }
   ],
   "source": [
    "model_df['rec_seq'] = model_df['rec_seq'].apply(lambda x: \" \".join(x))\n",
    "vectorizer = CountVectorizer(token_pattern=r'(?u)\\b\\w+\\b',ngram_range=(2,3))\n",
    "X = vectorizer.fit_transform(model_df['rec_seq'].tolist())\n",
    "vectorizer.get_feature_names_out()\n",
    "\n",
    "rec_n_grams_df = pd.DataFrame(X.toarray(),columns=list(s.replace(\" \", \"\") for s in vectorizer.get_feature_names_out()))\n",
    "print(\"model_df: {:,} x {:,}\".format(*model_df.shape))\n",
    "model_df = pd.concat([model_df,rec_n_grams_df],axis=1)\n",
    "print(\"rec_n_grams_df: {:,} x {:,}\".format(*rec_n_grams_df.shape))\n",
    "print(\"model_df: {:,} x {:,}\".format(*model_df.shape), \"(added n-gram features)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_df = model_df.drop([\"rec_seq\"],axis=1)\n",
    "model_df.drop(model_df.tail(4).index, inplace = True)\n",
    "resp_col = model_df.pop('resp_categorical')\n",
    "model_df['resp_categorical'] = resp_col #to add to end"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Model Matrices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "955\n",
      "236\n",
      "484\n",
      "801\n",
      "724\n",
      "605\n",
      "355\n",
      "954\n",
      "560\n",
      "529\n",
      "223\n",
      "6\n",
      "683\n",
      "486\n",
      "682\n",
      "870\n",
      "389\n",
      "928\n",
      "783\n",
      "543\n",
      "584\n",
      "732\n",
      "238\n",
      "680\n",
      "827\n",
      "267\n",
      "233\n",
      "813\n",
      "502\n",
      "613\n",
      "194\n",
      "710\n",
      "527\n",
      "133\n",
      "950\n",
      "90\n",
      "649\n",
      "969\n",
      "888\n",
      "154\n",
      "119\n",
      "597\n",
      "41\n",
      "553\n",
      "464\n",
      "824\n",
      "34\n",
      "744\n",
      "35\n",
      "935\n",
      "335\n",
      "94\n",
      "120\n",
      "823\n",
      "370\n",
      "374\n",
      "957\n",
      "78\n",
      "915\n",
      "940\n",
      "931\n",
      "805\n",
      "559\n",
      "654\n",
      "722\n",
      "17\n",
      "201\n",
      "650\n",
      "951\n",
      "632\n",
      "774\n",
      "65\n",
      "186\n",
      "268\n",
      "429\n",
      "979\n",
      "800\n",
      "148\n",
      "216\n",
      "180\n",
      "504\n",
      "845\n",
      "101\n",
      "159\n",
      "340\n",
      "404\n",
      "104\n",
      "873\n",
      "799\n",
      "599\n",
      "867\n",
      "205\n",
      "679\n",
      "697\n",
      "175\n",
      "108\n",
      "13\n",
      "156\n",
      "306\n",
      "676\n",
      "368\n",
      "981\n",
      "516\n",
      "479\n",
      "142\n",
      "446\n",
      "241\n",
      "531\n",
      "59\n",
      "856\n",
      "567\n",
      "455\n",
      "318\n",
      "600\n",
      "364\n",
      "248\n",
      "307\n",
      "3\n",
      "189\n",
      "672\n",
      "320\n",
      "686\n",
      "439\n",
      "528\n",
      "819\n",
      "862\n",
      "480\n",
      "574\n",
      "788\n",
      "388\n",
      "2\n",
      "483\n",
      "331\n",
      "430\n",
      "403\n",
      "704\n",
      "647\n",
      "73\n",
      "117\n",
      "372\n",
      "893\n",
      "111\n",
      "664\n",
      "385\n",
      "810\n",
      "759\n",
      "571\n",
      "880\n",
      "550\n",
      "8\n",
      "345\n",
      "56\n",
      "745\n",
      "457\n",
      "244\n",
      "84\n",
      "415\n",
      "200\n",
      "125\n",
      "590\n",
      "875\n",
      "874\n",
      "781\n",
      "904\n",
      "786\n",
      "848\n",
      "961\n",
      "81\n",
      "541\n",
      "435\n",
      "228\n",
      "947\n",
      "806\n",
      "563\n",
      "274\n",
      "663\n",
      "50\n",
      "191\n",
      "555\n",
      "523\n",
      "482\n",
      "195\n",
      "258\n",
      "702\n",
      "578\n",
      "982\n",
      "978\n",
      "638\n",
      "899\n",
      "347\n",
      "422\n",
      "579\n",
      "851\n",
      "700\n",
      "250\n",
      "339\n",
      "255\n",
      "314\n",
      "47\n",
      "941\n",
      "934\n",
      "19\n",
      "832\n",
      "207\n",
      "906\n",
      "912\n",
      "262\n",
      "474\n",
      "812\n",
      "242\n",
      "349\n",
      "177\n",
      "643\n",
      "60\n",
      "691\n",
      "939\n",
      "719\n",
      "902\n",
      "749\n",
      "309\n",
      "735\n",
      "975\n",
      "882\n",
      "99\n",
      "762\n",
      "298\n",
      "896\n",
      "608\n",
      "756\n",
      "181\n",
      "12\n",
      "604\n",
      "85\n",
      "496\n",
      "640\n",
      "741\n",
      "925\n",
      "878\n",
      "551\n",
      "135\n",
      "582\n",
      "592\n",
      "160\n",
      "937\n",
      "311\n",
      "983\n",
      "853\n",
      "894\n",
      "487\n",
      "107\n",
      "202\n",
      "49\n",
      "701\n",
      "787\n",
      "881\n",
      "383\n",
      "121\n",
      "720\n",
      "187\n",
      "16\n",
      "623\n",
      "443\n",
      "964\n",
      "301\n",
      "459\n",
      "106\n",
      "814\n",
      "23\n",
      "820\n",
      "69\n",
      "629\n",
      "432\n",
      "304\n",
      "970\n",
      "634\n",
      "472\n",
      "525\n",
      "861\n",
      "74\n",
      "218\n",
      "517\n",
      "265\n",
      "625\n",
      "110\n",
      "821\n",
      "967\n",
      "537\n",
      "185\n",
      "286\n",
      "573\n",
      "354\n",
      "436\n",
      "408\n",
      "61\n",
      "395\n",
      "764\n"
     ]
    }
   ],
   "source": [
    "train,test = train_test_split(model_df, random_state=1, test_size=.3)\n",
    "#preventing data leakage\n",
    "train_smis = train['Smiles'].tolist()\n",
    "drop_indices = []\n",
    "rows = []\n",
    "for index, row in test.iterrows():\n",
    "    print(index)\n",
    "    if row['Smiles'] in train_smis:\n",
    "        drop_indices.append(index)\n",
    "test.drop(drop_indices, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "nfeatures: 9,078\n",
      "x train before: 689 x 9,077 (x df) \n",
      "\n"
     ]
    }
   ],
   "source": [
    "features = train.columns.tolist()[1:-1]  #dropping \"resp_categorical, resp still there\"\n",
    "print(f\"nfeatures: {len(features):,}\")\n",
    "\n",
    "X_train = train.drop([\"resp_categorical\",\"resp\",\"Smiles\"], axis=1).values\n",
    "print(\"x train before: {:,} x {:,}\".format(*X_train.shape), \"(x df) \\n\")\n",
    "y_train = train[\"resp_categorical\"].values\n",
    "y_train = np.array([int(x) for x in y_train])\n",
    "\n",
    "X_test = test.drop([\"resp_categorical\",\"resp\",\"Smiles\"], axis=1).values\n",
    "y_test = test[\"resp_categorical\"].values\n",
    "y_test = np.array([int(x) for x in y_test])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[train] X: 689 x 9,077\n",
      "[test] X: 144 x 9,077\n"
     ]
    }
   ],
   "source": [
    "print(\"[train] X: {:,} x {:,}\".format(*X_train.shape,y_train.shape))\n",
    "print(\"[test] X: {:,} x {:,}\".format(*X_test.shape,y_train.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hyperparameter Optimization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best trial (20): 0 (0.730) 1 (0.733) 5 (0.753) 10 (0.826) 14 (0.832) "
     ]
    }
   ],
   "source": [
    "class LightGBM:\n",
    "  \n",
    "  def __init__(self):\n",
    "    self.best_booster = None\n",
    "    self._booster = None\n",
    "    \n",
    "    self.X = X_train\n",
    "    self.y = y_train\n",
    "    self.nclass = 2\n",
    "    #self.wts = class_weight.compute_sample_weight(class_weight=\"balanced\", y=self.y)\n",
    "    self.prm_lgb = {      \n",
    "      \"objective\": \"multiclass\" if self.nclass>2 else \"binary\",\n",
    "      \"metric\": None, \n",
    "      \"verbosity\": -1,\n",
    "      #\"num_class\": self.nclass,\n",
    "      \"boosting_type\": \"gbdt\",\n",
    "      \"force_row_wise\": True,\n",
    "      \"min_gain_to_split\": .5,\n",
    "    }\n",
    "  \n",
    "  def __call__(self, trial):    \n",
    "    dtrain = lgb.Dataset(self.X, self.y)\n",
    "    prm_lgb = {\n",
    "      **self.prm_lgb\n",
    "      , **{\n",
    "        \"learning_rate\": trial.suggest_loguniform(\"learning_rate\", .01, .3),\n",
    "        \"lambda_l1\": trial.suggest_loguniform(\"lambda_l1\", 1E-3, 1.),\n",
    "        \"lambda_l2\": trial.suggest_loguniform(\"lambda_l2\", .5, 3.),\n",
    "        \"num_leaves\": trial.suggest_int(\"num_leaves\", 8, 32),\n",
    "        \"min_data_in_leaf\": trial.suggest_int(\"min_data_in_leaf\", 50, 100),\n",
    "        \"feature_fraction\": trial.suggest_uniform(\"feature_fraction\", .3, .6),\n",
    "        \"bagging_fraction\": trial.suggest_uniform(\"bagging_fraction\", .4, 1.),\n",
    "        \"bagging_freq\": trial.suggest_int(\"bagging_freq\", 2, 6),\n",
    "        \"boosting_type\":trial.suggest_categorical(\"boosting_type\",[\"gbdt\",\"rf\"])\n",
    "      }\n",
    "    }\n",
    "    \n",
    "    if self.nclass>2:\n",
    "      def f1_eval(preds,dataset):\n",
    "        f_score = f1_score(dataset.get_label(),preds.reshape(-1,self.nclass).argmax(axis=1),average=\"micro\")\n",
    "        return \"f1_score\",f_score,True\n",
    "    else:\n",
    "      def f1_eval(preds,dataset):\n",
    "        pred1 = np.zeros(dataset.get_label().shape[0],dtype=int); pred1[-dataset.get_label().sum().astype(int):] = 1\n",
    "        f_score = f1_score(dataset.get_label()[preds.argsort()],pred1,average=\"micro\")\n",
    "        return \"f1_score\",f_score,True\n",
    "    \n",
    "    early_stop = lgb.early_stopping(20, first_metric_only=False, verbose=False)\n",
    "    eval_hist = lgb.cv(prm_lgb, dtrain, nfold=5, seed=12345, num_boost_round=prm[\"NBOOST\"],\n",
    "                           feval=f1_eval, callbacks=[early_stop])\n",
    "    \n",
    "    return eval_hist[\"f1_score-mean\"][-1]\n",
    "  \n",
    "  def callback(self, study, trial):\n",
    "    if study.best_trial==trial:\n",
    "      print(f\"{study.best_trial.number} ({study.best_trial.values[0]:.3f})\", end=\" \", flush=True)\n",
    "      self.best_booster = self._booster\n",
    "      return\n",
    "    if trial.number%20==0:\n",
    "      print(f\"{trial.number}\", end=\" \", flush=True)\n",
    "\n",
    "  \n",
    "objective = LightGBM()\n",
    "print(f\"best trial ({prm['NTRIALS']}):\", end=\" \")\n",
    "study = optuna.create_study(pruner=optuna.pruners.MedianPruner(n_warmup_steps=10), \n",
    "                              sampler=optuna.samplers.TPESampler(seed=1234), direction=\"maximize\")\n",
    "study.optimize(objective, n_trials=prm[\"NTRIALS\"], callbacks=[objective.callback])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "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>bagging_fraction</th>\n",
       "      <th>bagging_freq</th>\n",
       "      <th>boosting_type</th>\n",
       "      <th>feature_fraction</th>\n",
       "      <th>lambda_l1</th>\n",
       "      <th>lambda_l2</th>\n",
       "      <th>learning_rate</th>\n",
       "      <th>min_data_in_leaf</th>\n",
       "      <th>num_leaves</th>\n",
       "      <th>value</th>\n",
       "      <th>duration</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.8876</td>\n",
       "      <td>3</td>\n",
       "      <td>gbdt</td>\n",
       "      <td>0.5412</td>\n",
       "      <td>0.2799</td>\n",
       "      <td>0.7884</td>\n",
       "      <td>0.2299</td>\n",
       "      <td>58</td>\n",
       "      <td>19</td>\n",
       "      <td>0.8316</td>\n",
       "      <td>1.7113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.8153</td>\n",
       "      <td>3</td>\n",
       "      <td>gbdt</td>\n",
       "      <td>0.4045</td>\n",
       "      <td>0.3054</td>\n",
       "      <td>0.6487</td>\n",
       "      <td>0.2936</td>\n",
       "      <td>60</td>\n",
       "      <td>21</td>\n",
       "      <td>0.8288</td>\n",
       "      <td>2.9213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.8784</td>\n",
       "      <td>3</td>\n",
       "      <td>gbdt</td>\n",
       "      <td>0.3990</td>\n",
       "      <td>0.2883</td>\n",
       "      <td>0.6946</td>\n",
       "      <td>0.2957</td>\n",
       "      <td>59</td>\n",
       "      <td>20</td>\n",
       "      <td>0.8287</td>\n",
       "      <td>3.0234</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.6383</td>\n",
       "      <td>5</td>\n",
       "      <td>rf</td>\n",
       "      <td>0.4954</td>\n",
       "      <td>0.0702</td>\n",
       "      <td>0.5723</td>\n",
       "      <td>0.0346</td>\n",
       "      <td>97</td>\n",
       "      <td>17</td>\n",
       "      <td>0.7126</td>\n",
       "      <td>1.1619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.4260</td>\n",
       "      <td>4</td>\n",
       "      <td>rf</td>\n",
       "      <td>0.4600</td>\n",
       "      <td>0.0036</td>\n",
       "      <td>0.5443</td>\n",
       "      <td>0.0123</td>\n",
       "      <td>80</td>\n",
       "      <td>24</td>\n",
       "      <td>0.6981</td>\n",
       "      <td>1.9441</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.5313</td>\n",
       "      <td>6</td>\n",
       "      <td>rf</td>\n",
       "      <td>0.5114</td>\n",
       "      <td>0.0203</td>\n",
       "      <td>2.1046</td>\n",
       "      <td>0.1922</td>\n",
       "      <td>85</td>\n",
       "      <td>11</td>\n",
       "      <td>0.6953</td>\n",
       "      <td>1.0099</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20 rows × 11 columns</p>\n",
       "</div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  duration: 0.7min\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>meanv</th>\n",
       "      <th>sdv</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>boosting_type</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>gbdt</th>\n",
       "      <td>0.800110</td>\n",
       "      <td>0.039404</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf</th>\n",
       "      <td>0.716027</td>\n",
       "      <td>0.016772</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  meanv       sdv\n",
       "boosting_type                    \n",
       "gbdt           0.800110  0.039404\n",
       "rf             0.716027  0.016772"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trials_df = study.trials_dataframe(attrs=(\"params\",\"user_attrs\",\"value\",\"duration\"))\n",
    "for y in [\"params\",\"user_attrs\"]:\n",
    "  trials_df.columns = [x[1+len(y):] if x.startswith(y) else x for x in trials_df.columns]\n",
    "trials_df[\"duration\"] = trials_df[\"duration\"].apply(lambda x:x.total_seconds())\n",
    "with pd.option_context(\"display.max_rows\",6, \"display.float_format\",\"{:.4f}\".format):\n",
    "  display_html(trials_df.sort_values(\"value\", ascending=False))\n",
    "print(f\"  duration: {trials_df['duration'].sum()/60:.1f}min\")\n",
    "\n",
    "trials_df.groupby(\"boosting_type\").agg(meanv=(\"value\",\"mean\"),sdv=(\"value\",\"std\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "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>bagging_fraction</th>\n",
       "      <th>bagging_freq</th>\n",
       "      <th>boosting_type</th>\n",
       "      <th>feature_fraction</th>\n",
       "      <th>lambda_l1</th>\n",
       "      <th>lambda_l2</th>\n",
       "      <th>learning_rate</th>\n",
       "      <th>min_data_in_leaf</th>\n",
       "      <th>num_leaves</th>\n",
       "      <th>value</th>\n",
       "      <th>duration</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.8876</td>\n",
       "      <td>3</td>\n",
       "      <td>gbdt</td>\n",
       "      <td>0.5412</td>\n",
       "      <td>0.2799</td>\n",
       "      <td>0.7884</td>\n",
       "      <td>0.2299</td>\n",
       "      <td>58</td>\n",
       "      <td>19</td>\n",
       "      <td>0.8316</td>\n",
       "      <td>1.7113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.8153</td>\n",
       "      <td>3</td>\n",
       "      <td>gbdt</td>\n",
       "      <td>0.4045</td>\n",
       "      <td>0.3054</td>\n",
       "      <td>0.6487</td>\n",
       "      <td>0.2936</td>\n",
       "      <td>60</td>\n",
       "      <td>21</td>\n",
       "      <td>0.8288</td>\n",
       "      <td>2.9213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.8784</td>\n",
       "      <td>3</td>\n",
       "      <td>gbdt</td>\n",
       "      <td>0.3990</td>\n",
       "      <td>0.2883</td>\n",
       "      <td>0.6946</td>\n",
       "      <td>0.2957</td>\n",
       "      <td>59</td>\n",
       "      <td>20</td>\n",
       "      <td>0.8287</td>\n",
       "      <td>3.0234</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.9998</td>\n",
       "      <td>2</td>\n",
       "      <td>gbdt</td>\n",
       "      <td>0.5844</td>\n",
       "      <td>0.6521</td>\n",
       "      <td>0.7608</td>\n",
       "      <td>0.0860</td>\n",
       "      <td>52</td>\n",
       "      <td>17</td>\n",
       "      <td>0.8258</td>\n",
       "      <td>1.7178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.8147</td>\n",
       "      <td>3</td>\n",
       "      <td>gbdt</td>\n",
       "      <td>0.4066</td>\n",
       "      <td>0.2149</td>\n",
       "      <td>0.9751</td>\n",
       "      <td>0.2409</td>\n",
       "      <td>58</td>\n",
       "      <td>21</td>\n",
       "      <td>0.8258</td>\n",
       "      <td>2.9295</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.5659</td>\n",
       "      <td>6</td>\n",
       "      <td>gbdt</td>\n",
       "      <td>0.3818</td>\n",
       "      <td>0.0735</td>\n",
       "      <td>1.0954</td>\n",
       "      <td>0.0192</td>\n",
       "      <td>89</td>\n",
       "      <td>27</td>\n",
       "      <td>0.7300</td>\n",
       "      <td>2.0412</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.7221</td>\n",
       "      <td>6</td>\n",
       "      <td>rf</td>\n",
       "      <td>0.4508</td>\n",
       "      <td>0.0507</td>\n",
       "      <td>1.2883</td>\n",
       "      <td>0.0168</td>\n",
       "      <td>74</td>\n",
       "      <td>31</td>\n",
       "      <td>0.7243</td>\n",
       "      <td>2.0561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.6383</td>\n",
       "      <td>5</td>\n",
       "      <td>rf</td>\n",
       "      <td>0.4954</td>\n",
       "      <td>0.0702</td>\n",
       "      <td>0.5723</td>\n",
       "      <td>0.0346</td>\n",
       "      <td>97</td>\n",
       "      <td>17</td>\n",
       "      <td>0.7126</td>\n",
       "      <td>1.1619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.4260</td>\n",
       "      <td>4</td>\n",
       "      <td>rf</td>\n",
       "      <td>0.4600</td>\n",
       "      <td>0.0036</td>\n",
       "      <td>0.5443</td>\n",
       "      <td>0.0123</td>\n",
       "      <td>80</td>\n",
       "      <td>24</td>\n",
       "      <td>0.6981</td>\n",
       "      <td>1.9441</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.5313</td>\n",
       "      <td>6</td>\n",
       "      <td>rf</td>\n",
       "      <td>0.5114</td>\n",
       "      <td>0.0203</td>\n",
       "      <td>2.1046</td>\n",
       "      <td>0.1922</td>\n",
       "      <td>85</td>\n",
       "      <td>11</td>\n",
       "      <td>0.6953</td>\n",
       "      <td>1.0099</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20 rows × 11 columns</p>\n",
       "</div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with pd.option_context(\"display.max_rows\",10, \"display.float_format\",\"{:.4f}\".format):\n",
    "  display_html(trials_df.sort_values(\"value\", ascending=False))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "ind = np.argmax(trials_df[\"value\"].values)\n",
    "lgb_prm = study.trials[ind].params\n",
    "lgb_prm = {**lgb_prm, \n",
    "  **{\n",
    "    \"objective\": \"multiclass\",\n",
    "    \"metric\": None,\n",
    "    \"num_class\": 2,\n",
    "    \"force_row_wise\": True, \n",
    "    \"verbosity\": -1,\n",
    "    \"min_gain_to_split\": .5,\n",
    "  }\n",
    "}\n",
    "\n",
    "def f1_eval(preds, dtrain):\n",
    "  preds = preds.reshape(prm['NCLASS'],-1).T\n",
    "  f_score = f1_score(dtrain.get_label(),preds.argmax(axis=1),average=\"micro\")\n",
    "  return \"f1_score\",f_score,True\n",
    "\n",
    "wt = class_weight.compute_sample_weight(class_weight=\"balanced\", y=y_train)\n",
    "dtrain = lgb.Dataset(X_train, y_train, weight=wt)\n",
    "#early_stop = lgb.early_stopping(20, first_metric_only=False, verbose=True)\n",
    "model = lgb.train(lgb_prm, dtrain, feval=f1_eval,\n",
    "                      num_boost_round=prm[\"NBOOST\"])\n",
    "\n",
    "del ind, lgb_prm, wt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Metrics on Hold-Out Test Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(144,) (144,)\n"
     ]
    }
   ],
   "source": [
    "y_hat = model.predict(X_test)\n",
    "y_hat = np.array([val[1] for val in y_hat])\n",
    "print(y_hat.shape, y_test.shape)\n",
    "scores_df = pd.DataFrame({'label':list(y_test),'score':list(y_hat)})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [],
   "source": [
    "blm = BinaryLabelMetrics()\n",
    "blm.add_model('binary_gbm', scores_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1008x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "correct pred pos [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, -4, -2, -5, -5, -6, -8, -4, -11, -19, -9]\n",
      "incorrect pred pos [0 0 0 0 0 0 0 0 0 0 1 1 2 3 0 3 0 0 1 1]\n",
      "correct pred neg [13  2  6 12  2  0  1  3  3  0  0  0  0  0  0  0  0  0  0  0]\n",
      "incorrect pred neg [0, 0, -2, -1, -4, -1, -2, -2, -1, -4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1008x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "blm.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "blm.plot_roc()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "l = blm._f1[0]\n",
    "l = [x for x in l if math.isnan(x) == False]\n",
    "original = ['original',blm._auc,blm._prrec,max(l)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Feature Importances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [],
   "source": [
    "def shuffle(var, df):\n",
    "    if var == 'cellType':\n",
    "        cols = test.columns.tolist()\n",
    "        ct_cols = [c for c in cols if 'ct' in c]\n",
    "        new_df = df.copy(deep=True)\n",
    "        for col in ct_cols:\n",
    "            new_df[col] = new_df[col].sample(frac=1).values\n",
    "        return new_df\n",
    "    elif var == 'e3':\n",
    "        cols = test.columns.tolist()\n",
    "        ct_cols = [c for c in cols if 'e3' in c]\n",
    "        new_df = df.copy(deep=True)\n",
    "        for col in ct_cols:\n",
    "            new_df[col] = new_df[col].sample(frac=1).values\n",
    "        return new_df\n",
    "    elif var == 'ligand':\n",
    "        cols = test.columns.tolist()\n",
    "        ct_cols = [c for c in cols if 'sm' in c]\n",
    "        new_df = df.copy(deep=True)\n",
    "        for col in ct_cols:\n",
    "            new_df[col] = new_df[col].sample(frac=1).values\n",
    "        return new_df\n",
    "    elif var == 'receptor':\n",
    "        cols = test.columns.tolist()\n",
    "        ct_cols = [c for c in cols if (len(c)==2 or len(c)==3) and 'sm' not in c]\n",
    "        new_df = df.copy(deep=True)\n",
    "        for col in ct_cols:\n",
    "            new_df[col] = new_df[col].sample(frac=1).values\n",
    "        return new_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(144,) (144,)\n",
      "(144,) (144,)\n",
      "(144,) (144,)\n",
      "(144,) (144,)\n"
     ]
    }
   ],
   "source": [
    "import math\n",
    "metrics = [['shuffled_var','auc','pr_rec','f1']]\n",
    "for var in ['cellType','e3','ligand','receptor']:\n",
    "    new_df = shuffle(var, test)\n",
    "    X_test = new_df.drop([\"resp_categorical\",\"resp\",\"Smiles\"], axis=1).values\n",
    "    y_hat = model.predict(X_test)\n",
    "    y_hat = np.array([val[1] for val in y_hat])\n",
    "    print(y_hat.shape, y_test.shape)\n",
    "    scores_df = pd.DataFrame({'label':list(y_test),'score':list(y_hat)})\n",
    "    blm = BinaryLabelMetrics()\n",
    "    blm.add_model('binary_gbm', scores_df)\n",
    "    lst = blm._f1[0]\n",
    "    newlist = [x for x in lst if math.isnan(x) == False]\n",
    "    metrics.append([var, blm._auc, blm._prrec,max(newlist)])\n",
    "metrics.append(original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['shuffled_var', 'auc', 'pr_rec', 'f1'],\n",
       " ['cellType', [0.8766460905349794], [0.8962226813923659], 0.8787878787878788],\n",
       " ['e3', [0.8775720164609053], [0.8963702921587283], 0.8743718592964823],\n",
       " ['ligand', [0.7294238683127572], [0.8085685948706341], 0.8148148148148148],\n",
       " ['receptor', [0.7150205761316872], [0.8115036868560717], 0.7713004484304933],\n",
       " ['original', [0.8766460905349794], [0.8962226813923659], 0.8787878787878788]]"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.12 ('my-rdkit-env')",
   "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": "3.9.12"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "ca09fa392acf7f4c9c27d29d2d1f288bcf246ad350262c369089ab5d60981f3e"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}