05_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pandas as pd \n",
    "import numpy as np\n",
    "import joblib\n",
    "\n",
    "from rdkit import Chem\n",
    "from rdkit.Chem import Crippen\n",
    "from mapchiral.mapchiral import encode\n",
    "\n",
    "from matplotlib.colors import ListedColormap, LogNorm \n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams['font.sans-serif'] = \"Menlo\"\n",
    "plt.rcParams['font.family'] = \"sans-serif\"\n",
    "plt.rcParams.update({'font.size': 22})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Datasets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Read data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "activity = joblib.load('data/activity_clean_features.pkl')\n",
    "hemolysis = joblib.load('data/hemolysis_clean_features.pkl')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Merge data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged = pd.merge(activity, hemolysis, on=['sequence', 'helicity', 'hydrophobic_moment'], how='left')\n",
    "merged.rename(columns={'label_x':'activity', 'label_y':'hemolysis', 'map4_x': 'map4'}, inplace=True)\n",
    "merged.fillna(-1, inplace=True)\n",
    "merged.sort_values(by=['activity', 'hemolysis'], inplace=True)\n",
    "merged = merged[['sequence', 'activity', 'hemolysis', 'helicity', 'hydrophobic_moment', 'map4']].copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calculate fingerprints and properties"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def FracC(mol):\n",
    "    smarts = Chem.MolFromSmarts('[C]')\n",
    "    return len(mol.GetSubstructMatches(smarts, uniquify=True))/len(mol.GetAtoms())\n",
    "\n",
    "merged['mol'] = merged['sequence'].apply(lambda x: Chem.MolFromSequence(x))\n",
    "merged['hac'] = merged['mol'].apply(lambda x: x.GetNumHeavyAtoms())\n",
    "merged['frac_c'] = merged['mol'].apply(FracC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged[['sequence', 'activity', 'hemolysis', 'hac', 'frac_c', 'helicity', 'hydrophobic_moment']].to_csv('tmap_data/merged.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Cross-Validations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot cross-validation results of the activity and hemolysis models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1800x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "filenames = os.listdir('results/')\n",
    "filenames.sort()\n",
    "\n",
    "metrics = ('ROC AUC', 'Accuracy', 'Precision', 'Recall', 'F1')\n",
    "colors = ['#648FFF', '#DC267F', '#D0D0D0', '#FFB000', '#648FFF', '#DC267F', '#FFB000','#D0D0D0', '#FFB000']\n",
    "\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(18, 6), sharey=True)\n",
    "\n",
    "activity_means = []\n",
    "activity_stds = []\n",
    "\n",
    "hemolysis_means = []\n",
    "hemolysis_stds = []\n",
    "\n",
    "for file in filenames:\n",
    "\n",
    "    df = pd.read_csv('results/' + file)\n",
    "\n",
    "    model_name = ' '.join(file.split('_')[:-1])\n",
    "    mean_values = df.iloc[-2, 1:].values.tolist()\n",
    "    std_values = df.iloc[-1, 1:].values.tolist()\n",
    "\n",
    "    if 'activity' in file:\n",
    "        activity_means.append(mean_values)\n",
    "        activity_stds.append(std_values)\n",
    "    \n",
    "    elif 'hemolysis' in file:\n",
    "        hemolysis_means.append(mean_values)\n",
    "        hemolysis_stds.append(std_values)\n",
    "\n",
    "x = np.arange(len(metrics))\n",
    "width = 0.2\n",
    "\n",
    "for i in range(len(activity_means)):\n",
    "    ax1.bar(x + i * width, activity_means[i], width, label=' '.join(filenames[i].split('_')[:-1]), yerr=activity_stds[i], capsize=5, color=colors[i])\n",
    "\n",
    "for i in range(len(hemolysis_means)):\n",
    "    ax2.bar(x + i * width, hemolysis_means[i], width, label=' '.join(filenames[i].split('_')[:-1]), yerr=hemolysis_stds[i], capsize=5, color=colors[i])\n",
    "\n",
    "ax1.set_xticks(x + 1.5 * width)\n",
    "ax2.set_xticks(x + 1.5 * width)\n",
    "\n",
    "ax1.set_xticklabels(metrics, rotation=90)\n",
    "ax2.set_xticklabels(metrics, rotation=90)\n",
    "\n",
    "ax1.set_ylim(0, 1)\n",
    "ax2.set_ylim(0, 1)\n",
    "\n",
    "ax1.title.set_text('Activity')\n",
    "ax2.title.set_text('Hemolysis')\n",
    "\n",
    "plt.savefig('figures/metrics.png', dpi=1200, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "openaienv",
   "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": "3.12.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}