Figures_Tables/main_text/Figure4.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Figure 4 \n",
    "## Distribution of IEAemph and IEAairway values grouped by previously published COPD K-means clustering subtypes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import sys\n",
    "sys.path.append(\"../../src/\")\n",
    "import utility\n",
    "import pandas as pd\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "from scipy.stats import pearsonr\n",
    "from scipy.stats import ttest_ind\n",
    "\n",
    "from scipy.stats import kruskal\n",
    "from statsmodels.stats.proportion import proportions_chisquare\n",
    "\n",
    "from mpl_toolkits.axes_grid1.inset_locator import inset_axes\n",
    "import matplotlib\n",
    "\n",
    "import statsmodels.api as sm\n",
    "\n",
    "from scipy import stats\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "import math\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Loading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/ocean/projects/asc170022p/juc91/anaconda/envs/IEA/lib/python3.7/site-packages/pandas/core/generic.py:8870: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  return self._update_inplace(result)\n",
      "../../src/utility.py:166: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cov[df_cov == \"\"] = \"nan\"\n",
      "/ocean/projects/asc170022p/juc91/anaconda/envs/IEA/lib/python3.7/site-packages/pandas/core/frame.py:3718: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self._where(-key, value, inplace=True)\n"
     ]
    }
   ],
   "source": [
    "df_IEA_train = pd.read_csv(\"../../output/IEA_train.csv\").set_index(\"sid\")\n",
    "df_IEA_test = pd.read_csv(\"../../output/IEA_test.csv\").set_index(\"sid\")\n",
    "\n",
    "df_data = utility.load_data()\n",
    "\n",
    "df_used = pd.merge(df_data, df_IEA_test, left_index = True, right_index = True)[ [\"IEA0\", \"IEA1\", \"KM_v2\"]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generating the box plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x396 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cluster_labels = [\"RRS\", \"UPE\", \"AP\", \"SE\"]\n",
    "\n",
    "df_used = df_used.rename(columns = {\"IEA0\":\"IEA$_{emph}$\", \"IEA1\":\"IEA$_{airway}$\", \"KM_v2\":\"KMeans Clusters\"})\n",
    "\n",
    "PROPS = {\n",
    "    'boxprops':{'facecolor':'none', 'edgecolor':'black'},\n",
    "    'medianprops':{'color':'black'},\n",
    "    'whiskerprops':{'color':'black'},\n",
    "}\n",
    "\n",
    "sns.set(font_scale = 1.5)\n",
    "sns.set_style('ticks', {'axes.edgecolor': '0',  \n",
    "                        'xtick.color': '0',\n",
    "                        'ytick.color': '0'})\n",
    "\n",
    "\n",
    "plt.figure(figsize = [15, 5.5])\n",
    "ax1 = plt.subplot(121)\n",
    "ax1 = sns.boxplot(x=\"KMeans Clusters\", y=\"IEA$_{emph}$\", data=df_used, **PROPS)\n",
    "ax1.set_xticklabels([\"RRS\", \"UPE\", \"AP\", \"SE\"])\n",
    "\n",
    "ax1.grid(False)\n",
    "ax1.set_facecolor((1, 1, 1))\n",
    "ax1.spines['bottom'].set_color('k')\n",
    "ax1.spines['top'].set_color('k') \n",
    "ax1.spines['right'].set_color('k')\n",
    "ax1.spines['left'].set_color('k')\n",
    "\n",
    "\n",
    "ax2 = plt.subplot(122)\n",
    "ax2 = sns.boxplot(x=\"KMeans Clusters\", y=\"IEA$_{airway}$\", data=df_used,  **PROPS)\n",
    "ax2.set_xticklabels(cluster_labels)\n",
    "\n",
    "ax2.grid(False)\n",
    "ax2.set_facecolor((1, 1, 1))\n",
    "ax2.spines['bottom'].set_color('k')\n",
    "ax2.spines['top'].set_color('k') \n",
    "ax2.spines['right'].set_color('k')\n",
    "ax2.spines['left'].set_color('k')\n",
    "\n",
    "\n",
    "_ = ax1.text(-.5, -5, \"RRS: Relatively resistant smokers; UPE: Mild upper zone-predominant emphysema;\", \n",
    "          fontsize = 18)\n",
    "_ = ax1.text(-.5, -5.5, \"AP: Airway-predominant; SE: Severe emphysema.\",\n",
    "          fontsize = 18)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# p-values with IEA$_{emph}$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "UPE vs RRS p-value:0.103\n",
      "AP vs RRS p-value:0.334\n",
      "AP vs UPE p-value:0.039\n",
      "SE vs RRS p-value:0.000\n",
      "SE vs UPE p-value:0.000\n",
      "SE vs AP p-value:0.000\n"
     ]
    }
   ],
   "source": [
    "for iii in range(1, 5):\n",
    "    for jjj in range(1, iii ):\n",
    "        \n",
    "        p_values = kruskal( \n",
    "                df_used.loc[df_used[\"KMeans Clusters\"] == iii, \"IEA$_{emph}$\"], \n",
    "                 df_used.loc[df_used[\"KMeans Clusters\"] == jjj, \"IEA$_{emph}$\"]\n",
    "               ).pvalue\n",
    "        \n",
    "        print( \"{} vs {} p-value:{:.3f}\".format(cluster_labels[iii - 1], cluster_labels[jjj - 1], p_values)\n",
    "        )\n",
    "        \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# p-values with IEA$_{airway}$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "UPE vs RRS p-value:0.065\n",
      "AP vs RRS p-value:0.002\n",
      "AP vs UPE p-value:0.348\n",
      "SE vs RRS p-value:0.252\n",
      "SE vs UPE p-value:0.020\n",
      "SE vs AP p-value:0.002\n"
     ]
    }
   ],
   "source": [
    "for iii in range(1, 5):\n",
    "    for jjj in range(1, iii ):\n",
    "        \n",
    "        p_values = kruskal( \n",
    "                df_used.loc[df_used[\"KMeans Clusters\"] == iii, \"IEA$_{airway}$\"], \n",
    "                 df_used.loc[df_used[\"KMeans Clusters\"] == jjj, \"IEA$_{airway}$\"]\n",
    "               ).pvalue\n",
    "        \n",
    "        print( \"{} vs {} p-value:{:.3f}\".format(cluster_labels[iii - 1], cluster_labels[jjj - 1], p_values)\n",
    "        )\n",
    "        \n"
   ]
  }
 ],
 "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": "3.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}