0.1_dataPrep_dHCP.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Neonatal brain measure estimation from dHCP data\n",
    "As dHCP is under data protection, no data is included in this repository. The code is provided as a reference for the analysis of the dHCP data. To reproduce the analyses, you need to apply for access to the dHCP dataset [here](https://nda.nih.gov/edit_collection.html?id=3955), and parcellate the data with the [MCRIBS atlas](https://github.com/DevelopmentalImagingMCRI/MCRIBS). Change the directory `mcribs_data_dir` accordingly. Data can be requested from the authors among reasonable request.\n",
    "\n",
    "I received parcellated data from my collaborators C. Adamson, G. Ball, and J. Seidlitz.\n",
    "\n",
    "This whole notebook should be run for cortical thickness (CT) or surface area (SA). Both measures were used in the original publication. Adjust the variable `brain_measure` to `CT` or `SA` accordingly. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load MRI related data\n",
    "## Rearrange individual files\n",
    "The output generated by MCRIBS is similar to FreeSurfer's ?.aparc.annot files. First, I have rearranged them as aparcstats2table would do (i.e., generate one file containing all subjects' brain_measure information).\n",
    "For measures of cerebral tissue volume, I have actually used asegstats2table, a FreeSurfer command. If you want to do so too, please make sure that FreeSurfer is installed on your machine and that the environment variable `FREESURFER_HOME` is correctly set (see [here](https://surfer.nmr.mgh.harvard.edu/fswiki/DownloadAndInstall)). Otherwise, just load the combined `aseg` file below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from os.path import join\n",
    "import glob\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import subprocess\n",
    "from abagen import fetch_desikan_killiany\n",
    "\n",
    "import sys\n",
    "sys.path.append('code')\n",
    "from preprocessing import dhcp_aparcstats2table, strip_dhcp_id, clean_line, read_dhcp_18months, reorder_vars\n",
    "\n",
    "\n",
    "mcribs_data_dir = 'data/dHCP/dhcp-MCRIBS'  # where MCRIBS parcellated data are saved\n",
    "\n",
    "dhcp_data_dir = 'data/dHCP'\n",
    "dhcp_freesurfer_outputs = join(dhcp_data_dir, 'freesurfer')  # where all freesurfer outputs are stored\n",
    "dhcp_out = join(dhcp_data_dir, 'derivatives')  # where the adapted data will be stored\n",
    "os.makedirs(dhcp_out, exist_ok=True)\n",
    "\n",
    "\n",
    "# adjust brain measurement for which the code should be run\n",
    "brain_measure = 'CT'  # CT or SA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# lh\n",
    "csv_files = glob.glob(mcribs_data_dir + \"/*lh.aparc.stats\")  # find all files with left hemisphere data\n",
    "aparc_lh = dhcp_aparcstats2table(csv_files, brain_measure)\n",
    "\n",
    "# rh\n",
    "csv_files = glob.glob(mcribs_data_dir + \"/*rh.aparc.stats\")\n",
    "aparc_rh = dhcp_aparcstats2table(csv_files, brain_measure)\n",
    "\n",
    "mri_data = pd.concat([aparc_lh, aparc_rh], axis=1)\n",
    "mri_data.index.name = 'participant'\n",
    "mri_data.columns.name = ''\n",
    "\n",
    "# separate session and participant id\n",
    "mri_data = strip_dhcp_id(mri_data)\n",
    "\n",
    "# get region names of the Desikan-Killiany atlas as now used in the dHCP data\n",
    "dhcp_idps_idx = mri_data.filter(regex='^ctx-').columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of IDPs: 68\n"
     ]
    }
   ],
   "source": [
    "# access DK atlas and save idp_labels\n",
    "atlas = fetch_desikan_killiany(surface=True)\n",
    "atlas = pd.read_csv(atlas['info'])\n",
    "atlas = atlas[(atlas['structure'] == 'cortex') & (atlas['hemisphere'] == 'L')]\n",
    "atlas_labels_l = ['L_'+label for label in atlas['label']]\n",
    "atlas_labels_r = ['R_'+label for label in atlas['label']]\n",
    "\n",
    "desikan_idps = atlas_labels_l + atlas_labels_r\n",
    "print(f\"Number of IDPs: {len(dhcp_idps_idx)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "cp: data/dHCP/dhcp-MCRIBS/sub-CC00597XX21-ses-190200_aseg.stats: No such file or directory\n",
      "cp: data/dHCP/dhcp-MCRIBS/sub-CC00649XX23-ses-191201_aseg.stats: No such file or directory\n",
      "./code/dhcp_asegstats2table.sh: line 35: 40378 Killed: 9               asegstats2table --subjectsfile=${MCRIBS_DIR}/subjectlist.txt -t ${DATA_DIR}/dhcp_aseg_mcribs.txt --skip --all-segs\n"
     ]
    }
   ],
   "source": [
    "# employ freesurfer's asegstats2table to combine CTV for all subjects into one csv file\n",
    "subprocess.run([\n",
    "    \"./code/dhcp_asegstats2table.sh\",\n",
    "    mcribs_data_dir,\n",
    "    dhcp_data_dir\n",
    "], check=True)\n",
    "\n",
    "aseg = pd.read_csv(join(dhcp_data_dir, 'dhcp_aseg_mcribs.txt'), sep='\\t', index_col=0)\n",
    "aseg = aseg[['SubCortGrayVol', 'CerebralWhiteMatterVol', 'CortexVol']]\n",
    "# separate session and participant id\n",
    "aseg = strip_dhcp_id(aseg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# meta\n",
    "meta = pd.read_csv(os.path.join(dhcp_data_dir, 'dHCP_add_info_data_release2.csv'), sep=';')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Adjust and rename variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# merge \n",
    "dhcp = mri_data.merge(aseg, on=['Subject_ID', 'Session_ID'])\n",
    "# dhcp = dhcp.merge(meta, on=['Subject_ID', 'Session_ID'])\n",
    "\n",
    "# rename variables\n",
    "dhcp = dhcp.rename(columns=dict(\n",
    "    SubCortGrayVol=\"sGMV\",\n",
    "    CerebralWhiteMatterVol=\"WMV\",\n",
    "    CortexVol=\"GMV\"\n",
    "))  \n",
    "dhcp = dhcp.rename(columns=dict(zip(dhcp_idps_idx, desikan_idps)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Filter for term-equivalent scans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "92 preterm subjects have a scan at term-equivalent age.\n",
      "Number of duplicates: 0\n",
      "377 term subjects are selected.\n",
      "Final number of subjects should be: 469.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/tk/3rqd03yd6rqb4_przznzwzjc0000gn/T/ipykernel_39204/3027250023.py:3: 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",
      "  preterm_scan_2['dx'] = 'preterm'\n",
      "/var/folders/tk/3rqd03yd6rqb4_przznzwzjc0000gn/T/ipykernel_39204/3027250023.py:11: 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",
      "  CN_scan_1['dx'] = 'CN'\n"
     ]
    }
   ],
   "source": [
    "# select the preterm subjects with two scans (i.e., one at birth, one at term-equivalent age)\n",
    "preterm_scan_2 = meta[(meta['birth_age'] < 37) & (meta['scan_age'] >= 37)]\n",
    "preterm_scan_2['dx'] = 'preterm'\n",
    "print(\"{0} preterm subjects have a scan at term-equivalent age.\".format(preterm_scan_2.shape[0]))\n",
    "\n",
    "# make sure that there are no subjects with 2 term-equivalent scans\n",
    "print(\"Number of duplicates: {0}\".format(preterm_scan_2.Subject_ID.duplicated().sum()))\n",
    "\n",
    "# select first scan for term subjects\n",
    "CN_scan_1 = meta[(meta['birth_age'] >= 37) & (meta['scan_number'] == 1)]\n",
    "CN_scan_1['dx'] = 'CN'\n",
    "print(\"{0} term subjects are selected.\".format(CN_scan_1.shape[0]))\n",
    "\n",
    "# combine the two\n",
    "filtered_meta = pd.concat([preterm_scan_2, CN_scan_1])\n",
    "filtered_meta.head()\n",
    "print(\"Final number of subjects should be: {0}.\".format(filtered_meta.shape[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataframe shape: (467, 82)\n"
     ]
    }
   ],
   "source": [
    "# combine all information\n",
    "dhcp = dhcp.merge(filtered_meta, on=['Subject_ID', 'Session_ID'])\n",
    "print('Dataframe shape:', dhcp.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# reorder variables\n",
    "dhcp = reorder_vars(['Subject_ID', 'Session_ID', 'dx','gender',\n",
    "        'birth_age', 'birth_weight', 'singleton', 'scan_age', 'scan_number',\n",
    "        'radiology_score', 'sedation'], dhcp, desikan_idps)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Average hemispheres"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "regions = [r[2:] for r in desikan_idps[:34]]\n",
    "\n",
    "# Combine L_ and R_ values\n",
    "for region in regions:\n",
    "    dhcp[f'{brain_measure}_{region}'] = dhcp[[f'L_{region}', f'R_{region}']].mean(axis=1)\n",
    "    # drop L_ and R_ columns\n",
    "    dhcp = dhcp.drop(columns=[f'L_{region}', f'R_{region}'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create region list bilateral\n",
    "desikan_idps_bilateral = [f'{brain_measure}_{region}' for region in regions]\n",
    "ctv_columns = ['GMV', 'WMV', 'sGMV']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Adapt df for BrainChart framework\n",
    "BrainChart needs a certain format of the data. We will adapt the data accordingly. More information can be found [here](https://brainchart.shinyapps.io/brainchart/)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(467, 51)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Index(['participant', 'Age', 'age_days', 'sex', 'study', 'fs_version',\n",
       "       'country', 'run', 'session', 'dx', 'Session_ID', 'GA', 'BW',\n",
       "       'CT_bankssts', 'CT_caudalanteriorcingulate', 'CT_caudalmiddlefrontal',\n",
       "       'CT_cuneus', 'CT_entorhinal', 'CT_fusiform', 'CT_inferiorparietal',\n",
       "       'CT_inferiortemporal', 'CT_isthmuscingulate', 'CT_lateraloccipital',\n",
       "       'CT_lateralorbitofrontal', 'CT_lingual', 'CT_medialorbitofrontal',\n",
       "       'CT_middletemporal', 'CT_parahippocampal', 'CT_paracentral',\n",
       "       'CT_parsopercularis', 'CT_parsorbitalis', 'CT_parstriangularis',\n",
       "       'CT_pericalcarine', 'CT_postcentral', 'CT_posteriorcingulate',\n",
       "       'CT_precentral', 'CT_precuneus', 'CT_rostralanteriorcingulate',\n",
       "       'CT_rostralmiddlefrontal', 'CT_superiorfrontal', 'CT_superiorparietal',\n",
       "       'CT_superiortemporal', 'CT_supramarginal', 'CT_frontalpole',\n",
       "       'CT_temporalpole', 'CT_transversetemporal', 'CT_insula', 'sGMV', 'GMV',\n",
       "       'WMV', 'sGMV'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# adapt for brainchart\n",
    "dhcp = dhcp.rename(columns=dict(\n",
    "    Subject_ID=\"participant\",\n",
    "    gender=\"sex\",\n",
    "    birth_age=\"GA\",\n",
    "    birth_weight=\"BW\"\n",
    ")) \n",
    "\n",
    "# age\n",
    "dhcp['age_days'] = dhcp['scan_age'] * 7  # because scan_age is in weeks\n",
    "dhcp['Age'] = (dhcp['age_days']-280) / 365.245  # from BrainChart instructions\n",
    "\n",
    "# other required info\n",
    "dhcp['study'] = 'dHCP_NewEstimation'  # as dHCP was used for original model fitting, we want to make sure that random effects of study are calculated again\n",
    "dhcp['fs_version'] = 'Custom' \n",
    "dhcp['country'] = 'Multisite'\n",
    "dhcp['run'] = 1\n",
    "dhcp['session'] = 1\n",
    "\n",
    "# reshape\n",
    "all_idps = desikan_idps_bilateral + ctv_columns\n",
    "dhcp_final = reorder_vars(['participant', 'Age', 'age_days', 'sex', 'study', 'fs_version','country', 'run', \n",
    "                            'session', 'dx'], dhcp, all_idps)\n",
    "dhcp_final.drop(columns=['scan_age', 'singleton', 'scan_number', 'radiology_score', 'sedation'], inplace=True)\n",
    "\n",
    "# save\n",
    "print(dhcp_final.shape)\n",
    "dhcp_final.to_csv(os.path.join(dhcp_out, f'dhcp_{brain_measure}_preprocessed.csv'), index=False)\n",
    "dhcp_final.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Stats\n",
    "Summary stats shown in Supp Table S1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Overall number of unique subjects with longitudinal data: n = 467\n",
      "Preterm stats: n =  92\n",
      "Male      55\n",
      "Female    37\n",
      "Name: sex, dtype: int64\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>age_weeks</th>\n",
       "      <th>GA</th>\n",
       "      <th>BW</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>92.00</td>\n",
       "      <td>92.00</td>\n",
       "      <td>92.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>40.85</td>\n",
       "      <td>32.21</td>\n",
       "      <td>1803.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.06</td>\n",
       "      <td>3.55</td>\n",
       "      <td>771.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>37.00</td>\n",
       "      <td>24.29</td>\n",
       "      <td>540.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>39.71</td>\n",
       "      <td>29.32</td>\n",
       "      <td>1185.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>40.86</td>\n",
       "      <td>32.43</td>\n",
       "      <td>1715.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>42.00</td>\n",
       "      <td>35.46</td>\n",
       "      <td>2400.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>45.14</td>\n",
       "      <td>36.86</td>\n",
       "      <td>4100.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       age_weeks     GA       BW\n",
       "count      92.00  92.00    92.00\n",
       "mean       40.85  32.21  1803.64\n",
       "std         2.06   3.55   771.28\n",
       "min        37.00  24.29   540.00\n",
       "25%        39.71  29.32  1185.00\n",
       "50%        40.86  32.43  1715.00\n",
       "75%        42.00  35.46  2400.00\n",
       "max        45.14  36.86  4100.00"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"Overall number of unique subjects with longitudinal data: n =\", len(dhcp_final))\n",
    "dhcp_final['age_weeks'] = dhcp_final['age_days'] / 7\n",
    "dhcp_final['BW'] = dhcp_final['BW'] * 1000  # convert to g\n",
    "\n",
    "dhcp_final_pt = dhcp_final[dhcp_final['dx'] == 'preterm']\n",
    "dhcp_final_cn = dhcp_final[dhcp_final['dx'] == 'CN']\n",
    "\n",
    "print('Preterm stats: n = ', len(dhcp_final_pt))\n",
    "print(dhcp_final_pt['sex'].value_counts())\n",
    "display(dhcp_final_pt[['age_weeks', 'GA', 'BW']].describe().round(2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Full-term stats: n =  375\n",
      "Male      204\n",
      "Female    171\n",
      "Name: sex, dtype: int64\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>age_weeks</th>\n",
       "      <th>GA</th>\n",
       "      <th>BW</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>375.00</td>\n",
       "      <td>375.00</td>\n",
       "      <td>375.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>41.08</td>\n",
       "      <td>39.96</td>\n",
       "      <td>3362.71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.72</td>\n",
       "      <td>1.23</td>\n",
       "      <td>548.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>37.43</td>\n",
       "      <td>37.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>39.86</td>\n",
       "      <td>39.14</td>\n",
       "      <td>3000.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>41.00</td>\n",
       "      <td>40.14</td>\n",
       "      <td>3400.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>42.14</td>\n",
       "      <td>40.86</td>\n",
       "      <td>3730.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>44.86</td>\n",
       "      <td>42.29</td>\n",
       "      <td>4800.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       age_weeks      GA       BW\n",
       "count     375.00  375.00   375.00\n",
       "mean       41.08   39.96  3362.71\n",
       "std         1.72    1.23   548.86\n",
       "min        37.43   37.00     0.00\n",
       "25%        39.86   39.14  3000.00\n",
       "50%        41.00   40.14  3400.00\n",
       "75%        42.14   40.86  3730.00\n",
       "max        44.86   42.29  4800.00"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Full-term stats: n = ', len(dhcp_final_cn))\n",
    "print(dhcp_final_cn['sex'].value_counts())\n",
    "display(dhcp_final_cn[['age_weeks', 'GA', 'BW']].describe().round(2))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}