{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Use Case 8: Outliers"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"When looking at data, we often want to identify outliers, extremely high or low data points. In this use case we will show you how to use the Blacksheep package to find these in the CPTAC data. For more detailed information about the Blacksheep package see [this](https://github.com/ruggleslab/blackSheep/) repository.\n",
"\n",
"In the CPTAC breast cancer study ([here](https://www.nature.com/articles/nature18003)) it was shown that tumors classified as HER-2 enriched are frequently outliers for high abundance of ERBB2 phosphorylation, protein and mRNA (see [figure 4](https://www.nature.com/articles/nature18003/figures/4) of the manuscript). In this use case we will show that same phenomena in an independent cohort of breast cancer tumors, whose data are included in the cptac package."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 1: Importing packages and setting up your notebook\n",
"\n",
"Before we begin performing the analysis, we must import the packages we will be using. In this first code block, we import the standard set of data science packages.\n",
"\n",
"We will need an external package called blacksheep. To install it run the following on your command line:\n",
"```\n",
"pip install blksheep\n",
"```"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In this next code block we import the blacksheep and cptac packages and grab our proteomic and clinical data."
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" \r"
]
}
],
"source": [
"import blacksheep\n",
"import cptac\n",
"brca = cptac.Brca()\n",
"clinical = brca.get_clinical()\n",
"proteomics = brca.get_proteomics()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 2: Binarize Data\n",
"\n",
"The Blacksheep package requires that annotations are a binary variable. Our cptac tumors are divided into 4 subtypes: LumA, LumB, Basal, and Her2. We will use the binarize_annotations function to create a binary table of these PAM50 tumor classifications. We will call this table 'annotations'."
]
},
{
"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>PAM50_LumA</th>\n",
" <th>PAM50_Basal</th>\n",
" <th>PAM50_LumB</th>\n",
" <th>PAM50_Her2</th>\n",
" <th>PAM50_Normal-like</th>\n",
" </tr>\n",
" <tr>\n",
" <th>Patient_ID</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>CPT000814</th>\n",
" <td>not-LumA</td>\n",
" <td>Basal</td>\n",
" <td>not-LumB</td>\n",
" <td>not-Her2</td>\n",
" <td>not-Normal-like</td>\n",
" </tr>\n",
" <tr>\n",
" <th>CPT001846</th>\n",
" <td>not-LumA</td>\n",
" <td>Basal</td>\n",
" <td>not-LumB</td>\n",
" <td>not-Her2</td>\n",
" <td>not-Normal-like</td>\n",
" </tr>\n",
" <tr>\n",
" <th>X01BR001</th>\n",
" <td>not-LumA</td>\n",
" <td>Basal</td>\n",
" <td>not-LumB</td>\n",
" <td>not-Her2</td>\n",
" <td>not-Normal-like</td>\n",
" </tr>\n",
" <tr>\n",
" <th>X01BR008</th>\n",
" <td>not-LumA</td>\n",
" <td>Basal</td>\n",
" <td>not-LumB</td>\n",
" <td>not-Her2</td>\n",
" <td>not-Normal-like</td>\n",
" </tr>\n",
" <tr>\n",
" <th>X01BR009</th>\n",
" <td>not-LumA</td>\n",
" <td>Basal</td>\n",
" <td>not-LumB</td>\n",
" <td>not-Her2</td>\n",
" <td>not-Normal-like</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" PAM50_LumA PAM50_Basal PAM50_LumB PAM50_Her2 PAM50_Normal-like\n",
"Patient_ID \n",
"CPT000814 not-LumA Basal not-LumB not-Her2 not-Normal-like\n",
"CPT001846 not-LumA Basal not-LumB not-Her2 not-Normal-like\n",
"X01BR001 not-LumA Basal not-LumB not-Her2 not-Normal-like\n",
"X01BR008 not-LumA Basal not-LumB not-Her2 not-Normal-like\n",
"X01BR009 not-LumA Basal not-LumB not-Her2 not-Normal-like"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"annotations = clinical[['PAM50']].copy()\n",
"annotations = blacksheep.binarize_annotations(annotations)\n",
"annotations.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 3: Perform Outlier Analysis\n",
"\n",
"Now that our dataframes are correctly formatted, we will start looking for outliers.\n",
"\n",
"We will start by using the deva function found in the blacksheep package. This function takes the proteomics data frame (which we transpose to fit the requirements of the function), and the annotations data frame that includes the binarized columns. We also indicate that we want to look for up regulated genes, and that we do not want to aggregate the data. The function returns two things:\n",
"1. A data object with a dataframe which states whether a sample is an outlier for a specific protein. In the code block below we named this 'outliers'\n",
"2. A data object with a dataframe with the Q Values showing if a gene shows an enrichment in outliers for a specific subset of tumors as defined in annotations. In the code block below, we named this 'qvalues'."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"07/29/2021 16:11:25:WARNING:No rows tested for fisherFDR_PAM50_LumA_not-LumA\n",
"07/29/2021 16:11:25:WARNING:No rows tested for fisherFDR_PAM50_LumA_LumA\n",
"07/29/2021 16:11:25:WARNING:No rows tested for fisherFDR_PAM50_Basal_not-Basal\n",
"07/29/2021 16:11:26:WARNING:No rows tested for fisherFDR_PAM50_LumB_not-LumB\n",
"07/29/2021 16:11:26:WARNING:No rows tested for fisherFDR_PAM50_Her2_not-Her2\n",
"07/29/2021 16:11:28:WARNING:No rows tested for fisherFDR_PAM50_Normal-like_not-Normal-like\n"
]
}
],
"source": [
"outliers, qvalues = blacksheep.deva(proteomics.transpose(),\n",
" annotations,\n",
" up_or_down='up',\n",
" aggregate=False,\n",
" frac_filter=0.3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 4: Inspect Results"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Because these two tables that are returned are quite complex, we will now look at each of these individually.\n",
"\n",
"The outliers table indicates whether each sample is an outlier for a particular gene. In this use case, we will focus on ERBB2. The first line below simplifies the index for each row by dropping the database id and leaving the gene name. We also only print off a portion of the table for brevity."
]
},
{
"cell_type": "code",
"execution_count": 5,
"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>CPT000814_outliers</th>\n",
" <th>CPT001846_outliers</th>\n",
" <th>X01BR001_outliers</th>\n",
" <th>X01BR008_outliers</th>\n",
" <th>X01BR009_outliers</th>\n",
" <th>X01BR010_outliers</th>\n",
" <th>X01BR015_outliers</th>\n",
" <th>X01BR017_outliers</th>\n",
" </tr>\n",
" <tr>\n",
" <th>Name</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>ERBB2</th>\n",
" <td>0.0</td>\n",
" <td>0.0</td>\n",
" <td>0.0</td>\n",
" <td>0.0</td>\n",
" <td>0.0</td>\n",
" <td>0.0</td>\n",
" <td>0.0</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" CPT000814_outliers CPT001846_outliers X01BR001_outliers \\\n",
"Name \n",
"ERBB2 0.0 0.0 0.0 \n",
"\n",
" X01BR008_outliers X01BR009_outliers X01BR010_outliers \\\n",
"Name \n",
"ERBB2 0.0 0.0 0.0 \n",
"\n",
" X01BR015_outliers X01BR017_outliers \n",
"Name \n",
"ERBB2 0.0 1.0 "
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"outliers.df.index = outliers.df.index.droplevel('Database_ID')\n",
"erbb2_outliers = outliers.df[outliers.df.index.str.match('ERBB2')]\n",
"erbb2_outliers.iloc[:, :8]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In the chart above you can see that most of the samples have 0, indiciating that the sample is not an outlier for ERBB2 protein abundance. X01BR017, however, has a 1, indicating that particular sample is an outlier.\n",
"\n",
"The Outliers table contains boolean columns for both outlier and notOutliers. The notOutliers columns are redundant so we will remove them."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"erbb2_outliers = erbb2_outliers.loc[:,~erbb2_outliers.columns.str.endswith('_notOutliers')]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We can now complile a list of all the samples that were considered to be outliers."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"['X01BR017_outliers', 'X05BR026_outliers', 'X09BR004_outliers', 'X09BR005_outliers', 'X11BR004_outliers', 'X11BR010_outliers', 'X11BR011_outliers', 'X11BR028_outliers', 'X11BR030_outliers', 'X11BR038_outliers', 'X11BR060_outliers', 'X18BR009_outliers', 'X21BR001_outliers', 'X22BR005_outliers']\n"
]
}
],
"source": [
"outlier_list = erbb2_outliers.columns[erbb2_outliers.isin([1.0]).all()].tolist()\n",
"print(outlier_list)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 5: Visualizing Outliers\n",
"To understand what this means, we will plot the proteomics data for the ERBB2 gene and label the outlier samples. Before we graph the result we will join the proteomics and clinical data, isolating the PAM50 subtype and ERBB2."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"combined_data = brca.join_metadata_to_omics(metadata_df_name=\"clinical\", \n",
" omics_df_name=\"proteomics\", \n",
" metadata_cols=[\"PAM50\"],\n",
" omics_genes=['ERBB2'])\n",
"combined_data.columns = combined_data.columns.droplevel(\"Database_ID\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We will now create the graph."
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 800x800 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.figure(figsize=(8, 8))\n",
"sns.set_palette('colorblind')\n",
"ax = sns.boxplot(data=combined_data, showfliers=False, y='ERBB2_proteomics', color='lightgray')\n",
"left = False\n",
"# This for loop labels all the specific outlier data points.\n",
"for sample in outlier_list:\n",
" if left:\n",
" position = -0.08\n",
" left = False\n",
" else:\n",
" position = 0.01\n",
" left = True\n",
" sample = sample.split(\"_\")[0]\n",
" ax.annotate(sample, (position, combined_data.transpose()[sample].values[1]))\n",
"ax = sns.swarmplot(data=combined_data, y='ERBB2_proteomics')\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As you can see from this graph, the samples we labeled, which had a 1.0 in the outliers table were all located at the top of the graph, indicating they are very highly expressed."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 6: Looking at the Qvalue table\n",
"\n",
"Let's now take a look at the Qvalues table. Remember that the qvalues table indicates the probability that a gene shows an enrichment in outliers for categories defined in our annotation dataframe."
]
},
{
"cell_type": "code",
"execution_count": 10,
"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></th>\n",
" <th>fisherFDR_PAM50_Basal_Basal</th>\n",
" <th>fisherFDR_PAM50_LumB_LumB</th>\n",
" <th>fisherFDR_PAM50_Her2_Her2</th>\n",
" <th>fisherFDR_PAM50_Normal-like_Normal-like</th>\n",
" </tr>\n",
" <tr>\n",
" <th>Name</th>\n",
" <th>Database_ID</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>A2ML1</th>\n",
" <th>NP_653271.2|NP_001269353.1</th>\n",
" <td>1.441146e-07</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>ABCC11</th>\n",
" <th>NP_149163.2|NP_660187.1|NP_150229.2</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.001545</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>ABCC5</th>\n",
" <th>NP_005679.2|NP_001306961.1|NP_001018881.1</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.093281</td>\n",
" </tr>\n",
" <tr>\n",
" <th>ACACB</th>\n",
" <th>NP_001084.3</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.068994</td>\n",
" </tr>\n",
" <tr>\n",
" <th>ACAD8</th>\n",
" <th>NP_055199.1</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.093281</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" fisherFDR_PAM50_Basal_Basal \\\n",
"Name Database_ID \n",
"A2ML1 NP_653271.2|NP_001269353.1 1.441146e-07 \n",
"ABCC11 NP_149163.2|NP_660187.1|NP_150229.2 NaN \n",
"ABCC5 NP_005679.2|NP_001306961.1|NP_001018881.1 NaN \n",
"ACACB NP_001084.3 NaN \n",
"ACAD8 NP_055199.1 NaN \n",
"\n",
" fisherFDR_PAM50_LumB_LumB \\\n",
"Name Database_ID \n",
"A2ML1 NP_653271.2|NP_001269353.1 NaN \n",
"ABCC11 NP_149163.2|NP_660187.1|NP_150229.2 NaN \n",
"ABCC5 NP_005679.2|NP_001306961.1|NP_001018881.1 NaN \n",
"ACACB NP_001084.3 NaN \n",
"ACAD8 NP_055199.1 NaN \n",
"\n",
" fisherFDR_PAM50_Her2_Her2 \\\n",
"Name Database_ID \n",
"A2ML1 NP_653271.2|NP_001269353.1 NaN \n",
"ABCC11 NP_149163.2|NP_660187.1|NP_150229.2 0.001545 \n",
"ABCC5 NP_005679.2|NP_001306961.1|NP_001018881.1 NaN \n",
"ACACB NP_001084.3 NaN \n",
"ACAD8 NP_055199.1 NaN \n",
"\n",
" fisherFDR_PAM50_Normal-like_Normal-like \n",
"Name Database_ID \n",
"A2ML1 NP_653271.2|NP_001269353.1 NaN \n",
"ABCC11 NP_149163.2|NP_660187.1|NP_150229.2 NaN \n",
"ABCC5 NP_005679.2|NP_001306961.1|NP_001018881.1 0.093281 \n",
"ACACB NP_001084.3 0.068994 \n",
"ACAD8 NP_055199.1 0.093281 "
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"qvalues.df.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This table includes all the q-values. Before really analyzing the table we will want to remove any insignificant q-values. For our purposes we will remove any q-values that are greater than 0.05."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"for col in qvalues.df.columns:\n",
" qvalues.df.loc[qvalues.df[col] > 0.05, col] = np.nan"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We will now isolate the ERBB2 gene."
]
},
{
"cell_type": "code",
"execution_count": 12,
"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>Name</th>\n",
" <th>fisherFDR_PAM50_Basal_Basal</th>\n",
" <th>fisherFDR_PAM50_LumB_LumB</th>\n",
" <th>fisherFDR_PAM50_Her2_Her2</th>\n",
" <th>fisherFDR_PAM50_Normal-like_Normal-like</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>ERBB2</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.000366</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Name fisherFDR_PAM50_Basal_Basal fisherFDR_PAM50_LumB_LumB \\\n",
"0 ERBB2 NaN NaN \n",
"\n",
" fisherFDR_PAM50_Her2_Her2 fisherFDR_PAM50_Normal-like_Normal-like \n",
"0 0.000366 NaN "
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"qvalues.df.index = qvalues.df.index.droplevel('Database_ID')\n",
"qvalues = qvalues.df[qvalues.df.index.str.match('ERBB2')]\n",
"erbb2_qvalues = qvalues.reset_index()['Name'] == 'ERBB2'\n",
"qvalues = qvalues.reset_index()[erbb2_qvalues]\n",
"qvalues.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Here we see that the only PAM50 subtype that has a significant enrichment is the Her2, which is exactly what is to be expected. To visualize this pattern, we will create a graph similiar to the one above, but with each of the categories in the PAM50 category differentially colored."
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 800x800 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.figure(figsize=(8, 8))\n",
"sns.set_palette('colorblind')\n",
"cols = {'Basal': 0, 'Her2':1, 'LumA':2, 'LumB':3, 'Normal-like':4}\n",
"ax = sns.boxplot(data=combined_data, y='ERBB2_proteomics', x='PAM50', color='lightgray')\n",
"ax = sns.swarmplot(data=combined_data, y='ERBB2_proteomics',x='PAM50', hue='PAM50')\n",
"for sample in outlier_list:\n",
" sample = sample.split(\"_\")[0]\n",
" ax.annotate(sample, (cols[combined_data.transpose()[sample].values[0]], combined_data.transpose()[sample].values[1]))\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Looking at the distribution of the graph you can see that distribution of the Her2 category is much different than the distributions of the other catgeories. The median of the proteomic data in the Her2 category is much higher than other categories, with many more data points in the upper portion of the graph."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Additional Applications"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We have just walked through one example of how you might use the Outlier Analysis. Using this same approach, you can run the outlier analysis on a number of different clinical attributes, cohorts, and omics data. For example, you may look for outliers within the transcriptomics of the Endometrial cancer type using the clinical attribute of Histological_type. You can also look at more than one clinical attribute at a time by appending more attributes to your annotations table, or you can look for downregulated omics by chaning the 'up_or_down' variable of the run_outliers function."
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"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": "3.9.5"
}
},
"nbformat": 4,
"nbformat_minor": 4
}