CMA_score.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CMA score\n",
    "Mathieu Bourdenx - Oct. 2020"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "+t vc \n",
    "import pandas as pd\n",
    "from tqdm import tqdm\n",
    "\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.rc('xtick', labelsize=14)\n",
    "plt.rc('ytick', labelsize=14)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:17: FutureWarning: pandas.core.index is deprecated and will be removed in a future version.  The public classes are available in the top-level namespace.\n",
      "  from pandas.core.index import RangeIndex\n"
     ]
    }
   ],
   "source": [
    "import scanpy as sc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load cluster ID (from Seurat)\n",
    "\n",
    "clusters = pd.read_excel('../../cluster_identity.xlsx')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load datasets (very very long - 1hr+)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-4-89961b0ed0dd>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m#Load full dataset\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0madata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msc\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_loom\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'../../cx_rnaAssay.loom'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/opt/anaconda3/lib/python3.7/site-packages/anndata/readwrite/read.py\u001b[0m in \u001b[0;36mread_loom\u001b[0;34m(filename, sparse, cleanup, X_name, obs_names, var_names, dtype, **kwargs)\u001b[0m\n\u001b[1;32m    158\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    159\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mX_name\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mlc\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayers\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkeys\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mX_name\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m''\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 160\u001b[0;31m         \u001b[0mX\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlc\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayers\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mX_name\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msparse\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mT\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtocsr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0msparse\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mlc\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayers\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mX_name\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mT\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    161\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    162\u001b[0m         \u001b[0mlayers\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mOrderedDict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda3/lib/python3.7/site-packages/loompy/loom_layer.py\u001b[0m in \u001b[0;36msparse\u001b[0;34m(self, rows, cols)\u001b[0m\n\u001b[1;32m    116\u001b[0m                 \u001b[0mcol\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mList\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndarray\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    117\u001b[0m                 \u001b[0mi\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 118\u001b[0;31m                 \u001b[0;32mfor\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mix\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mselection\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mview\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mds\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscan\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mitems\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcols\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlayers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwhat\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"layers\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    119\u001b[0m                         \u001b[0;32mif\u001b[0m \u001b[0mrows\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    120\u001b[0m                                 \u001b[0mvals\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mview\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayers\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mrows\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda3/lib/python3.7/site-packages/loompy/loompy.py\u001b[0m in \u001b[0;36mscan\u001b[0;34m(self, items, axis, layers, key, batch_size, what)\u001b[0m\n\u001b[1;32m    610\u001b[0m                                 \u001b[0;31m# Load the whole chunk from the file, then extract genes and cells using fancy indexing\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    611\u001b[0m                                 \u001b[0;32mfor\u001b[0m \u001b[0mlayer\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mlayers\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 612\u001b[0;31m                                         \u001b[0mtemp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayers\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlayer\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mix\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mix\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mcols_per_chunk\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    613\u001b[0m                                         \u001b[0mtemp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtemp\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mordering\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    614\u001b[0m                                         \u001b[0;32mif\u001b[0m \u001b[0mselection\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda3/lib/python3.7/site-packages/loompy/loom_layer.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, slice)\u001b[0m\n\u001b[1;32m     86\u001b[0m         \u001b[0;32mdef\u001b[0m \u001b[0m__getitem__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mslice\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mTuple\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mint\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mslice\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mint\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mslice\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndarray\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     87\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 88\u001b[0;31m                         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mds\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_file\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'/matrix'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__getitem__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mslice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     89\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mds\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_file\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'/layers/'\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__getitem__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mslice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     90\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32mh5py/_objects.pyx\u001b[0m in \u001b[0;36mh5py._objects.with_phil.wrapper\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;32mh5py/_objects.pyx\u001b[0m in \u001b[0;36mh5py._objects.with_phil.wrapper\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda3/lib/python3.7/site-packages/h5py/_hl/dataset.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, args)\u001b[0m\n\u001b[1;32m    571\u001b[0m         \u001b[0mmspace\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mh5s\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcreate_simple\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmshape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    572\u001b[0m         \u001b[0mfspace\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mselection\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mid\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 573\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mid\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmspace\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfspace\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0marr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdxpl\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dxpl\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    574\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    575\u001b[0m         \u001b[0;31m# Patch up the output for NumPy\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "#Load full dataset\n",
    "adata = sc.read_loom('../../cx_rnaAssay.loom')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load small dataset containing cell metadata\n",
    "small_adata = sc.read_loom('./cx_integratedAssay.loom')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Copy metadata \n",
    "adata.obs = small_adata.obs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a barcode dataframe\n",
    "\n",
    "barcode = small_adata.obs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n"
     ]
    }
   ],
   "source": [
    "# Normalize counts as CPM\n",
    "sc.pp.normalize_per_cell(adata, counts_per_cell_after=1e6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Log transform data\n",
    "sc.pp.log1p(adata)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## CMA score calculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load matrix file with weight and direction \n",
    "model_matrix = pd.read_excel('./activation_model.xlsx')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "cma_network = adata[:, model_matrix['Gene name Ms']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 18/18 [00:00<00:00, 348.61it/s]\n"
     ]
    }
   ],
   "source": [
    "cma_data_zs = cma_network.copy().X.todense().T\n",
    "\n",
    "for i in tqdm(np.arange(cma_data_zs.shape[0])):\n",
    "    \n",
    "    µ = np.mean(cma_data_zs[i, :])\n",
    "    sd = np.std(cma_data_zs[i, :])\n",
    "    \n",
    "    cma_data_zs[i, :] = (cma_data_zs[i, :] - µ)/sd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "142057it [12:04, 196.08it/s]\n"
     ]
    }
   ],
   "source": [
    "for i,j in tqdm(enumerate(barcode.index)):\n",
    "    \n",
    "    cell_matrix = model_matrix.copy()\n",
    "        \n",
    "    for g in cell_matrix.index:\n",
    "        \n",
    "        cell_matrix.loc[g, 'gene_count'] = cma_data_zs[g, i]\n",
    "    \n",
    "    cell_matrix['gene_score'] = cell_matrix['gene_count'] * cell_matrix['Direction'] * cell_matrix['Weight']\n",
    "    score = cell_matrix['gene_score'].sum()/np.sum(cell_matrix['Weight'])\n",
    "    \n",
    "    barcode.loc[j, 'score'] = score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "for barplotin tqdm(barcode.index):\n",
    "    \n",
    "    barcode.loc[i, 'broad.cell.type'] = clusters.loc[int(barcode.loc[i, 'seurat_clusters']), 'broad.cell.type']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fbd76a47f50>"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 6))\n",
    "\n",
    "sns.barplot(data=barcode, x=\"broad.cell.type\", y='score', hue='Condition')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/10 [00:00<?, ?it/s]/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:3: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n",
      "100%|██████████| 10/10 [00:47<00:00,  4.74s/it]\n"
     ]
    }
   ],
   "source": [
    "# Calculation of net score to WT 2m\n",
    "\n",
    "for maj_cell in tqdm(np.unique(barcode['broad.cell.type'])):\n",
    "    \n",
    "    µ = np.mean(barcode[barcode['broad.cell.type'] == maj_cell][barcode['Condition'] == 'CX_WT_2m']['score'])\n",
    "    \n",
    "    for cell_index in barcode[barcode['broad.cell.type'] == maj_cell].index:\n",
    "    \n",
    "        barcode.loc[cell_index, 'net_score_group'] = barcode.loc[cell_index, 'score'] - µ  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 142057/142057 [03:51<00:00, 614.96it/s]\n"
     ]
    }
   ],
   "source": [
    "# Create a new age category to align 6 and 8m\n",
    "\n",
    "for i in tqdm(barcode.index):\n",
    "    \n",
    "    if barcode.loc[i, 'Age'] == '2m': \n",
    "        barcode.loc[i, 'new_age'] = '2m'\n",
    "        \n",
    "    elif barcode.loc[i, 'Age'] == '6m': \n",
    "        barcode.loc[i, 'new_age'] = '8m'\n",
    "        \n",
    "    elif barcode.loc[i, 'Age'] == '8m': \n",
    "        barcode.loc[i, 'new_age'] = '8m'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOAAAADgCAYAAAAaLWrhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO2deXhV1bn/P28SQkAIKIiCYawis0ACiKgISnEWZbBaFbE4a4veotfh/qrX2v6sgmJLUbkWFSsoiF5wHrEqVGVQJKCiEKagEGQeM7z3j7UTTg7n5OwkZ0hO3s/z7Oecvaa99km+e639rrXeJaqKYRiJISXRFTCMuowJ0DASiAnQMBKICdAwEogJ0DASiAnQMBKICTCOnH322QrYUfeOsJgA40hBQUGiq2DUMEyAhpFATICGkUBMgIaRQNISXQGj9lJYWMiGDRvYv39/oqtSI8jIyCArK4t69er5zmMCNKrMhg0baNy4Me3atUNEEl2dhKKqbN26lQ0bNtC+fXvf+awLalSZ/fv306xZszovPgARoVmzZpXuDZgAjWph4jtEVX4LE6ARN3766Scuv/xyOnToQHZ2Nv379+eVV16J6jUee+wx9u7dG9UyY4kJ0IgLqsqwYcM4/fTTWb16NYsXL2bmzJls2LAhqtcxARpGCD744APS09O54YYbysLatm3LrbfeSnFxMePHj6dPnz706NGDJ598EoD58+dzxhlnMGLECDp16sSvf/1rSj04vP/++/Tq1Yvu3btzzTXXcODAAR5//HHy8/MZNGgQgwYN4umnn+a2224ru97UqVO5/fbbycvLo1OnTowePZoePXowYsSIMtEuXryYgQMHkp2dzdChQ9m0aVNsfxhVtSNOR3Z2tiYTK1as8J120qRJOm7cuJBxTz75pD7wwAOqqrp//37Nzs7W1atX64cffqiZmZm6fv16LS4u1pNPPlk//vhj3bdvn2ZlZem3336rqqpXXnmlPvroo6qq2rZtW92yZYuqqu7evVs7dOigBw8eVFXV/v3767Jly3TNmjUK6CeffKKqqmPGjNGHH35YDx48qP3799fNmzerqurMmTN1zJgx0fhNwv5P2DCEkRBuvvlmPvnkE9LT02nbti3Lli1j9uzZAOzYsYNVq1aRnp5O3759ycrKAqBnz57k5eXRuHFj2rdvT8eOHQEYPXo0kydPZty4ceWuccQRRzB48GBee+01OnfuTGFhId27dycvL4/WrVszYMAAAK644goef/xxzj77bJYvX86QIUMAKC4upmXLljH9HUyARlzo2rUrL7/8ctn55MmTKSgoICcnhzZt2vDXv/6VoUOHlsszf/586tevX3aemppKUVFRWTfUD2PHjuVPf/oTnTp1YsyYMWXhwRZLEUFV6dq1KwsXLqzs7VUZewc04sLgwYPZv38/U6ZMKQsrfe8aOnQoU6ZMobCwEIDvvvuOPXv2hC2rU6dO5OXl8f333wMwffp0Bg4cCEDjxo3ZtWtXWdp+/fqxfv16XnjhBS677LKy8HXr1pUJbcaMGZx66qmceOKJbNmypSy8sLCQ3NzcaNx+WEyARlwQEV599VU++ugj2rdvT9++fRk9ejQPPfQQY8eOpUuXLvTu3Ztu3bpx/fXXU1RUFLasjIwMpk2bxsiRI+nevTspKSllxp3rrruOc845h0GDBpWlHzVqFAMGDODII48sC+vcuTPPPvssPXr04Oeff+bGG28kPT2d2bNnc+edd3LSSSfRs2dPFixYELsfBcwIE8+jLhthEsl5552n7733Xtn5mjVrtGvXrjG5VmWNMNYCGknL9u3b6dixIw0aNODMM89MdHVCYkYYI2lp2rQp33333WHh7dq1Y/ny5Qmo0eEkfQsoIjeJyBoR2S8ii0XktAjpB3rp9ovIahG5IUSaliLyrIhs8dKtEJGBsbsLI1lJagGKyKXAJOBPQC9gAfCmiLQJk7498IaXrhfwZ+CvIjI8IE1T4FNAgPOAzsCtwObY3YmRrCR7F/R24BlVneqd3yoiZwM3AneFSH8DkK+qt3rnK0WkH/B7oHQQ6w5gk6peFZBvTfSrbtQFkrYFFJF0IBt4JyjqHeCUMNn6h0j/NpAjIqXLnIcBn4nIiyKyWUS+FJFbxNblGFUgaQUINAdSgZ+Cwn8Cjg2T59gw6dO88gA6ADcBq4GhuC7u/wduDlWgiFwnIotEZNGWLVsqew9GBG677TYee+yxsvOhQ4cyduzYsvPhw4eTmZlJz549Oeqoo2jfvj09e/bkrLPOSkR1DyPZu6BwuGNUCREWKX1geAqwSFVLu7BLReQEnAD/dlhhqk8BTwHk5OT4n0OVZBQVlzBnyUZeXLSeTdv30bJpAy7Nac3w7CxSU6reeTjllFOYNWsW48aNo6SkhIKCAnbu3FkWn5+fz7vvvku/fv24+uqrOf/88xkxYkQ0bikqJLMAC4BiDm/tWnB4K1fKj2HSFwFbvfNNwIqgNCuB31W5pklOUXEJt7ywlLdyfywLy9+xn8Vrt/HBN5v52+W9SEutWmdswIABZUuOcnNz6datG5s2bWLbtm00bNiQlStX0qtXr6jcRyxI2i6oqh4EFgNDgqKG4KycoVgIBPdNhuBavELv/FPgxKA0HYG1Va9tcjNnycZy4gvkrdwfmbN0Y5XLbtWqFWlpaaxbt44FCxbQv39/+vXrx8KFC1m0aBE9evQgPT29yuXHmqQVoMdE4GoRGSsinUVkEtAKeAJARJ4TkecC0j8BZInIY176scDVwCMBaR4FThaRe0TkeBEZCfwWmByPG6qNvLhofYXxL31RcXwkBgwYwIIFC8oE2L9//7LzU04JZ2+rGSRzFxRVfVFEmgH3Ai2B5cC5qlraWrUJSr9GRM7FiexGIB/4raq+HJDmCxEZhhtb/C9gnff591jfT21l0/Z9FcbnR4iPxCmnnMKCBQv4+uuv6datG61bt2bChAlkZmZyzTXXVKvsWJPUAgRQ1b8TRhyqekaIsI+A3hHKfB14PRr1qwu0bNqA/B3h3fW1atqgWuUPGDCACRMm0KFDB1JTUznqqKPYvn07ubm5TJ06NXIBCSTZu6BGDeDSnNYVxo/qU3F8JLp3705BQQEnn3xyubAmTZrQvHnzCnImnri2gCJSH/cO1gDYoqo2MFYHGJ6dxQffbA5piDm767EM751VrfJTU1PLDT0APPPMM4elCxWWaGIuQBFpDFwBXAb0BerhjcWJSD7wFvCUqn4R67oYiSE1Rfjb5b2Ys3QjL32xnvzt+2jVtAGj+rRmeO/qjQPWdmIqQBG5DWcAWQ3MBR7EGTb2AUcB3YDTgHdF5N/Araq6KpZ1MhJDWmoKo3JaMypCd7SuEesW8BRgoKqGW3z1OfAPb8nPb4CBgAnQqDPEVICqOtJnugOYGd+og8TdCioizUWkn2eQMYw6TdwEKCKNReQl3MLVBcBxXvgTInJfvOphGDWJeLaAD+FE1xtnhCnlNeDiONbDSCIaNWp0WNjEiRPp0qULPXr04Mwzz2Tt2vLTdB999FEyMjLYsWNHWdj8+fM5//zzy87vvfdehg4dyoEDBzh48CDjxo3jF7/4BSeccAIXXXRR1DaViacALwTGqeqXlF/ysxK3xs5IZoqLYMl0ePqXMLGr+1wyHUqKo36pXr16sWjRIpYtW8aIESO44447ysXPmDGDPn36hN0a7cEHH+TTTz/l1VdfpX79+tx9993s2rWL7777jlWrVjFs2DAuueSSSnnoDkc8BXgkh5b0BNIYt2zISFaKi2D21TD3Flj/Gezc4D7n3gKzRrv4KDJo0CAaNmwIwMknn1yutfrhhx/YvXs3f/zjH5kxY8ZheSdMmMAbb7zBvHnzaNCgAXv37mXatGk8+uijpKamAjBmzBjq16/PBx98UO26xlOAX+BawVJKHx/XE355kJEMfDUDVs4LHbdyHiybGbNLP/3005xzzjll5zNmzOCyyy7jtNNO49tvv2Xz5kO+tD799FOeeOIJ3nzzzbKu7ffff0+bNm3IzMwsV25OTk5U3NbHU4B3Aw+IyFTc8MftIvIBcCVusN5IVpZOrzh+SYT4KvL888+zaNEixo8fXxY2c+ZMfvWrX5GSksIll1zCrFmzyuKOP/54VJV33jnkFkhVQ249HS68ssRNgKq6ADcwnw78AJyJmxXTX1WXxKseRgLYEWHB7Y7o7pIL8N577/Hggw8yd+7csh2Wli1bxqpVqxgyZAjt2rVj5syZ5bqhxxxzDG+88Qa33XYbH374IeBEuXbt2nIbvgAsWbKELl26VLuecR0HVNWvVXW0qnZT1S6qeoWqfh3POhgJoMlxEeKrNxk7mKVLl3L99dczd+5cWrRoURY+Y8YM7rvvPvLy8sjLyyM/P5+NGzeWs5J27NiROXPmcMUVV/Dll19yxBFHMHr0aG6//XaKi52p4rnnnmPv3r0MHjy42nWN5zjgSBG5KET4RSJSc7zkGNGn15UVx/eOEF8Be/fuJSsrq+yYOHEi48ePZ/fu3YwcOZKePXty4YXO9DBz5kwuvrj8iNfFF1/MzJnl30H79OnDtGnTuPDCC/nhhx/485//TEZGBh07duSEE05g1qxZvPLKK1Hpgko0TKm+LiSSC9yuqm8HhZ8FPKaq3eJSkQSSk5OjixYtSnQ1osbKlSvp3Llz5IQlxc7aGcoQ0/kCGPkspKRGv4IJIMxvElap8VwP2AH4NkT499g4YHKTkgojnnHWziXT3TtfkyzX8p10WdKIryrEU4DbgBOAvKDwjsCuw1IbyUVqGvS6wh1GGfE0wvwv8KiIdCwNEJETcZ7LXo1jPQyjxhBPAd4B7ABWiMh6EVkP5AI7gfEV5jRqLPGyIdQGqvJbVLoLKiLH4Py5lFQmn6ruAgaIyBCgJ+7FdAnwvtpfsVaSkZHB1q1badasWVQsgrUZVWXr1q1kZGRUKp8vAXo7Az2I85XZAPfetlpEHgLWeq7//Fb0XeDdStXSqJFkZWWxYcMGbNMZR0ZGBllZlRvT9NsC/gG4AOdc6YWA8M+BO/G5mt3ba+9M3H4L5bq/qvpbn3Uxagj16tWjffv2ia5GrcavAC8DrlHVj0QksOu5HNcaRkREfg/8BTfskE/5JUnWBTXqJH4F2IrQm4+kVaKM3+HcvB+2hZdh1FX8WkFzgdNDhI/C7UDkh0zc/uuGYXj4bb3uB54Xkda4XWdHikgn4HLgPJ9lzADOxryfGUYZvgSoqvNEZBRuTV8JziizBLhAVd/zea31wP0iMgBYBhQGRqrqRN+1NowkIaIARSQN+CXwmaoOrMa1xgK7cWsCgzdtU9yMGMOoU0QUoKoWicgcoBOhfbr4QlXNXm0YQfg1wnwFHB+ti4rIMSJiW6MZdR6/IrgPmCAiw0SktYgcFXj4KUBE6onIX0RkF7ARaOeFPyQiN1Wl8oZR2/ErwNeB7sAc3HKiLd5R4H36IXA2zYGA8M9x+7AbRp3D7zDEoChcq9qzaQwj2fA7DPFRFK4Vjdk0hpFU+P7H95Yh3Qx0wQ0b5AJTVPUnn0WUzqbJCwqvzGwaw0gq/C5HGoDbSvonYKEXfAXOue5QVV0YNvMhojGbxjCSCr8t4CO4qWQ3lC7E9YYRngAmcPjA+mFEaTaNYSQVfgXYE7g6cBW8qpaIyERgaaTMUZxNYxhJhd9hiB1AqJks7YHtkTKrahFuCKOx/6oZRvLjV4AzgadF5Nci0l5E2onIFcBUXNfUD1GdTWMYyYDfLugdOCdK/wjIUwhMAf7TZxn34WbT/AFn9dwTGKmqP/ssxzCShkq5pheRhsAvcGL8XlX3ViJv4OB74EUFUFVNevfIyeaa3vBN9VzTi8ixQJqqbgC+DgjPAgp9jgVGYzaNYcSFK5/+jA3b9pF1ZAOm/6ZfzK7jtws6HXgJ984XyFDgUpyFs0KiNJvGMGJKUXEJc5Zs5PM1P3OgqIT87ft46Yv1DM/OIjUl+r5P/Rph+gD/ChH+MZDj92Ii0l1E/iYib4pISy9smIj08luGYcSKouISbnlhKXe8vIwDRe6N6UBRCXe8vIyb/7mEouJK+aL2hV8BpgH1Q4RnhAk/DBH5JW6f+OOAwTgHv+DeKf/gsx6GETPmLNnIW7k/hox7K/dH5iyNsNNvFfArwM9wXrGDuRknKj88gNsf8GLgYED4fKCvzzIMIybs2FvI/3yyusI0L32xPurX9fsOeA/wgYicBLzvhQ0GegFn+SyjK6HdEv4M+FrUaxjVRVXZtGM/ufk7WZG/k9z8HazYtJMN2/ZFzJu/PXKayuJ3OdK/RaQ/bhejSzi0scpNqvqVz2ttw3U/84LCewMbfJZhGL4pKi5hTcEeJ7ZNntjyd7Jtb2HkzCFo1bRB5ESVxPdyJE9o1dld8QXgYW9CtgJpIjIQN9F7WjXKNQz2HSxm5Y+lrZoT3DebdpYZUyriuKYN6NIqE1XlvZWbw6Yb1ad1NKsM+B8H7AIUq+q33vkQYDRujd9fVLXYRzH3As/gFuUKsML7fAG385Jh+OLnPQfJzd9Rrhu5pmAPJRHmlKSmCMcf3YgurTLp2iqTLi0z6dIqk6YN0wEoLlFu/ueSkIaYs7sey/Deldv5yA++ZsKIyEJgkqrO9Abfv8MZT3oA01X1rjD52gDrA/f/E5EOuG5nCrBUVVdV+y5qCTYTpnKoKut/3seKTYFi28mPO/dHzNugXiqdWzb2xNaErq0y6XhMYzLqVTzhqqi4hDlLN/Jfry7nQFEJ9dNSeGBYN4b3rtY4YPVmwgCdce98ACNxy4rOFZFBuO5jSAECa4CWwGYR+QC4RFVXAxWbm4w6x8GiEr7fvLvMKJKbv5OV+TvZdaAoYt5mR6SXCa20dWvX7IgqCSYtNYVROa2ZMv8H1hTsoVXTBozKiX7Xs+x6PtOlcmjo4EwOWTN/AI6pIN8uoDmwGTgDqFf5KhrJxq79hazctIsVpd3ITTtZ9dNuDvoY6G7brGFZ97FUcC0a14/6Dr1ZRzYo9xkr/ApwOXCjiLyGE2Bpi3cczjVhON7DDV+s9M5fEZGDoRKq6mCfdTFqCarKll0HDrNC5m2NPIe/XqpwQovGTmxe69a5ZWMaZ8TnGR7L+Z+B+BXgncCrwO+BZ1W1dEL2hTi/nuG4ErgGtw5wIPAt4HsFhVF7KClR1mzdU84KuSJ/BwW7Qz5vy9G4fhqdy1o1J7gTWjQmPS35naf7HQf8l4gcDWSq6raAqCepWFBHA39XVRWRnsB/qGrEFfRGzWZ/YTHf/bSrnBXymx93sfdgZGP4sZkZ5ayQXVs1IevIBqTEYKJzbaAy44DFuMH0wLC8CNnKjDDYNtS1kh17C8nd5LqOpa3b91t2UxzB5i8CHZofQRfPAllq8m/eyNfU4TpDrB3iBhphBmJGmBqLqpK/Yz+5Gw9ZIVfk72Sjj+lX9dNS6HRsY7oEWCE7HduYhunmbzkSsf6FAo0wghlhqk00FooWFZewumBPmVGk9J1tu48pWk0a1KNrq8xyxpEOzY8gLTX539diQawFmHAjjLfz0nhcVzgXGKeqH1eQfiBus9CuQD5ups8T1SkzmmzYto81BXsiJ/TYe7DImfw9o8iK/J188+Mu31O0AoXWpVUmrZpkRN3kX5eptgBF5HhV/T5UnKruAyZ76eJuhBGRS4FJwE3AJ97nmyLSRVXXhUjfHjfG+Q/cvNdTgb+LyBZVfbkqZUaL0pXapTPyQ63U3rrbmfwDzf5rCvYQabJTaopwQotGZe9pXVpl0rVlE5o0tDeGWFMpp0xlmUQygBHAtcCpNdWhkoh8BixT1WsDwlYBs0NNnxORh3CzdU4ICPsfoKuq9q9KmYFUdSpa6UrtUHMUOzQ/grbNGrJi005+2nkgRO7yNExPpXPL8iZ/P1O0jGpR7alorhTnOmIsbquxA8ArOHeD4dI/Dtylqnu872FR1d9Wpi6REJF0IBu32iKQdwjvSr+/Fx/I28BoEamH+yErW2a1qWil9uqCPawO0yVt3ii9nBWya6tM2lZxipYRGyIKUESaAL/GtXbHA/+L83B9qqquiJC9O4csn90rSBeLIYrmuCl0wR7bfiL8IuJjcYaj4PRpXnlShTLLyM/P5/777w8Zd/7555OdnQ3A4sWLee2118rFjwmYETVt3yE3PBfUX0HzlNCv1b1P6s0FFwwpu/bUv08IW7drr72WVq1aATBv3jyWLFkSMl3Lli257rrrys7D3Y+fewrkD3845JXkqaeeYtOmTSHT9e7dmwsuuADw7mlqsJ+wQ9SkewpHhQIUkenAMJxLisdw3aw93ntQRFR1UKjvcSZY3BIiLFL60nCpIE3IMkXkOuA6gCZNmkSqq1HHqPAdUESKgIeARwJnwIhIIXCSjxYwYXhd0L3AZao6KyB8MtAt1CYxIvIv4GtVvTkgbCRuzWJDnNAqVWYgVX0HHD5lAYvXbgsbn9P2SGbfGLMesFF9qvwOOAr4DbBBRN4BngPCt7nBVxX5h9+0qnqN37Q+yzsoIouBIcCsgKghwMthsi3EtfiBDAEWqWohQBXKrDaX5rSuUICxWKltxIcKR09VdY6qngeciFsP+Ajwo5evl0QeEDo66BgOXIx7lzwe989+Ce79KhZMBK4WkbEi0llEJuG2yn4CQESeE5HnAtI/AWSJyGNe+rHA1ZQ3ulRYZiwYnp3F2V2PDRkXq5XaRpxQ1UodOGPDTGAfzvgwxWe+u3DetY8ICDvCK+ueytajEvW9CecI6gBuU5jTA+LmA/OD0g/EPWwO4Oay3lCZMis6srOztaoUFhXri1+s0473vKFt73xNO97zhr74xTotKi6pcplG3Aj7P1GlcUAAETkSuAq4RlVP8pF+E3CmBr03ikhX4H1VDf2ITyKi4ZIiXnsWGFElOuOAgagzykzyDj80wnXVgg03LXEGDsMHJrrkItIwxO0+ylBVfdRHupeBaSIyHvi3F3Yyzso6x0d+w0g6IrWAj+BcTuwmfDOqgB8B3ghMwLkmLB2cLwKexq20N4w6RyQBLgK6AK8DT6vqJ1W9kLqJ2Td5LWDgJp/+p/YbRpIRaRiiL9APtxJ+joh8KyJ3iEhFntAqRFX3qOoyVf3KxGfUdSKuolTVXFW9HecB7R6ce8E8EflfETH/AoZRDSrjE6YQmC0iO3FWy/Nwe/xFXgNjGEZIfPkREJF2IvLfIrIWt031x8AJah7ODKNaVChAEblcRN7Hjd2dCFwPtFPV/1LVNZW5kIi0CTV1TRxtKlOWYSQLkbqgzwPrcEuRCnAW0S7BOlLViT6uFeiiMJCjvDhbkm3UOSIJcB1unO+yCtIoboJyJMKtmWsERN7uxjCSkAoFqKrtqnuBAFcUCvxZRAKXb6fi9of/srrXMYzaSDw8p5a6ohDcNmeBfkEPcmiZk2HUOSLNBT0HmIJb/b4jKK4J8BVwnaoGOzIqQz1XFCIyDfidqu6sdq0NI0mINAxxC/BwsPgAvLCHgN/5uZCqjlHVnSLSXET62SC+YUQWYA8O9xIWyAdAxLWAACLSSERewllBF+Bm1iAiT4jIfX7KMIxkI5IAjwYq8mGuQDOf1/oLTnS9cavpS3kN56bC8MNzw+Dx3u7TqPVEMsJswLWCq8LE9wA2+rzWhcDFqvqliAQOR6wEOvgso+5SXARfzYB1C6FoP+zcCEumQ8/LIcWGUKPOc8Ng+zpo2gauejVml4nUAr4OPCAih22ULSINgf/20vjhSGBriPDGQOSdHesyxUUw+2qYe4sTH7jPubfArNEu3ogOxUXuwbZuIfz8g/tcMh1KYvMvGskvaAtgKa6r+VfgGy+qM85AI0BvVQ32FB2qrPnAq6r6mIjsAnqo6hoRmQK0VdVzq3UntYAq+4RZMt2JLRxZfeHojpBSD1LrQWo6pKS57yn1IDXN+0wP+F4vIH1wuqD4SGWkJMnWZKUPupXzDo/rfAGMeMbde+Wpmk8YVd0sIqfghiL+FFCQ4vZMuMmP+DzuBt72nDClAbd73/sCp/sso26ydHrF8Rs+d0eikBRP9EEiTknzBFsv6IEQLPC0Kjw40sJcs4oPlZRU18UPJT5w4ctmQq8rovrTRZSzqq4FzvW8oB2PE+EqLb9XfERUdYEn5t8DPwBn4gbh+6vq15WueV1ih9/X7AShJV7XuLbPKIzg5nbJ9PgLsBRPcF9U52Ke0EZXp4w6SZPjYOeG8PGtcuDyGVBcCMUHoaTIfS8pdN2qEi+8uPDwuOKD3vfguMIwZRQFpA8uoyiorAjXLKlp764RXHTuqOBvUEXivom3iLQCWhBkAFLV0FvXGNDrSlj/Wfj4PmOgUYv41SdaqFYg+gDhVuqhEurBEakM7/u6f8OBCiZqNYm+B/K4CdDbW/B5oBOHt/WKLUcKT8/LYdXb4Y0DJ1W0WKUGI3Lo3awmEMnY1fvKqF8ynuarp4D1wGm4cb/2AYeNA1ZESqqzwF00GdIyXFhahjsf+ayNA0aLnpe7B1ooYvSgi2cXtAvQS1W/i+M1k4fUNGcA+Hr2oQHiKBsE6jylD7plM+H1/3CGpbQMOG+CE18MHnTxFODXuB1oTYDVIYazMgwOPeg+nugG4jOPi+mDLp4CvBv4i4jcixNjYWCkqv4cx7oYRsU0bVP+M0bEU4Clqyreoby9t9RVhb3IGDWHOPU04inARO0Rbxg1lrgJUFU/ite1DKO2ENNhCBFpX4m0IiK22blRp4j1OOBCEXlaRPqHSyAiR4rIjTjnvxfFuD6GUaOIdRe0E25Dl9dFpBi3n/om3KzdI3Fjg52Bz4Fxqvp2jOtjGDWKmLaAqrpdVcfjXFHciFtP2BQ3+6UIeBY3OD/AxGfUReJihPE255ztHYZheCTJUmbDqJ2YAA0jgZgADSOBmAANI4GYAA0jgZgADSOBmAANI4GYAA0jgZgADSOBmAANI4GYAA0jgZgADSOBmAANI4GYAA0jgZgADSOBmAANI4GYAA0jgZgADSOBmAANI4GYAA0jgZgADSOBmAANI4GYAA0jgZgADSOBmAANI4GYAA0jgSStAEWkvoj8VUQKRGSPiMwVkSwf+W4SkTUisl9EFovIaSHS9BWRd0Vkt4jsEpEFItI8NndiJDNJK0DgMaHOQ1MAAAY0SURBVGA4cBlwGpAJvCYiYbfCFpFLgUnAn4BewALgTRFpE5CmH26b7fnAyUA28AhBe94bhh9EVSOnqmWISBNgCzBGVf/phbUG1gLnhNuJSUQ+A5ap6rUBYauA2ap6l3e+APhQVe+pbL1ycnJ00aJFlb4fo9Yj4SKStQXMBurhWioAVHU9sBI4JVQGEUn38r0TFPVOaR4RaQH0BzaJyCci8pOIfCwiZ0b/Foy6QNz2iI8zxwLFQEFQ+E9eXCiaA6lemuA8Z3nfO3if9wPjgaXASOBtEclW1a+CCxWR64DrvNPdIvJtJe4jHM05/N6M6BOt3/ktVT07VEStEqCI/BG3425FDKqoCCBSnzs4PjBPaY/hSVX9h/d9qYicAdyA24S0fGGqTwFPRbhmpRCRRaqaE80yjcOJx+9cqwSIM6w8HyHNOpxxJBX3BNsSENcC+FeYfAW4VjO4hWzBoVZxk/e5IijNSqANhlFJapUAVbUAH10CEVmMs0oOAV7wwrJw+9EvCFP2QS/fEGBWQNQQ4GXvex6QD5wYlL0j8LXf+zCMMlQ1KQ9gCrAR9/7WC/gQ+BJIDUjzDXBLwPmlwEFgLE6sk4DdQNuANOOAHbh3v+OBu3FiPymO93Zdon/funDE43euVS1gJbkNKAJeBBoA7wNXqWpxQJoTcd1UAFT1RRFpBtwLtASWA+eq6tqANI95FtMJQDMgFze0cZgBJlaoe680Ykw8fuekHAc0jNpCso4DGkatwARYgxCRu0TkCxHZKSJbRGSeiHRLdL2SDRFJFZEHAub8rhGRP4pI3F/JTIA1izOAv+Nm3gzGvcO+JyJHJbJSScidwM3Ab4FOwO+887viXRETYA1CVYeq6jRVXa6qXwNXAkcDA0rTiEieiPw/EXnGW4mxXkQuFZGmIjLTW6GxSkR+mbAbqfmcAsxT1Xmqmqeqc4G5QL/SBPH6nU2ANZvGuL/RtqDwccDnQG/gJeBZ3HjnG0BP3GSD50UkI35VrVV8AgwSkU4AItIF1+N4Iyhd7H/nRI+12FHhONRLuPmmgWOXecCMgPNGuKlyjweEtfPCchJ9DzXxwE0vfBAowY3hKvDHoDRx+Z2TeRywViMiE4FTgVO1/NglwLLSL6q6W0T2Un4mTunUuRaxrWWt5VLgKuBy3DhuT2CSiKxR1acD0sX8dzYB1kBE5FHgV8AgVV0dIknw4l8NCguePG6U52HgEVWd6Z1/LSJtcUaYQAHG/Hc2AdYwRGQSTnxnqOo3ia5PktIQN/E+kGIS8MAyAdYgRGQyzvI5DNgmIqUrM3ar6u7E1SzpmAf8p4iswXVBewG3A8/FuyImwJrFTd7n+0Hh9wP3xbcqSc2twAO4MdcWuGVmU4H/jndFbC6oYSQQe0k3jARiAjSMBGICNIwEYgI0jARiAjSMBGICNIwEYgI0jARiAjSqhYj0EpFiEfk00XWpjZgAjepyLW5GSTcR6ZzoytQ2TIBGlRGRBrglPVOB2cBvguL7icgSz+/KUhE5V0TUc+VfmqaLiLzurTrfLCIzAubAJj0mQKM6jADWquoyYDpwlYjUAxCRRsBrOOfH2cAduGVAZYhIS9yq8uVAX5wT5UbAXBGpE/+bdeImjZgxFic8gI+AvcCF3vmvcftz/EZVc1X1Xdwq9EBuBL5S1TtVdaUn5KuAPkCd2HzGBGhUCRE5Hucs6gUAdbP6/4kTJThvY8tVdV9Ats+CiskGTvccHO0Wkd3Aei/uFzGrfA3CliMZVWUsroVbJ1K2AaxA2W7EfraCSwFeB34fIi54n8akxARoVBrPge1onAuH14KipwNjcFu2XSUiDQJawb5BaZcAo3DvkcHuH+oE1gU1qsJ5uE1tpqrzYVp2ADOBa4AZODcPUz1L51m4naTgUMs4GWgCvOhZTDuIyFki8pSINI7vLSUGE6BRFX4DfKiqW0PEzQLa4jZJvQDoinOt+DCHVvXvB1DVfNx7ZAnwFs49xGTggHckPbYi3ogbInIR8ArQQt1mq3Ueewc0YoaIjAZW4yyb3XBbjM8z8R3CBGjEkmNwDqVaAj/iLJ53JrRGNQzrghpGAjEjjGEkEBOgYSQQE6BhJBAToGEkEBOgYSQQE6BhJJD/Aw2V971DRntNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(3,3), constrained_layout=True)\n",
    "ax.spines['right'].set_visible(False)\n",
    "ax.spines['top'].set_visible(False)\n",
    "\n",
    "sns.pointplot(x='new_age', y='net_score_group', data=barcode[barcode['broad.cell.type'] == \"excit.\"], hue='Genotype', order=['2m', '8m'] , hue_order=['WT', \"L2AKO\"])\n",
    "\n",
    "plt.axhline(y=0, linestyle='dashed', linewidth=2, color='gray', zorder=1)\n",
    "\n",
    "#plt.legend(title='Cell type', loc='upper left')\n",
    "plt.ylabel('CMA score \\n(net difference)', fontdict={'size': 14})\n",
    "plt.xlabel('Age', fontdict={'size': 14})\n",
    "\n",
    "plt.yticks([-0.06, 0, 0.06])\n",
    "\n",
    "plt.ylim(-0.06, 0.08)\n",
    "plt.xlim(-.2, 1.2)\n",
    "\n",
    "plt.savefig('./plots/excitatory_net.png', dpi=300)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(3,3), constrained_layout=True)\n",
    "ax.spines['right'].set_visible(False)\n",
    "ax.spines['top'].set_visible(False)\n",
    "\n",
    "sns.pointplot(x='new_age', y='net_score_group', data=barcode[barcode['broad.cell.type'] == \"inhib.\"], hue='Genotype', order=['2m', '8m'] , hue_order=['WT', \"L2AKO\"])\n",
    "\n",
    "plt.axhline(y=0, linestyle='dashed', linewidth=2, color='gray', zorder=1)\n",
    "\n",
    "#plt.legend(title='Cell type', loc='upper left')\n",
    "plt.ylabel('CMA score \\n(net difference)', fontdict={'size': 14})\n",
    "plt.xlabel('Age', fontdict={'size': 14})\n",
    "\n",
    "plt.yticks([-0.06, 0, 0.06])\n",
    "\n",
    "plt.ylim(-0.06, 0.08)\n",
    "plt.xlim(-.2, 1.2)\n",
    "\n",
    "plt.savefig('./plots/inhib_net.png', dpi=300)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOAAAADgCAYAAAAaLWrhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO2dd3iUVfbHPyehBKSDKBiq9CYhIE1FQAQ7SrEiotjLImv56equLuIWRcCVtbAsCC6gILKAuFZ0VVwVgnQBhQAhEQjSIaRwfn/cN5PJZDIzSaZkJvfzPPMk723vfYd8ue8t5xxRVSwWS2SIi3QHLJaKjBWgxRJBrAAtlghiBWixRBArQIslglgBWiwRxAowjAwZMkQB+6l4n2KxAgwjmZmZke6CpZxhBWixRBArQIslglgBWiwRpFKkO1DRycnJIS0tjaysrEh3pVyQkJBAYmIilStXjnRXwoIVYIRJS0ujZs2aNG/eHBGJdHciiqpy4MAB0tLSaNGiRaS7ExbsK2iEycrKon79+hVefAAiQv369SvU24AVYDnAiq+AivZdWAFGGXv37uWmm26iZcuWJCcn07t3b957772g3mPKlCmcOHEiqG1avGMFGEWoKkOHDuWiiy5i+/btrF69mvnz55OWlhbU+1gBhg8rwCjis88+o0qVKtxzzz2utGbNmvHggw+Sl5fHo48+So8ePejSpQuvv/46AJ9//jkXX3wxw4cPp127dtx8883ke0H49NNPSUpKonPnztx+++2cOnWKl19+mfT0dPr370///v2ZMWMGDz/8sOt+06dPZ/z48aSmptKuXTtGjx5Nly5dGD58uEu0q1evpl+/fiQnJzN48GAyMjLC+C1FGapqP2H6JCcnqyebNm0qklYcU6dO1XHjxnnNe/3113XChAmqqpqVlaXJycm6fft2XbFihdaqVUt3796teXl52qtXL/3yyy/15MmTmpiYqFu2bFFV1VGjRunkyZNVVbVZs2a6f/9+VVU9duyYtmzZUrOzs1VVtXfv3rpu3TrdsWOHAvrVV1+pquqYMWP0hRde0OzsbO3du7fu27dPVVXnz5+vY8aMCfgZS/qdRAnF/k3YbYgo5v777+err76iSpUqNGvWjHXr1rFw4UIADh8+zLZt26hSpQrnn38+iYmJAHTt2pXU1FRq1qxJixYtaNOmDQCjR49m2rRpjBs3rtA9zjjjDAYMGMCyZcto3749OTk5dO7cmdTUVJo0aULfvn0BuOWWW3j55ZcZMmQIGzZsYNCgQQDk5eXRqFGjcH0lUYcVYBTRsWNH3n33Xdf1tGnTyMzMpHv37jRt2pS//e1vDB48uFCdzz//nKpVq7qu4+Pjyc3Ndb2GBsLYsWN5/vnnadeuHWPGjHGle65YigiqSseOHfnmm29K+ngVEjsHjCIGDBhAVlYWr776qistf941ePBgXn31VXJycgDYunUrx48fL7atdu3akZqayk8//QTAnDlz6NevHwA1a9bk6NGjrrI9e/Zk9+7dzJ07lxtvvNGVvmvXLpfQ5s2bxwUXXEDbtm3Zv3+/Kz0nJ4eNGzcG4/FjEivAKEJEWLx4MV988QUtWrTg/PPPZ/To0fzlL39h7NixdOjQgW7dutGpUyfuvvtucnNzi20rISGBmTNnMmLECDp37kxcXJxrceeuu+7isssuo3///q7yI0eOpG/fvtStW9eV1r59e9588026dOnCr7/+yr333kuVKlVYuHAhjz/+OOeddx5du3Zl5cqVoftSoh1fE0T7KV+LMJHkiiuu0E8++cR1vWPHDu3YsWNI7hUt30kJKPZvwo6AFp8cOnSINm3aUK1aNQYOHBjp7sQcdhHG4pM6deqwdevWIunNmzdnw4YNEehRbBHzI6CI3CciO0QkS0RWi8iFfsr3c8plich2EbnHS5lGIvKmiOx3ym0SkX6hewpLrBLTAhSR64GpwPNAErAS+EBEmhZTvgWw3CmXBPwJ+JuIDHMrUwf4GhDgCqA98CCwL3RPYolVYv0VdDwwS1WnO9cPisgQ4F7gCS/l7wHSVfVB53qziPQEHgHyN+AeAzJU9Va3ejuC33VLJBk141vSDp4ksW415tzRM2T3idkRUESqAMnARx5ZHwF9iqnW20v5D4HuIpJvoj0U+FZE3haRfSLyg4g8IBXNjibGSTt4kh2Zx0k7eDKk94lZAQINgHhgr0f6XuDsYuqcXUz5Sk57AC2B+4DtwGDMK+6fgfu9NSgid4nIKhFZtX///pI+Q8h5+OGHmTJliut68ODBjB071nU9bNgwatWqRdeuXalXrx4tWrSga9euXHLJJZHobswR66+gUNQxqnhJ81fePT0OWKWq+a+wa0SkNUaArxRpTPUN4A2A7t27B37+ywu5eadZlLKHt1ftJuPQSRrVqcb13ZswLDmR+LjSDcB9+vRhwYIFjBs3jtOnT5OZmcmRI0dc+enp6Xz88cf07NmT2267jSuvvJLhw4eX5TEsbsSyADOBPIqOdg0pOsrl80sx5XOBA851BrDJo8xm4Del7mkA5Oad5oG5a/jPxl9caemHs1i98yCf/biPV25KolJ8yV9o+vbt6zI32rhxI506dSIjI4ODBw9SvXp1Nm/eTFJSUtCew1KYmH0FVdVsYDUwyCNrEGaV0xvfAJ7vVoMwI16Oc/010NajTBtgZ+l7659FKXsKic+d/2z8hUVr9pSq3caNG1OpUiV27drFypUr6d27Nz179uSbb75h1apVdOnShSpVqpSl6xYfxKwAHV4CbhORsSLSXkSmAo2B1wBEZLaIzHYr/xqQKCJTnPJjgduAF93KTAZ6icjvRKSViIwAHgKmhfJB3l6122f+O9/7zvdF3759WblypUuAvXv3dl336VPcepUlGMTyKyiq+raI1AeeAhoBG4DLVTV/tGrqUX6HiFyOEdm9QDrwkKq+61bmexEZitlbfBrY5fz8eyifJeOQ79W4dD/5vujTpw8rV65k/fr1dOrUiSZNmjBp0iRq1arF7bffXup2Lf6JaQECqOrfKUYcqnqxl7QvgG5+2nwfeD8Y/QuURnWqkX64eHd9jetUK3Xbffv2ZdKkSbRs2ZL4+Hjq1avHoUOH2LhxI9OnT/ffgKXUxPoraMxwffcmPvNH9vCd74vOnTuTmZlJr169CqXVrl2bBg0a+KhpKSthHQFFpCpmDlYN2K+q5W9jrJwyLDmRz37c53UhZkjHsxnWLbHUbcfHxxfaegCYNWtWkXLe0ixlI+QCFJGawC3AjcD5QGWcvTgRSQf+A7yhqt+Hui/RTHyc8MpNSSxas4d3vt9N+qGTNK5TjZE9mjCsW+n3AS2RJaQCFJGHMQsg24ElwETMwsZJoB7QCbgQ+FhE/gc8qKrbQtmnaKZSfBwjuzdhpJ/XUUv0EOoRsA/QT1WLMxz7DvinY/JzB9APsAK0VBhCKkBVHRFguVOEeBnfYimPhH0VVEQaiEhPZ0HGYqnQhE2AIlJTRN7BGK6uBM5x0l8TkWfC1Q+LpTwRzhHwLxjRdcMswuSzDLg2jP2weFCjRo0iaS+99BIdOnSgS5cuDBw4kJ07Cx91nTx5MgkJCRw+fNiV9vnnn3PllVe6rp966ikGDx7MqVOnyM7OZty4cZx77rm0bt2aa665JuhBZYJBbt5p1yozmBNG73y/m7zTZTJkKZZwCvBqYJyq/kBhk5/NGBs7iz/yciFlDsy4FF7qaH6mzIHTeUG/VVJSEqtWrWLdunUMHz6cxx57rFD+vHnz6NGjR7Gh0SZOnMjXX3/N4sWLqVq1Kk8++SRHjx5l69atbNu2jaFDh3LdddeVyEN3qMm3OHns3XWcyj0NwKnc0zz27jru/1cKuXmng37PcAqwLgUmPe7UxJgNWXyRlwsLb4MlD8Dub+FImvm55AFYMNrkB5H+/ftTvXp1AHr16lVotPr55585duwYzz33HPPmzStSd9KkSSxfvpylS5dSrVo1Tpw4wcyZM5k8eTLx8fEAjBkzhqpVq/LZZ58Ftd9lIVQWJ74IpwC/x4yC+eT/13c3xZsHWfJZOw82L/Wet3kprJsfslvPmDGDyy67zHU9b948brzxRi688EK2bNnCvn0F/qi+/vprXnvtNT744APXq+1PP/1E06ZNqVWrVqF2u3fvXq7c1ofS4qQ4winAJ4EJIjIds/0xXkQ+A0ZhNustvlgzx3d+ip/8UvLWW2+xatUqHn30UVfa/PnzueGGG4iLi+O6665jwYIFrrxWrVqhqnz0UYFrHVX1Gnq6uPRIcDQrhx8zjvgsUxaLk+IImwBVdSVmY74K8DMwEHMqpreqpoSrH1HLYT+vP4eDv6DxySefMHHiRJYsWeKKsLRu3Tq2bdvGoEGDaN68OfPnzy/0GnrWWWexfPlyHn74YVasWAEYUe7cubNQwBeAlJQUOnToEPR+lwRVZcnadAZO+oLj2b5nQmWxOCmOsO4Dqup6VR2tqp1UtYOq3qKq68PZh6il9jl+8kt/GNsba9as4e6772bJkiU0bNjQlT5v3jyeeeYZUlNTSU1NJT09nT179hRaJW3Tpg2LFi3illtu4YcffuCMM85g9OjRjB8/nrw880c+e/ZsTpw4wYABA4La75Lw8/5j3DLjWx6at4Z9R0/5LV8Wi5PiCJs1hGM5nq2q//ZIvwaorKoLw9WXqCRplFl0KY5uo0rd9IkTJ1wBPAHGjx/P8uXLOXbsGCNGmMNMTZs2ZcmSJcyfP58PPvigUP1rr72W+fPn07Nngf/MHj16MHPmTK6++mpWrFjBn/70Jx555BHatGlDXFwc7dq147333ovIK+jJ7DxeWbGNN/67nZy8glXYkd2bkHn0FJ9tKepjuawWJ8Uh4VoGFpGNwHhV/dAj/RJgiqp2CktHIkj37t111apVhdI2b95M+/bt/Vc+nWdWO70txLS/Cka8CXHxQeppZAn4OykFH2/ayzNLNrLHbT7XoVEtnru2E92a1jWe59bs4enFGziVe5qqleKYMLRTWS1Oiq0YTnvAlsAWL+k/YfcB/RMXD8NnmdXOlDlmzlc70Yx8590YM+ILFbt/PcEzSzby6Y8Fo1vNqpX47aVtuKVXM5dHuXyLk1c//5kdmceNyVcIrU/CKcCDQGsg1SO9DXC0SGlLUeIrQdIt5mMJiFO5ebzxxXZeWfGTa3MdYGjXxjx5RXsa1kyIYO/CK8B/A5NF5DpV3QogIm0xnssWh7EflgrCf7fu5w9LNrIjsyBUd6uGNZhwTSd6n1s/gj0rIJwCfAxj/b5JRDKctEYYm8BHi61VAShP+2GRJhhrEr8czmLCsk28vz7DlVatcjy/uaQ1t/dtQZVK5ccVUokFKCJnYfy5lOhgnKoeBfqKyCCgK2ZimgJ8quXpQGCYSUhI4MCBA9SvX7/Ci1BVOXDgAAkJpXstzMk7zayvU5nyydZCe3pDOp7N01d14JwQ7OOVlYAE6EQGmojxlVkNM2/bLiJ/AXY6rv8CQlU/Bj4uRV9jksTERNLS0iiPgVsiQUJCQqEtkUD5bsevPL14A1v2FiwnNKtfnWeu7kj/tg191IwsgY6AfwCuwjhXmuuW/h3wOAFaszux9gZi4i0Ueg9Q1YcC7EtMUblyZVq0aBHpbkQtmcdO8fzyzSxKKTgpVKVSHPddfC739DuXhMrle3U4UAHeCNyuql+IiPur5wbMaOgXEXkE+Ctm2yGdwiZJFfYV1FI68k4rc7/dyQsfbuFIVoElyMVtz+TZqzvSrP4ZEexd4AQqwMZ4Dz5SqQRt/Abj5r1ICC+LpSSs3X2IpxZvYP2eAmPgxrUT+P1VHRnc8ayomksHKp6NwEUU3cMbiYlAFAi1MPHXLZZScehENi98uIW53+0if9muUpww9sKWPDSwFdWrRF+khUB7/Czwlog0wUSdHSEi7YCbgCsCbGMeMATr/cxSQk6fVhampPHnD37k1+PZrvReLesx4ZpOtD6rZgR7VzYCEqCqLhWRkRibvtOYRZkU4CpV/STAe+0GnhWRvsA6IMc9U1VfCrjXlgrD5owjPL14A6t2HnSlNahRlaevbM/V5zWOqtdNb/gVoIhUAi4FvlXVfmW411jgGMYm0DPonGJOxFgsgDGQnfLJNmatTHU5RIoTuLV3c8Zf2oZaCZUj3MPg4FeAqporIouAdnj36RIQqmrX2i1+UVWWrsvguWWbCtnoJTWtw4RrOtHpnNoR7F3wCXQOuBZoRdFFmFJR2tM0ltjm5/3H+P2/N/D1TwX/z9epXpknLmvHiOQmxMVgAJpABfgMMElE/oBZ9Tzunqmqv/prIJinaSyxRXEGsjf0aMJjQ9pR74zYjVEfqADzo8EuovCmuTjXgRw3CMppGkts4c9ANlIk1q1W6GeoCFSA/YNwrzKfprHEDoEayEaKOXf09F8oCAS6DfFFEO4VjNM0liinvBvIhpuA//CdhZP7gQ6Y186NwKuqujfAJoJxmsYSxUSDgWy4CdQcqS/GmHYv8I2TfAvGue5gVf2m2MoFBOM0jSUKiSYD2XAT6Aj4IuYo2T35WwciEge8Bkyi6MZ6EYJ0msYSRUSjgWy4CVSAXYHb3PftVPW0iLwErPFXOYinaSxRQrQayIabQAV4GGhBUbeCLYBD/ioH6zSNpfyTeewUf1r+I++mFLjKjyYD2XATqADnAzNE5DFMJCMFLgD+jHk1DYSgnqaxlC9ixUA23AQqwMcwm+7/dKuTA7wK/F+AbTxDGU/TWMonsWQgG25K5JpeRKoD52LE+JOqnihBXffN9yKnaVQ15t9NvLmmj2Zi0UA2RJTNNb2InA1UUtU0YL1beiKQE+BeYDBO01jKAbFsIBtuAv0vag7wDjDdI30wcD1mhdMnQTpNU+EZNeNb0g6eJLFutbAdl3In1g1kw02gAuwBPOAl/UvghUBvJiKdMSGpz8WcC80QkaEYawi/2xkWSDt4stBJknBRUQxkw02gAqwEVPWSnlBMehFE5FJgCfABMABjkgRGjLcBQwPsiyWMVDQD2XATqAC/xdjx3euRfj/wfYBtTMDEB/y7iLhHQ/oc+G2AbVjCSEU0kA03gQrwd8BnInIe8KmTNgBIAi4JsI2OeHdL+CtQL8A2LGGgIhvIhptAzZH+JyK9MVGMrqMgsMp9qro2wHsdBM6h6EZ8NyCtSGlLRPBmINuxcS0mDI2sgWysEvBGjSO0skSGnAu84BzIVqCSiPTDHPSeWYZ2LUGgvBvIxiqB7gN2APJUdYtzPQgYjbHx+6uq5vmq7/AUMAtjlCvAJufnXIyvGEsEsAaykSXQEXAGMBXY4my+/xuzeHI/xuX8E94qiUhTYLcacoCbReRpzGtnHLBGVbeV7REspeXLbfv5/b8LG8i2bliDP1ZgA9lwE6gA22PmfAAjMGZFl4tIf8zro1cBAjswUXD3ichnwHWquh3YXoY+W8rIL4ezmPD+Jt5fZw1kI02gAowH8s8cDaRgNfNn4Cwf9Y4CDYB9wMWA3a2NINZAtvwRqAA3APeKyDKMAPNHvHOATB/1PsFsX2x2rt8TkWxvBVV1QIB9sZQCayBbPglUgI8Di4FHgDdVNf9A9tUYv57FMQq4HWMH2A9j0BuwBYWl7FgD2fJNoPuA/xWRM4FaqnrQLet1fAvqTODvqqoi0hX4rar6taC3lB1rIFtGZg+FQ7ugTlO4dXHIblOSfcA8zGa6e1qqn2quRRhsGOqwYQ1kg8ChXfDrzyG/TagtJt0XYfphF2FCijWQjT5C/S/ivggj2EWYkGANZKOXUG/4jMJYQfzgXG/BnJ7x9gkJInKfiOwQkSwRWS0iF/op388plyUi20XknrK2GUo2Zxxh5Ovf8NjCdS7xNahRlak3dGXenb2s+Mo5ZR4BRaSVqv7kLU9VTwLTnHJhX4QRkesxJ3juA75yfn4gIh1UdZeX8i0we5z/xJx7vQD4u4jsV9V3S9NmsMjNO82ilD2kO4ek0w+dZNQ/vmXl9gPWQDaKKZFTJlclkQRgOHAncEF5dagkIt8C61T1Tre0bcBCVS1yeseJVXidqrZ2S/sH0FFVe5emTXdK65QpN+80D8xdw382/lJsGWsgG2Re7mYWYeqdCw+l+C/vm7I5ZXK1IpKEifV+I3AKeA/jbrC48i8DT6jqcef3YlHVh0rSF3+ISBUgGWNt4c5HFO9Kv7eT786HwGgnwKiUos0ysyhlj0/xDU9O5K/DulgD2SjErwBFpDZwM2a0a4U5iF0TM/Jt8lO9MwUrn519lAvFFkUDzBE6T49teyneiPhszMKRZ/lKTntSijZdpKen8+yzz3rNu/LKK0lOTgZg9erVLFu2rFD+GLdTYjNPdnf9flXVTdTctIoJm4ruVXXr1o2rrrrKde/p0z19ahVw55130rhxYwCWLl1KSor3//UbNWrEXXfd5bou7nkCeSZ3/vCHP7h+f+ONN8jIyPBaLlzP9IAeoD5w4NcDvOL2jKV9puLwKUARmYPx1fItMAXzmnXcmQf5RVX7e/s9zHiKW7yk+Sufny4+ynhtU0TuAu4CqF3bvh5aCuNzDigiucBfgBfdT8CISA5wXgAjYMRwXkFPADeq6gK39GlAJ29BYkTkv8B6Vb3fLW0ExmaxOkZoJWrTndLOAYe9upLVOw8Wm9+9WV0W3huyN+CKSTmZA44E7gDSROQjYDZQ/JjreVeRfwZaVlVvD7RsgO1li8hqYBCwwC1rEPBuMdW+oah3tkHAKseekVK0WWau797EpwBH9mgSqltbQozPfUBVXaSqVwBtMfaALwK/OPWSxP+ZpjM9PsOAazFzyVaYP/brMPOrUPAScJuIjBWR9iIyFRMq+zUAEZktIrPdyr8GJIrIFKf8WIzLxBcDbTMUDEtOZEjHs73mDel4NsO6JYbq1pYQE+hh7DTMhvoEEbkEsxL6D+AlEVmkqp7uCvPrXZX/u4g8AZwExqjqcSftDIy1/Xpv9cuKqr4tIvUx7jAaYcyqLlfV/Fj1TT3K7xCRy4HJGBeM6cBD+XuAAbYZdOLjhFduSmLRmj08vXgDp3JPU7VSHBOGdmJYt0Ti7epn1FKqfUAAEakL3IrxcH1eAOUzgIGe80YR6Qh8qqre/4uPIYIRnKX/i5+zI/M4LRqcwYpHLg5OxyxFCdMcsNRH0VT1oKpODUR8DjUwr2qeNMIscFgsFQ5/2xDjA2hDVXVyAOXeBWaKyKPA/5y0XphV1kUB1LdYYg5/c8AXMS4njlH8MKqYOZM/7gUmYVwT5m/O52LmgI8EUN9iiTn8CXAV0AF4H5ihql+V9kbOwez7nBHQPchn+EP9WCzlBH/bEOcDPTGW8ItEZIuIPCYivjyh+URVj6vqOlVda8Vnqej4XYRR1Y2qOh7jAe13GPeCqSLybxEJKDSZxWLxTkl8wuQAC0XkCGbV8gpMjL9TPitaLNFEXi6snQdH9pjrI3sgZQ50vQnigm91F9A2hIg0F5E/ishOTJjqL4HW1sOZJabIy4WFt8GSByA3y6TlZpnrBaNNfpDxKUARuUlEPsUEUmmLCS/dXFWfVtUdJbmRiDT1dnRNDE291bFYwsraebB5qfe8zUth3fyg39LfK+hbwC6MKVImZkW0g6eOVPWlAO7l7qLQnXpOXrm0qrdUEPJy4OupvsukzIGkskToK4o/Ae7C7PPd6KOMYg4o+6M4m7kaQFYA9S2W4HMwFVJmw5q34JinnbUHh4MfR9anAFW1eVlv4OaKQoE/iYi7J+144HwKvKZZLKEnLwe2LIfVs+DnFQTskKF28K1OwuGpNd8VhWDCnLn7Bc2mwMzJYgktv+6AlDdhzb/guMdMqEpNaHQe7PRx1qTbqKB3yd9Z0MuAVzHW74c98moDa4G7VNXTkZGLfFcUIjIT+I2qHilzryswiXWrFfpp8UNudsFot31F0fxzkiH5Nuh4HVSuZlY7vS3EtL8KzvM1Eysd/lxSvA8sV9VpxeTfC1zpGO0GdkORBpijaD+oaoXaQwyGOZIlQA78bOZ2P/wLju8vnFe1FnQZCd1GQ6MuhfPycs1q5/u/NVsQlRLgiklGfKXfByy1S4ougC+LiM8wp2P890CkBsbh7XDMS3drYLuIvAb8oqrPBNKOxVIsudmw5X1YNRN2fFE0/5zuZrTrdB1UKSY6VHwls9L55UvGHrDWOUFf+XTHnwDPBE77yFcg0GDif8UcZ+uG8SidzzJgIj78i1osPjnwc8Hc7oRHvNiqtaDL9ZA8Gs725RkzMvgTYBpmFNxWTH4XYE+A97oauFZVfxAR9/fezUDLANuwWAy5p+DHZWZut+O/RfMTe0DyGOg4tPjRrhzgT4DvY/zALHfMiVyISHXgj06ZQKgLHPCSXhPI85JusRQl8ydImQU/zIUTHn9OVWvDedebud3ZnSLSvZLiT4ATMXO2bSLyN+BHJ7098ABmcvl8gPf6HjMKTnGu80fBu4GVgXbYUgHJPWVWJlfPgtQvi+Y36Wnmdh2GQpXo8m7ibyN+n4j0wWxFPE9hz9AfAvepqp/jAy6eBD50nDBVAsY7v58PXFSazltinMxtRnQ/zIWTvxbOS6htVia7jYazOkSke8HA70a8427vcscLWiuMCLd5xIr3i6qudMT8CPAzMBCzCd9bVUPiltASheRkFYx23jbFm/RyRrtrom6080ZJ7AEPYl4jS40jtNFlacMSo+zfalYyvY52dcxolzwaGraPTP9CRNiDhotIY6AhHqZQqlpm54sVgtlD4dAuqNMUbi0aESmqyMmCzUuc0e7rovlN+zij3dXmlEoMEjYBOrEF3wLaUfRkgGLNkQLj0C6zQRzN7PvRjHZr58FJj5lMQh1jfZ58G5zZNiLdCyfhHAHfAHZj4gymE5qYgJbySs5J2LQEVs+EXd8UzW/W14iu/dVQOSHs3YsU4RRgByBJVbeG8Z6WSLNvM6x2RrssDw8m1epC15vNSuaZbSLTvwgTTgGux0SgtQKMdXJOwsbFZm63+39F85tdAN3HQLsrK9Ro541wCvBJ4K8i8hRGjDnumar6q9daluhh7yYjunXzIetw4bxq9Qrmdg1aR6J35ZJwCjA/9vpHFJ7/5buqsIsw0Uj2Cdi02FggpH1XNL/5hc7c7iqoZH2TiDAAAAlGSURBVN3IehJOAUYqRrwlFOzdaEa7tW/DKY/Rrnr9grldg1YR6V60EDYBqqoXAy1LVJF9HDa+Z4SX5uVMRouLzGjX7ko72gVISAUoIi0C9R/q+AxNVNXdoeyTpRT8ssGZ270Npzw8ilRvAEnOaFf/3Ih0L5oJ9Qj4jePW4h+q6mXzxxVp9wbgIWAa8EqI+2QJhOzjsGGREd4eL240Wl5sRru2V0ClKuHtWwwRagG2w7iseF9E8oDVQAbGD2hdzN5ge+A7YJyqfhji/lj8kbHOiG79gqKj3Rlnmrld8mioZ22og0FIBejEjnhURH6PCeZyAdAME9QlE3gT+FBVN4SyHxY/nDoGGxeZlcx0L0dyW/Z3RrvLK85oV6dp4Z8hIiyLMI41/ULnYykvZKx15nYLIPto4bwzGhpnRN1GVczRLkwH3cNuDWGJMKeOwoZ3jfDS1xTNP3eA8aXS9jKIr1w03xJUrAArCuk/FMztso8VzqtxljPa3Qp1m0eidxUWK8BY5tRRWL/QCC/DM/yGQKuBZm7XZogd7SKEFWAssifFGe0WQs7xwnk1znYb7ZpFpHuWAqwAY4WsI7Ahf7Rb65Ep0OoSZ7QbbEe7coQVYLTgLXb56tnGR8qa2bD+3aKjXc1GkDTKrGSGeDndUjqsAKOB/Njl7lF7crNg6YNeCgu0HmRWMltfamIdWMot9l8nGvAVuzyfmo3NSJc0Cuo0CU+/LGXGCjAaWDPHd36DtnDvSjvaRSFx/otYIs5hP/Fvso9b8UUpVoDRQO1z/OQHP3a5JTxYAUYDSX5ik4cgdrklPFgBRgNdbzI+VbwRotjllvBgBRgNxMXD8FlwzTQTsxzMz2umwYg3yxK73BJhrACjhfzY5bWc+WB+7HIrvqjGCtBiiSBWgBZLBLECtFgiiBWgxRJBrAAtlghiBWixRBArQIslglgBWiwRxArQYokgVoAWSwSxArRYIkjMClBEqorI30QkU0SOi8gSEfFrOCci94nIDhHJEpHVInKhlzLni8jHInJMRI6KyEoRaRCaJ7HEMjErQGAKMAy4EbgQqAUsE5FiTy+LyPXAVOB5IAlYCXwgIk3dyvTEhNn+HOgFJAMv4hHz3mIJhJj0YyAitYE7gDGq+rGTNgrYCVwCFBcGbTwwS1WnO9cPisgQ4F7gCSdtMjBNVSe61dsa5EewVBBidQRMBipjRioAnMi7m4E+3iqISBWn3kceWR/l1xGRhkBvIENEvhKRvSLypYgMDP4jWCoCMTkCAmcDeZgYhO7sdfK80QCId8p41rnE+T0/TtezwKPAGmAE8KGIJKuqp0tqROQu4C7n8piIbCnBcxRHA1iTyW8kCE1ZfNCAon9DpeE/qjrEW0ZUCVBEnsNE3PVFf19NAOqnvme+e538N4bXVfWfzu9rRORi4B7Mq2rhxlTfAN7wc88SISKrVLV7MNu0FCUc33NUCRCzsPKWnzK7MIsj8Zj/wfa75TUE/ltMvUzMqOk5QjakYFTMcH5u8iizGbC+3y0lJqoEqKqZBPBKICKrMauSg4C5TloiJh79ymLaznbqDQIWuGUNAt51fk8F0oG2HtXbAOsDfQ6LxYWqxuQHeBXYg5m/JQErgB+AeLcyPwIPuF1fD2QDYzFinQocA5q5lRkHHMbM/VoBT2LEfl4Yn+2uSH+/FeETju85qkbAEvIwkAu8DVQDPgVuVdU8tzJtMa+pAKjq2yJSH3gKaARsAC5X1Z1uZaY4K6aTgPrARuAy9bIAEyrUzCstISYc37M4SrdYLBEgVvcBLZaowAqwHCEiT4jI9yJyRET2i8hSEekU6X7FGiISLyIT3M787hCR50Qk7FMyK8DyxcXA3zEnbwZg5rCfiEi9SHYqBnkcuB94CGgH/Ma5fsJXpVBgBViOUNXBqjpTVTeo6npgFHAm0De/jIikisjvRWSWY4mxW0SuF5E6IjLfsdDYJiKXRuxByj99gKWqulRVU1V1CbAE6JlfIFzfsxVg+aYm5t/ooEf6OOA7oBvwDvAmZr9zOdAVc9jgLRFJCF9Xo4qvgP4i0g5ARDpg3jiWe5QL/fcc6b0W+/G5D/UO5ryp+95lKjDP7boG5qjcy25pzZ207pF+hvL4wRwvnAicxuzhKvCcR5mwfM+xvA8Y1YjIS8AFwAVaeO8SYF3+L6p6TEROUPgkTv7RuYah7WXUcj1wK3ATZh+3KzBVRHao6gy3ciH/nq0AyyEiMhm4Aeivqtu9FPE0/lWPNM/D45bCvAC8qKrznev1ItIMswjjLsCQf89WgOUMEZmKEd/FqvpjpPsTo1THHLx3J48I/IdlBViOEJFpmJXPocBBEcm3zDimqsci17OYYynwfyKyA/MKmoTxhjA73B2xAixf3Of8/NQj/VngmfB2JaZ5EJiA2XNtiDEzmw78MdwdsWdBLZYIYifpFksEsQK0WCKIFaDFEkGsAC2WCGIFaLFEECtAiyWCWAFaLBHECtBSJkQkSUTyROTrSPclGrECtJSVOzEnSjqJSPtIdybasAK0lBoRqYYx6ZkOLMREpHLP7ykiKY7flTUicrmIqOPKP79MBxF537E63yci89zOwMY8VoCWsjAc2Kmq64A5wK0iUhlARGoAyzDOj5OBxzBmQC5EpBHGqnwDcD7GiXINYImIVIi/zQrxkJaQMRYjPIAvgBPA1c71zZj4HHeo6kY1cRonetS/F1irqo+r6mZHyLcCPYAKEXzGCtBSKkSkFcZZ1FwANaf6/4URJRhvYxtU9aRbtW89mkkGLnIcHB0TkWPAbifv3JB1vhxhzZEspWUsZoTbJeKKUygAItKEwELBxQHvA494yfOM0xiTWAFaSozjwHY0xoXDMo/sOcAYTMi2W0WkmtsoeL5H2RRgJGYe6en+oUJgX0EtpeEKTFCb6Wp8mLo+wHzgdmAexs3DdGel8xJMJCkoGBmnAbWBt50V05YicomIvCEiNcP7SJHBCtBSGu4AVqjqAS95C4BmmCCpVwEdMa4VX6DAqj8LQFXTMfPI08B/MO4hpgGnnE/MYy3iLWFDRK4B3gMaqgm2WuGxc0BLyBCR0cB2zMpmJ0yI8aVWfAVYAVpCyVkYh1KNgF8wK56PR7RH5Qz7CmqxRBC7CGOxRBArQIslglgBWiwRxArQYokgVoAWSwSxArRYIsj/A1AGs36o8MtAAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(3,3), constrained_layout=True)\n",
    "ax.spines['right'].set_visible(False)\n",
    "ax.spines['top'].set_visible(False)\n",
    "\n",
    "sns.pointplot(x='new_age', y='net_score_group', data=barcode[barcode['broad.cell.type'] == \"astro.\"], hue='Genotype', order=['2m', '8m'] , hue_order=['WT', \"L2AKO\"])\n",
    "\n",
    "plt.axhline(y=0, linestyle='dashed', linewidth=2, color='gray', zorder=1)\n",
    "\n",
    "#plt.legend(title='Cell type', loc='upper left')\n",
    "plt.ylabel('CMA score \\n(net difference)', fontdict={'size': 14})\n",
    "plt.xlabel('Age', fontdict={'size': 14})\n",
    "\n",
    "plt.yticks([-0.06, 0, 0.06])\n",
    "\n",
    "plt.ylim(-0.06, 0.08)\n",
    "plt.xlim(-.2, 1.2)\n",
    "\n",
    "plt.savefig('./plots/astro_net.png', dpi=300)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(3,3), constrained_layout=True)\n",
    "ax.spines['right'].set_visible(False)\n",
    "ax.spines['top'].set_visible(False)\n",
    "\n",
    "sns.pointplot(x='new_age', y='net_score_group', data=barcode[barcode['broad.cell.type'] == \"microglia\"], hue='Genotype', order=['2m', '8m'] , hue_order=['WT', \"L2AKO\"])\n",
    "\n",
    "plt.axhline(y=0, linestyle='dashed', linewidth=2, color='gray', zorder=1)\n",
    "\n",
    "#plt.legend(title='Cell type', loc='upper left')\n",
    "plt.ylabel('CMA score \\n(net difference)', fontdict={'size': 14})\n",
    "plt.xlabel('Age', fontdict={'size': 14})\n",
    "\n",
    "plt.yticks([-0.06, 0, 0.06])\n",
    "\n",
    "plt.ylim(-0.06, 0.08)\n",
    "plt.xlim(-.2, 1.2)\n",
    "\n",
    "plt.savefig('./plots/microglia_net.png', dpi=300)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOAAAADgCAYAAAAaLWrhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO2dd3hVRfrHP28CIXQIRcBQgogBQg1IdREQwQoK6uKKiLJ2XWQtv7WsuljWtWBZ18K6qLiCiqiAFQVsuCpFaaEJAUIQCNXQ0t7fH3Nyc3Nzb+5NcktyM5/nOc/NmZkzZ85Nvpk5M/O+r6gqFoslMsREugEWS3XGCtBiiSBWgBZLBLECtFgiiBWgxRJBrAAtlghiBRhGRo4cqYA9qt/hEyvAMJKVlRXpJlgqGVaAFksEsQK0WCKIFaDFEkFqRLoB1Z3c3FwyMjI4fvx4pJtSKYiPjycxMZGaNWtGuilhwQowwmRkZFC/fn3atWuHiES6ORFFVdm3bx8ZGRkkJSVFujlhwQ5BI8zx48dp0qRJtRcfgIjQpEmTajUasAKsBFjxFVHdvgsrwCrG7t27ufzyy2nfvj2pqan079+f9957L6j3ePrppzl69GhQ67R4xwqwCqGqjB49mt/97nds2bKF5cuXM3v2bDIyMoJ6HyvA8GEFWIVYtGgRcXFxXH/99a60tm3bcsstt5Cfn88dd9xBnz596NatGy+99BIAS5Ys4cwzz2Ts2LEkJyfzhz/8gUIvCF988QU9e/aka9euXH311Zw4cYJnn32WzMxMhgwZwpAhQ3jllVe47bbbXPebPn06U6ZMIT09neTkZCZMmEC3bt0YO3asS7TLly9n8ODBpKamMmLECHbt2hXGb6mKoar2CNORmpqqnqxbt65Emi+eeeYZnTx5ste8l156SadOnaqqqsePH9fU1FTdsmWLLl68WBs0aKA7duzQ/Px87devn3799dd67NgxTUxM1A0bNqiq6vjx43XatGmqqtq2bVvdu3evqqpmZ2dr+/btNScnR1VV+/fvr6tWrdKtW7cqoN98842qqk6cOFEff/xxzcnJ0f79++uePXtUVXX27Nk6ceLEgJ+xrN9JFcHn34RdhqjC3HTTTXzzzTfExcXRtm1bVq1axZw5cwA4dOgQmzZtIi4ujtNPP53ExEQAevToQXp6OvXr1ycpKYmOHTsCMGHCBJ5//nkmT55c7B5169Zl6NChLFiwgE6dOpGbm0vXrl1JT0+ndevWDBw4EIArrriCZ599lpEjR7JmzRqGDx8OQH5+Pi1btgzXV1LlsAKsQnTp0oV3333Xdf7888+TlZVF7969adOmDc899xwjRowods2SJUuoVauW6zw2Npa8vDzXMDQQJk2axCOPPEJycjITJ050pXvOWIoIqkqXLl347rvvyvp41RL7DliFGDp0KMePH+eFF15wpRW+d40YMYIXXniB3NxcADZu3MiRI0d81pWcnEx6ejqbN28GYObMmQwePBiA+vXr89tvv7nK9u3blx07dvDmm28ybtw4V/r27dtdQps1axaDBg3itNNOY+/eva703Nxc1q5dG4zHj0qsAKsQIsL777/Pl19+SVJSEqeffjoTJkzgscceY9KkSXTu3JlevXqRkpLCddddR15ens+64uPjmTFjBpdccgldu3YlJibGNblz7bXXcs455zBkyBBX+UsvvZSBAwfSuHFjV1qnTp147bXX6NatG/v37+eGG24gLi6OOXPmcNddd9G9e3d69OjB0qVLQ/elVHVKe0G0R+WahIkk5513nn7++eeu861bt2qXLl1Ccq+q8p2UAZ9/E7YHtJTKwYMH6dixI7Vr12bYsGGRbk7UYSdhLKXSqFEjNm7cWCK9Xbt2rFmzJgItii6ivgcUkRtFZKuIHBeR5SJyhp/yg51yx0Vki4hc76VMSxF5TUT2OuXWicjg0D2FJVqJagGKyGXAM8AjQE9gKfCxiLTxUT4J+Mgp1xN4FHhORMa4lWkEfAsIcB7QCbgF2BO6J7FEK9E+BJ0CvKqq053zW0RkJHAD8Bcv5a8HMlX1Fuc8TUT6ArcDhQtwdwK7VPVKt+u2Br/plupA1PaAIhIHpAKfeWR9BgzwcVl/L+U/BXqLSKGJ9mjgexF5S0T2iMhPInKzVDc7GktQiFoBAk2BWGC3R/puoIWPa1r4KF/DqQ+gPXAjsAUYgRni/h24yVuFInKtiCwTkWV79+4t6zOEnNtuu42nn37adT5ixAgmTZrkOh8zZgwNGjSgR48eJCQkkJSURI8ePTjrrLMi0dyoI9qHoFDSMap4SfNX3j09BlimqoVD2JUicipGgP8sUZnqy8DLAL179w58/5cX8vILmLtiJ28t28Gug8do2ag2l/VuzZjURGJjytcBDxgwgHfeeYfJkydTUFBAVlYWhw8fduVnZmaycOFC+vbty1VXXcX555/P2LFjK/IYFjeiWYBZQD4le7vmlOzlCvnVR/k8YJ9zvgtY51EmDfhTuVsaAHn5Bdz85ko+WfurKy3z0HGWbzvAovV7+OflPakRW/YBzcCBA13mRmvXriUlJYVdu3Zx4MAB6tSpQ1paGj179gzac1iKE7VDUFXNAZYDwz2yhmNmOb3xHeA5thqO6fFynfNvgdM8ynQEtpW/tf6Zu2JnMfG588naX5m7cme56m3VqhU1atRg+/btLF26lP79+9O3b1++++47li1bRrdu3YiLi6tI0y2lELUCdHgKuEpEJolIJxF5BmgFvAggIq+LyOtu5V8EEkXkaaf8JOAq4Am3MtOAfiJyj4h0EJFLgFuB50P5IG8t21Fq/ts/lp5fGgMHDmTp0qUuAfbv3991PmCAr/kqSzCI5iEoqvqWiDQB7gVaAmuAc1W1sLdq41F+q4icixHZDUAmcKuqvutW5kcRGY1ZW7wP2O58/iuUz7Lr4LFS8zP95JfGgAEDWLp0KatXryYlJYXWrVvz5JNP0qBBA66++upy12vxT1QLEEBV/4UPcajqmV7SvgR6+anzQ+DDYLQvUFo2qk3mId/u+lo1ql3uugcOHMiTTz5J+/btiY2NJSEhgYMHD7J27VqmT5/uvwJLuYn2IWjUcFnv1qXmX9qn9PzS6Nq1K1lZWfTr169YWsOGDWnatGkpV1oqSlh7QBGphXkHqw3sVdXKtzBWSRmTmsii9Xu8TsSM7NKCMb0Sy113bGxssaUHgFdffbVEOW9p0cr4V74n48AxEhvXZuY1fUN2n5ALUETqA1cA44DTgZo4a3Eikgl8Arysqj+Gui1VmdgY4Z+X92Tuyp28/eMOMg8eo1Wj2lzapzVjepV/HdDinYwDx9ia5dujQLAIqQBF5DbMBMgWYB7wMGZi4xiQAKQAZwALReR/wC2quimUbarK1IiN4dLerbnUz3DUUnUIdQ84ABisqr4Mx34A/uOY/FwDDAasAC3VhpAKUFUvCbDcCUI8jW+xVEbCPgsqIk1FpK8zIWOxVGvCJkARqS8ib2MMV5cCJzvpL4rIA+Fqh8VSmQhnD/gYRnS9MJMwhSwALgpjOywe1KtXr0TaU089RefOnenWrRvDhg1j27biW12nTZtGfHw8hw4dcqUtWbKE888/33V+7733MmLECE6cOEFOTg6TJ0/mlFNO4dRTT2XUqFFBDyoTDPLyC1yzzGB2GL394w7yCypkyOKTcArwQmCyqv5EcZOfNIyNncUf+XmwYia8cjY81cV8rpgJBflBv1XPnj1ZtmwZq1atYuzYsdx5553F8mfNmkWfPn18hkZ7+OGH+fbbb3n//fepVasWd999N7/99hsbN25k06ZNjB49mosvvrhMHrpDTaHFyZ3vruJEXgEAJ/IKuPPdVdz03xXk5RcE/Z7hFGBjikx63KmPMRuylEZ+Hsy5CubdDDu+h8MZ5nPezfDOBJMfRIYMGUKdOnUA6NevX7He6pdffiE7O5uHHnqIWbNmlbj2ySef5KOPPmL+/PnUrl2bo0ePMmPGDKZNm0ZsbCwAEydOpFatWixatCio7a4IobI4KY1wCvBHTC9YSOG/vuvwbR5kKeTnWZA233te2nxYNTtkt37llVc455xzXOezZs1i3LhxnHHGGWzYsIE9e4r8UX377be8+OKLfPzxx66h7ebNm2nTpg0NGjQoVm/v3r0rldv6UFqc+CKcArwbmCoi0zHLH1NEZBEwHrNYbymNlTNLz1/hJ7+cvPHGGyxbtow77rjDlTZ79mx+//vfExMTw8UXX8w777zjyuvQoQOqymefFbnWUVWvoad9pUeCIyfyWL/rcKllKmJx4ouwCVBVl2IW5uOAX4BhmF0x/VV1RbjaUWU55Gf4cyj4Exqff/45Dz/8MPPmzXNFWFq1ahWbNm1i+PDhtGvXjtmzZxcbhp500kl89NFH3HbbbSxevBgwoty2bVuxgC8AK1asoHPnzkFvd1lQVT5avYuznvqSIzmlvwlVxOLEF2FdB1TV1ao6QVVTVLWzql6hqqvD2YYqS8OT/eSXfzO2N1auXMl1113HvHnzaN68uSt91qxZPPDAA6Snp5Oenk5mZiY7d+4sNkvasWNH5s6dyxVXXMFPP/1E3bp1mTBhAlOmTCE/3/yRv/766xw9epShQ4cGtd1lYWvWESbM+JEb/7uCXaWYehVSEYsTX4TNGsKxHM9R1Q880kcBNVV1TrjaUiXpOd5Muvii1/hyV3306FFXAE+AKVOm8NFHH5Gdnc0ll5jNTG3atGHevHnMnj2bjz/+uNj1F110EbNnz6Zv3yKrgT59+jBjxgwuvPBCFi9ezKOPPsrtt99Ox44diYmJITk5mffeey8iQ9Djufn8a8kvvLjkF3LcZjbHpiayPzuHRRtK+liuqMWJLyRc08AishaYoqqfeqSfBTytqilhaUgE6d27ty5btqxYWlpaGp06dfJ/cUG+me30NhHT6QK45DWIiQ1SSyNLwN9JOVi0fjf3z1vLjv1F73PJLerz0OgUerdLMJ7nVu7kvvfXcCKvgFo1Ypg6OqWiFic+LwynPWB7YIOX9M3YdUD/xMTC2FfNbOeKmeadr2Gi6fm6j4sa8YWKjANHeXD+OhauK3KIVzcultuGd+SqAe1cHuUKLU5eWPILW7OOGJOvEFqfhFOAB4BTgXSP9I7AbyVKW0oSWwN6XmEOS0Dk5BXw72+28OwXmzieWzTcvKB7K+49rxMnNYiPYOvCK8APgGkicrGqbgQQkdMwnsveD2M7LNWEpZuzuO+DNfyyt8iwtn2zukwdlcLADpXD1UY4BXgnxvp9nYjsctJaYmwC7/B5VTWgMq2HRZpgzEnsOXychz5MY97Pma60+Jox3DL0VCadkUStGpVnuF5mAYrISRh/LmXaGKeqvwEDRWQ40APzYroC+EIr04bAMBMfH8++ffto0qRJtRehqrJv3z7i48s3LMzLL+C177YxbeFGsk8Ubc0b3vkk/np+Z1on1AlWU4NGQAJ0IgM9jPGVWRvz3rZFRB4Dtjmu/wJCVRcCC8vR1qgkMTGRjIwMKmPglkgQHx9fbEkkUJZv2889761h/a9F0wmJjWvz4IVdGNbppDLXl9i4drHPUBFoD3g/cAHGudKbbuk/AHcRoDW7E2tvGCbeQrFNAKp6a4BtiSpq1qxJUlJSpJtRZdmXfYLHPlnP28uKdgLFxcZw/eD23DikA/E1yzfcDKUnNHcCFeA44GpV/VJE3IeeazC9oV9E5HbgH5hlh0yKmyRV2yGopXwUFCizftzOPz7ZwKFjua70M05tyt9GpZDUtG4EWxc4gQqwFd6Dj9QoQx1/wrh5LxHCy2IpC6szDnHvB2v4ecdBV1qLBvHcd35nzu3aokq9SwcqnrXA7yi5hncpJgJRIDTAxF+3WMrFoaO5PPHZBt74fhuF03axMcI1g5K4ddip1KtV9SItBNriB4E3RKQ1JursJSKSDFwOnBdgHbOAkVjvZ5YyoqrMXbGTRz9OIys7x5V+ersEpo5O4bQW9SPYuooRkABVdb6IXIqx6SvATMqsAC5Q1c8DvNcO4EERGQisAnLdM1X1qYBbbak2bPj1N+77YA0/bN3vSmtaL467z+3ERT1PrlLDTW/4FaCI1ADOBr5X1cEVuNckIBtjE+gZdE4xO2IsFgCyT+TxzOcb+c+36S6HSCIwvl9b/nz2aTSsXTPCLQwOfgWoqnkiMhdIxrtPl4BQVTvXbvGLMZD9lakL1vHr4SIbve6tG/HQqBS6JjaMYOuCT6DvgD8DHSg5CVMuyrubxhLdbNmbzf3z1vL1pixXWsPaNblrZDK/79OamCgMQBOoAB8AnhSR+zGznsXCxqjqfm8XuRPM3TSW6OJ4bj7PL97MS19uKWYge2nvRO4amUyTetHrRD1QARZGg51L8UVzcc4D2W4QlN00lujii7TdPDC/pIHswxelkNo2IYItCw+BCnBIEO5V4d00luhhx35jIPt5WpGBbL1aNZgyvCNX9m/rMpCNdgJdhvgyCPcKxm4aSxXnRF4+//56K88tKm4ge2H3VtxTCQxkw03Af/jOxMlNQGfMsHMt8IKq7i71wiKCsZvGUoX51jGQ3eJmIHuKYyA7oJIYyIabQM2RBmKMaXcD3znJV2Cc645Q1e98XlxEMHbTWKoguw8fZ+qCdSxYtcuVVrtmLLcM68CkQe2Jq1E9hpveCLQHfAKzlez6wqUDEYkBXgSepOTCegmCtJvGUoXIyy/g1aXpPP35pmIGsmd3Pom/XtCZxMaVz0A23AQqwB7AVe7rdqpaICJPASv9XRzE3TSWKsKy9P3c+35xA9nWCcZAdmhy2Q1ko5VABXgISKKkW8Ek4GDJ4sUJ1m4aS+VnX/YJHv14PXOWexjInnkKN555SrkNZKOVQAU4G3hFRO7ERDJSYBDwd8zQNBCCupvGUrnIL1Bm/bCdxz8tbiD7u47NePDCLlXGQDbcBCrAOzGL7v9xuyYXeAH4vwDreIAK7qaxVE5WZRzk3vfXsCqjKFpuiwbx3H9BZ0amVC0D2XBTJtf0IlIHOAUjxs2qerQM17ovvpfYTaOqUT828eaavipz6Gguj3+2nv9+v91lIFvDzUC2bhU0kA0RFXNNLyItgBqqmgGsdktPBHIDXAsMxm4aSyVAVXl3xU4e/SiNfUfcDGSTEnhodAodT6q6BrLhJtB/UTOBt4HpHukjgMswM5ylEqTdNJYIs/7Xw9z3/hp+TD/gSmtarxb3nJfM6B5V30A23AQqwD7AzV7SvwYeD/RmItIVE5L6FMy+0F0iMhpjDeF3OcMSObJP5PH0wo3MWFpkIBvjGMhOiSID2XATqABrAN5sQuJ9pJdARM4G5gEfA0MxJklgxHgVMDrAtlRvXh8NB7dDozZwZehDaqgqH67exdQF69h9+IQrvXvrRjw8OoWUk6PLQDbcBCrA7zF2fDd4pN8E/BhgHVMx8QH/JSLu0ZCWAH8OsA7Lwe2w/5ew3MqbgWyjOsZA9rLe0WkgG24CFeA9wCIR6Q584aQNBXoCZwVYRxe8uyXcD0S/4VcV4liOMZB9+aviBrKX9W7NXeckk1A3LoKtiy4CNUf6n4j0x0QxupiiwCo3qurPAd7rAHAyJRfiewEZJUpbIsLn64yBbMaBIgPZzi0bMHV0CqltG0ewZdFJwAs1jtAqEhnyTeBxZ0O2AjVEZDBmo/eMCtRrCQLGQHYtn6cVxUevX6sGU87uyPh+1cdANtwEug7YGchX1Q3O+XBgAsbG7x+qmh9ANfcCr2KMcgVY53y+ifEVY4kAJ/Lymf7VFp5btJkTeUXDzVE9WnHPuZ1oXs0MZMNNoD3gK8AzwAZn8f0DzOTJTRiX83/xdpGItAF2qCEX+IOI3IcZdsYAK1V1U8UeoZqQnwc/z4LDO8354Z0mVnyPy8sdH/6bTVn89YM1bMkq2hXYoXk9/jaqCwNOqZ4GsuEmUAF2wrzzAVyCMSs6V0SGYIaPXgUIbMVEwd0jIouAi1V1C7ClAm2ufuTnwZyrIG1+UVrecZh3M2z6FMa+auLHB8ivh44z9cN1fOhhIHvrsFO5ZlBStTaQDTeB/tZigcI9R8Moms38BSjNuOs3oCmwBzgTsKu15eHnWcXF507afPjf89D/FogpXTi5+QW8tjSdaQs3ciSn6K1hRJeT+OsFXTi5UWiDUVpKEqgA1wA3iMgCjAALe7yTgSyfV8HnmOWLNOf8PRHJ8VZQVYcG2Jbqx8qZpecv/Css+Ts06QDNToOmp0GzjuYzoT3UiOPH9P3c+94aNuwuWoJtk1CHBy/swpDk5iF+AIsvAhXgXcD7wO3Aa6pauCH7QoxfT1+MB67G2AEOxhj0BmxBYXE4tNN/mdyj8Osqc7ihEsvemiez/1hzLtRWbI45mW0xiQwdNIhJw7paA9kIE7A5kojEAg1U9YBbWjvgqKru8XGNaxJGRBYDF6mqXwv6aKXc5kivnA07vvedX7eZ6en2boDjZfh6GyQW9ZSuz9Ogrp2ACTIVM0cCcJYaDnikpfu5zDUJgw1DXX56ji9dgGc9AD2vAFU4spfN65az8MuviD/0Cx1kJx1iMmkpXuydD2eY45dFxdNrJzhD2Y7Fh7QNEv2+Z1rKRqgtJt0nYQZjJ2HKR4/LzWynt4mYThdA93EAHDqWxz8W7ubNHwpQHQQMMgayA5K4deBJ1D28FbI2mJ4ya6P5PLAVPGPkHNsP278zhzs160DTU0v2mAntIdb+astDmSziy1y5yByM75g0jACXUjSbWozqMAlTIYv4/DxYNZsTH9xGLXI4QRy1Rk2D7uMoIIZ3V2Tw94/XFzOQ7esYyJ5amoFs7nGzudtdlFkbIWsT5J/wfZ07MTWMCD17zKYdIc76gqGUIWioBVibokmYP2HWDL1OwqjqLSFqw42YPawtMTt3Jqvq16WUH4wJFtoFyMTs9HmxInUWEgyXFDse7ERrzWSHtKL1/Wmk7TIGssu2FTeQvfe8Tozq0ar8BrIF+XBwG+zd6PSabp8nDvm/vpCGrd2E6SbQuk3K165wEVyzr4q/A/qsWaSDqm72lqeqx4DnnXI9gD+HcxJGRC7D7OC5EfjG+fxYRDqr6nYv5ZMwa5z/wex7HQT8S0T2quq75akzWOTlFzB3xU76FBSAQF5BAeP//T1Lt+wrZiB7Zf92TDm7Iw3iKzgkjIk1vVpCezhtZFG6KmTv9ugxHWFm/1qynkM7zPHLF8XT6zTxGMo6nw0TTSjcSBMms69y9YAiEg+MBf4IDKqsDpVE5Htglar+0S1tEzBHVUvs3nFiFV6sqqe6pf0b6KKq/ctTpzvl7QHz8gu4+c2VfLL2VxbFTaF9zK9sKWjB0JyiqN492zRi6qgIG8geO2iGriXeM9MJeA6uZl3znunZYyYkhfc989leRoAJp8CtK/yXL53g9IAi0hMT630ccAJ4D+Nu0Ff5Z4G/qOoR52efqOqtZWmLP0QkDkjFWFu48xm+Xen3d/Ld+RSY4AQYlXLUWWHmrtjJJ2u99C4OY1JP5vEx3SNvIFu7EbTuYw53co/Dvs0lh7L7NkG+x5RA7hHY9ZM53ImpaXrjZh2hWXJRr9nkVIirui7u/QpQRBoCf8D0dh0wG7HrY3q+dX4u70rRzGfXUsqF4kW0KWYLnafHtt34NiJugdm941m+hlOflKNOF5mZmTz44INe884//3xSU1MBWL58OQsWLCiWP9HZJaZaiy0FLcjQZgBcUGsdDdYtY+q6D0rU2atXLy644ALXvadP9/SpVcQf//hHWrVqBcD8+fNZscL7f/2WLVty7bXXus59PU+xZ6oZz/KdJ1iwYI2TkwQkITqcRhyiGfsZd1Yvp8dcT07mGuLUYwKoINcIN2tDsdlgBQ7SgCwSyKIJe0kgiwT2ksD4a28t9zPdrPtoAuzbv49/uj2jv9+TO/fff7/PvEJKFaCIzMT4avkeeBozzDrivAf5RVWHePs5zHiKW7yk+StfmC6llPFap4hcC1wL0LBhxYeHb8gYZuT0rnA9lQGVGA7QmAM0hkGTXemvvvQS2bs20ZT9NGM/TdlPU/bRjP3U85jDE6Axh2nMYU71sPXOn/k2NO8EzTqStCeXg3qUvSTwG/V8v2fm59FD19AAs2WvAb/RQ9fwM51RCf4aaKnvgCKSBzwGPOGxAyYX6B5ADxgxnCHoUWCcqr7jlv48kOItSIyIfAWsVtWb3NIuwdgs1sH8vstUpzvlfQcc88JSlm874DO/d9vGzLkhZCPgysWxA27DWLf3zIPbCXggFVff+3tmw0SYO8n3emsZrU7cKPc74KXANUCGiHwGvA747nM97yryn0DLqurVgZYNsL4cEVkODAfeccsaDrzr47LvKOmdbTiwzLFnpBx1VpjLercuVYCX9mkdqltXPmo3hjZ9zeFOzlHnPdNjZnbfZjN8LVb2N8hcYQ53YmpAQR5eSZsPq2abHUdBpFQBqupcYK5jhDsRM/nwb4wxbU8RSdPSp1GbeZz/DhMbsHAzd4pT11flaHsgPAXMFJEfgG+B6zGhsl8EEJHXAVT1Sqf8i8DNIvI08BIwEOMycVygdYaCMamJLFq/x+tEzMguLRjTKzFUt646xNWBlt3M4U5+npmF9ewxszZCTnbxsr7EV8iKmUEXYJmXIUTkLMxM6CjgMDBXVT3dFXq77i8YL2oTVfWIk1YXY22/WlVD4pbCWTS/E7Novga4TVW/cvKWAKjqmW7lBwPTKFqIf8zHQrzXOkujIgvxefkFzF25k/veX8OJvAJq1Yhh6ugUxvRKJDbSs59VEVU4nFl8ZnblGyVnZd1pkAhT1pbnbsHfCSMijYErMR6uuwdQfhcwzPO9UUS6AF+oaotyNaQKEYydMEOeWMLWrCMkNa3L4tvPDE7DLAZ/Viet+8E1n5anZp8CLPe0jqoeUNVnAhGfQz3MUM2TlpgJDoslsvQcX3p+Lz/55cDfMsSUAOpQVZ0WQLl3gRkicgfwPyetH2aWdW4A11ssoSVAq5Ng4m8W9AmMy4lsfHejinln8scNwJMY14SFi/N5mHfA2wO43mIJLTGxZqlh1Wz48M/G8VWNeDjvSSO+cnqfKw1/AlwGdAY+BF5R1W/KeyNnY/aNTg/oHuTzSOlXWtxJbOXzgtQAAAqNSURBVFy72KclyMTWMDOdXz9l9oI2ODnoM5/u+FuGON2ZJLkGsxxxANNjvRZgUE5vdR4BVvktaPHKzGv6+i9kqTL4nYRR1bWqOgXjAe0ejHvBdBH5QEQCCk1msVi8UxafMLnAHBE5jJm1PA8T4y9As2mLxeJJoLEh2mEs2yc4Sa9jFtSrrYczS5TTqE3xzxDhbxnicsz7X39gPia89Kd+tp/5qsvlotAjXYDWobQmt1jKTBiiD4P/HvANYDvGFCkLMyPa2dPPiKo+VfLSEri7KHQnwcmrlFb1Fkso8SfAQhuP0lYgFbNB2R++bObqAccDuN5iiTr8LUO0q+gN3FxRKPCoiLhbVMYCpwM/lbjQYqkGhNoxLxS5ohBMmDP37eY5mLBnnj5WLJZqgb9JmHOAFzDW74c88hoCPwPXqqqnIyMXha4oRGQG8CdVPVzhVlssUYK/hfibgcc9xQfgpD2GcbjrF1WdqKqHRaSpiPS1i/gWi38BdqOklzB3FgEBmSOJSD0ReRszC7oUs7MGEXlRRB4IpA6LJdrwJ8BmGBcSvlAgUB/j/8CIrhdwzC19AXBRgHVYLFGFPwFmYHpBX3QDAogeCZhgnpNV9SeKL0ekAe0DrMNiiSr8CfBDYKoTZKUYIlIH+JtTJhAaA/u8pNcH8r2kWyxRjz8BPgw0BDaJyF0iMso5/g/Y6OQ9EuC9fsT0goUU9oLXYd4JLZZqh7+F+D0iMgCzFPEIxT1DfwrcWAa7wLuBTx37whrAFOfn0zHuCi2WaoffhXhV3Qac63hB64AR4SZ3T9mBoKpLHTHfDvwCDMMswvdX1dWlXmyxRCllsQc8gBlGlhtHaBP8FrRYqgnh2IpWDBFpBTTH4/1TVSschM1iqWqETYBObME3gGRKelhTrDmSpRoSzh7wZWAHJs5gJqGJCWixVCnCKcDOQE9V3RjGe1oslZrgRxz0zWpMBFqLxeIQTgHeDfxDRM4SkZNEJMH9CGM7LJZKQziHoIVWFZ9R/P2v0FWFnYSxVDvCKcBIxYi3WCotYROgqn4ZrntZLFWFkL4DikhSGcqKiFSjYOcWS+gnYb4TkVdEpL+vAiLSWERuANZhwl5bLNWGUA9BkzEBXT4UkXxgObAL4we0MWZtsBPwA8ZYt1zxfy2WqkpIe0BVPaiqd2BcUdwArAcaAUmY4JyvYRbnB1rxWaojYZmEcYJzznEOi8XiEM6FeIvF4oEVoMUSQawALZYIYgVosUQQK0CLJYJYAVosEcQK0GKJIFaAFksEsQK0WCKIFaDFEkGsAC2WCGIFaLFEECtAiyWCWAFaLBHECtBiiSBWgBZLBLECtFgiiBWgxRJBrAAtlghiBWixRBArQIslglgBWiwRxArQYokgVoAWSwSxArRYIogVoMUSQaJWgCJSS0SeE5EsETkiIvNEJDGA624Uka0iclxElovIGV7KnC4iC0UkW0R+E5GlItI0NE9iiWaiVoDA08AYYBxwBtAAWCAiPkNhi8hlwDPAI0BPYCnwsYi0cSvTFxNmewnQD0gFngByQ/IUlqhGVNV/qSqGiDQE9gITVfW/TlprYBtwjq9ITCLyPbBKVf/olrYJmKOqf3HOlwKLVfWesrard+/eumzZsjI/j6XKI74yorUHTAVqYnoqAFR1B5AGDPB2gYjEOdd95pH1WeE1ItIc6A/sEpFvRGS3iHwtIsOC/wiW6kDYYsSHmRZAPpDlkb7byfNGUyDWKeN5zVnOz+2dzweBO4CVwCXApyKSqqo/e1YqItcC1zqn2SKyoQzP4YumlHw2S/AJ1vf8iaqO9JZRpQQoIg9hIu6WxpDSqgD8jbk9892vKRwxvKSq/3F+XikiZwLXY4KQFq9M9WXgZT/3LBMiskxVewezTktJwvE9VykBYiZW3vBTZjtmciQW8x9sr1tec+ArH9dlYXpNzx6yOUW94i7nc51HmTSgDRZLGalSAlTVLAIYEojIcsys5HDgTSctEROPfqmPunOc64YD77hlDQfedX5OBzKB0zwu7wisDvQ5LBYXqhqVB/ACsBPz/tYTWAz8BMS6lVkP3Ox2fhmQA0zCiPUZIBto61ZmMnAI8+7XAbgbI/buYXy2ayP9/VaHIxzfc5XqAcvIbUAe8BZQG/gCuFJV893KnIYZpgKgqm+JSBPgXqAlsAY4V1W3uZV52pkxfRJoAqzFLG2UmIAJFWreKy0hJhzfc1SuA1osVYVoXQe0WKoEVoCVCBH5i4j8KCKHRWSviMwXkZRItyvaEJFYEZnqtud3q4g8JCJhfyWzAqxcnAn8C7PzZijmHfZzEUmIZKOikLuAm4BbgWTgT875X8LdECvASoSqjlDVGaq6RlVXA+OBZsDAwjIiki4ifxWRVx1LjB0icpmINBKR2Y6FxiYROTtiD1L5GQDMV9X5qpquqvOAeUDfwgLh+p6tACs39TG/owMe6ZOBH4BewNvAa5j1zo+AHpjNBm+ISHz4mlql+AYYIiLJACLSGTPi+MijXOi/50ivtdij1HWotzH7Td3XLtOBWW7n9TBb5Z51S2vnpPWO9DNUxgOzvfBhoACzhqvAQx5lwvI9R/M6YJVGRJ4CBgGDtPjaJcCqwh9UNVtEjlJ8J07h1rnmoW1lleUy4Ergcsw6bg/gGRHZqqqvuJUL+fdsBVgJEZFpwO+BIaq6xUsRT+Nf9Ujz3DxuKc7jwBOqOts5Xy0ibTGTMO4CDPn3bAVYyRCRZzDiO1NV10e6PVFKHczGe3fyicA/LCvASoSIPI+Z+RwNHBCRQsuMbFXNjlzLoo75wP+JyFbMELQnMAV4PdwNsQKsXNzofH7hkf4g8EB4mxLV3AJMxay5NseYmU0H/hbuhti9oBZLBLEv6RZLBLECtFgiiBWgxRJBrAAtlghiBWixRBArQIslglgBWiwRxArQUiFEpKeI5IvIt5FuS1XECtBSUf6I2VGSIiKdIt2YqoYVoKXciEhtjEnPdGAOcI1Hfl8RWeH4XVkpIueKiDqu/AvLdBaRDx2r8z0iMsttD2zUYwVoqQhjgW2qugqYCVwpIjUBRKQesADj/DgVuBNjBuRCRFpirMrXAKdjnCjXA+aJSLX426wWD2kJGZMwwgP4EjgKXOic/wETn+MaVV2rqgsxVuju3AD8rKp3qWqaI+QrgT5AtQg+YwVoKRci0gHjLOpNADW7+v+LESUYb2NrVPWY22Xfe1STCvzOcXCULSLZwA4n75SQNb4SYc2RLOVlEqaH2y7iCgAr4IpGHEgouBjgQ+B2L3mecRqjEitAS5lxHNhOwLhwWOCRPROYiAnZdqWI1HbrBU/3KLsCuBTzHunp/qFaYIeglvJwHiaozXQ1PkxdBzAbuBqYhXHzMN2Z6TwLE0kKinrG54GGwFvOjGl7ETlLRF4WkfrhfaTIYAVoKQ/XAItVdZ+XvHeAtpggqRcAXTCuFR+nyKr/OICqZmLeIwuATzDuIZ4HTjhH1GMt4i1hQ0RGAe8BzdUEW6322HdAS8gQkQnAFszMZgomxPh8K74irAAtoeQkjEOplsCvmBnPuyLaokqGHYJaLBHETsJYLBHECtBiiSBWgBZLBLECtFgiiBWgxRJBrAAtlgjy//8cpuJITq52AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(3,3), constrained_layout=True)\n",
    "ax.spines['right'].set_visible(False)\n",
    "ax.spines['top'].set_visible(False)\n",
    "\n",
    "sns.pointplot(x='new_age', y='net_score_group', data=barcode[barcode['broad.cell.type'] == \"oligo.\"], hue='Genotype', order=['2m', '8m'] , hue_order=['WT', \"L2AKO\"])\n",
    "\n",
    "plt.axhline(y=0, linestyle='dashed', linewidth=2, color='gray', zorder=1)\n",
    "\n",
    "#plt.legend(title='Cell type', loc='upper left')\n",
    "plt.ylabel('CMA score \\n(net difference)', fontdict={'size': 14})\n",
    "plt.xlabel('Age', fontdict={'size': 14})\n",
    "\n",
    "plt.yticks([-0.06, 0, 0.06])\n",
    "\n",
    "plt.ylim(-0.06, 0.08)\n",
    "plt.xlim(-.2, 1.2)\n",
    "\n",
    "plt.savefig('./plots/oligo_net.png', dpi=300)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plots with 3 groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_plots(cellpop): \n",
    "    \n",
    "    fig, ax = plt.subplots(figsize=(4,3), constrained_layout=True)\n",
    "    ax.spines['right'].set_visible(False)\n",
    "    ax.spines['top'].set_visible(False)\n",
    "\n",
    "    sns.pointplot(x='new_age', y='net_score_group', data=barcode[barcode['broad.cell.type'] == cellpop], hue='Genotype', order=['2m', '8m'] , hue_order=['WT', \"L2AKO\", 'PD'])\n",
    "\n",
    "    plt.axhline(y=0, linestyle='dashed', linewidth=2, color='gray', zorder=1)\n",
    "\n",
    "    plt.legend(bbox_to_anchor=(1,1))\n",
    "    plt.ylabel('CMA score \\n(net difference)', fontdict={'size': 14})\n",
    "    plt.xlabel('Age', fontdict={'size': 14})\n",
    "\n",
    "    plt.yticks([-0.06, 0, 0.06])\n",
    "\n",
    "    plt.ylim(-0.06, 0.08)\n",
    "    plt.xlim(-.2, 1.2)\n",
    "\n",
    "    plt.savefig('./plots_3groups/{}_netscore.png'.format(cellpop), dpi=300)\n",
    "\n",
    "    plt.show()\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {},
   "outputs": [],
   "source": [
    "cell_to_plot = ['excit.', \"oligo.\", 'astro.', 'microglia', 'inhib.', 'OPCs']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in cell_to_plot:\n",
    "    make_plots(cellpop=i)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CMA component heatmap"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### *Trial function on excitatory neurons*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [],
   "source": [
    "neuron_matrix = np.zeros((18, 6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:2: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:4: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  after removing the cwd from sys.path.\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:5: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \"\"\"\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:7: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  import sys\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:8: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "wt_2m_index = list(barcode[barcode['broad.cell.type'] == 'excit.'][barcode['Genotype'] == 'WT'][barcode['new_age'] == '2m'].index)\n",
    "wt_8m_index = list(barcode[barcode['broad.cell.type'] == 'excit.'][barcode['Genotype'] == 'WT'][barcode['new_age'] == '8m'].index)\n",
    "\n",
    "ko_2m_index = list(barcode[barcode['broad.cell.type'] == 'excit.'][barcode['Genotype'] == 'L2AKO'][barcode['new_age'] == '2m'].index)\n",
    "ko_8m_index = list(barcode[barcode['broad.cell.type'] == 'excit.'][barcode['Genotype'] == 'L2AKO'][barcode['new_age'] == '8m'].index)\n",
    "\n",
    "pd_2m_index = list(barcode[barcode['broad.cell.type'] == 'excit.'][barcode['Genotype'] == 'PD'][barcode['new_age'] == '2m'].index)\n",
    "pd_8m_index = list(barcode[barcode['broad.cell.type'] == 'excit.'][barcode['Genotype'] == 'PD'][barcode['new_age'] == '8m'].index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/18 [00:00<?, ?it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "  6%|▌         | 1/18 [00:00<00:03,  5.15it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 11%|█         | 2/18 [00:00<00:02,  5.49it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 17%|█▋        | 3/18 [00:00<00:02,  5.69it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 22%|██▏       | 4/18 [00:00<00:02,  5.51it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 28%|██▊       | 5/18 [00:00<00:02,  5.53it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 33%|███▎      | 6/18 [00:01<00:02,  5.50it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 39%|███▉      | 7/18 [00:01<00:01,  5.69it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 44%|████▍     | 8/18 [00:01<00:01,  5.90it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 50%|█████     | 9/18 [00:01<00:01,  5.79it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 56%|█████▌    | 10/18 [00:01<00:01,  6.13it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 61%|██████    | 11/18 [00:01<00:01,  6.46it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 67%|██████▋   | 12/18 [00:02<00:00,  6.03it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 72%|███████▏  | 13/18 [00:02<00:00,  6.44it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 78%|███████▊  | 14/18 [00:02<00:00,  6.17it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 83%|████████▎ | 15/18 [00:02<00:00,  6.14it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 89%|████████▉ | 16/18 [00:02<00:00,  6.15it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 94%|█████████▍| 17/18 [00:02<00:00,  5.74it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "100%|██████████| 18/18 [00:03<00:00,  5.98it/s]\n"
     ]
    }
   ],
   "source": [
    "for rank in tqdm(np.arange(neuron_matrix.shape[0])):\n",
    "    \n",
    "    neuron_matrix[rank, 0] = np.mean(cma_network[wt_2m_index, :].X.todense()[:, rank])\n",
    "    neuron_matrix[rank, 1] = np.mean(cma_network[wt_8m_index, :].X.todense()[:, rank])\n",
    "    neuron_matrix[rank, 2] = np.mean(cma_network[ko_2m_index, :].X.todense()[:, rank])\n",
    "    neuron_matrix[rank, 3] = np.mean(cma_network[ko_8m_index, :].X.todense()[:, rank])\n",
    "    neuron_matrix[rank, 4] = np.mean(cma_network[pd_2m_index, :].X.todense()[:, rank])\n",
    "    neuron_matrix[rank, 5] = np.mean(cma_network[pd_8m_index, :].X.todense()[:, rank])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [],
   "source": [
    "neuron_matrix_zs = neuron_matrix.copy()\n",
    "\n",
    "for i in np.arange(neuron_matrix_zs.shape[0]):\n",
    "    \n",
    "    µ = np.mean(neuron_matrix_zs[i, :])\n",
    "    sd = np.std(neuron_matrix_zs[i, :])\n",
    "    \n",
    "    neuron_matrix_zs[i, :] = (neuron_matrix_zs[i, :] - µ) / sd   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6, 6))\n",
    "plt.imshow(neuron_matrix_zs, cmap='viridis', vmin=-1, vmax=1)\n",
    "plt.colorbar(shrink=.5, label='Gene z score')\n",
    "\n",
    "plt.yticks(np.arange(18), model_matrix['Gene name'])\n",
    "\n",
    "plt.ylim(17.5, -0.5)\n",
    "\n",
    "plt.xticks(np.arange(6), ['WT 2m', 'WT 8m', 'KO 2m', 'KO 8m', 'PD 2m', 'PD 8m'], rotation='vertical')\n",
    "\n",
    "plt.savefig('./heatmaps_3groups/ex_neurons.png', dpi=300)\n",
    "plt.savefig('./heatmaps_3groups/ex_neurons.pdf')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 183,
   "metadata": {},
   "outputs": [],
   "source": [
    "neuron_matrix_2group = neuron_matrix[:, :-2]\n",
    "\n",
    "for i in np.arange(neuron_matrix_2group.shape[0]):\n",
    "    \n",
    "    µ = np.mean(neuron_matrix_2group[i, :])\n",
    "    sd = np.std(neuron_matrix_2group[i, :])\n",
    "    \n",
    "    neuron_matrix_2group[i, :] = (neuron_matrix_2group[i, :] - µ) / sd   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAN4AAAF+CAYAAADtDKTtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO2dedxd073/3x9BjCExExrFNTamFKU119BrrCJRLZ3S9mq1lKpbv9vc9rq02uqgA22JukoNpWgVRUpLEEQGSgQlooKgiESS5/v7Y62T7JycYZ9n77P3fp5836/Xfj3nrLXX96yTPN9nDXt9P1+ZGY7jFMtyZXfAcZZF3PEcpwTc8RynBNzxHKcE3PEcpwTc8RynBJYvuwNVYvCQ5WyDodn/SVZRPn/Pnpi0Si52ADRwxcw23p7/Ou8sfFs5dGeZxx0vwQZDl+c3N62X2c7wFVfKoTdw4IY75GIHYPmhwzLbuGfGZdk74gA+1XScUnDHc5wScMdznBIo1PEkvdmi7hFJV9SVjZU0R9LqibIfSjJJa8f3CyVNlDRF0tWSVpG0saQ7JT0maaqkL3XvWzlO51RixJO0NaEve0pata76SeDweN9ywD7A84n6t81sBzPbDngH+BywAPiKmW0N7AacJGmbLn8Nx0lNJRwPOA64DLgVOKyu7grg2Ph6b+BvBMdqxN3A5mb2gpk9BGBmbwCPARvl3GfH6TVVcbxjgd8SnGxUXd00YB1Jg2PdlY0MSFoeOBiYXFc+DNgRuC/XHjtOBkp3PEnvBV4ys38AtwM7RSdL8jtgJLArYVRLsrKkicAE4FngVwnbqwHXAl82s381+fzRkiZImvDa7J5cvpPjtKMKD9BHAVtJeia+HwQcBfwycc+VwEPApWbWIy1xeOJtM1vqSbOkFQhOd7mZ/a7Zh5vZRcBFANsMX9Gjgp1CKHXEi5slRwPDzWyYmQ0jbKQsMd00s2eBrwM/TWlXhJHvMTP7fq6ddpwcKHrEW0XSjMT77wPPm1lyl/IuYBtJGyQbmtmFHXzOHsDHgMlxGgrwn2b2x9502nHyplDHM7NGI+z36+5ZCNSc7sQmdoYlXq/WoP6vgB/mdSpL6ZsrjrMs4o7nOCXgjuc4JVCFxwmVYfqr63HUVadktjPt+J/l0Jt8efyk9TPbmPv9FXLoiQM+4jlOKbjjOU4JuOM5Tgm44zlOCbjjOU4JFO549VHokk6UdEF8vaWkcTGi/DFJF8XyvSW9LunhWP6NOhs/lPR8PPtZK1tD0o0xsn2qpE8U8f0cJw1VG/F+BJwfI8q3Bn6cqLvbzHYERgDHS9oZFh20PhJ4Dtgzcf9JwKNmtj0hgPZ7krKLSzpODlTN8TYAFh2iNrPJ9TeY2VvAg8BmsWgfYArwM5aMajBg9RipsBowm+aR645TKGU43spxKjkxRg58M1F3PnCHpJslnSJpzfrGktYi6KhMjUWjCJHr1wGHxDg8gAuArYGZhKj0L5mZR7o6laAMx6uJE+0QA1j/q1ZhZpcQnOVqwvRwvKSBsfoDkh4m6LKca2ZT49TxQ8D1McL8PuCAeP+BwERgQ2AH4AJJg+o7k4xA73nzrW58X8dZisodGTOzmcDFwMWSpgDbxaq7zeyQutsPAtYgxN0BrALMAf4AfILgoAY8KelpYCvg/rrPWxSBPnDjjT0C3SmESq3xJB1UmypKWh9YiyWl/OoZBXw6Eb2+KXCApFUI+iv7RVvrAVsCT3Wx+46Tmko5HmGaOEXSI8AtwOlm9s9GN0bnOpAwugGLNl7+ChwKfAvYXdJkgojSGWb2cpf77zipKHyqWR8xbmZjgbHx9anAqQ3ajAPG1ZXNAYY0uPfDibcH1Nc7ThWo2ojnOMsE7niOUwLueI5TApV7nFAmA2e8xbu/em92Q8dnNwEw/bu75WMI2Ghc9rMDL73hT1vywkc8xykBdzzHKQF3PMcpAXc8xykBdzzHKYG2jpdnxLikFSVdImlyjAzfO2F351j+pKQfKZGLS9Lykl6WdE6D/q0jab6kz9aVny3puVZ51x2nLLKOeJ1GjH8GwMzeA3yQEBVe68PPgNHAFvE6KGHrAOBx4JikQ0aOBsazdCbZG4FdMn4/x+kKWR2v04jxbQgHljGzWcBrwIiYkmuQmd0bw3h+DRyRMDMK+CEh4qD+4dYo4CvAUEmL8pyb2XgzeyHj93OcrpDG8fKMGH8EODxOHTcFdgY2BjYi4cDx9Uax/cqE8J6bqMuRLmljYH0zux+4ipBL3XEqTxrHyy1inBDgOoOQr/wHwD0EHZRGuexqxyQOAe6M0QjXAkdKGhDrRhIcDkK65vrpZluSEejzmddpc8fpFZmPjHUSMW5mC4BFWUEk3QNMA14FhiZuHUrQSoHgTHskcqSvRRA4+nOsW0/SR2PdhpK2MLNpHfR/UQT6IA3xM1FOIWRa43UaMS5pFUmrxtcfBBaY2aNxLfaGpN3i5snHgd9HjZT3A5skosxPAkZJ2hJY1cw2StSdQxgFHafSZN1cSR0xHlkXeEjSY8AZhDzlNT4P/BJ4EpgO3Ax8GLjDzJJzwN8DhxHSNF9XZ/9a4nRT0ncU8q2vImmGpDG9+oaO0wUUNhEdCFPNXbVfZju3zJyYQ29g8998Lhc7ABv9JXt0wsS//JA3XpvhueVzwE+uOE4JuOM5Tgm44zlOCXgEeoVZ5YX8/i7+c5fsS7P5E3x5lxc+4jlOCbjjOU4JuOM5Tgm44zlOCbjjOU4JuOM5Tgn0aemH+NmPJz5/dKLOpR+cytIfpB8+GuME9wC+rZAlFlz6wakw/UH6ocZqwFvAwmjfpR+cypLm5MrKUfKhxhDghvi6Jv1wDyHS/BIzey3ZOCH98C1gTYL0w5UEyYea9EMP7aUfPhvbjwKSCQ4ulzSPMEp+2cwWpvhOyf6NJoy0rMQqnTR1nF7T16UfIEw1hwObAKdJeleK77T4Q8wuMrMRZjZiBQa2b+A4OdDXpR+Stl+S9BCwK/CPrN/LcbpJn5V+aGQb2JEQve44labPSj8kprSXxzXog8BYM3sQXPrBqTYu/ZCgatIPw7/3H7nYAZi3Zvb/5+d+ej5zn3/OY4NywE+uOE4JuOM5Tgm44zlOCbj0Q4VZad+XcrM16JIhmW288HYOHXEAH/EcpxTc8RynBNzxHKcE3PEcpwTc8RynBLrmeJIWxsjwqTHa/NRa0GuMUDdJhybuv6kuIn0dSfMlfbbO7jOS1q77jEckPSRp98R9f5L0mqSbuvUdneKRdLGkWfFAfqN6RQWDJyVNkrRTou4ESdPidUJxvV6abo54tXCibQnR5h8CvpGonwF8vUX7o4HxtM7yWvuM7YEzCfnxapzHkmdBnf7BWJZUJ6jnYBarGIwmKBsgaQjh929XgjLBNyQN7mpPW1DIVDNGm48GvpCQbngEeD1GKTRiFPAVYKikjVJ8zCBCeFHtM28H3uh9r50qYmZ3AbNb3HI48GsLjAfWjAoHBwK3mdlsM3sVuI3WDtxVCnuAbmZPxanmuoni/4nXbcl7JW0MrG9m90u6CjgW+H4Ds7Xo+JUIMhT7dtovj0AvlgP3WdVemd1cJODBSfOmAnMTRRfFdNlp2Qh4LvG+pmbQrLwUij65ssTJdjO7WxKSPlB330jgqvj6SuBXNHa8t2NUPJLeB/xa0nbWQciF50AvlpdnL+S+W4Y2rV9hg+lzzWxEho9opmbQSuWgcArb1ZT0boIQ0ay6qrNZeq03CjgxRp3fAGwvaYtW9s3sXmBtYJ1cOux0CWOh9TS9cmAGQcenRk3NoFl5KRTieJLWAX4OXFA/GpnZrcBgYPt475bAqma2USLq/BzCKNjqM7YCBgCv5P8NnLwwYAELm145cAPw8bi7uRvwelQ4uAU4QNLguKlyQCwrhW5ONWvrrxUIgkaX0Xi6CGHU+318PQq4rq7+WsKU81uEPtci0pMKaAJOqKmMSbob2ApYLUaif8rMSvuHdgKGsTBD8LWkKwjCWmvH/9dvEH7HMLOfA38k7KA/CcwBPhHrZkv6FvBANPVNM2u1SdNVuuZ4ZjagRd04YFzi/Q0snoOPa3D/JGCbOHLKzN5I8Rn160anAhgwn95PKc2s1eMl4ozqpCZ1FxOEuUqnz5xckXQYcDfheZ3TRzFgvvU0vZYV+kw8XhwVb2h7o1N5lh33ak6fcTynf2BmvOMCW+54SWz1VZi/286Z7Ry4YQ6dAe6feXU+hoD/WLdZyon0TJsyJ7MNw0c8cMdzCsYQ880VAt3xnMJZ2PAQybKFO55TKGFXs89spncNdzynUHoQ79D08esygzueUzg9vsarluNJWo+Q7HI3QmzdO8B34uvfA0/HW182s/0T7R4BHk2eapA0FtgLeJ2wkXZSPEjtlIgh3ml+4GiZoTKOFwNkrwcuNbPjYtm7gMMIjrdUvr14z9aEEzh7Slo1pn6ucbqZXSPpAOBCYHi3v4fTmvA4wdd4VfoX2Bd4Jx50BcDM/mFmP27T7jjCAexbCU7aiLuAzXPppZMJszDiNbvSEPMyPh51Vb7WoP78qMUzUdITkl5L1C1M1JV2EqoyIx6wLfBQi/oPJCIRrjazs+PrYwmaLlsCXwCuaND2UGByI6PJCPSBK63Zi247ndKT4XFCTMP9E8L/+QzgAUk3mNmjtXvMLJl1+IuEhKU1FgVPl0mVHG8JJP2EkA32HeB0Gkw1Jb0XeMnM/hFDRC6WNDhqagCcJ+ks4CXgU40+JxmBvvqgoX6WqcuENV6mX7tdgCfN7CkASVcSdFYebXL/KJYU2aoEVZpqTgUWSbGZ2UnAfrSOKB8FbBUj1acTBI+OStSfHlXIPmhmDeXgnGIJz/EGNL1SkFo7Je4RbArckSheSdIESeMlHdHLr5GZKjneHYR/lM8nypqqD0XhpKOB4YlI9cNpLQfolIwhFrJc04sQ4DohcY2uM9GJdspI4JpacHRkk6jpchzwA0mbZf5SvaAyU00zs/gX6HxJXyVMD98i5EpvxJ7A82b2fKLsLkLA7Abd7a3TW8KI1/LX7uU2YkedaKeMpC4o1sxmxp9PSRpHWP9Nb93r/KmM4wFEbYxm2irj6u4dR3jelyxbSJD5Azgx3945eWCIhdkeoD8AbCFpU+B5wu/LcfU3Re2ewcC9ibLBwBwzmxfVyPcgPCcunEo5ntP/MWs74rVpbwskfYEgVDQAuNjMpkr6JjAhBkxDWHJcWSeutTVwoaQewjLr3ORuaJG44zkFo0yPEwDM7I8EUaNk2X/VvR/ToN09wHsyfXhOuOM5hWKQ9XFCv8D/BRJstuksrr2k3UGZ9gwekI8U/IEb5vecd7ntN81s451nBma2EQJh/aymO55TKAb0eDyeO55TLD7iBdzxnMJx6Qd3PKdgzMT8Hv+1838Bp1BCPJ6PeF1d5dbFPk2sxU5JGhfjqWrl18TyMZKeT5SfG8u/EGOvLJ44qP+c30u6t65sT4W86AskfaSb39NJjyHm9wxoei0rdHvEaxX79FEzm9Cg/Hwz+25d2d+Am2iQ0ETSmoSohjclbWpmNXmIZwnHxk7rTced7uCbK4E+MdU0s4cBFqdPX4KjgBuBFwnn9s6JbZ6JbVy4uGK49EP3w4JWrptqHpuouzxRfl6i/JRE+YEpPmMUIer8CnoREiRpdC0E5ZVX3Ee7jRnM71mu6bWs0Femmg2JqmSbA3+NYUULYg701EGvyQj0HbZf0SPQu4whf4BOtQJhe8OxhNCPp2MU+jDapGx2yqWmJN3sSkMKsaMTJb2UmDl9OlF3gqRp8Tohv2/WGX1ijdeCUcBBNb3MGKN1G3BWqb1yWpBtxEsjdhT5rZl9oa7tEIL+ygjC34AHY9tXKZii13jnJuqSa7w/tzIi6eQoZjQUmCTpl5KGAZsA42v3xR3Nf0naVdJ7Y5ujCTFYU/P+ck7nhHi8TCPeIrEjM3sHqIkdpeFA4DYzmx2d7TbgoF59kYx0dcRrlqPczPZuUj6mSfmPgB81qFpK5MbMdkq8Hdq2k06hGGJB6+d1a0tKrv0viuvwGo3EjnZtYOcoSXsCTwCnmNlzTdo2FErqNn19qun0QdqcXGmnuZJG7OhG4Ioo8fA54FKCYHInQkldpa9vrjh9DAMW9AxoeqWgrdiRmb1iZvPi218AO6dtWxTueE6xmOhpcaVgkdiRpBUJu9hLSLHXqcwdBjwWX98CHCBpcBQ+OiCWFY5PNRM8+ey6HP7FL2W2c9dPL2p/UwqOemxWLnYArt06uw2zudltAAsy7GqmFDs6WdJhwAJgNlFxzsxmS/oWwXkBvmlms3vdmQy44zmFEiLQuyt2ZGZnAmc2aXsxcHGmDuSAO55TKGFX01c47nhOsVi2qWZ/wR3PKZQ8ppr9AXc8p1B8qhkoKgJ9iqSrJa0Sy9+su+9ESRfE12MkLRW82sJWs/KLJc2S5Om5KoaZml59DUkrxzwNHdHtPz1vx/x02xESTH6uC7aalY+lpHN4TnMsrvGaXX0JSYcCE4E/xfc7KGV65yK/6d3kl4e8ma1F5WZ2F+EZjlMx+tGIN4ZwaPs1ADObSAhNa0shazxJywMHE/8yEKMWErcMoe70QQe2WpY7VUMs7D9rvAVm9noTSZKWdNvxkg52N/Cr+HqJyHRJJxJipHpjq1l5KhQyjo4GGLjymp00dXqBGSzs6XMjWzOmSDoOGCBpC+Bk4J40DcuUfsjLVqbPSEo/rDZ4qEs/FEA/0tX8IvB1YB7wG8Ixtv9J09AfJziFYv1kqhkj4f/bzE4nOF9HVPVf4CxJM2pXbwxIuoKQhnfLaOdT+XbR6S1mza80pNBcOVXSo5ImSbpd0rsSdUmR5VT7Co2/gy1kcbhRx3Q7An21NOVmNpaw/V+LQh/TW1uJ8o6l/pzuYwY9GUa8lJorDwMjzGyOpM8T8pzXpCXzXP48HJ33auCtWqGZ/a5dQ59qOoWT8cjYIs0VAEk1zZVFjmdmdybuHw8cn+UDWzAEeIUQ3b7o4wF3PKd69LTe1cxLc6XGp4CbE+9XivYXAOea2fXper00ZvaJ3rZ1x3MKJQjadl1zJdwoHU94TLVXongTM5sp6d3AHZImm9n0dv1uYn8o8GNgj9iHvwJfMrO2+xLueAmWe20OK19/f2Y7mx3V6z+ESzB9v0tysQPw3f9Nq4DXnHcuGN/+pnYYWU+opNJNkbQ/Ybdxr4T+CmY2M/58StI4YEegV44HXEJ4jHB0fH98LPtgu4ZV3dV0+jHWo6ZXCtJoruwIXAgcZmazEuWDJQ2Mr9cmjFT1QridsI6ZXWJmC+I1FlgnTUN3PKdwsjxOMLMFQE1z5THgqprmStRZATgPWA24uu6xwdbABEmPAHcS1nhZHO9lScdLGhCv4wmbLW3xqaZTKGZgGR+gp9Bc2b9Ju3uA92T68CX5JHABcD5hjXdPLGuLO55TOGkflFcdM3uWIB/YMT7VdAqm+fou5RqvMki6VCEjce39YEmpFMwq43h1keQ3Jr9QrD9F0lxJa9SV7yLprniE6O8xockqkraSdK+keY0i2p2SsMybK1ViuJm9VnsTE6HsmKZhZRyPJSPJZwMn1dWPIuxoHVkrUEhMeTVwhpltSVg8/wlYPdo4GUiV5NIpEGtx9S2Wi4rUwKI0YKmWb1Vd490LDK+9kbQZYZfqdOA/iec6Cc55aS0/npkZcE3CzixJ/15Eh50O6HuR5s34HnCPpNrv3NHA2WkaVmnEAxYdgt2PJZ/N1PKc302INlg3lm8HPFhsD53M9JMRz8x+DRwFvAjMAj5sZpelaVslx6tFkr9COHx6W6JuJHClmfUQDqAe3aB9r5A0WtIESRPmM699Aycb/WiNF2di083sAmAysH/93kQzquR4tXCNdwErEtd4koYDWwC3KeQ5H0kYAQGmkiEmCkIEupmNMLMRKzAwiyknLf1kxAOuBRZK2hz4JbAp4QhZW6rkeACY2euETZHTJK1AcLIxZjYsXhsCG8XgxguAEyQtOp0eTxKsX0rnnVSoR02vPkZPPEnzYeCHZnYKsEGbNkAFHQ/AzB4GHiGMbiOB6+puuQ4YaWYvxvrvxscJjwEfIORBXz9Gr5/K4oj2QcV9C6chJuhpcfUt5ksaBXwcuCmWrZCmYWV2NRtEpR8aXy61WDWzUxOv7yU4Wz1z8Bzo1aTvTSmb8QmCgPLZZva0pE2B/0vTsJIjntPP6WlxpSCF5spASb+N9fdJGpaoOzOWPy7pwCxfw8weNbOTzeyK+P5pMzs3TVt3PKdYjDDdbHa1IaG5cjCwDTBK0jZ1t30KeNXMNiccYP52bLsNYWmyLUHe/6fRXuG44zmFo57mVwoWaa6Y2TtATXMlyeHApfH1NcB+CnLPhxMeS80zs6eBJ6O9wqnMGq8KrLPdPEZf91RmO2N+sXsOvYEDP5aXGBY8MfNnmW3scsVLOfQkM2k0VxbdE3Omvw6sFcvH17XdqLcdkfQB4J4o9Vcr28nMHmrX1kc8p3DaPE5Yu3agIV6j65s3MFm/XdPsntR6LSm5haDbsl6i7JdpGvqI5xSL0W4TpZ3YURrNldo9MxSS2axBODSfSq+lAx4nRLuPk/SpGGib6pmIj3hO4ciaXyloq7kS358QX38EuCMeoL8BGBl3PTclnIjKom5lZnYTIRj2AklfIOUI6iOeUzwpHxs0Iq7ZaporA4CLa5orwAQzu4GQMeoySU8SRrqRse1USVcRBI4WACcl12e9QNHuNEnvJ0TNDG/ZIuKO5xRKByNbU1JorsylyUF6MzublKE7KfqxY+L1HOAYSZukaVtkDvRFUeWShimRm7xJFPlJWpxc4h1Jk+Prc2ObIxSSUvw91h2RsDdW0tPx/kck7dfN7+l0SP85MrYUUYelLYXlx5N0KSHiYIm/Noko8pFmdm983nIUcLeZ/STe8wywj5m9HN9vT4gs/2DiqM5tkp4ys0nR9Olmdo2kfQj577bo8nd1UpJ1xOsPFLm5ci+Nn5ksFUVuZtfEA9DNOA343/gQlPjzHEKEetrPdcrAMj9A7xcU4nhNospr9CaKfNsGbSbE8noOAnqdmMLpAhnPalYFSetJ+pWkm+P7bZQyD2O3Ha9VVHkWROOHpsmy8yQ9RTgt/r9NDSUi0P81e0FO3XNakfFxQpUYS9hd3TC+fwL4cpqG3Xa8hlHldfQminwqIQtMkp1YUgf/dGBz4CwWn9tbimQE+qAhvslbCP0nAn1tM7uKOFbHoNhUjycKmWo2iCpP0pso8u8CZ9bCPeLP/ySoPiU/twf4IUGGLVMIiJMTLUa7PjjivSVpLeKfDEm7Aa+naVjYn3gze1ghWcRIglpYrfxFSbUo8nUJfz3uokVWTTObKOkM4MboyPOBr5rZxAb3mqT/Ab5KmBY4ZdPH1nItOJWwb7GZpL8RMgV9JE3DQnOgJ6LKIWyq1MqbRZHX6oc1KPsdTZzTzE6se38tQZjGKRnRJ0e2hpjZQ5L2ArYkfLXHzWx+mra+qHGKxfrdY4NdgGEEX9pJUk1vsyXueE7x9JMRT9JlwGbARBZvqhjgjudUj3404o0AtomRDx3hYUFOsdTi8brwAF3SEEm3SZoWfw5ucM8OClmkpsazvscm6pJnfCdKaicBMAXolYarj3gJXpoykIv+7d2Z7Uye+dMcegM7v/75XOwA/GnOY5ltvJ7TIeYubq58DbjdzM6N6mNfA86ou2cO8PEYyrMh8KCkWxLptk43s2tIx9rAo5Luh8X6/2bWNlmlO55TOF2cah4O7B1fXwqMo87xzOyJxOuZkmYRHgO8RueM6U0nwaeaThm0PrnSTnOlFeuZ2QsA8ee6rW6WtAvhRNX0RPHZcQp6vqSWyTTM7C/AM8AK8fUDQFuhI/ARzymYFCdUWmquSPozjddVX++oH9IGBJXyE+IJJ4AzgX8SnPEiwmj5zRY2PgOMJpxD3owQBfNzQkBAS9zxnOLJsMYzs/2b1Ul6UdIGZvZCdKxZTe4bBPwBOMvMFsn91UZLYJ6kSwjhZ604ifAc777YfpoW525sSeWmmpKOlGSStorv66PVPyPpIUk/iTtPj0p6O7ET9RFJR8ddqx5JrRSrnBLoYjxeUuToBOD3S312EEi6Dvi1mV1dV7dB/CngCMKuZSvmRVHdWvvl6cNiR6OAvxLOdI5JVkj6GPBFYN9ENPow4KZapHss25qQOunCQnrspKe7J1fOBa6KMXHPEnVX4h/fz5nZp4FjgD2BtSSdGNudGM/5Xi5pHcLxr4mEhCSt+Iuk/ySEv30Q+A/gxjQdrZTjSVoN2APYh/DXa0yi7hjC9vB+Nadrhpk9Ftt0ra9OBrr0OMHMXqHB+srMJgCfjq//jyYZfcxs3w4/8muEPA2Tgc8SBJj6pKDtEcCfzOwJSbMl7USQZ6slodzRzP5Zag+dzPSXkytxU+YX8eqIqq3xRhGSUBB/1lIuv0SYOhyT9wfKc6AXTl+Px5N0uKSTEu/vk/RUvBrKCtZTmREvBhTuC2wnyQhipQb8lHDa4GDgr5JmmdnleX2umV1E2DpmkIb0kf/6Pkx7Cfe+wFeJIrmRgcB7gVWBSwiqeS2pjOMRAgh/bWafrRVI+gsxq6uZvSTpIIJO/ctm5kGtfZB+Eo+3opklMxb9Na4vX5G0ahoDVZpqjmLpXOfXEiQdgEUyfocBFyelIuqJjyRmAO8D/iDJnbRCqMeaXn2EJQ5fm9kXEm/XSWOgMiOeme3doOxHwI/qyh4hoZNpZs+QiGaPZdextBM7VaB/BMLeJ+kzZrbEpoqkz5IyCUplHM9ZhugzA1tTTgGul3Qci89m7kxY6x3RtFUCdzyncPr6iGdms4DdJe3LYhHlP5jZHWltuOM5xdKHHhu0IzpaamdL4o7nFIro+yNeHrjjJdDKK7HcllvnYGkpec9eYQNyMQPA+cdlP3vw4tM5HX3tXKKk31GlxwnOskAXswWl0VyJ99XyNk6UdEOifNN4CmWapN/GSIau4I7nFI4WNr8yUtNc2QK4Pb5vxNtmtkO8kvoo3wbOj+1fJRyA7grueE7hdPGs5uEsTlBzKSm39mFRDN6+QE3oqKP2neKO5xSLtT25UoTmykrR9ngtTuG9FvBazPgDMIMuJjT1zRWneK9Rz+MAABRtSURBVMrXXNkkKoy9G7hD0mTgXx33NAOVczxJRxKSkWxtZn9PRJhvF+s/A3yekGJ5D4IwzabA49HE/wC7AocC7xAUpD6R0E10SkSW7UxmHporZjYz/nxK0jhgR8K54DUlLR9HvaHAzF53tA1VnGompR+WICH9cICZnRTlHj4ETE8slq8hZJ7dzsyGE7J0nllc9512dHGNl0ZzZXBNtk/S2oQ/3o9GGfY7WZxmq2H7vKiU4yWkHz5FneMlpB8OSCH9cGtirj6eGFrkVIMuih2dC3xQ0jTgg/E9kkZIqkkybA1MUMjVeCdwrpnVMgmfAZwq6UnCmu9XmXvUhKpNNbsh/fBJ4LfNKuPifTTASius0bteO+kxYGF3lk4pNVfuAd7TpP1TBLm+rlOpEY+cpR8kfR1YADSNWE/mQF9x+VU677HTMX1d+iEPKjPi5S39IOkE4BCCKtky9F9affpQwGvXqIzjkaP0Q7zvDGAvM5vT5X47nbA4R8IyTZWmmrlJPxDWg6sDt8XzeD/Pu7NO7xCghdb0WlaozIiXs/TD5l3ppJML8pl/dRzPWUYwA1/jueM5xeObK+54TtH0D5WxzLjjJZi79gAe/2T2h+jHPNU2L2Eq1vnZvbnYAbhlZvao+F0OfCWHnuBTTdzxnBLwzRV3PKdounhkrC9Rped4zjKAsBAa1OTKZDuF5oqkfRJ6KxMlza0Fw0oaK+npRN0OS39KPrjjOcXT09P8ykZbzRUzu7MWQkY4ojgHuDVxy+mJELN85OIa4I7nFIt19eRKp5orHwFuLuNYYaUcT9L6kq6UNF3So5LulDQnDvuzE9OAP0taTtKPJE2RNFnSA5I2TdjaUZJJOrDM7+Q0wKz5lY20mis1RgJX1JWdLWmSpPNrAbPdoDKbK1Hl6TrgUjMbGct2AFY3s7sljSVIQFwT60YBGwLDzaxH0lDgrYTJWiT7KMDTdFUFs3ZTyrUlTUi8vygmDwVy01whSkO8hyV/N84E/kmQE7mIcND+m53YTUtlHA/YB5hvZosONLeZY28AvBDzUGNmM2oV0Yk/QohCvlvSSmY2tzvddjqm9VKupdhRHporkWOA68xsfsL2C/HlPEmXAKe17GkGqjTV3A54sIP7rwIOjVPP70naMVG3B/C0mU0HxhF0WRqiRA70hW++1ew2J0fU09P0ykhbzZUEo6ibZkZnrf3hPgKYkrVDzaiS43VEHOG2JEwPeoDbJdWOjDSLZG9kZ1EE+oDVUmXRdbJghJMrza5spNFcISrXbQz8pa795VHqbzKwNkGxritUaao5lcUKT6kws3nAzcDNkl4EjohybUcBh0XpBwFrSVrdzN7Iuc9Ox7Rd4/XecgrNlfj+GRqI1ZrZvl3pWAOqNOLdAQyMupkASHqvpL0a3SxpJ0kbxtfLAcOBfwD7A4+Y2cZmNszM3kUIqO2aHLfTAQYs7Gl+LSNUxvGiLsqRhKnCdElTgTE0FxVdF7hR0hRgEkHU6AKaR7If141+O51iYD3Nr2WEKk01awq/DZXEzOzEuvd/Av7U4NYT6wvM7AbCwtspm9qIt4xTKcdzlhE8OsEdzyma7m2u9CXc8ZxiMdzxcMdbgoHPvcUWpzyQ2c7rPdlTmwJcNSO/CHRYOUdbGXHHc8dzCsYMW5jPH6a+jDueUzyuueKO5xSMGfiI547nlIA/TnDHc4rG13jQpSNjMfL7e4n3p0kaE1+PkfR8QlDm3MR960iaLymZMei+eN+zkl5KtBsmaTVJF9aOmEm6q5bMRNLFkmbFI2VOVehidIKko+PvQY+kpjF9kg6S9LikJyV9LVG+afx9mybpt5JWzNShFnTrrOY84MMKOaYbcX5CUCYpSHM0IXXyojAeM9s1CtP8F/DbRLtngF8SMsZuYWbbEo6L1T5zLHBQjt/JyQEDbOHCpldGpgAfBu5qdoOkAcBPCPkWtwFGSdomVn+b8Lu5BfAqISV4V+iW4y0ghM6f0mG7UcBXgKGSlgrbSCJpM2BX4KxEFPpTZvaH+PouglM6VSI+TuiG45nZY2b2eJvbdgGejL8r7xDiNQ+Pwa/7AtfE+9KIJfWabq7xfgJMkvSdBnWnSDo+vj7DzG6RtDGwvpndL+kq4Fjg+y3sbwtMNLNM/1tK5kDHUzF3mzd49ZY/91zVbCYEsFIrzZUc2Ah4LvF+BuEP+FrAa2a2IFHe8o9/FrrmeGb2L0m/Bk4G3q6rPt/MvltXNpIg5wDhr9CvaO14uRD/Uy8CGKQhvt3WZcws0/S/ldiRmbWSelhkolG3WpR3hW7vav4AeAi4JMW9o4D1JH00vt9Q0hZmNq3J/VOB7SUtV5tqOv2fVmJHKZlBkH2oMZQQ8/kysKak5eOoVyvvCl0NhDWz2YRRrOUiVdKWwKpmtlGMGh8GnEMYBZvZng5MAP47zs+RtIWkw/Pqv9MveQDYIu5grkj4HbshBmLfyWL5kXZiSZkoIgL9eyzeaWxGs6jxpiJFkU8Tph1PRpGaXxD/Skm6ArgX2FLSDEld26FyqoGkIyXNAN4H/EHSLbF8Q0l/BIij2RcIepqPAVeZ2dRo4gzgVElPEtZ8v+paX81PESxikIbYrgMOyG6ogtEJayyXPTphlwOfY8IjcxuthZwOqYzmiuMsS7jjOU4JuOM5Tgn4IekEmw5/k8v+2PS0UWrm5rRuPmbo+3OxA/DKZ96X2cbfZ56fQ08c8BHPcUrBHc9xSsAdz3FKwB3PcUrAHc9xSqAQx2sVkR7ff1whl/lUhdznp8VySTorRgQ/oZATfdtEu2cU8p9PkvQXSe9K1HkEulNZihrxmkakSzoY+DJwQIwi3wl4PVafBOwObG9m/0Y4OH2DpJUSJvYxs+GEzK9nJcrH4hHoTkUpyvFaRaSfCZwWMwVhZnPN7Bex7gzgi2Y2J9bdCtwDfLSBnXtJBC56BLpTZYpc4/0E+KikNerKG+Y+lzSIECo0va5qAiH6vJ6DgOs77ZQSOdBfecXD+pxiKMzxzOxfQC0iPQtiycjgOyXNImSC/U0v+rUoB/paa/lek1MMRf+m/YAQFLtqomwqsHP9jdFR35L07rqqnYBHE+/3Ad4V7Xwz1946Tpco1PGaRKSfA3xH0voAkgZKqo2K5wE/krRyrNsfeD91I5uZvU3YoPm4pCHd/RaOk50y5lZLRKSb2R8J678/x7znD7L48PaPCaH6kyU9Dvw/4PDoaEtgZi8AVxB2Qj0C3ak0hUQnmNlqidcvwpI6emZ2CQ0EkaIOxn/Hq5HdYXXvv5h43U42wnFKw3cTHKcE3PEcpwTc8RynBDwCPcHTk1bjYxvvkdnOLTMn5tAbmHfrsFzsAAw+e25mG8vP9QMGeeEjnuOUgDue45SAO57jlIA7nuOUgDue45RAVx2vN7nQJY2L+alr5dfE8lNjdPokSbfXos0VcqEvFWUu6TxJf4/3XydpzW5+V8fphG6PeL3Nhf7RRHktbdLDwIgYbX4N0CjTbJLbgO3i/U8QAm4dpxJ02/F6mwt9KczszlokOjCekDiw1f23JtLqtr3fcYqkiDVes8hzCLnQa1PKAxPllyfKz2vQ7lPAzR304ZMd3u84XaXrJ1d6kQsdwlRzQoNyJB0PjAD2SvP5kr5OGHkvb1I/GhgNsNKSQROO0zWK2tVsFHneMTEQ9uvAYWY2L8X9JwCHEBy5YSaRpPTDCgzM0j3HSU0hjpc2F3orJO0IXEhwulkp7j+IoFJ2WGJt6DiVoMhD0t8j5J5Ow+WSatPSl81sf4IMxGrA1ZIAnjWzw+I9W8bc1zVOIUhKDARui/ePN7PPZfwOjpMLXXW8VpHnZjamSZu9m5Tv36T8GWCFBlVXp++p4xSLn1xxnBJwx3OcEnDHc5wScMdznBJw6YcusNvp+Wyejj/v57nYATj5e+/NbGPy8dnlI5yAj3iOUwLueI5TAu54jlMC7niOUwK5O56khTGcZ4qkG+sjvyWdImluMkxI0t6SXo/tJkn6s6R1Y91Wku6VNK+WGz3Rrmmec0nLS3pZ0jl5f0fHyUo3Rry3Y+T4doRUyCfV1Y8iZAA6sq787thueKyvtZtNCClqFD40luZ5zg8AHgeOUTys6ThVodtTzSXykkvajHDQ+SyCAy5FdJLVgVcBzGyWmT0AzK+/t02e81HAD4Fngd16/xUcJ3+65niSBgD7ATckikcRctjdTYgoWDdR9wFJEwmOsj9wcYbPXjl+9k3x8zxll1MpuuF4K0cHegUYQhAdqjESuNLMeoDfAUcn6mpTzY0JufLaiRm14hCgptFyLXBk/EOwFJJGS5ogacJ82sbWOk4udG2NR8hLviKLM7QOB7YgxMc9Q3DCZiPRDcCeGfowCtg/fs6DwFqEXOlL4RHoThl0bappZq8TNkVOk7QCwRnGmNmweG0IbFTTx6zj/cD03nyupEGx/Sa1zyI4v083ncrQ7UDYhyU9QhjdRgIH191yXSy/j8VrPAGvA58GkLQ+MAEYBPRI+jKwTRRRugLYG1g7RqB/A1gI3FGnyfJ74DuSBqbRanGcbpO74yWjzuP7Q+PLyxrce2ribSP5P8zsnzTRxGyR53xs3X2zgXWa3Os4heMnVxynBNzxHKcE3PEcpwTc8RynBDwCvQvkFTn+7ms/m4sdgE1u7sls4+2Zd+XQEwd8xHOcUnDHc5wScMdznBJwx3OcEqiE42WMWn845jpfKlBW0u8l3VvEd3CcTqiE45Etan1HYEfgEEl71Cqi8+4ErClp0+513XE6pyqOl6TjqHUzexuYmGwHHAXcCFxJOIjtOJWhUo7Xi6j1WrvBhFi/uxq08wh0p3JUxfF6G7X+AUmTgH8CN8VIBiStB2wO/NXMngAWSNqu0Qd7BLpTBlVxvN5Grd8dVcneA3xe0g6x/FhgMPB0bDeMJtNNj0B3yqAqjgf0Pmo9jmrnEHKeE9sdlIhA3xlf5zkVolKOByFqHUhGrV9Xd0star2enwN7xh3MTYDxCZtPA/+StGtXOu04HVKJQ9IZotbHJcrfZvGuZnJ3s1a/U+aOOk5OVG7Ec5xlAXc8xykBdzzHKQF3PMcpAZlZ2X2oDJJeAv7R5ra1gZdz+Li+aOddZuYyiTngjtchkiaY2Qi342TBp5qOUwLueI5TAu54nXOR23Gy4ms8xykBH/EcpwTc8RynBCpxSLrqxAj3jUn8e5nZQ720NajOzuzMHXT6HO54bZD0LeBEQoba2oLYgH07tPNZ4JvA23V23t2hnTWBjxOCe5MOfHKHdgYA/97Azvc7seP0Dne89hwDbGZm72S0cxqwrZllPWXyR0Ks4WQgS0KEG4G5OdhxeoE7XnumAGsCszLamQ7Myd4dVqqLSewtQ6NshlMC/jihDZJGEHKoT4HFakhmdliHdnYELiHke0/a6XSKeArwJnBTnZ2O1oqSvg3cbma3dtLOyQcf8dpzKfBtsk/JLgTuyMHOO8B5wNfJsFYkTFevk7QcMB8QYGY2KEPfnJT4iNcGSX8xs71ysHOPme2eg53pwK5Z14qSngKOACab/xIUjo947XlQ0jkEkd3k1K7Txwl3ShpN2NTo9RQRmEo+a8VpwBR3unLwEa8Nku5sUGxm1unjhKeb2On0ccJ1wLbAnWRbK44lTE9vrrPjjxMKwEe8NpjZPjnZyStxyvXxysrT8VoxXk6B+IjXAklbEaQC7zOzNxPlB5nZn1LaWJGgA/q8md0u6Thgd+BR4BdmNr8LXXcqjp/VbIKkkwmPEb4ITJF0eKL6fzswdQnhhMiXJV1GyP1wH7AL8MsO+jNI0jmSLovOm6z7aQd2Rki6U9L/SdpY0m2SXpP0QEIC3+k2ZuZXg4uw7b9afD0MmAB8Kb5/uAM7k+LP5YEXgQHxvWp1Ke1cC5xL2Im8Ib4fGOse6sDO/cDBBJn754CPxPL9gHvL/ndfVi4f8ZozwOL00syeAfYGDpb0fYLTpGW5ON1cHVgFqGW1HQis0IGdzczsa2Z2vYWH9w8Bd0haqwMbACuY2c1mdgVhc+cawovbgZU6tOX0Et9cac4/Je1gZhMBzOxNSYcAFxOyE6XlV8DfgQGEh95Xx2douxGSZqZloKTlLKQrw8zOljSDkBNwtdZNl2CupAMIfwBM0hFmdr2kvYCFHdhxMuCbK02QNBRYYDHnXl3dHmb2tw5sbQhgZjNjdMH+wLNmdn8HNr4D3Gpmf64rPwj4sZltkdLO9sB3CKdnTgE+D5wAPA98xszuSdsnp/e44zlOCfgaz3FKwB3PcUrAHa8JknIJl8nLjtO/8F3N5uSVIyC3XAOS1iXkh9+WEAr0KPBTM3uxDDtO73HHa84akj7crNLMflekHUl7AL8BxgK/JjxL3Am4T9JH0+6y5mXHyYbvajZB0iuEI2ONHpabmX2yYDvjgc9byBGfLN8BuNDMUuV3z8uOkw0f8ZrzbFqnKMjOoHpnATCziZJWL8GOkwHfXOk7KOp71hcOobP/x7zsOBnwf+jmXCfpvZKyzgrysnM+cKukvSStHq+9CYGsPyjBjpMBX+M1QdJ3CXFzWwGTgHuAvxFO8KeWa8jLTrR1CPBVltyNPM/MbizDjtN73PHaECMLRhCc533xes3MtinSjqShZjajSd2haZ0mLztONnyq2Z6VgUGE0/xrADMJgaxF27ld0rD6QkmfpLMpYl52nAz4iNcESRcRpmJvEBxkPDDezF4tyc6HgB8CHzKzabHsTOA44OBmo1i37DjZ8McJzdmEEKw6jRAyMwN4rSw7ZvZHSfOAmyUdAXwaeC+wZydOnJcdJxs+4rVAkgij1e7x2g6YTdgY+UbRdqKt9xNUxu4BjjGzuZ20z9uO0zvc8VIQg2L3IDjNIcBaZrZmkXYkvUHYgRRhBJ1PiBjvSHo9LztONtzxmhBVxnYnOMp84iOA+HNyTYKhKDtO/8LXeM0ZBlwDnGJmL1TAjtOP8BHPcUrAn+M5Tgm44zlOCbjjOU4JuOM5Tgm44zlOCfx/zCpPXawjW/gAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6, 6))\n",
    "plt.imshow(neuron_matrix_2group, cmap='viridis', vmin=-1, vmax=1)\n",
    "plt.colorbar(shrink=.5, label='Gene z score')\n",
    "\n",
    "plt.yticks(np.arange(18), model_matrix['Gene name'])\n",
    "\n",
    "plt.ylim(17.5, -0.5)\n",
    "\n",
    "plt.xticks(np.arange(4), ['WT 2m', 'WT 8m', 'KO 2m', 'KO 8m'], rotation='vertical')\n",
    "\n",
    "plt.savefig('./heatmaps/ex_neurons.png', dpi=300)\n",
    "plt.savefig('./heatmaps/ex_neurons.pdf')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### *All cells function*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_heatmaps(cellpop):\n",
    "    \n",
    "    # Prepare empty matrix\n",
    "    matrix = np.zeros((18, 6))\n",
    "    \n",
    "    #Find cell indices for each condition\n",
    "    wt_2m_index = list(barcode[barcode['broad.cell.type'] == cellpop][barcode['Genotype'] == 'WT'][barcode['new_age'] == '2m'].index)\n",
    "    wt_8m_index = list(barcode[barcode['broad.cell.type'] == cellpop][barcode['Genotype'] == 'WT'][barcode['new_age'] == '8m'].index)\n",
    "\n",
    "    ko_2m_index = list(barcode[barcode['broad.cell.type'] == cellpop][barcode['Genotype'] == 'L2AKO'][barcode['new_age'] == '2m'].index)\n",
    "    ko_8m_index = list(barcode[barcode['broad.cell.type'] == cellpop][barcode['Genotype'] == 'L2AKO'][barcode['new_age'] == '8m'].index)\n",
    "\n",
    "    pd_2m_index = list(barcode[barcode['broad.cell.type'] == cellpop][barcode['Genotype'] == 'PD'][barcode['new_age'] == '2m'].index)\n",
    "    pd_8m_index = list(barcode[barcode['broad.cell.type'] == cellpop][barcode['Genotype'] == 'PD'][barcode['new_age'] == '8m'].index)\n",
    "    \n",
    "    #Calculate mean per gene for every condition\n",
    "    for rank in tqdm(np.arange(matrix.shape[0])):\n",
    "    \n",
    "        matrix[rank, 0] = np.mean(cma_network[wt_2m_index, :].X.todense()[:, rank])\n",
    "        matrix[rank, 1] = np.mean(cma_network[wt_8m_index, :].X.todense()[:, rank])\n",
    "        matrix[rank, 2] = np.mean(cma_network[ko_2m_index, :].X.todense()[:, rank])\n",
    "        matrix[rank, 3] = np.mean(cma_network[ko_8m_index, :].X.todense()[:, rank])\n",
    "        matrix[rank, 4] = np.mean(cma_network[pd_2m_index, :].X.todense()[:, rank])\n",
    "        matrix[rank, 5] = np.mean(cma_network[pd_8m_index, :].X.todense()[:, rank])\n",
    "        \n",
    "    #Perform z-scoring on each row\n",
    "    matrix_zs = matrix.copy()\n",
    "\n",
    "    for i in np.arange(matrix_zs.shape[0]):\n",
    "\n",
    "        µ = np.mean(matrix_zs[i, :])\n",
    "        sd = np.std(matrix_zs[i, :])\n",
    "\n",
    "        matrix_zs[i, :] = (matrix_zs[i, :] - µ) / sd   \n",
    "        \n",
    "    #Plot heatmap including all conditions\n",
    "    plt.figure(figsize=(6, 6))\n",
    "    plt.imshow(matrix_zs, cmap='viridis', vmin=-1, vmax=1)\n",
    "    plt.colorbar(shrink=.5, label='Gene z score')\n",
    "\n",
    "    plt.yticks(np.arange(18), model_matrix['Gene name'])\n",
    "\n",
    "    plt.ylim(17.5, -0.5)\n",
    "\n",
    "    plt.xticks(np.arange(6), ['WT 2m', 'WT 8m', 'KO 2m', 'KO 8m', 'PD 2m', 'PD 8m'], rotation='vertical')\n",
    "\n",
    "    plt.savefig('./heatmaps_3groups/{}.png'.format(cellpop), dpi=300)\n",
    "    plt.savefig('./heatmaps_3groups/{}.pdf'.format(cellpop))\n",
    "\n",
    "    plt.show()\n",
    "    \n",
    "    \n",
    "    #Perform z-scoring on only 2 groups\n",
    "    \n",
    "    matrix_2group = matrix[:, :-2]\n",
    "\n",
    "    for i in np.arange(matrix_2group.shape[0]):\n",
    "\n",
    "        µ = np.mean(matrix_2group[i, :])\n",
    "        sd = np.std(matrix_2group[i, :])\n",
    "\n",
    "        matrix_2group[i, :] = (matrix_2group[i, :] - µ) / sd   \n",
    "\n",
    "        \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {},
   "outputs": [],
   "source": [
    "cell_to_plot = ['excit.', \"oligo.\", 'astro.', 'microglia', 'inhib.', 'OPCs']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:7: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  import sys\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:8: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:10: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # Remove the CWD from sys.path while we load stuff.\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:11: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # This is added back by InteractiveShellApp.init_path()\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:13: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  del sys.path[0]\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:14: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "  0%|          | 0/18 [00:00<?, ?it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "  6%|▌         | 1/18 [00:00<00:02,  6.72it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 11%|█         | 2/18 [00:00<00:02,  6.49it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 17%|█▋        | 3/18 [00:00<00:02,  6.24it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 22%|██▏       | 4/18 [00:00<00:02,  6.25it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 28%|██▊       | 5/18 [00:00<00:02,  5.95it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 33%|███▎      | 6/18 [00:00<00:01,  6.29it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 39%|███▉      | 7/18 [00:01<00:01,  5.99it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 44%|████▍     | 8/18 [00:01<00:01,  5.85it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 50%|█████     | 9/18 [00:01<00:01,  5.87it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 56%|█████▌    | 10/18 [00:01<00:01,  5.42it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 61%|██████    | 11/18 [00:01<00:01,  5.63it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 67%|██████▋   | 12/18 [00:02<00:01,  5.69it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 72%|███████▏  | 13/18 [00:02<00:00,  5.51it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 78%|███████▊  | 14/18 [00:02<00:00,  5.80it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 83%|████████▎ | 15/18 [00:02<00:00,  5.41it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 89%|████████▉ | 16/18 [00:02<00:00,  5.50it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 94%|█████████▍| 17/18 [00:02<00:00,  5.66it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "100%|██████████| 18/18 [00:03<00:00,  5.68it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:7: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  import sys\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:8: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:10: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # Remove the CWD from sys.path while we load stuff.\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:11: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # This is added back by InteractiveShellApp.init_path()\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:13: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  del sys.path[0]\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:14: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "  0%|          | 0/18 [00:00<?, ?it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 11%|█         | 2/18 [00:00<00:00, 17.75it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 22%|██▏       | 4/18 [00:00<00:00, 16.83it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 33%|███▎      | 6/18 [00:00<00:00, 16.45it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 44%|████▍     | 8/18 [00:00<00:00, 16.73it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 56%|█████▌    | 10/18 [00:00<00:00, 17.41it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 67%|██████▋   | 12/18 [00:00<00:00, 17.79it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 78%|███████▊  | 14/18 [00:00<00:00, 16.97it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 89%|████████▉ | 16/18 [00:00<00:00, 16.61it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "100%|██████████| 18/18 [00:01<00:00, 16.92it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:7: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  import sys\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:8: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:10: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # Remove the CWD from sys.path while we load stuff.\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:11: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # This is added back by InteractiveShellApp.init_path()\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:13: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  del sys.path[0]\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:14: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "  0%|          | 0/18 [00:00<?, ?it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 17%|█▋        | 3/18 [00:00<00:00, 29.30it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 33%|███▎      | 6/18 [00:00<00:00, 29.41it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 50%|█████     | 9/18 [00:00<00:00, 29.24it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 67%|██████▋   | 12/18 [00:00<00:00, 29.14it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 83%|████████▎ | 15/18 [00:00<00:00, 29.38it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "100%|██████████| 18/18 [00:00<00:00, 29.08it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:7: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  import sys\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:8: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:10: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # Remove the CWD from sys.path while we load stuff.\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:11: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # This is added back by InteractiveShellApp.init_path()\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:13: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  del sys.path[0]\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:14: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "  0%|          | 0/18 [00:00<?, ?it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 22%|██▏       | 4/18 [00:00<00:00, 39.14it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 44%|████▍     | 8/18 [00:00<00:00, 39.19it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 67%|██████▋   | 12/18 [00:00<00:00, 39.08it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 89%|████████▉ | 16/18 [00:00<00:00, 38.32it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "100%|██████████| 18/18 [00:00<00:00, 38.31it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:7: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  import sys\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:8: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:10: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # Remove the CWD from sys.path while we load stuff.\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:11: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # This is added back by InteractiveShellApp.init_path()\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:13: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  del sys.path[0]\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:14: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "  0%|          | 0/18 [00:00<?, ?it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 17%|█▋        | 3/18 [00:00<00:00, 20.08it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 33%|███▎      | 6/18 [00:00<00:00, 20.25it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 44%|████▍     | 8/18 [00:00<00:00, 19.92it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 61%|██████    | 11/18 [00:00<00:00, 20.17it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 78%|███████▊  | 14/18 [00:00<00:00, 20.61it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 94%|█████████▍| 17/18 [00:00<00:00, 21.27it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "100%|██████████| 18/18 [00:00<00:00, 20.94it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:7: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  import sys\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:8: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:10: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # Remove the CWD from sys.path while we load stuff.\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:11: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  # This is added back by InteractiveShellApp.init_path()\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:13: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  del sys.path[0]\n",
      "/opt/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:14: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  \n",
      "  0%|          | 0/18 [00:00<?, ?it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 22%|██▏       | 4/18 [00:00<00:00, 38.03it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 44%|████▍     | 8/18 [00:00<00:00, 37.89it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 67%|██████▋   | 12/18 [00:00<00:00, 37.74it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      " 83%|████████▎ | 15/18 [00:00<00:00, 34.01it/s]/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "/opt/anaconda3/lib/python3.7/site-packages/anndata/core/anndata.py:1239: FutureWarning: is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead\n",
      "  if is_categorical(df_full[k]):\n",
      "100%|██████████| 18/18 [00:00<00:00, 34.17it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQMAAAF+CAYAAACGbX/UAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO2debxd47nHvz+RQRBjTDFEUXOIxtyax9asSHArrYr20l5cqq7eW1evS2lplbbSltCrlJiVxpiKJkEQiVCSoMQUxBwynPPcP953Jys7e1j7rLXP2fvs5/v5rM9Z6x2e9a599nr2O/5emRmO4zjLdHUBHMdpDNwZOI4DuDNwHCfizsBxHMCdgeM4EXcGjuMAsGxXF6CR6KXe1oflM9v54qC5OZQGXpzSNxc7C9bM/kwF2vu252Jn6xXfy2zjldcW8O6cNuVQHAd3BkvQh+XZUXtntjNmzOQcSgP7r7NtLnbeOm6XXOwAzN0+H0f3+B6jMtvYYf/XshfEWYQ3ExzHAdwZOI4TcWfgOA7Qyc5A0icV4p6RdENR2ChJcyWtmAj7pSSTtHq8bpM0WdKzkm6W1FfSepIelvS8pGmS/q1+T+U43YOGqBlI2pxQlt0kFXd9zwAOjemWAfYEXk/Ef2Zm25rZVsB84DvAQuDfzWxzYCfgFElb1PkxHKepaQhnABwL/BG4DzikKO4G4Jh4vgfwd8LLXopxwMZm9qaZPQVgZh8DzwMDci6z43QrGsUZHAP8mfDiDyuKmw70l7RKjLuxlAFJywIHAlOLwgcCg4HHci2x43QzutwZSNoeeMfM/gk8CGwXX/wktwJDgR0Jv/5JlpM0GZgEvAr8IWF7BeAW4DQz+6jM/UdImiRp0gLm5fJMjtOMNMKko2HAZpJeidf9gCOB3yfS3Ag8BVxrZu3SEpPOPjOzpWbnSOpJcATXm9mt5W5uZiOBkQD9tKorvTgtS5fWDGKH4FHAIDMbaGYDCZ2FSzQVzOxV4Fzg1yntilBDeN7MLs210I7TTensmkFfSbMS15cCr5tZcnTgEWALSWsnM5rZVTXcZ1fgX4CpsQkB8B9mdk9HCu04rUCnOgMzK1UTubQoTRtQcATDy9gZmDhfoUT8o4AvYHGcGujyDkTHcRoDdwaO4wDuDBzHiTTC0GLD0HOzHvS/ZuXMdnLTITgtHx2CHntmFxIpcO4XH87FzpT5n2e2MdfyEVpxAl4zcBwHcGfgOE7EnYHjOIA7A8dxIu4MHMcBusAZFKsdSRou6Yp4vqmksVG56HlJI2P4HpI+lPR0DP9xkY1fSno9rnUohK0k6a6ooDRN0jc74/kcp1lptJrB5cBlUbloc+BXibhxZjYYGAIcL+lLsGix0+HAa8BuifSnAM+Z2TYEUZSfS+rVCc/gOE1JozmDtYFFC5nMbGpxAjP7FHgS2CgG7Qk8C/yGJVc7GrBiXMG4AjCH8gpJjtPydIUzWC42AybHFYXnJ+IuAx6SdK+k0yUtNQNI0moEXcNpMWgYQSHpNuCgqGMAcAWwOfAGQf3o38x8lorjlKMrnEFBwHTbKEryX4UIM7uG8ALfTKjaT5TUO0Z/RdLTBJ3Ei8xsWqz2fxW4PSoZPQbsF9PvD0wG1gG2Ba6Q1K+4MEmlo88/yD4rznGalYabjmxmbwBXA1dLehbYKkaNM7ODipIfAKxE0C0A6AvMBf4CfJPgNAyYIellYDPg8aL7LVI6Wm3z/q505LQsDdVnIOmAQjVf0lrAaiwpi17MMODbCZWkDYH9JPUl6CHuHW2tCWwKvFTH4jtOU9NQzoBQxX9W0jPAGOAsM3urVML4wu9PqAUAizoXHwUOBn4C7CJpKkFo9Wwze7fO5XecpqXTmwnFykRmNgoYFc/PAM4okWcsMLYobC6waom0RyQu9yuOdxynNI1WM3Acp4twZ+A4DuDOwHGcSMMNLXYlC/7Rxju7fJDZzlu3b55DaaDfdW252On7v72rJ0rJA5fms3/t8IGzM9voK/8tyxP/NB3HAdwZOI4TcWfgOA7gzsBxnIg7A8dxgBTOIE9lIkm9JF0jaWpUINojYfdLMXyGpMuV2Hdd0rKS3pV0YYny9Ze0QNLJReEXSHqtuPyO45Qma82gVmWikwDMbGtgX4L6UKEMvwFGAJvE44CErf2AF4Cjk04ichQwkaJt3IG7gB0yPp/jtAxZnUGtykRbEBYNYWazgQ+AIXH79X5mNiEuOb4OOCxhZhjwS8JKxJ2KbjEM+HdgXUkDEvedaGZvZnw+x2kZ0jiDPJWJngEOjdX+DYEvAesBA0g4lXg+IOZfjrAU+W6CotGwhO31gLXM7HHgJuCYlM/tOE4RaZxBbspEBNGSWcAk4BfAeIIuYXHVH4KGIcBBwMNxleItwOGSesS4oQQnAHAjSzcVqpJUOlrAvFqzO063IfN05FqUicxsIXB64VrSeGA68D6wbiLpugTtQggv+K6SXonXqxFEUB+IcWtKOi7GrSNpEzObXkP5Fykd9dOqrnTktCyZ+gxqVSaS1FfS8vF8X2ChmT0X2/YfS9opdhB+A7gjahZ+GVg/oWZ0CjBM0qbA8mY2IBF3IaG24DhOjWTtQEytTBRZA3hK0vPA2cC/JOK+C/wemAHMBO4FjgAeMrNk/f0O4BBgOEEROcktxKaCpIslzQL6Spol6bwOPaHjtAgKnfcOhGbCjto7s538Vi0uJebcIfq+8VkudgD6X/pqLnb+b+DYzDZ22P81Jj3zean+JqcD+AxEx3EAdwaO40TcGTiOA7jSUV14ZocbcrHztdMOycXO2/sOqJ4oJXN+l09/CBeMzceOkxteM3AcB3Bn4DhOxJ2B4ziAOwPHcSLuDBzHAdwZOI4TaWrZs3jvFxL3H5GIc9kzx6mB7iB7dlzUWdgV+KmkXjHcZc8cpwa6g+xZgRWAT4G2aN9lzxynBtLMQFwuyp0VWBW4M54XZM/GExSNrjGzJTYrTMie/QRYmSB7diNB7qwge9ZOddmzk2P+YcCERNrrJc0j1CZOM7OaNiiMTYsRAH3oW0tWx+lWNLvsGYRmwiBgfeBMSRukeKbFNzEbaWZDzGxIT/LboNRxmo1mlz1L2n5H0lPAjsA/sz6X47QaTSt7Vso2MJigkuQ4To00rexZojlyfezTeBIYZWZPgsueOU6tuOxZgrxkz8a8Mbl6ohR8bZfGW8KshfnYeeKC32S24bJn+eIzEB3HAdwZOI4TcWfgOA7gsmdL8MVBcxkzJp/2fh4sXGuprSs7xGq/m1A9UUraH1wvFzvbXPyvmW3MeOvSHEriFPCageM4gDsDx3Ei7gwcxwHcGTiOE3Fn4DgOUEdnIKktKhBNi6pGZxSETKISkkk6OJH+7iLlo/6SFkg6ucjuK5JWL7rHM5KekrRLIt1fJX0g6e56PaPjSLpa0uy4SK9UvKJy1wxJUyRtl4g7QdL0eJzQeaUuTT1rBoWlz1sSVI2+Cvw4ET8LOLdC/qOAiZRYlFTiHtsA5wBJWbRLWHLtg+PUg1EsqcpVzIEsVu8aQVD0QtKqhPdhR4Ii148lrVLXklahU5oJUdVoBHBqQrbsGeDDuHqxFMOAfwfWlZRmcn0/wlLowj0fBD7ueKkdpzpm9ggwp0KSQ4HrLDARWDkqe+0P3G9mc8zsfeB+KjuVutNpk47M7KXYTFgjEfw/8bg/mVbSesBaZva4pJuAY4BSM0wKKkx9CBJse9VarqTS0foDfA5Wq7H/nsvbe3PKi2M9OWXeNODzRNBIMxtZwy0GAK8lrgsqXuXCu4zO/vYvscLMzMZJQtJXitINBW6K5zcCf6C0M/gsqi8haWfgOklbWQ1LMeM/diTAkG36+BLOFuPdOW08NmbdsvE91575uZkNyXCLcipeldS9uoROG02Q9AWCWOnsoqgLWLrvYBgwPKob3QlsI2mTSvbNbAKwOtA/lwI7LYLRZu1ljxyYRdD5LFBQ8SoX3mV0ijOQ1B/4LXBF8a+2md0HrAJsE9NuCixvZgMS6kYXEmoLle6xGdADeC//J3C6KwYspK3skQN3At+Iowo7AR9GZa8xwH6SVokdh/vFsC6jns2EQnu+J0H09I+UrupDqB3cEc+HAbcVxd9CaC78hFDmgvJRUrlZwAkFdWRJ44DNgBWi4tGJZtalH7bTeBhGWwaBH0k3EMSAV4/fsx8TvvOY2W+BewgjaTOAucA3Y9wcST8BnoimzjezSh2RdaduzsDMelSIGwuMTVzfyeI21NgS6acAW8Qahszs4xT3KO6HcJylMGABHW8OmFmloW9iTfiUMnFXE8SEG4KmmYEo6RBgHGE+gePkggELrL3s0Uo0zVharD3cWTWh49RIa73y5WkaZ+A49cDMmO+iwIA7gyVYSDvvt83NbOeUV7+WQ2lgzK3X5WLnC7edXD1RSjbZ+7Fc7Ky586qZbfzz0+y/6YbXDAq4M3BaGkMsMFdbB3cGjkNbycmArYc7A6elCaMJTTOoVlfcGTgtTTtiPmWnq7QU7gyclqfd+wyABnMGktYELgN2ImgTzAcujud3AC/HpO+a2T6JfM8AzyVng0kaBewOfEjoMD4lLmZynEUYYn75iawtRcM4gyh6cjtwrZkdG8M2AA4hOINxZnZQiXybE2ZS7iZpeTP7NBF9lpmNlrQfcBUwqN7P4TQXYWjR+wygsaYj7wXMj4s7ADCzf5rZr6rkO5awCOo+guMoxSPAxrmU0ulWmIWaQbkjDZIOkPRC1Dn8YYn4y6JW52RJL0r6IBHXlojr0hm2DVMzALYEnqoQ/5XECsWbzeyCeH4MQWNxU+BU4IYSeQ8GppYymlQ6WneAVxdbkfYMQ4uSegBXEr6Ds4AnJN1pZs8V0pjZ6Yn03wMGJ0wsEujpahqpZrAEkq6MqseFJZ7jovjptgVHIGl74B0z+yfwILBdkajkJdGBjABOLHUfMxtpZkPMbMhqqzXsx+HUidBnsGzZIwU7ADPM7CUzm09Yan9ohfTDKP2D1eU00rd/GrBIRtrMTgH2prJy0TBgs6iINJMginpkIv6s6Dz2NbOSUtZOaxPmGfQoe6QgtZZh7APbEHgoEdxH0iRJEyUd1sHHyIVGcgYPET6Y7ybC+pZLHMVVjwIGJRSRDqWytLrjLIEh2lim7EEQLZmUOEYUmahFy3AoMLogwBNZP2osHgv8QtJGmR+qgzRMn4GZWfSMl0n6AfAO8ClwdpksuwGvm9nribBHCCIoa9e3tE53IdQMKr4G71YRRK1Fy3AoRUInZvZG/PuSpLGE/oSZlUtdHxrGGQBEbbhyWodji9KOJcxHSIa1ESTTAYbnWzqnO2KItmyTjp4ANpG0IfA64ft7bHGiqO25CjAhEbYKMNfM5sVdwnYlzKvpEhrKGThOZ2NWtWZQJb8tlHQqQcy0B3C1mU2TdD4wKYryQGi+3lgkCLw5cJWkdkKT/aLkKERn487AaXGUaWgRwMzuIQifJsP+q+j6vBL5xgNbZ7p5jrgzcFoag7RDiN0e/xQSzPxsVQ5/fqnmXs2svtwnOZQG9l8nn7koXxzyafVEKbnslfG52OmjRzPbOOxr72a2EcRNfLIZuDNwWhwD2l3PAHBn4LQ4XjNYjDsDp+Vx2bOAOwOnpTETC9r9NQB3Bk6LE/QMvGYAdV6bULRWe3JhrbeksXH9dyF8dAw/T9LrifCLYvipca24xZlaxfe5Q9KEorDdJD0laaGkr9fzOZ3mxRAL2nuUPVqJetcMKq3VPs7MJpUIv8zMflYU9nfgbkpsyippZcJqx08kbWhmBWm0VwlTks/sSMGd1sA7EBfTFM0EM3saICijLcWRwF3A24R54RfGPK/EPL5hjlMRlz0L1PtTWK6omXBMIu76RPglifDTE+H7p7hHQSziBjqwfFnSiMLy1AUfZt9azWkuzGBB+zJlj1aiWZoJJYlqyhsDj8Yl0AslbVWLkImZjQRGAqz4xbV8B84Ww5BPOoo0+6dwDGFZ6MtR7Wgg5ZdAO85SFHZUKnekIYUg6nBJ7yRqvN9OxJ0gaXo8TsjvyWqnKfoMKjAMOKCwH0JcU34/8KMuLZXTRGSrGaQRRI382cxOLcq7KvBjYAjBLz0Z877f4QJloLP7DC5KxCX7DB6oZETS9yXNIqjITJH0e0kDgfWBiYV0cSThI0k7Sto+5jmKsGZ8Wt4P5zQ/Qc8gU82gVkHUJPsD95vZnOgA7gcO6NCD5EBdawZmpcdszGyPMuHnlQm/HLi8RNRSwpNmtl3ict2qhXRaGkMsrDyfYHVJyb6tkbGfqUApQdQdS9g5UtJuwIvA6Wb2Wpm8JcVUO4NmbyY4TmaqzECspoGYRhD1LuCGKG/2HeBawqZBtYip1p1m70B0nEwYsLC9R9kjBVUFUc3sPTObFy9/B3wpbd7OxJ2B09qYaK9wpGCRIKqkXoTRrCW2SStS6z4EeD6ejwH2k7RKFEfdL4Z1Cd5MSLDg0568+Vh2lfVlxi3IoTTQd2zPXOy8OH7FXOwAnDZwl1zsjHljcvVEVeit7J3uBizMMJqQUhD1+5IOARYCc4jK3WY2R9JPCA4F4Hwzm9PhwmTEnYHT0gSlo/oKoprZOcA5ZfJeDVydqQA54c7AaWnCaIK3lsGdgdPqWLZmQnfCnYHT0uTRTOguuDNwWhpvJiyms5SOnpV0s6S+MfyTonTDJV0Rz8+TtJQgSQVb5cKvljRbkm/F7lTETGWPZkPScnFfx5qpt0v8zMy2NbOtgPnAd+pgq1z4KLpwnrfTHFjsMyh3NBOSDgYmA3+N19tKurNyrsV05tOOI2gP1NPWonAze4Qwpus4FelGNYPzCAunPgAws8mEZf2p6JQ+A0nLAgcSPRZxNWMiyaoUzdqqwVbFcMepjGjrPn0GC83swzLygFWptzNIvvTjgD/E8yUUkCQNJ6zp7oitcuGpkDQCGAGw7Eqr1JLV6QaYQVt709UAyvGspGOBHpI2Ab4PpN4csytlz/KylekeSdmzPgPWc9mzFqQb7ZvwPeBcYB7wJ8IU6f9Jm9mHFp2WxrpJMyEqLv23mZ1FcAg106ifwo8kzSocHTEg6QZgArBptHNivkV0ugtm5Y80pNBAPEPSc5KmSHpQ0gaJuORGQ6l7/pd+Bmtj8dLoDlFvpaMV0oSb2SjCUGBB7ei8jtpKhNcsm+60HmbQnqFmkFID8WlgiJnNlfRd4GKCmC/k25R+OjqUm4FPC4FmdmuazN5McFqejNORF2kgAkgqaCAucgZm9nAi/UTg+Cw3rMCqwHsEFaVFtwfcGThOGtorjybkpYFY4ETg3sR1n2h/IXCRmd2ertRLY2bf7GhecGfgtDhhE5W6ayCGhNLxhCH03RPB65vZG5K+ADwkaaqZzaxW7jL21wV+Beway/Ao8G9mlqrfzZ1Bgl5vfMrA/5xQPWEVfv3PR3MoDQx7NpOjX8SLw3+Tix2AnadmmVG+mL2Pz9TXBcALL1+RvSBG1pmGqXQMJe1D6OXfPaGHiJm9Ef++JGksMBjokDMAriEMKR4Vr4+PYfumydyoowmO02lYu8oeKUijgTgYuAo4xMxmJ8JXkdQ7nq9O+EUv3nylFvqb2TVmtjAeo4D+aTO7M3BanixDi2a2EChoID4P3FTQQIy6hwCXACsANxcNIW4OTJL0DPAwoc8gizN4V9LxknrE43hCh2IqvJngtDRmYBknHaXQQNynTL7xwNaZbr4k3wKuAC4j9BmMj2GpcGfgtDxpJxc1Omb2KkGKvUN4M8Fpccr3F6TsM2gYJF0raeXE9SqSUisvN4wzKFIsuiv5UDH+dEmfS1qpKHwHSY/E6aD/iJuy9pW0maQJkuaVUk5yHCCMJnQTZwAMMrMPChdxM9fBaTM3jDNgScWiOcApRfHDCD23hxcCJK1JmHp5tpltSuiQ+SuwYrTxfeBnnVB2p5mxCkdzsUzcmQlYtOV76q6ARu0zmAAMKlxI2ojQG3sW8B/EdQwEh3GtmU0AMDMDRifszJb0tc4osNPENJ+iUTl+DoyXVHgHjgIuSJu5kWoGwKKFH3uz5FjtMOAGgnjJppLWiOFbAU92bgmdbkc3qRmY2XXAkcDbwGzgCDP7Y9r8jeQMCopF7xEWXNyfiBsK3Ghm7YRFF0eVyN8hJI2QNEnSpAXMq57B6V50oz6DWIOeaWZXAFOBfYr73irRSM6gsJRzA6AXsc9A0iBgE+B+Sa8QHENhefI0Mq7hNrORZjbEzIb0pHcWU06z0k1qBsAtQJukjYHfAxsSpienopGcAQBm9iGh4+9MST0JL/55ZjYwHusAA6JAxBXACZIWrRKLM7DW6pLCO02J2lX2aDLa44zII4BfmtnpQOptxRvOGQCY2dPAM4RawFDgtqIktwFDzeztGP+zOLT4PPAV4CNJa0WVpDNYrJzUr/OewmkKTNBe4WguFkgaBnwDuDuG9UybuWFGE0qoHx0cT5fqADGzMxLnEwgOoJi5hBVkjlOZ5msOlOObhE2ELjCzlyVtCPxf2swNWTNwnE6lvcKRghQaiL0l/TnGPyZpYCLunBj+gqT9szyGmT1nZt83sxvi9ctmdlHa/O4MnNbGCE2FckcVEhqIBwJbAMMkbVGU7ETgfTPbmLCI6Kcx7xaEZu6WhK0Afx3tdQnuDJyWR+3ljxQs0kA0s/lAQQMxyaHAtfF8NLC3wrZHhxKGzOeZ2cvAjGivS2iYPoNG4IuD5jJmzOTqCasw9rN8+ikPW29KLnb2Xycv8V1YuW8+ZZr+uw5tFLwE819qiMZ+Gg3ERWnMbKGkD4HVYvjEorwDOloQSV8BxkfZ9ELYdmb2VJr8XjNwWp4qQ4urFyalxWNEcfYSJou9VLk0qfUTUzKGoKO4ZiLs92kze83AaW2Mah2F1QRR02ggFtLMUtggeCXCQrpU+ok18AJBVWmspBOjeErq8VGvGTgtj6z8kYKqGojx+oR4/nXgobio7k5gaBxt2JAw0/bxDI9iZnY3QeDkCkmnUkNNw2sGjpNyCLEUsQ+goIHYA7i6oIEITDKzOwk7g/9R0gxCjWBozDtN0k0EEdSFwCnJ9n4HULQ7XdKXCat7B1XMkcCdgdPS1FADKEsKDcTPKbO4zswuoIZlxlXKMThxPhc4WtL6afPXtZlQTr1I0kBJzybSlVIrOkWLN6ScL2lqPL8o5jlMYSPLf8S4wxL2Rkl6OaZ/RtLe9XxOp8npPtORlyLqIqai3jWDRZtKSrqWsBJxCS+YUCsaamYT4vjrkcA4M7sypnkF2NPM3o3X2xAUjPZNTLu8X9JLZlYY+zrLzEZL2hMYSWiPOc5SZK0ZdBc6swNxAqXHUJdSKzKz0XERUjnOBP43TtQg/r2QoISU9r6OA5Z50lG3oVOcQRn1ogIdUSvaskSeSTG8mAOADm9m6bQAGdcmNAqS1pT0B0n3xustJJ2YNn+9nUEl9aIsiNITO5Jhl0h6ibBq63/LGkooHb3zXpaOXKdZyTi02EiMIoxqrBOvXwROS5u53s6gpHpRER1RK5pG2M02yXYsuU/dWcDGwI9YPC98KZJKR/1X67I1Ik5X0n2UjlY3s5uIdZoodJL6F65Tmgkl1IuSdESt6GfAOYWloPHvfxDUYZP3bQd+SZCQzrQ81OmmVKgVNGHN4FNJqxHdmKSdgA/TZu60eQZm9rTCBpNDCSrHhfC3JRXUitYgeLVHCMKn5WxNlnQ2cFd0LguAH5jZUquMzMwk/Q/wA0IVynGWpMn6BipwBqFfbiNJfyfswPz1tJnr6gwqqBdB6DgshJdTKyrEDywRditlHIaZDS+6voUgFuk4SyCasgZQEjN7StLuwKaER3vBzBakze8zEJ3WxrrdEOIOwEDCu72dpMJ+ClVxZ+A43aRmIOmPwEbAZBZ3HBrgzsBx0tCNagZDgC3iisia8SXMTmtT0DOow6QjSatKul/S9Ph3lRJptlXYLXxaXGtzTCIuucZmsqRqklXPAh3eM8RrBgmmzunPRjd+J7OdPu/k42N7fZyLGZ5+49f5GCI/CbXVxvTJbOPtj/L5nOvYgfhD4EEzuyiqJv8QOLsozVzgG3HZ8TrAk5LGJLZWP8vMRpOO1YHnJD0Oi/cKNLND0mR2Z+C0PHVsJhwK7BHPrwXGUuQMzOzFxPkbkmYThgQ/oHbO60ghC3gzwXEqz0CspoFYiTXN7E2A+HeNSokl7UCYqTszEXxBbD5cJqniZqBm9jfgFaBnPH8CSCWGCl4zcFqcFDMNK2ogSnqA0u30c2sqh7Q2YfewE+LMWYBzgLcIDmIkoVZxfgUbJwEjCOuANiKs1v0tYZFgVdwZOE6GPgMz26dcnKS3Ja1tZm/Gl312mXT9gL8APzKzRdLphVoFME/SNYSl+5U4hTDP4LGYf3qc1ZuKhmsmSDpckknaLF4XqyKdJOkpSVfGHtbnJH2W6HH9uqSjYu9su6RKyraOU089g6QQ6gnAHUvdO4io3gZcZ2Y3F8WtHf8KOIwwWlCJeXEjl0L+ZWlyQdRhwKOENQznJSMk/QvwPWCvhOrRQODugqJSDNucsC31VZ1SYqd5qe8MxIuAm6KmwKtEHcT4A/UdM/s2cDSwG7CapOEx3/C4zuZ6Sf0JU4snEzZVrcTfJP0HQTpgX+BfgbvSFrahnIGkFYBdgT0JXvW8RNzRhKGZvQuOoBxm9nzMU7eyOt2IOg0tmtl7lGivm9kk4Nvx/P8os1Oyme1V4y1/SNjXcSpwMkGktWk3UTkM+KuZvShpjqTtCNLSGxCWOg82s7e6tIROt6O7zECMHY+/i0fNNFqfwTDCxpXEv8Pi+TuEatbRed8wqXTU9umneZt3moBm1zOQdKikUxLXj0l6KR4lJdpL0TA1gyjKsBewlSQjbEhhwK8Js7QOBB6VNNvMrs/rvmY2kjBsQ+/11muSf7+TG9W3V2sGfkDcmCXSG9geWB64hqA+XpWGcQYEEYbrzOzkQoCkvxH2n8PM3pF0AGEfuXfNzIVKnMx0Ez2DXmaW3An60dhf8Z6k5dMaaaRmwjDCEEuSWwhyZsAiSfRDgKuTMmnFxOHJWcDOwF8kueNwyqJ2K3s0CUssgDKzUxOX/dMaaZiagZntUSLscuDyorBnSOyDYGavkFBNimG3sbRjcZyl6R7iJidSoqEAABhnSURBVI9JOsnMlug4lHQyNWzk2jDOwHG6jKapAJTldOB2SceyeC3Clwh9B4eVzVWEOwOn5Wn2moGZzQZ2kbQXizcS+ouZPVSLHXcGTmvTREOI1Ygvf00OIIk7A6elEc1fM8gLdwYJen9obHjn/OoJq/Dyob1yKA385/A/52KnEflkQPap4u3F2/F0lI5JBnY7Gmlo0XE6nzruwpxGAzGma0usur0zEb5hnE04XdKf4wrHuuHOwGl51Fb+yEhBA3ET4MF4XYrPzGzbeCT1Cn8KXBbzv09YhFQ33Bk4LU8d1yYcyuJNf6+lhmG+qGGwF1AQQ60pf0dwZ+C0NlZ1BmJnaCD2ibYnSiq88KsBH8SdlAFmkZhsVw+8A9Fxul4Dcf2ojPwF4CFJU4GPai5pRhrOGUg6nLCh6uZm9o+EktFWMf4k4LvABIIQSi9gQ+CFaOJ/gB2Bg4H5BKXZbyZ06B1nEbJsaxDy0EA0szfi35ckjQUGE9blrCxp2Vg7WBd4o8MFTUEjNhOSsmdLkJA928/MTolSZ18FZiY6YEYD9wNbmdkg4EWCyqzjlKSOfQZpNBBXKUigS1qd8AP3XNwi7WEWb6leMn+eNJQzSMienUiRM0jInu2XQvbsvkRbayJxGbTjlKKOgqgXAftKmg7sG6+RNERSQY5sc2CSpGcIL/9FZvZcjDsbOEPSDEIfwh8yl6gCjdZMqIfs2beAsrN3YofQCIDevVfuWKmd5sWAtvo0xVNqII4Hti6T/yWC9Hmn0FA1A3KWPZN0LrAQKKuMZGYjzWyImQ3p1Su1DoTTjWh22bO8aJiaQd6yZ5JOAA4iqCm32L/VqYUmEjGpKw3jDMhR9iymOxvY3czm1rncTjOzeE/FlqeRmgm5yZ4R+hdWBO6P871/m3dhne6BALVZ2aOVaJiaQc6yZxvXpZBOt0TeigQayBk4TpdgBt5nALgzcBzvQIy4M3Bam+6hjpwL7gySfDyXHmOfqp6uCjP/NDmHwsCOZ383FzuXDq04YbMmNv57Prbmz309s41ZoxfkUBK8mRBxZ+C0PN6BGHBn4LQ2dZyO3Gw00jwDx+l0hIVlzGWOTLZTaCBK2jOhfzhZ0ucFgRNJoyS9nIjbNlOBquDOwHHa28sf2aiqgWhmDxeW3xOm488F7kskOSuxPD+fzqgyuDNwWhur6wzEWjUQvw7c21VT6BvKGUhaS9KNkmZKek7Sw5LmxirSnESV6QFJy0i6XNKzkqZKekLShglbgyWZpP278pmcJsCs/JGNtBqIBYYCNxSFXSBpiqTLCiIo9aJhOhCjGuxtwLVmNjSGbQusaGbjJI0iyJ+NjnHDgHWAQWbWLmld4NOEyYJi0jDAt2R3SmNWrTmwuqRJieuRZjaycJGTBiJRFm1rlvyungO8RZD2G0lYfHd+LXZroWGcAbAnsMDMFi0qqtJGWht408zaY9pZhYjoWL5OUJcZJ6mPmX1en2I7TU/lroGKgqh5aCBGjgZuM7NFkycKtQpgnqRrgDMrljQjjdRM2Ap4sob0NwEHx2bDzyUNTsTtCrxsZjOBsQSdxJJIGlGQwV7AvI6U22ly1N5e9shIVQ3EBMMoaiJEB1L4cTsMeDZrgSrRSM6gJmJNYFNCVaodeFBSQWKqnGJSKTuLlI56UtcmmdOIGGEGYrkjG2k0EIkK4OsBfyvKf32UTZ8KrE5Q/q4bjdRMmMZiJdhUmNk84F7gXklvA4dFqekjgUOi7JmA1SStaGYf51xmp+mp2mfQccspNBDj9SuU2CDFzPaqS8HK0Eg1g4eA3nFfBAAkbS9p91KJJW0naZ14vgwwCPgnsA/wjJmtZ2YDzWwDgkhKXbemcpoUA9rayx8tRMM4g6hTeDihWjVT0jTgPMpvHLEGcJekZ4EpBOHTKyivmHRsPcrtNDsG1l7+aCEaqZlQ2FmmpAKymQ0vuv4r8NcSSYcXB5jZnYTOHMdZkkLNwGksZ+A4XYKvWgTcGTgtT/06EJsNdwZOa2O4M4i4M6gDt3zSLxc7C3La4OmpwTfnYwj48vdOrp4oBa9/tS2zjfnzcvr6ujMA3Bk4rY4Z1pbdMXUH3Bk4jmsgAu4MnFbHDLxmALgzcBwfWoy4M3BaHO8zKFCX6chRYejnieszJZ0Xz8+T9HpC5PGiRLr+khZISu7E/FhM96qkdxL5BkpaQdJVhenLkh4pbMgq6WpJs+N0ZccpTR1XLUo6Kn4v2yWV1USQdICkFyTNkPTDRPiG8fs/XdKfJfXKVKAq1GttwjzgCEmrl4m/LCHymBSJPAqYSGLJsZntGMUi/wv4cyLfK8DvgTnAJma2JWEqcuGeo4ADcnwmpxtigLW1lT0y8ixwBPBIuQSSegBXAgcCWwDDJG0Ro39KeFc2Ad4HTsxaoErUyxksJMg0nV5jvmHAvwPrSlpqSWcSSRsBOwI/SqgdvWRmf4nnjxAcheOUJw4t1sMZmNnzZvZClWQ7ADPid3c+QX/j0ChoshcwOqZLI6iaiXr2GVwJTJF0cYm40yUdH8/PNrMxktYD1jKzxyXdBBwDXFrB/pbAZDPL9B+TNAIYAdCHvllMOU3Ix7w/5oH2m8rVYAH6VNJAzIEBwGuJ61mEH7nVgA/MbGEivOIPZFbq5gzM7CNJ1wHfBz4rir7MzH5WFDaUIGUGwTv+gcrOIBfiP3YkQD+t6t3KLYaZZWpKVhJENbNKMmeLTJQqVoXwulHv0YRfAE8B16RIOwxYU9Jx8XodSZuY2fQy6acB20haptBMcJzOppIgakpmESTPCqxL0PB4F1hZ0rKxdlAIrxt1FTcxszmEX/uKHR+SNgWWN7MBUZ1oIHAhobZQzvZMYBLw37F9haRNJB2aV/kdpxN4Atgkjhz0Inzn74xiPw+zWAqwmqBqZjpD6ejnLO7hL0c5daKyQqaRbxOqaDOicOTviN5T0g3ABGBTSbMk1bUn1nGKkXS4pFnAzsBfJI2J4etIugcg/uqfStgv4XngJjObFk2cDZwhaQahD+EP9SxvXZoJZrZC4vxtWNwzZ2bnlUhfKmwKYailcD2KMFyYTPMRcBIlMLNqjsRx6oqZ3cbSP3IFRa+vJq7vAe4pke4lwmhDp9AwGoiO43Qt7gwcxwHcGTiOE5H5iq1FbDmol914d7WNcqtz6M1n5FAa2OisCbnYyZPpV+6Yi52XDr8qs40d9n+NSc98Xmo83ukAXjNwHAdwZ+A4TsSdgeM4gDsDx3Ei7gwcxwE6yRlUUj6K19+Q9GxUhXlO0pkxXJJ+FJVeXpT0sKQtE/lekTRV0hRJf5O0QSLOlY4cpwY6q2ZQVvlI0oHAacB+Ua1oO+DDGH0KsAuwjZl9kbB46U5JfRIm9jSzQcBY4EeJ8FG40pHjpKaznEEl5aNzgDPjfG3M7HMz+12MOxv4npnNjXH3AeOB40rYmUBC/MGVjhynNjqzz+BK4DhJKxWFbwU8WZxYUj/CsuaZRVGTCCpHxRwA3F5roSSNkDRJ0qT357gsgtO6dJoziCsMC8pHWRBLKr48LGk2sA/wpw6Ua6SZDTGzIaus6v2pTuvS2d/+XxCETpJbik4DvlScMDqPTyV9oShqO+C5xPWewAbRzvm5ltZxWohOdQZllI8uBC6WtBaApN6SCrWHS4DLJS0X4/YBvkxRDcDMPiN0Qn5D0qr1fQrH6Z50Rb14CeWjKOxwJfCApGmE/oOC6MqvCLJQUyW9APwncGh8+ZfAzN4EbiCMQLjSkePUSKdsr1ZJ+SiGXUMJ0dSoA/ff8Shld2DR9fcS56505Dg14D1mjuMA7gwcx4m4M3AcB/At2ZfgtakrcMbAnTPbmfHGb3IoDewy6Tu52FnrlOJ5Wx1n4x9+noudDXuVFLWuibc+vDyHkjgFvGbgOA7gzsBxnIg7A8dxAHcGjuNE3Bk4jgPU2RlUUjiSdJ6k1yVNjsdFMXyspBcS4aNj+BlRBWmKpAcLqkaSBpZSM5J0iaR/xPS3SVq5ns/qOM1OvWsGZRWOIpeZ2bbx+GEi/LhEeGFL6qeBIVHVaDRwcZV73w9sFdO/SBBRcRynDPV2BpUUjmrCzB4uKB4BE4F1q6S/L253nSq947Q6ndFnUE7hCOD0RHNg/0T49YnwS0rkOxG4t4YyfKvG9I7TctR9BqKZfSSpoHBUvPT4MjP7WYlsx5nZpFL2JB0PDAF2T3N/SecSaijXl4kfAYwA6LPkYkrHaSk6azShlMJRzURxk3OBQ8xsXor0JwAHEZxLyR1mk7JnPemdpXiO09R0ijMoo3BUE5IGA1cRHMHsFOkPIKgrH5Loa3AcpwyduVDp58CpKdNeL6nQpHjXzPYhSKCtANwsCeBVMzskptlU0qxE/tMJcmq9gftj+olmls/KH8fphtTVGVRSODKz88rk2aNM+D5lwl8BepaIujl9SR3H8RmIjuMA7gwcx4m4M3AcB3Bn4DhOxGXPEizsvzyzj94ls52v7bpB9UQp+OTwfHz1rRvfn4sdgP3Hb5uLnZN+u9TWFzXzm+U/yaEkTgGvGTiOA7gzcBwn4s7AcRzAnYHjOJHcnYGktrj0+FlJdxUrDEk6XdLnySXNkvaQ9GHMN0XSA5LWiHGbSZogaZ6kM4tsXS1pdhmlo2UlvSvpwryf0XG6I/WoGXwWFYq2AuYQd0VOMIyws/LhReHjYr5BMb6Qbw5h+XOppc6jgAPKlGM/4AXgaMXFCY7jlKfezYQJwIDChaSNCIuNfkRwCksRX9wVgfcBzGy2mT0BLChOa2aPEJxFKYYBvwReBXbq+CM4TmtQN2cgqQewN3BnIngYcAMwjrDScI1E3FckTSa8vPsAV2e493Lx3nfH+/n27I5ThXo4g+XiS/0esCpBmLTAUOBGM2sHbgWOSsQVmgnrAddQXfC0EgcBBc3EW4DDo3NaCkkjJE2SNGnhZ59muKXjNDd16zMANgB6Edv+kgYBmxD0BV4hOIZyv9h3ArtlKMMwYJ94nyeB1YA9SyVMKh0tu1wmISbHaWrq1kwwsw8JHX9nSupJeEHPM7OB8VgHGFDY/6CILwMd2jpYUr+Yf/3CvQgOyZsKjlOBeoubPC3pGUItYChwYFGS22L4YyzuMxDwIfBtAElrAZOAfkC7pNOALaLQ6g3AHsDqUenox0Ab8FCRRuIdwMWSeqfRTnScViR3Z5BUN4rXB8fTP5ZIe0bispSUOmb2FmX2PDCzcr/2o4rSzQH6l0nrOA4+A9FxnIg7A8dxAHcGjuNE3Bk4jgO40tESLPvOp6zx6/GZ7fzljck5lAb2XycfVSHOrJ4kLTP+b3Audl55vi2zjfc+fzqHkjgFvGbgOA7gzsBxnIg7A8dxAHcGjuNEGsIZZFRHelrSPyQtJX4i6Q5JEzrjGRyn2WkIZ0A2daTBwGDgIEm7FiKiQ9kOWFnShvUruuN0DxrFGSSpWR3JzD4DJifzAUcCdwE3EhZDOY5TgYZyBh1QRyrkW4WglfBIiXyudOQ4KWgUZ9BRdaSvSJoCvAXcHVc4ImlNYGPgUTN7EVgoaatSN04qHS3AVzc7rUujOIOOqiONi2rKWwPflVSYsncMsArwcsw3kDJNhaTSUU965/1cjtM0NIozADqujhR//S8Ezo5Bw4ADEkpHX8L7DRynIg3lDCCoIwFJdaTbipIU1JGK+S2wWxw5WB+YmLD5MvCRpB3rUmjH6QY0xEKlDOpIYxPhn7F4NCE5qlCI3y5zQR2nG9NwNQPHcboGdwaO4wDuDBzHibgzcBwHAJlZV5ehYZD0DvDPKslWB97N4XZuJ7udDczMJfBzwp1BjUiaZGZD3E5z2HHS480Ex3EAdwaO40TcGdTOSLfTVHaclHifgeM4gNcMHMeJuDNwHAdokIVKjU5UUlqPxOdlZk910Fa/IjtzMhewAeiuz9VKuDOogqSfAMOBmUChg8WAvWq0czJwPvBZkZ0v1GhnZeAbBMGW5Mv3/Rrt9AC+VsLOpTXayeu5DgJ+QhC4WRZQKI71q8WO03G8A7EKkl4Atjaz+RntTAd2NrNMs/MkjSdoNUwF2gvhZnZtjXbuAT4vYee/a7ST13PNAI4Appp/KbsErxlU51lgZWB2RjszgbnZi0OfIk2HjrJulIzLSl7P9RrwrDuCrsNrBlWQNAS4g+AUFimmmtkhNdoZDFwDPFZkp9bq/enAJ8DdRXZqaqNL+inwoJndV0u+Enbyeq7tCc2EvxXZqanZ4nQcrxlU51rgpxRVpzvAVcBDOdiZD1wCnEuGNjqhqXGbpGWABXS8jZ7Xc11AcHJ9CKK4TifjNYMqSPqbme2eg53xZrZLDnZmAjvm0EZ/CTiMjG30HJ/LFyZ1MT7PoDpPSrpQ0s6StiscHbDzcNyjYW1JqxaODtiZRj5t9Onk00bP67kekLRfxrI4GfCaQRUkPVwi2Mys1qHFl8vYqXUI7jZgS+BhsrXRRxGaFveSoY2e43N9DCwfy7IwYceHFjsJ7zOogpntmZOdvDZ/vT0eWXk5Hr3I0EbP67nMbMU87Dgdx2sGFZC0GUF2/TEz+yQRfoCZ/TWljV6EfR5eN7MHJR0L7AI8B/zOzBbUoeidQpx12N/MZhaFDzKzKTXY2Q1428xekPRlYCfgOTO7J98SO5VwZ1AGSd8nbPP2PLAt8G9mdkeMeyrtPgySrifUwPoCHxB2lL6VsMGszOyElHb6AecA6wL3mtmfEnG/NrN/TWlnCGE04vVo72pge0IfwklmNjmlnaOBXxDmX/QEhpvZEzGuls/nF8AOhM9oDOFzuRfYHZhsZmemsePkgJn5UeIgDJWtEM8HApMIDgHg6RrsTIl/lwXeBnrEaxXiUtq5BbiIMAJwZ7zuHeOeqsHO48CBhC3oXgO+HsP3BibUYGcysHY83wH4B3BEBz6fafGz6Au8D/SN4T0JHZxd/l1olcP7DMrTw2LTwMxekbQHMDru86ga7CwTmwrLE77wKwFzgN6EL3xaNjKzI+P57ZLOBR6SVNPkJ6Cnmd0LYeKRmY0GsNCE+VkNdnqY2Zsx7+OS9gTulrQui+c/pMHMzCQV5igU8rbjo12dijuD8rwlaVuL1WYz+yQuprmasOtzWv5A+NXsQZgodHMc498JuLEGO70lLWNha3rM7AJJs4BHCE2PtHweh/BWAkzSYWZ2u6TdgbYa7HwsaSOL/QVm9mZ0mLcTRjvS8hdJ4wiTjX4P3CRpIqGZ8EgNdpyMeJ9BGeIv3EIze6tE3K5m9vcabK0DYGZvxFWH+wCvmtnjNdi4GLjPzB4oCj8A+JWZbZLSzjbAxYRf3tOB7wInEPoQTjKz8TXY+dTMZhSF9wSONrPr09iJeXYm1BAmStoIOBx4FRhdcH5O/XFn4DgO4G0yx3Ei7gwcxwHcGZRFUqalvXnbaWQk9ZeUeZuzvOw4HcNHE8qT15cyty+3pDUIE6G2JAzBPQf82sze7mw7kgT8GDiVMNS6jKSFhM7M8zvbjpMddwblWUnSEeUizezWzrQjaVfgT8Ao4DrCi7Md8Jik49KObuRlBzgN2BXY3sxejra/APxG0ulmdlkn23Ey4qMJZZD0HkHhqNQEIzOzb3WynYnAd83s6aLwbYGrzGzHTrbzNLCvFekqxGr+fWY2uDPtONnxmkF5Xk37onaSnX7FLzCAmU2WVMuKv7zs9Cx+gaOdd+Jcg86242TEOxCbByns31AcuCq1/R/zslNJLboWJem87DgZcWdQntskbS8pa+0pLzuXAfdJ2l3SivHYg7DC7xddYGcbSR9J+jgeHxWuqW26dl52nIx4n0EZ4qKdXYDNgCnAeODvhJV9qZWI87ITbR0E/IAlRwEuMbO7usKO071wZ1CFuOJwCOGF3jkeH5jZFp1pR9K6ZjarTNzBaV/kHO30Ab4DbExwcleb2cLKuepnx8mONxOqsxzQj7DKbyXgDcIeAZ1t50FJA4sDJX2L2qr3edm5luDcpgJfBX5eQ9562HEy4jWDMkgaSahGf0x4aScCE83s/S6y81Xgl8BXzWx6DDsHOBY4sNyvfR3tTDWzreP5ssDjllLdqB52nOz40GJ51icIkEwnLO+dRZAt6xI7ZnaPpHnAvZIOA75NkCvbrRbHkpcdwsYrBZsLw0TCDpGXHScjXjOoQJwquyWhnb8LsBVBpWiCmf24s+1EW18mCIiMJ+gGfF5L/rzsSGoDPi1cEppBc6lxZ6a87DjZcWeQgih0sivhRT4IWM3MVu5MO3GozQgvSW/CL2obtb98udhxuh/uDMqgoI68C+HlXUAcDox/p6ZV4MnLjuPUG+8zKM9AYDRwekH4s4vtOE5d8ZqB4ziAzzNwHCfizsBxHMCdgeM4EXcGjuMA7gwcx4n8P21oZ6HKAnYTAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in cell_to_plot:\n",
    "    make_heatmaps(cellpop=i)"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}