SingleCellRNASeq/Scanpy/scRNAseq_DE_Part1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "102ad5c4-8f4d-4438-8309-f9643d336cf6",
   "metadata": {},
   "source": [
    "# Set up Python and R environments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "38253729-b7db-4f9f-8ff8-f9d3ff87e928",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install anndata2ri\n",
    "!pip install sc_toolbox"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "010c55c4-0c27-4dc8-9645-a3a0b1f03c53",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/8p/9jl2wyz136b60m4bwf608msr0000gn/T/ipykernel_3141/538058082.py:2: DeprecationWarning: The global conversion available with activate() is deprecated and will be removed in the next major release. Use a local converter.\n",
      "  anndata2ri.activate() # Activate the conversion of AnnData objects to R data type\n"
     ]
    }
   ],
   "source": [
    "import anndata2ri\n",
    "anndata2ri.activate() # Activate the conversion of AnnData objects to R data type\n",
    "%load_ext rpy2.ipython"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c641608d-346b-4a77-af83-689400f977da",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.5.14\n"
     ]
    }
   ],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\") # Suppress warnings\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import scanpy as sc\n",
    "import anndata as ad\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import random\n",
    "import sc_toolbox\n",
    "import rpy2.rinterface_lib.callbacks\n",
    "import rpy2\n",
    "print(rpy2.__version__)\n",
    "import anndata2ri\n",
    "import logging\n",
    "from rpy2.robjects import pandas2ri\n",
    "from rpy2.robjects import r\n",
    "\n",
    "sc.settings.verbosity = 0 # Configure scanpy to not print any output messages\n",
    "rpy2.rinterface_lib.callbacks.logger.setLevel(logging.ERROR) # Reduce logging messages from the R interface\n",
    "pandas2ri.activate() # Activate the conversion of pandas DataFrames to R data type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f55e8c8f-2fcd-47e0-ac53-d3ef226b41e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# %%R\n",
    "# if (!require('BiocManager', quietly = TRUE))\n",
    "#     install.packages('BiocManager')\n",
    "\n",
    "# BiocManager::install(c('edgeR', 'MAST'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "74f43902-9ec4-498e-af7c-ea012d346de1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Loading required package: limma\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R\n",
    "library(edgeR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "da9e0de1-95ed-4bf6-b2ec-f6d0a1498ddd",
   "metadata": {},
   "outputs": [],
   "source": [
    "def count_entries(df, *args):\n",
    "    \"\"\"Return a dictionary with counts of occurrences as value for each key.\"\"\"\n",
    "    cols_count = {}\n",
    "    for col_name in args:\n",
    "        col = df[col_name]\n",
    "        for entry in col:\n",
    "            if entry in cols_count.keys():\n",
    "                cols_count[entry] += 1\n",
    "            else:\n",
    "                cols_count[entry] = 1\n",
    "    return cols_count"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6554b918-d0c2-4888-98b5-a9b1d0251222",
   "metadata": {},
   "source": [
    "# Prepare the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c68fa409-6502-460d-ad96-48471fb35a1c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/anniliu/Desktop\n"
     ]
    }
   ],
   "source": [
    "!pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d6cf9a8e-7193-478d-aa22-fa8b31d93ae1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# This is a preprocessed data that filter cells with less than 200 genes and genes found in less than 3 cells\n",
    "adata = ad.read_h5ad('adata_v2.h5ad')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "6077be36-e321-4504-9538-ed324e4a10ef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AnnData object with n_obs × n_vars = 62186 × 21231\n",
       "    obs: 'library_id', 'plate_name', 'well', 'row', 'col', 'cell_id', 'donor_id', 'cell_type', 'sm_lincs_id', 'sm_name', 'SMILES', 'dose_uM', 'timepoint_hr', 'control', 'sm_cluster_3', 'sm_cluster_5', 'sm_cluster_7', 'sm_cluster_9', 'n_genes', 'n_genes_by_counts', 'log1p_n_genes_by_counts', 'total_counts', 'log1p_total_counts', 'pct_counts_in_top_20_genes', 'total_counts_mt', 'log1p_total_counts_mt', 'pct_counts_mt', 'outlier', 'mt_outlier'\n",
       "    var: 'n_cells', 'mt', 'n_cells_by_counts', 'mean_counts', 'log1p_mean_counts', 'pct_dropout_by_counts', 'total_counts', 'log1p_total_counts'"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "07eb4602-c685-468d-bed9-f63beca5008b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>gene</th>\n",
       "      <th>A1BG</th>\n",
       "      <th>A1BG-AS1</th>\n",
       "      <th>A2M</th>\n",
       "      <th>A2M-AS1</th>\n",
       "      <th>A2MP1</th>\n",
       "      <th>A4GALT</th>\n",
       "      <th>AAAS</th>\n",
       "      <th>AACS</th>\n",
       "      <th>AAGAB</th>\n",
       "      <th>AAK1</th>\n",
       "      <th>...</th>\n",
       "      <th>ABLIM2</th>\n",
       "      <th>ABO</th>\n",
       "      <th>ABR</th>\n",
       "      <th>ABRA</th>\n",
       "      <th>ABRACL</th>\n",
       "      <th>ABRAXAS1</th>\n",
       "      <th>ABRAXAS2</th>\n",
       "      <th>ABT1</th>\n",
       "      <th>ABTB1</th>\n",
       "      <th>ABTB2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>obs_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0000233976e3cd37</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0001533c5e876362</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00022f989630d14b</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0002560bd38ce03e</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0002771c64e94903</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0053c28a68667d33</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>005439062a521170</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0056d9b9e10266a6</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>005b2a17b9b4c8c4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>005c9e2ea5404302</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 100 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "gene              A1BG  A1BG-AS1  A2M  A2M-AS1  A2MP1  A4GALT  AAAS  AACS  \\\n",
       "obs_id                                                                      \n",
       "0000233976e3cd37   0.0       0.0  0.0      0.0    0.0     0.0   0.0   0.0   \n",
       "0001533c5e876362   0.0       0.0  0.0      0.0    0.0     0.0   0.0   0.0   \n",
       "00022f989630d14b   0.0       0.0  0.0      0.0    0.0     0.0   0.0   0.0   \n",
       "0002560bd38ce03e   0.0       0.0  0.0      0.0    0.0     0.0   0.0   0.0   \n",
       "0002771c64e94903   0.0       0.0  0.0      0.0    0.0     0.0   0.0   0.0   \n",
       "...                ...       ...  ...      ...    ...     ...   ...   ...   \n",
       "0053c28a68667d33   0.0       0.0  0.0      0.0    0.0     0.0   0.0   0.0   \n",
       "005439062a521170   0.0       0.0  0.0      0.0    0.0     0.0   0.0   0.0   \n",
       "0056d9b9e10266a6   0.0       0.0  0.0      0.0    0.0     0.0   0.0   0.0   \n",
       "005b2a17b9b4c8c4   0.0       0.0  0.0      0.0    0.0     0.0   1.0   0.0   \n",
       "005c9e2ea5404302   0.0       0.0  0.0      0.0    0.0     0.0   0.0   0.0   \n",
       "\n",
       "gene              AAGAB  AAK1  ...  ABLIM2  ABO  ABR  ABRA  ABRACL  ABRAXAS1  \\\n",
       "obs_id                         ...                                             \n",
       "0000233976e3cd37    0.0   1.0  ...     0.0  0.0  0.0   0.0     1.0       0.0   \n",
       "0001533c5e876362    0.0   0.0  ...     0.0  0.0  0.0   0.0     0.0       0.0   \n",
       "00022f989630d14b    0.0   0.0  ...     0.0  0.0  0.0   0.0     1.0       0.0   \n",
       "0002560bd38ce03e    0.0   3.0  ...     0.0  0.0  0.0   0.0     0.0       0.0   \n",
       "0002771c64e94903    0.0   3.0  ...     0.0  0.0  0.0   0.0     4.0       0.0   \n",
       "...                 ...   ...  ...     ...  ...  ...   ...     ...       ...   \n",
       "0053c28a68667d33    0.0   0.0  ...     0.0  0.0  0.0   0.0     1.0       0.0   \n",
       "005439062a521170    0.0   0.0  ...     0.0  0.0  0.0   0.0     0.0       0.0   \n",
       "0056d9b9e10266a6    0.0   2.0  ...     0.0  0.0  0.0   0.0     1.0       0.0   \n",
       "005b2a17b9b4c8c4    1.0   2.0  ...     0.0  0.0  0.0   0.0     2.0       2.0   \n",
       "005c9e2ea5404302    1.0   0.0  ...     0.0  0.0  0.0   0.0     3.0       0.0   \n",
       "\n",
       "gene              ABRAXAS2  ABT1  ABTB1  ABTB2  \n",
       "obs_id                                          \n",
       "0000233976e3cd37       1.0   0.0    0.0    0.0  \n",
       "0001533c5e876362       0.0   0.0    1.0    0.0  \n",
       "00022f989630d14b       2.0   0.0    0.0    0.0  \n",
       "0002560bd38ce03e       0.0   0.0    0.0    0.0  \n",
       "0002771c64e94903       0.0   0.0    0.0    0.0  \n",
       "...                    ...   ...    ...    ...  \n",
       "0053c28a68667d33       0.0   0.0    1.0    0.0  \n",
       "005439062a521170       0.0   0.0    0.0    0.0  \n",
       "0056d9b9e10266a6       1.0   0.0    0.0    0.0  \n",
       "005b2a17b9b4c8c4       0.0   1.0    0.0    0.0  \n",
       "005c9e2ea5404302       0.0   0.0    0.0    0.0  \n",
       "\n",
       "[100 rows x 100 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata[0:100,0:100].to_df()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "104fd690-32e0-448f-9741-69caa1002aa4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2008.0"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.max(adata.X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7a7b905f-8a4c-4e7c-9dd9-2cb4d7967d59",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.min(adata.X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "d2244ec9-679b-4650-a247-bae55b98f924",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>library_id</th>\n",
       "      <th>plate_name</th>\n",
       "      <th>well</th>\n",
       "      <th>row</th>\n",
       "      <th>col</th>\n",
       "      <th>cell_id</th>\n",
       "      <th>donor_id</th>\n",
       "      <th>cell_type</th>\n",
       "      <th>sm_lincs_id</th>\n",
       "      <th>sm_name</th>\n",
       "      <th>...</th>\n",
       "      <th>n_genes_by_counts</th>\n",
       "      <th>log1p_n_genes_by_counts</th>\n",
       "      <th>total_counts</th>\n",
       "      <th>log1p_total_counts</th>\n",
       "      <th>pct_counts_in_top_20_genes</th>\n",
       "      <th>total_counts_mt</th>\n",
       "      <th>log1p_total_counts_mt</th>\n",
       "      <th>pct_counts_mt</th>\n",
       "      <th>outlier</th>\n",
       "      <th>mt_outlier</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>obs_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0000233976e3cd37</th>\n",
       "      <td>library_0</td>\n",
       "      <td>plate_3</td>\n",
       "      <td>D4</td>\n",
       "      <td>D</td>\n",
       "      <td>4</td>\n",
       "      <td>PBMC</td>\n",
       "      <td>donor_1</td>\n",
       "      <td>T cells CD4+</td>\n",
       "      <td>LSM-46203</td>\n",
       "      <td>BMS-265246</td>\n",
       "      <td>...</td>\n",
       "      <td>1602</td>\n",
       "      <td>7.379632</td>\n",
       "      <td>5553.0</td>\n",
       "      <td>8.622274</td>\n",
       "      <td>27.174500</td>\n",
       "      <td>341.0</td>\n",
       "      <td>5.834811</td>\n",
       "      <td>6.140825</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0001533c5e876362</th>\n",
       "      <td>library_2</td>\n",
       "      <td>plate_0</td>\n",
       "      <td>B11</td>\n",
       "      <td>B</td>\n",
       "      <td>11</td>\n",
       "      <td>PBMC</td>\n",
       "      <td>donor_0</td>\n",
       "      <td>T regulatory cells</td>\n",
       "      <td>LSM-45663</td>\n",
       "      <td>Resminostat</td>\n",
       "      <td>...</td>\n",
       "      <td>1431</td>\n",
       "      <td>7.266827</td>\n",
       "      <td>4051.0</td>\n",
       "      <td>8.306966</td>\n",
       "      <td>24.487781</td>\n",
       "      <td>281.0</td>\n",
       "      <td>5.641907</td>\n",
       "      <td>6.936559</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00022f989630d14b</th>\n",
       "      <td>library_35</td>\n",
       "      <td>plate_2</td>\n",
       "      <td>E6</td>\n",
       "      <td>E</td>\n",
       "      <td>6</td>\n",
       "      <td>PBMC</td>\n",
       "      <td>donor_0</td>\n",
       "      <td>T cells CD4+</td>\n",
       "      <td>LSM-43216</td>\n",
       "      <td>FK 866</td>\n",
       "      <td>...</td>\n",
       "      <td>1430</td>\n",
       "      <td>7.266129</td>\n",
       "      <td>4510.0</td>\n",
       "      <td>8.414274</td>\n",
       "      <td>26.629712</td>\n",
       "      <td>343.0</td>\n",
       "      <td>5.840642</td>\n",
       "      <td>7.605322</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0002560bd38ce03e</th>\n",
       "      <td>library_22</td>\n",
       "      <td>plate_4</td>\n",
       "      <td>B6</td>\n",
       "      <td>B</td>\n",
       "      <td>6</td>\n",
       "      <td>PBMC</td>\n",
       "      <td>donor_2</td>\n",
       "      <td>T cells CD4+</td>\n",
       "      <td>LSM-1099</td>\n",
       "      <td>Nilotinib</td>\n",
       "      <td>...</td>\n",
       "      <td>1331</td>\n",
       "      <td>7.194437</td>\n",
       "      <td>4004.0</td>\n",
       "      <td>8.295299</td>\n",
       "      <td>28.796204</td>\n",
       "      <td>312.0</td>\n",
       "      <td>5.746203</td>\n",
       "      <td>7.792208</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0002771c64e94903</th>\n",
       "      <td>library_0</td>\n",
       "      <td>plate_3</td>\n",
       "      <td>D1</td>\n",
       "      <td>D</td>\n",
       "      <td>1</td>\n",
       "      <td>PBMC</td>\n",
       "      <td>donor_1</td>\n",
       "      <td>T cells CD8+</td>\n",
       "      <td>LSM-43181</td>\n",
       "      <td>Belinostat</td>\n",
       "      <td>...</td>\n",
       "      <td>1860</td>\n",
       "      <td>7.528869</td>\n",
       "      <td>6833.0</td>\n",
       "      <td>8.829665</td>\n",
       "      <td>26.240304</td>\n",
       "      <td>453.0</td>\n",
       "      <td>6.118097</td>\n",
       "      <td>6.629592</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fff9ef161a5bfa33</th>\n",
       "      <td>library_14</td>\n",
       "      <td>plate_5</td>\n",
       "      <td>C4</td>\n",
       "      <td>C</td>\n",
       "      <td>4</td>\n",
       "      <td>PBMC</td>\n",
       "      <td>donor_1</td>\n",
       "      <td>T cells CD4+</td>\n",
       "      <td>LSM-1172</td>\n",
       "      <td>TGX 221</td>\n",
       "      <td>...</td>\n",
       "      <td>2046</td>\n",
       "      <td>7.624131</td>\n",
       "      <td>8830.0</td>\n",
       "      <td>9.086024</td>\n",
       "      <td>24.824462</td>\n",
       "      <td>650.0</td>\n",
       "      <td>6.478510</td>\n",
       "      <td>7.361268</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fffab23ff3de1e9a</th>\n",
       "      <td>library_26</td>\n",
       "      <td>plate_5</td>\n",
       "      <td>F7</td>\n",
       "      <td>F</td>\n",
       "      <td>7</td>\n",
       "      <td>PBMC</td>\n",
       "      <td>donor_1</td>\n",
       "      <td>T cells CD8+</td>\n",
       "      <td>LSM-4944</td>\n",
       "      <td>MLN 2238</td>\n",
       "      <td>...</td>\n",
       "      <td>1480</td>\n",
       "      <td>7.300473</td>\n",
       "      <td>4648.0</td>\n",
       "      <td>8.444407</td>\n",
       "      <td>30.873494</td>\n",
       "      <td>265.0</td>\n",
       "      <td>5.583496</td>\n",
       "      <td>5.701377</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fffb7ee424643703</th>\n",
       "      <td>library_0</td>\n",
       "      <td>plate_3</td>\n",
       "      <td>D12</td>\n",
       "      <td>D</td>\n",
       "      <td>12</td>\n",
       "      <td>PBMC</td>\n",
       "      <td>donor_1</td>\n",
       "      <td>T cells CD4+</td>\n",
       "      <td>LSM-3828</td>\n",
       "      <td>Vorinostat</td>\n",
       "      <td>...</td>\n",
       "      <td>1569</td>\n",
       "      <td>7.358831</td>\n",
       "      <td>6408.0</td>\n",
       "      <td>8.765459</td>\n",
       "      <td>29.822097</td>\n",
       "      <td>380.0</td>\n",
       "      <td>5.942799</td>\n",
       "      <td>5.930087</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fffc0f266011f710</th>\n",
       "      <td>library_8</td>\n",
       "      <td>plate_3</td>\n",
       "      <td>C10</td>\n",
       "      <td>C</td>\n",
       "      <td>10</td>\n",
       "      <td>PBMC</td>\n",
       "      <td>donor_1</td>\n",
       "      <td>T regulatory cells</td>\n",
       "      <td>LSM-1132</td>\n",
       "      <td>BMS-387032</td>\n",
       "      <td>...</td>\n",
       "      <td>1394</td>\n",
       "      <td>7.240650</td>\n",
       "      <td>5230.0</td>\n",
       "      <td>8.562358</td>\n",
       "      <td>32.906310</td>\n",
       "      <td>220.0</td>\n",
       "      <td>5.398163</td>\n",
       "      <td>4.206501</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fffdafaabb2b2c4d</th>\n",
       "      <td>library_28</td>\n",
       "      <td>plate_5</td>\n",
       "      <td>B11</td>\n",
       "      <td>B</td>\n",
       "      <td>11</td>\n",
       "      <td>PBMC</td>\n",
       "      <td>donor_1</td>\n",
       "      <td>T cells CD4+</td>\n",
       "      <td>LSM-45663</td>\n",
       "      <td>Resminostat</td>\n",
       "      <td>...</td>\n",
       "      <td>1275</td>\n",
       "      <td>7.151485</td>\n",
       "      <td>4203.0</td>\n",
       "      <td>8.343792</td>\n",
       "      <td>28.527242</td>\n",
       "      <td>306.0</td>\n",
       "      <td>5.726848</td>\n",
       "      <td>7.280514</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>62186 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  library_id plate_name well row  col cell_id donor_id  \\\n",
       "obs_id                                                                   \n",
       "0000233976e3cd37   library_0    plate_3   D4   D    4    PBMC  donor_1   \n",
       "0001533c5e876362   library_2    plate_0  B11   B   11    PBMC  donor_0   \n",
       "00022f989630d14b  library_35    plate_2   E6   E    6    PBMC  donor_0   \n",
       "0002560bd38ce03e  library_22    plate_4   B6   B    6    PBMC  donor_2   \n",
       "0002771c64e94903   library_0    plate_3   D1   D    1    PBMC  donor_1   \n",
       "...                      ...        ...  ...  ..  ...     ...      ...   \n",
       "fff9ef161a5bfa33  library_14    plate_5   C4   C    4    PBMC  donor_1   \n",
       "fffab23ff3de1e9a  library_26    plate_5   F7   F    7    PBMC  donor_1   \n",
       "fffb7ee424643703   library_0    plate_3  D12   D   12    PBMC  donor_1   \n",
       "fffc0f266011f710   library_8    plate_3  C10   C   10    PBMC  donor_1   \n",
       "fffdafaabb2b2c4d  library_28    plate_5  B11   B   11    PBMC  donor_1   \n",
       "\n",
       "                           cell_type sm_lincs_id      sm_name  ...  \\\n",
       "obs_id                                                         ...   \n",
       "0000233976e3cd37        T cells CD4+   LSM-46203   BMS-265246  ...   \n",
       "0001533c5e876362  T regulatory cells   LSM-45663  Resminostat  ...   \n",
       "00022f989630d14b        T cells CD4+   LSM-43216       FK 866  ...   \n",
       "0002560bd38ce03e        T cells CD4+    LSM-1099    Nilotinib  ...   \n",
       "0002771c64e94903        T cells CD8+   LSM-43181   Belinostat  ...   \n",
       "...                              ...         ...          ...  ...   \n",
       "fff9ef161a5bfa33        T cells CD4+    LSM-1172      TGX 221  ...   \n",
       "fffab23ff3de1e9a        T cells CD8+    LSM-4944     MLN 2238  ...   \n",
       "fffb7ee424643703        T cells CD4+    LSM-3828   Vorinostat  ...   \n",
       "fffc0f266011f710  T regulatory cells    LSM-1132   BMS-387032  ...   \n",
       "fffdafaabb2b2c4d        T cells CD4+   LSM-45663  Resminostat  ...   \n",
       "\n",
       "                 n_genes_by_counts  log1p_n_genes_by_counts  total_counts  \\\n",
       "obs_id                                                                      \n",
       "0000233976e3cd37              1602                 7.379632        5553.0   \n",
       "0001533c5e876362              1431                 7.266827        4051.0   \n",
       "00022f989630d14b              1430                 7.266129        4510.0   \n",
       "0002560bd38ce03e              1331                 7.194437        4004.0   \n",
       "0002771c64e94903              1860                 7.528869        6833.0   \n",
       "...                            ...                      ...           ...   \n",
       "fff9ef161a5bfa33              2046                 7.624131        8830.0   \n",
       "fffab23ff3de1e9a              1480                 7.300473        4648.0   \n",
       "fffb7ee424643703              1569                 7.358831        6408.0   \n",
       "fffc0f266011f710              1394                 7.240650        5230.0   \n",
       "fffdafaabb2b2c4d              1275                 7.151485        4203.0   \n",
       "\n",
       "                  log1p_total_counts pct_counts_in_top_20_genes  \\\n",
       "obs_id                                                            \n",
       "0000233976e3cd37            8.622274                  27.174500   \n",
       "0001533c5e876362            8.306966                  24.487781   \n",
       "00022f989630d14b            8.414274                  26.629712   \n",
       "0002560bd38ce03e            8.295299                  28.796204   \n",
       "0002771c64e94903            8.829665                  26.240304   \n",
       "...                              ...                        ...   \n",
       "fff9ef161a5bfa33            9.086024                  24.824462   \n",
       "fffab23ff3de1e9a            8.444407                  30.873494   \n",
       "fffb7ee424643703            8.765459                  29.822097   \n",
       "fffc0f266011f710            8.562358                  32.906310   \n",
       "fffdafaabb2b2c4d            8.343792                  28.527242   \n",
       "\n",
       "                 total_counts_mt log1p_total_counts_mt pct_counts_mt  outlier  \\\n",
       "obs_id                                                                          \n",
       "0000233976e3cd37           341.0              5.834811      6.140825    False   \n",
       "0001533c5e876362           281.0              5.641907      6.936559    False   \n",
       "00022f989630d14b           343.0              5.840642      7.605322    False   \n",
       "0002560bd38ce03e           312.0              5.746203      7.792208    False   \n",
       "0002771c64e94903           453.0              6.118097      6.629592    False   \n",
       "...                          ...                   ...           ...      ...   \n",
       "fff9ef161a5bfa33           650.0              6.478510      7.361268    False   \n",
       "fffab23ff3de1e9a           265.0              5.583496      5.701377    False   \n",
       "fffb7ee424643703           380.0              5.942799      5.930087    False   \n",
       "fffc0f266011f710           220.0              5.398163      4.206501    False   \n",
       "fffdafaabb2b2c4d           306.0              5.726848      7.280514    False   \n",
       "\n",
       "                  mt_outlier  \n",
       "obs_id                        \n",
       "0000233976e3cd37       False  \n",
       "0001533c5e876362       False  \n",
       "00022f989630d14b       False  \n",
       "0002560bd38ce03e       False  \n",
       "0002771c64e94903       False  \n",
       "...                      ...  \n",
       "fff9ef161a5bfa33       False  \n",
       "fffab23ff3de1e9a       False  \n",
       "fffb7ee424643703       False  \n",
       "fffc0f266011f710       False  \n",
       "fffdafaabb2b2c4d       False  \n",
       "\n",
       "[62186 rows x 29 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata.obs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "f4a9c5b3-d17b-4cea-8a0f-8c59319aeffa",
   "metadata": {},
   "outputs": [],
   "source": [
    "adata.layers['counts'] = adata.X.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "3c5624a6-ae40-49fd-aae3-b4f946f348e3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AnnData object with n_obs × n_vars = 62186 × 21231\n",
       "    obs: 'library_id', 'plate_name', 'well', 'row', 'col', 'cell_id', 'donor_id', 'cell_type', 'sm_lincs_id', 'sm_name', 'SMILES', 'dose_uM', 'timepoint_hr', 'control', 'sm_cluster_3', 'sm_cluster_5', 'sm_cluster_7', 'sm_cluster_9', 'n_genes', 'n_genes_by_counts', 'log1p_n_genes_by_counts', 'total_counts', 'log1p_total_counts', 'pct_counts_in_top_20_genes', 'total_counts_mt', 'log1p_total_counts_mt', 'pct_counts_mt', 'outlier', 'mt_outlier'\n",
       "    var: 'n_cells', 'mt', 'n_cells_by_counts', 'mean_counts', 'log1p_mean_counts', 'pct_dropout_by_counts', 'total_counts', 'log1p_total_counts'\n",
       "    layers: 'counts'"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b2367855-a177-42a7-be49-426925109acf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "146"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Treat sm_name as a categorical variable - cat, \n",
    "# get the distinct categories - categories, \n",
    "# and calculate the number of categories - len\n",
    "len(adata.obs.sm_name.cat.categories)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "890804c4-6b15-476d-ab92-8aec2c54554d",
   "metadata": {},
   "outputs": [],
   "source": [
    "adata.obs['condition'] = ['perturbated' if sm_i == 'Belinostat' else 'control' for sm_i in adata.obs['sm_name']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "e7be9e16-aaa9-4c5b-b717-98cc3444085a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert the condition into categories\n",
    "adata.obs.condition = adata.obs.condition.astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "64d24291-7e8f-474b-99f1-7fbe947a9614",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "obs_id\n",
       "0000233976e3cd37        control\n",
       "0001533c5e876362        control\n",
       "00022f989630d14b        control\n",
       "0002560bd38ce03e        control\n",
       "0002771c64e94903    perturbated\n",
       "                       ...     \n",
       "fff9ef161a5bfa33        control\n",
       "fffab23ff3de1e9a        control\n",
       "fffb7ee424643703        control\n",
       "fffc0f266011f710        control\n",
       "fffdafaabb2b2c4d        control\n",
       "Name: condition, Length: 62186, dtype: category\n",
       "Categories (2, object): ['control', 'perturbated']"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata.obs.condition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "30d39fed-42d9-41cf-9ca6-8b85514b6d2b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['control', 'perturbated'], dtype='object')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata.obs.condition.cat.categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "7e954ff8-b55a-46dc-8fb2-1ef711f23bd6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'control': 56328, 'perturbated': 5858}"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count_entries(adata.obs, 'condition')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "467f79ea-ff09-49bc-9bfc-0e1dd2c7c58a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'plate_3': 16518,\n",
       " 'plate_0': 10291,\n",
       " 'plate_2': 10722,\n",
       " 'plate_4': 6099,\n",
       " 'plate_5': 15719,\n",
       " 'plate_1': 2837}"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count_entries(adata.obs, 'plate_name')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "a25ec353-a92a-4e49-b5ca-6c04f765b64d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "plate_name   plate_0  plate_1  plate_2  plate_3  plate_4  plate_5\n",
      "condition                                                        \n",
      "control         9407     2617     9712    14786     5551    14255\n",
      "perturbated      884      220     1010     1732      548     1464\n"
     ]
    }
   ],
   "source": [
    "df = adata.obs\n",
    "crosstab = pd.crosstab(df['condition'], df['plate_name'], margins=False)\n",
    "print(crosstab)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "0a0ec996-b4fc-4b6f-9869-39dd9d9af8dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "con_1 = (df['condition'] == 'perturbated') & (df['plate_name'].isin(['plate_2', 'plate_3', 'plate_5']))\n",
    "con_2 = (df['condition'] == 'control') & (df['plate_name'].isin(['plate_0', 'plate_1', 'plate_4']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "008c0e8b-0d7f-4d28-8c34-2f7a7e6b2557",
   "metadata": {},
   "outputs": [],
   "source": [
    "bdata = adata[con_1 | con_2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "a2a01f4b-04e8-4aae-b5bf-7b394bfa70e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>plate_name</th>\n",
       "      <th>plate_0</th>\n",
       "      <th>plate_1</th>\n",
       "      <th>plate_2</th>\n",
       "      <th>plate_3</th>\n",
       "      <th>plate_4</th>\n",
       "      <th>plate_5</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>condition</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>control</th>\n",
       "      <td>9407</td>\n",
       "      <td>2617</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5551</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>perturbated</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1010</td>\n",
       "      <td>1732</td>\n",
       "      <td>0</td>\n",
       "      <td>1464</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "plate_name   plate_0  plate_1  plate_2  plate_3  plate_4  plate_5\n",
       "condition                                                        \n",
       "control         9407     2617        0        0     5551        0\n",
       "perturbated        0        0     1010     1732        0     1464"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_2 = bdata.obs\n",
    "pd.crosstab(df_2['condition'], df_2['plate_name'], margins=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "63be9fb0-5352-44a6-9cec-66a34a188611",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>plate_name</th>\n",
       "      <th>plate_0</th>\n",
       "      <th>plate_1</th>\n",
       "      <th>plate_2</th>\n",
       "      <th>plate_3</th>\n",
       "      <th>plate_4</th>\n",
       "      <th>plate_5</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cell_type</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>B cells</th>\n",
       "      <td>64</td>\n",
       "      <td>6</td>\n",
       "      <td>37</td>\n",
       "      <td>77</td>\n",
       "      <td>37</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Myeloid cells</th>\n",
       "      <td>23</td>\n",
       "      <td>16</td>\n",
       "      <td>3</td>\n",
       "      <td>9</td>\n",
       "      <td>58</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NK cells</th>\n",
       "      <td>815</td>\n",
       "      <td>513</td>\n",
       "      <td>63</td>\n",
       "      <td>61</td>\n",
       "      <td>1011</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>T cells CD4+</th>\n",
       "      <td>7659</td>\n",
       "      <td>1763</td>\n",
       "      <td>906</td>\n",
       "      <td>1025</td>\n",
       "      <td>3788</td>\n",
       "      <td>1328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>T cells CD8+</th>\n",
       "      <td>321</td>\n",
       "      <td>39</td>\n",
       "      <td>0</td>\n",
       "      <td>560</td>\n",
       "      <td>110</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>T regulatory cells</th>\n",
       "      <td>525</td>\n",
       "      <td>280</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>547</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "plate_name          plate_0  plate_1  plate_2  plate_3  plate_4  plate_5\n",
       "cell_type                                                               \n",
       "B cells                  64        6       37       77       37       41\n",
       "Myeloid cells            23       16        3        9       58        6\n",
       "NK cells                815      513       63       61     1011       87\n",
       "T cells CD4+           7659     1763      906     1025     3788     1328\n",
       "T cells CD8+            321       39        0      560      110        1\n",
       "T regulatory cells      525      280        1        0      547        1"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(df_2['cell_type'], df_2['plate_name'], margins=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "724fac5a-1006-450b-afcc-580999f4038a",
   "metadata": {},
   "outputs": [],
   "source": [
    "cdata = bdata[bdata.obs.cell_type.isin(['B cells', 'NK cells', 'T cells CD4+'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "073f6dfe-3739-4147-b462-c7c952387e65",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>plate_name</th>\n",
       "      <th>plate_0</th>\n",
       "      <th>plate_1</th>\n",
       "      <th>plate_2</th>\n",
       "      <th>plate_3</th>\n",
       "      <th>plate_4</th>\n",
       "      <th>plate_5</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cell_type</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>B cells</th>\n",
       "      <td>64</td>\n",
       "      <td>6</td>\n",
       "      <td>37</td>\n",
       "      <td>77</td>\n",
       "      <td>37</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NK cells</th>\n",
       "      <td>815</td>\n",
       "      <td>513</td>\n",
       "      <td>63</td>\n",
       "      <td>61</td>\n",
       "      <td>1011</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>T cells CD4+</th>\n",
       "      <td>7659</td>\n",
       "      <td>1763</td>\n",
       "      <td>906</td>\n",
       "      <td>1025</td>\n",
       "      <td>3788</td>\n",
       "      <td>1328</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "plate_name    plate_0  plate_1  plate_2  plate_3  plate_4  plate_5\n",
       "cell_type                                                         \n",
       "B cells            64        6       37       77       37       41\n",
       "NK cells          815      513       63       61     1011       87\n",
       "T cells CD4+     7659     1763      906     1025     3788     1328"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_3 = cdata.obs\n",
    "pd.crosstab(df_3['cell_type'], df_3['plate_name'], margins=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "26642d96-f37f-414a-a7a1-c44be04a0c33",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>plate_name</th>\n",
       "      <th>plate_0</th>\n",
       "      <th>plate_1</th>\n",
       "      <th>plate_2</th>\n",
       "      <th>plate_3</th>\n",
       "      <th>plate_4</th>\n",
       "      <th>plate_5</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>condition</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>control</th>\n",
       "      <td>8538</td>\n",
       "      <td>2282</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4836</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>perturbated</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1006</td>\n",
       "      <td>1163</td>\n",
       "      <td>0</td>\n",
       "      <td>1456</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "plate_name   plate_0  plate_1  plate_2  plate_3  plate_4  plate_5\n",
       "condition                                                        \n",
       "control         8538     2282        0        0     4836        0\n",
       "perturbated        0        0     1006     1163        0     1456"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(df_3['condition'], df_3['plate_name'], margins=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf867055-59ed-4801-91e7-b8855bbff9b3",
   "metadata": {},
   "source": [
    "# Create the pseudobulk from the single-cell RNA dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd248284-5bf1-447b-81dc-29f5fba299cf",
   "metadata": {},
   "source": [
    "For each plate, one pseudobulk sample per cell type is created. This is done by aggregating the cell of each type using the summation of gene expression within that cell type."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "5b2b727b-f30a-4bee-b9fe-82b29dcbd73b",
   "metadata": {},
   "outputs": [],
   "source": [
    "cdata.obs['sample'] = [\n",
    "    f'{p}_{c}' for p, c in zip(cdata.obs.plate_name, cdata.obs.condition)\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "3d2cc191-7663-4c56-8824-0aadb58ac785",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "obs_id\n",
       "0002560bd38ce03e        plate_4_control\n",
       "0004ce8793cc33e4    plate_3_perturbated\n",
       "0009dd2a269fc9ce        plate_1_control\n",
       "0019c2b227447e12        plate_0_control\n",
       "001e0b925f1b35cc        plate_4_control\n",
       "                           ...         \n",
       "ffed9bdf747330e3        plate_4_control\n",
       "fff1a72e22675ad6        plate_0_control\n",
       "fff31a28495ca441        plate_0_control\n",
       "fff3c497006b77dd        plate_0_control\n",
       "fff4eafbf448b5ec        plate_0_control\n",
       "Name: sample, Length: 19281, dtype: object"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata.obs['sample']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "894be78f-a58b-48ac-92cf-9bd01a1dbfef",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Replace spaces with _; remove +\n",
    "cdata.obs['cell_type'] = [ct.replace(\" \", \"_\") for ct in cdata.obs['cell_type']]\n",
    "cdata.obs['cell_type'] = [ct.replace(\"+\", \"\") for ct in cdata.obs['cell_type']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "93231281-a6e0-4e78-9edd-0df3edf28eaa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'T_cells_CD4': 16469, 'NK_cells': 2550, 'B_cells': 262}"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count_entries(cdata.obs, 'cell_type')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "193a9704-eacd-4d47-a688-2e446c2c8769",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Force the following variables to be categorical\n",
    "cat_var = ['plate_name', 'condition', 'sample', 'cell_type']\n",
    "cdata.obs[cat_var] = cdata.obs[cat_var].astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "7e34328f-c787-49e6-854e-b9a78af2b0b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Original function\n",
    "NUM_OF_CELL_PER_DONOR = 30 # Filter out donors that have fewer than 30 cells for the specified population\n",
    "\n",
    "def aggregate_and_filter(\n",
    "    adata,\n",
    "    cell_identity,\n",
    "    donor_key='sample',\n",
    "    condition_key='condition',\n",
    "    cell_identity_key='cell_type',\n",
    "    obs_to_keep=[], # Which additional metadata to keep, e.g. gender, age, etc.\n",
    "    replicates_per_patient=1,\n",
    "):\n",
    "    \"\"\"\n",
    "    Aggregate and filter single-cell data from an AnnData object based on specified cell identity, \n",
    "    with additional donor-based filtering and data aggregation.\n",
    "\n",
    "    This function filters cells based on the specified cell identity, then aggregates data for each donor,\n",
    "    and further divides the data into specified numbers of replicates per donor. It then aggregates \n",
    "    this data (summing gene expression) and retains specified metadata.\n",
    "\n",
    "    Parameters:\n",
    "    -----------\n",
    "    adata : AnnData\n",
    "        The AnnData object containing the single-cell dataset.\n",
    "    cell_identity : str\n",
    "        The cell identity to filter on. Only cells with this identity are considered.\n",
    "    donor_key : str, optional\n",
    "        The key in adata.obs used to identify different donors. Defaults to 'sample'.\n",
    "    condition_key : str, optional\n",
    "        The key in adata.obs used to identify different conditions. Defaults to 'condition'.\n",
    "    cell_identity_key : str, optional\n",
    "        The key in adata.obs used to identify cell types. Defaults to 'cell_type'.\n",
    "    obs_to_keep : list of str, optional\n",
    "        A list of additional metadata fields to keep from adata.obs. Defaults to an empty list.\n",
    "    replicates_per_patient : int, optional\n",
    "        The number of replicates to create for each donor. Defaults to 1.\n",
    "\n",
    "    Returns:\n",
    "    --------\n",
    "    AnnData\n",
    "        An aggregated AnnData object with summed gene expression and retained metadata, \n",
    "        filtered for specified cell identity and donors meeting the cell count criteria.\n",
    "\n",
    "    Notes:\n",
    "    ------\n",
    "    - The function filters out donors with fewer than 30 cells for the specified cell identity.\n",
    "    - It assumes that the entire filtered dataset fits into memory.\n",
    "    - Random shuffling is used for creating replicates, which introduces a random element to the output.\n",
    "\n",
    "    Example:\n",
    "    --------\n",
    "    aggregated_data = aggregate_and_filter(adata, 'T-cell', obs_to_keep=['age', 'gender'])\n",
    "    \"\"\"\n",
    "    \n",
    "    # Filter the adata object to include only cells of a specified type \n",
    "    adata_cell_pop = adata[adata.obs[cell_identity_key] == cell_identity].copy()\n",
    "    # Calculate the number of cells per donor \n",
    "    size_by_donor = adata_cell_pop.obs.groupby([donor_key]).size()\n",
    "    # Filter out donors with fewer cells than NUM_OF_CELL_PER_DONOR\n",
    "    donors_to_drop = [\n",
    "        donor\n",
    "        for donor in size_by_donor.index\n",
    "        if size_by_donor[donor] <= NUM_OF_CELL_PER_DONOR\n",
    "    ]\n",
    "    if len(donors_to_drop) > 0:\n",
    "        print('Drop the following samples:')\n",
    "        print(donors_to_drop)\n",
    "\n",
    "    # Initialize a DataFrame df to store aggregated data\n",
    "    df = pd.DataFrame(columns=[*adata_cell_pop.var_names, *obs_to_keep])\n",
    "    # Categorize donor_key\n",
    "    adata_cell_pop.obs[donor_key] = adata_cell_pop.obs[donor_key].astype('category')\n",
    "    # Iterate over each donor (which is represented as a categorical level)\n",
    "    for i, donor in enumerate(donors := adata_cell_pop.obs[donor_key].cat.categories):\n",
    "        print(f'\\tProcess donor {i+1} out of {len(donors)}...', end='\\r')\n",
    "\n",
    "        # Check if the donor is not in the filtered list\n",
    "        if donor not in donors_to_drop:\n",
    "            adata_donor = adata_cell_pop[adata_cell_pop.obs[donor_key] == donor]\n",
    "            # Split the data for each donor into replicates\n",
    "            indices = list(adata_donor.obs_names) # Extract the unique identifiers of each cell in an AnnData object (adata_donor) and store them in a Python list\n",
    "            random.shuffle(indices) # Randomly shuffle (reorder) the elements in the indices list in-place \n",
    "            indices = np.array_split(np.array(indices), replicates_per_patient) # Split an array (or list) of indices into multiple subarrays or sublists, where the number of subarrays is determined by the replicates_per_patient variable\n",
    "\n",
    "            # For each replicate, it aggregates data (sums gene expression and keeps additional metadata)\n",
    "            for i, rep_idx in enumerate(indices):\n",
    "                adata_replicate = adata_donor[rep_idx] # rep_idx is a sublist of cell identifiers for a specific replicate; if replicate per donor is 1, then rep_idx is a whole list of cell identifiers for that donor\n",
    "                agg_dict = {gene: 'sum' for gene in adata_replicate.var_names} # The 'sum' specifies the aggregation method to compute the sum of gene expression values when aggregating data for each gene within the adata_replicate \n",
    "                for obs in obs_to_keep:\n",
    "                    agg_dict[obs] = 'first' # For the metadata variables associated with each replicate, it should take the first value encountered\n",
    "                # Build a df with all genes, donor and metadata\n",
    "                df_donor = pd.DataFrame(adata_replicate.X.A) # adata_replicate.X typically contains the matrix of gene expression values, and .A is used to access the underlying NumPy array representation of the data\n",
    "                df_donor.index = adata_replicate.obs_names # The row (index) labels of the df_donor DataFrame are set to the observation names from adata_replicate\n",
    "                df_donor.columns = adata_replicate.var_names # Set the column names of the df_donor DataFrame to the gene names (variable names) from adata_replicate\n",
    "                df_donor = df_donor.join(adata_replicate.obs[obs_to_keep]) # Add additional metadata columns to the df_donor DataFrame \n",
    "                # Aggregate data within a specific donor \n",
    "                df_donor = df_donor.groupby(donor_key).agg(agg_dict) # Apply the specified aggregation methods (summation) to each replicate belonging to the same donor\n",
    "                df_donor[donor_key] = donor # A new column named donor_key is added to the df_donor DataFrame, and its value is set to the donor identifier\n",
    "                df.loc[f\"donor_{donor}_{i}\"] = df_donor.loc[donor] # It assigns the aggregated data for the current donor (identified by donor) to a new row in df. The row label is constructed using an f-string, which includes the donor identifier (donor) and a unique identifier (i) for the current replicate\n",
    "    print('\\n')\n",
    "    # Convert the aggregated DataFrame back into an AnnData object for further analysis\n",
    "    adata_cell_pop = sc.AnnData(\n",
    "        df[adata_cell_pop.var_names], obs=df.drop(columns=adata_cell_pop.var_names) # It assigns the columns from the DataFrame df that are not included in adata_cell_pop.var_names as the observation metadata. In other words, it includes any metadata columns from df that are not related to gene expression\n",
    "    )\n",
    "    return adata_cell_pop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "2dd80aed-3923-40c7-81c1-9855d2778195",
   "metadata": {},
   "outputs": [],
   "source": [
    "obs_to_keep = ['condition', 'cell_type', 'sample']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "45452a39-771e-461c-8c65-9b198d1b06e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "cdata.X = cdata.layers[\"counts\"].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "6d374b57-10ca-4f78-b6b5-7bdb89f3621d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "obs_id\n",
       "0002560bd38ce03e    T_cells_CD4\n",
       "0004ce8793cc33e4    T_cells_CD4\n",
       "0009dd2a269fc9ce    T_cells_CD4\n",
       "0019c2b227447e12    T_cells_CD4\n",
       "001e0b925f1b35cc       NK_cells\n",
       "                       ...     \n",
       "ffed9bdf747330e3    T_cells_CD4\n",
       "fff1a72e22675ad6    T_cells_CD4\n",
       "fff31a28495ca441       NK_cells\n",
       "fff3c497006b77dd    T_cells_CD4\n",
       "fff4eafbf448b5ec    T_cells_CD4\n",
       "Name: cell_type, Length: 19281, dtype: category\n",
       "Categories (3, object): ['B_cells', 'NK_cells', 'T_cells_CD4']"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata.obs['cell_type']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "7dbdddca-258c-4eb0-80aa-e5a1c0f21c45",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing B_cells (1 out of 3)...\n",
      "Drop the following samples:\n",
      "['plate_1_control']\n",
      "\tProcess donor 6 out of 6...\n",
      "\n",
      "Processing NK_cells (2 out of 3)...\n",
      "\tProcess donor 6 out of 6...\n",
      "\n",
      "Processing T_cells_CD4 (3 out of 3)...\n",
      "\tProcess donor 6 out of 6...\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Get the list of cell categories\n",
    "cell_categories = cdata.obs[\"cell_type\"].cat.categories\n",
    "\n",
    "# Initialize the concatenated data object\n",
    "cdata_pb = None\n",
    "\n",
    "# Iterate through cell types and aggregate/concatenate\n",
    "for i, cell_type in enumerate(cell_categories):\n",
    "    print(\n",
    "        f'Processing {cell_type} ({i+1} out of {len(cell_categories)})...'\n",
    "    )\n",
    "    cdata_cell_type = aggregate_and_filter(cdata, cell_type, obs_to_keep=obs_to_keep)\n",
    "    if cdata_pb is None:\n",
    "        cdata_pb = cdata_cell_type\n",
    "    else:\n",
    "        cdata_pb = cdata_pb.concatenate(cdata_cell_type)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "8e13906f-295a-4e09-a094-16e6e2296e10",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>A1BG</th>\n",
       "      <th>A1BG-AS1</th>\n",
       "      <th>A2M</th>\n",
       "      <th>A2M-AS1</th>\n",
       "      <th>A2MP1</th>\n",
       "      <th>A4GALT</th>\n",
       "      <th>AAAS</th>\n",
       "      <th>AACS</th>\n",
       "      <th>AAGAB</th>\n",
       "      <th>AAK1</th>\n",
       "      <th>...</th>\n",
       "      <th>ZW10</th>\n",
       "      <th>ZWILCH</th>\n",
       "      <th>ZWINT</th>\n",
       "      <th>ZXDA</th>\n",
       "      <th>ZXDB</th>\n",
       "      <th>ZXDC</th>\n",
       "      <th>ZYG11A</th>\n",
       "      <th>ZYG11B</th>\n",
       "      <th>ZYX</th>\n",
       "      <th>ZZEF1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>donor_plate_0_control_0-0-0</th>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_2_perturbated_0-0-0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_3_perturbated_0-0-0</th>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_4_control_0-0-0</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_5_perturbated_0-0-0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_0_control_0-1-0</th>\n",
       "      <td>5.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>59.0</td>\n",
       "      <td>144.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>86.0</td>\n",
       "      <td>476.0</td>\n",
       "      <td>...</td>\n",
       "      <td>34.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>117.0</td>\n",
       "      <td>105.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_1_control_0-1-0</th>\n",
       "      <td>4.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>49.0</td>\n",
       "      <td>130.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>56.0</td>\n",
       "      <td>413.0</td>\n",
       "      <td>...</td>\n",
       "      <td>16.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>78.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_2_perturbated_0-1-0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_3_perturbated_0-1-0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>56.0</td>\n",
       "      <td>...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_4_control_0-1-0</th>\n",
       "      <td>6.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>161.0</td>\n",
       "      <td>295.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>54.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>47.0</td>\n",
       "      <td>974.0</td>\n",
       "      <td>...</td>\n",
       "      <td>19.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>47.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>174.0</td>\n",
       "      <td>123.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_5_perturbated_0-1-0</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>84.0</td>\n",
       "      <td>...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_0_control_0-1</th>\n",
       "      <td>99.0</td>\n",
       "      <td>183.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>261.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>368.0</td>\n",
       "      <td>129.0</td>\n",
       "      <td>634.0</td>\n",
       "      <td>6507.0</td>\n",
       "      <td>...</td>\n",
       "      <td>202.0</td>\n",
       "      <td>175.0</td>\n",
       "      <td>91.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>169.0</td>\n",
       "      <td>396.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>863.0</td>\n",
       "      <td>1105.0</td>\n",
       "      <td>855.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_1_control_0-1</th>\n",
       "      <td>27.0</td>\n",
       "      <td>48.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>112.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>123.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>153.0</td>\n",
       "      <td>1773.0</td>\n",
       "      <td>...</td>\n",
       "      <td>49.0</td>\n",
       "      <td>52.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>95.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>186.0</td>\n",
       "      <td>391.0</td>\n",
       "      <td>203.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_2_perturbated_0-1</th>\n",
       "      <td>16.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>47.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>117.0</td>\n",
       "      <td>860.0</td>\n",
       "      <td>...</td>\n",
       "      <td>33.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>102.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_3_perturbated_0-1</th>\n",
       "      <td>13.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>62.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>116.0</td>\n",
       "      <td>979.0</td>\n",
       "      <td>...</td>\n",
       "      <td>34.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>52.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>194.0</td>\n",
       "      <td>85.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_4_control_0-1</th>\n",
       "      <td>41.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>177.0</td>\n",
       "      <td>288.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>163.0</td>\n",
       "      <td>72.0</td>\n",
       "      <td>303.0</td>\n",
       "      <td>4194.0</td>\n",
       "      <td>...</td>\n",
       "      <td>93.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>383.0</td>\n",
       "      <td>871.0</td>\n",
       "      <td>372.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_5_perturbated_0-1</th>\n",
       "      <td>24.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>156.0</td>\n",
       "      <td>1559.0</td>\n",
       "      <td>...</td>\n",
       "      <td>51.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>62.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>308.0</td>\n",
       "      <td>162.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>17 rows × 21231 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 A1BG A1BG-AS1    A2M A2M-AS1 A2MP1 A4GALT  \\\n",
       "donor_plate_0_control_0-0-0       2.0      4.0    0.0     0.0   0.0    0.0   \n",
       "donor_plate_2_perturbated_0-0-0   0.0      2.0    0.0     0.0   0.0    0.0   \n",
       "donor_plate_3_perturbated_0-0-0   2.0      6.0    0.0     0.0   0.0    0.0   \n",
       "donor_plate_4_control_0-0-0       2.0      0.0    0.0     0.0   0.0    0.0   \n",
       "donor_plate_5_perturbated_0-0-0   1.0      1.0    0.0     0.0   0.0    0.0   \n",
       "donor_plate_0_control_0-1-0       5.0     21.0   59.0   144.0  11.0    0.0   \n",
       "donor_plate_1_control_0-1-0       4.0      6.0   49.0   130.0   5.0    0.0   \n",
       "donor_plate_2_perturbated_0-1-0   0.0      1.0    0.0     1.0   0.0    0.0   \n",
       "donor_plate_3_perturbated_0-1-0   0.0      1.0    1.0     3.0   1.0    0.0   \n",
       "donor_plate_4_control_0-1-0       6.0     16.0  161.0   295.0  15.0    0.0   \n",
       "donor_plate_5_perturbated_0-1-0   2.0      3.0    9.0    22.0   0.0    0.0   \n",
       "donor_plate_0_control_0-1        99.0    183.0   89.0   261.0  18.0    0.0   \n",
       "donor_plate_1_control_0-1        27.0     48.0   55.0   112.0   8.0    0.0   \n",
       "donor_plate_2_perturbated_0-1    16.0     43.0    7.0    18.0   3.0    0.0   \n",
       "donor_plate_3_perturbated_0-1    13.0     24.0    2.0     9.0   0.0    0.0   \n",
       "donor_plate_4_control_0-1        41.0     99.0  177.0   288.0  21.0    0.0   \n",
       "donor_plate_5_perturbated_0-1    24.0     40.0    8.0    31.0   2.0    0.0   \n",
       "\n",
       "                                  AAAS   AACS  AAGAB    AAK1  ...   ZW10  \\\n",
       "donor_plate_0_control_0-0-0        9.0    1.0    7.0     2.0  ...    5.0   \n",
       "donor_plate_2_perturbated_0-0-0    4.0    2.0    6.0     2.0  ...    1.0   \n",
       "donor_plate_3_perturbated_0-0-0    6.0    1.0   10.0     7.0  ...    3.0   \n",
       "donor_plate_4_control_0-0-0        2.0    0.0    2.0     1.0  ...    1.0   \n",
       "donor_plate_5_perturbated_0-0-0    1.0    1.0    8.0     3.0  ...    2.0   \n",
       "donor_plate_0_control_0-1-0       28.0   27.0   86.0   476.0  ...   34.0   \n",
       "donor_plate_1_control_0-1-0       26.0   11.0   56.0   413.0  ...   16.0   \n",
       "donor_plate_2_perturbated_0-1-0    4.0    1.0   11.0    61.0  ...    4.0   \n",
       "donor_plate_3_perturbated_0-1-0    2.0    2.0   11.0    56.0  ...    6.0   \n",
       "donor_plate_4_control_0-1-0       54.0   21.0   47.0   974.0  ...   19.0   \n",
       "donor_plate_5_perturbated_0-1-0    3.0    2.0    9.0    84.0  ...    3.0   \n",
       "donor_plate_0_control_0-1        368.0  129.0  634.0  6507.0  ...  202.0   \n",
       "donor_plate_1_control_0-1        123.0   34.0  153.0  1773.0  ...   49.0   \n",
       "donor_plate_2_perturbated_0-1     47.0   20.0  117.0   860.0  ...   33.0   \n",
       "donor_plate_3_perturbated_0-1     62.0   23.0  116.0   979.0  ...   34.0   \n",
       "donor_plate_4_control_0-1        163.0   72.0  303.0  4194.0  ...   93.0   \n",
       "donor_plate_5_perturbated_0-1     80.0   35.0  156.0  1559.0  ...   51.0   \n",
       "\n",
       "                                ZWILCH ZWINT  ZXDA   ZXDB   ZXDC ZYG11A  \\\n",
       "donor_plate_0_control_0-0-0        4.0   0.0   2.0    1.0    4.0    0.0   \n",
       "donor_plate_2_perturbated_0-0-0    0.0   0.0   1.0    1.0    0.0    0.0   \n",
       "donor_plate_3_perturbated_0-0-0    0.0   0.0   0.0    1.0    3.0    0.0   \n",
       "donor_plate_4_control_0-0-0        4.0   0.0   1.0    0.0    1.0    0.0   \n",
       "donor_plate_5_perturbated_0-0-0    0.0   0.0   1.0    2.0    1.0    0.0   \n",
       "donor_plate_0_control_0-1-0       15.0  12.0   4.0   12.0   31.0    0.0   \n",
       "donor_plate_1_control_0-1-0       12.0   6.0   5.0    4.0   19.0    0.0   \n",
       "donor_plate_2_perturbated_0-1-0    0.0   0.0   0.0    2.0    4.0    0.0   \n",
       "donor_plate_3_perturbated_0-1-0    0.0   0.0   1.0    0.0    1.0    0.0   \n",
       "donor_plate_4_control_0-1-0       33.0   3.0  11.0   15.0   47.0    0.0   \n",
       "donor_plate_5_perturbated_0-1-0    2.0   0.0   0.0    1.0    7.0    0.0   \n",
       "donor_plate_0_control_0-1        175.0  91.0  88.0  169.0  396.0    0.0   \n",
       "donor_plate_1_control_0-1         52.0  18.0  32.0   39.0   95.0    0.0   \n",
       "donor_plate_2_perturbated_0-1     13.0  12.0   5.0   12.0   41.0    0.0   \n",
       "donor_plate_3_perturbated_0-1     21.0  11.0   8.0   10.0   52.0    0.0   \n",
       "donor_plate_4_control_0-1         99.0  27.0  40.0   61.0  180.0    0.0   \n",
       "donor_plate_5_perturbated_0-1     36.0  22.0   6.0   25.0   62.0    0.0   \n",
       "\n",
       "                                ZYG11B     ZYX  ZZEF1  \n",
       "donor_plate_0_control_0-0-0       11.0     3.0    7.0  \n",
       "donor_plate_2_perturbated_0-0-0    3.0     1.0    2.0  \n",
       "donor_plate_3_perturbated_0-0-0   13.0     9.0    4.0  \n",
       "donor_plate_4_control_0-0-0        2.0     1.0    1.0  \n",
       "donor_plate_5_perturbated_0-0-0    5.0     3.0    2.0  \n",
       "donor_plate_0_control_0-1-0       35.0   117.0  105.0  \n",
       "donor_plate_1_control_0-1-0       34.0    83.0   78.0  \n",
       "donor_plate_2_perturbated_0-1-0    5.0    16.0    5.0  \n",
       "donor_plate_3_perturbated_0-1-0    7.0     7.0    5.0  \n",
       "donor_plate_4_control_0-1-0       43.0   174.0  123.0  \n",
       "donor_plate_5_perturbated_0-1-0    3.0    16.0   13.0  \n",
       "donor_plate_0_control_0-1        863.0  1105.0  855.0  \n",
       "donor_plate_1_control_0-1        186.0   391.0  203.0  \n",
       "donor_plate_2_perturbated_0-1    120.0   150.0  102.0  \n",
       "donor_plate_3_perturbated_0-1    100.0   194.0   85.0  \n",
       "donor_plate_4_control_0-1        383.0   871.0  372.0  \n",
       "donor_plate_5_perturbated_0-1    150.0   308.0  162.0  \n",
       "\n",
       "[17 rows x 21231 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_pb.to_df()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c2f8836-0279-4b45-bfaf-78f75c60044b",
   "metadata": {},
   "source": [
    "# Perform EDA on the created pseudobulks"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80d4792b-e7e6-4e33-9713-d604f081d1b4",
   "metadata": {},
   "source": [
    "Purpose: if we need to exclude some outlier patients that could bias the DE results."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9178b823-9af3-46b7-accf-5e4ceb46f0d5",
   "metadata": {},
   "source": [
    "[Guide of design matrices](https://bioconductor.org/packages/release/workflows/vignettes/RNAseq123/inst/doc/designmatrices.html#design-matrices-with-and-without-intercept-term)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "521db952-840c-41c9-8e9a-2b47f906b145",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[2.0, 4.0, 0.0, ..., 11.0, 3.0, 7.0],\n",
       "       [0.0, 2.0, 0.0, ..., 3.0, 1.0, 2.0],\n",
       "       [2.0, 6.0, 0.0, ..., 13.0, 9.0, 4.0],\n",
       "       ...,\n",
       "       [13.0, 24.0, 2.0, ..., 100.0, 194.0, 85.0],\n",
       "       [41.0, 99.0, 177.0, ..., 383.0, 871.0, 372.0],\n",
       "       [24.0, 40.0, 8.0, ..., 150.0, 308.0, 162.0]], dtype=object)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_pb.X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "0ca7caf8-294c-406a-bfef-4ffa7d463813",
   "metadata": {},
   "outputs": [],
   "source": [
    "cdata_pb.layers['counts'] = cdata_pb.X.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "e4e2eadd-3cbd-45bc-8905-d84a0c394f14",
   "metadata": {},
   "outputs": [],
   "source": [
    "sc.pp.normalize_total(cdata_pb, target_sum=1e6)\n",
    "sc.pp.log1p(cdata_pb)\n",
    "sc.pp.pca(cdata_pb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "3266bcd0-d8f9-4221-8832-8e110370f8c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "cdata_pb.obs['lib_size'] = np.sum(cdata_pb.layers['counts'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "04113993-e9b5-434e-af4d-3deb04186a49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "donor_plate_0_control_0-0-0          356966.0\n",
       "donor_plate_2_perturbated_0-0-0      240979.0\n",
       "donor_plate_3_perturbated_0-0-0      410903.0\n",
       "donor_plate_4_control_0-0-0          186590.0\n",
       "donor_plate_5_perturbated_0-0-0      235308.0\n",
       "donor_plate_0_control_0-1-0         3020734.0\n",
       "donor_plate_1_control_0-1-0         2173727.0\n",
       "donor_plate_2_perturbated_0-1-0      267785.0\n",
       "donor_plate_3_perturbated_0-1-0      246949.0\n",
       "donor_plate_4_control_0-1-0         3421251.0\n",
       "donor_plate_5_perturbated_0-1-0      396553.0\n",
       "donor_plate_0_control_0-1          41624636.0\n",
       "donor_plate_1_control_0-1          11336428.0\n",
       "donor_plate_2_perturbated_0-1       5791472.0\n",
       "donor_plate_3_perturbated_0-1       6216921.0\n",
       "donor_plate_4_control_0-1          22339083.0\n",
       "donor_plate_5_perturbated_0-1       8803763.0\n",
       "Name: lib_size, dtype: object"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_pb.obs['lib_size']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "b3c6f5f4-f281-44cf-ac6b-ed4ca772f0ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "# https://stackoverflow.com/questions/47208473/attributeerror-numpy-float64-object-has-no-attribute-log10\n",
    "cdata_pb.obs['log_lib_size'] = np.log(cdata_pb.obs['lib_size'].astype(np.float64))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "bae28b5f-23b8-4210-8ff9-353617bd8890",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "donor_plate_0_control_0-0-0        12.785396\n",
       "donor_plate_2_perturbated_0-0-0    12.392465\n",
       "donor_plate_3_perturbated_0-0-0    12.926112\n",
       "donor_plate_4_control_0-0-0        12.136669\n",
       "donor_plate_5_perturbated_0-0-0    12.368651\n",
       "donor_plate_0_control_0-1-0        14.921010\n",
       "donor_plate_1_control_0-1-0        14.591954\n",
       "donor_plate_2_perturbated_0-1-0    12.497940\n",
       "donor_plate_3_perturbated_0-1-0    12.416937\n",
       "donor_plate_4_control_0-1-0        15.045517\n",
       "donor_plate_5_perturbated_0-1-0    12.890565\n",
       "donor_plate_0_control_0-1          17.544203\n",
       "donor_plate_1_control_0-1          16.243532\n",
       "donor_plate_2_perturbated_0-1      15.571897\n",
       "donor_plate_3_perturbated_0-1      15.642785\n",
       "donor_plate_4_control_0-1          16.921848\n",
       "donor_plate_5_perturbated_0-1      15.990690\n",
       "Name: log_lib_size, dtype: float64"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_pb.obs['log_lib_size']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "5ab905f3-984a-4df8-9534-c41deca9d2fa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>condition</th>\n",
       "      <th>cell_type</th>\n",
       "      <th>sample</th>\n",
       "      <th>batch</th>\n",
       "      <th>lib_size</th>\n",
       "      <th>log_lib_size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>donor_plate_0_control_0-0-0</th>\n",
       "      <td>control</td>\n",
       "      <td>B_cells</td>\n",
       "      <td>plate_0_control</td>\n",
       "      <td>0</td>\n",
       "      <td>356966.0</td>\n",
       "      <td>12.785396</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_2_perturbated_0-0-0</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>B_cells</td>\n",
       "      <td>plate_2_perturbated</td>\n",
       "      <td>0</td>\n",
       "      <td>240979.0</td>\n",
       "      <td>12.392465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_3_perturbated_0-0-0</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>B_cells</td>\n",
       "      <td>plate_3_perturbated</td>\n",
       "      <td>0</td>\n",
       "      <td>410903.0</td>\n",
       "      <td>12.926112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_4_control_0-0-0</th>\n",
       "      <td>control</td>\n",
       "      <td>B_cells</td>\n",
       "      <td>plate_4_control</td>\n",
       "      <td>0</td>\n",
       "      <td>186590.0</td>\n",
       "      <td>12.136669</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_5_perturbated_0-0-0</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>B_cells</td>\n",
       "      <td>plate_5_perturbated</td>\n",
       "      <td>0</td>\n",
       "      <td>235308.0</td>\n",
       "      <td>12.368651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_0_control_0-1-0</th>\n",
       "      <td>control</td>\n",
       "      <td>NK_cells</td>\n",
       "      <td>plate_0_control</td>\n",
       "      <td>0</td>\n",
       "      <td>3020734.0</td>\n",
       "      <td>14.921010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_1_control_0-1-0</th>\n",
       "      <td>control</td>\n",
       "      <td>NK_cells</td>\n",
       "      <td>plate_1_control</td>\n",
       "      <td>0</td>\n",
       "      <td>2173727.0</td>\n",
       "      <td>14.591954</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_2_perturbated_0-1-0</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>NK_cells</td>\n",
       "      <td>plate_2_perturbated</td>\n",
       "      <td>0</td>\n",
       "      <td>267785.0</td>\n",
       "      <td>12.497940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_3_perturbated_0-1-0</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>NK_cells</td>\n",
       "      <td>plate_3_perturbated</td>\n",
       "      <td>0</td>\n",
       "      <td>246949.0</td>\n",
       "      <td>12.416937</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_4_control_0-1-0</th>\n",
       "      <td>control</td>\n",
       "      <td>NK_cells</td>\n",
       "      <td>plate_4_control</td>\n",
       "      <td>0</td>\n",
       "      <td>3421251.0</td>\n",
       "      <td>15.045517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_5_perturbated_0-1-0</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>NK_cells</td>\n",
       "      <td>plate_5_perturbated</td>\n",
       "      <td>0</td>\n",
       "      <td>396553.0</td>\n",
       "      <td>12.890565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_0_control_0-1</th>\n",
       "      <td>control</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_0_control</td>\n",
       "      <td>1</td>\n",
       "      <td>41624636.0</td>\n",
       "      <td>17.544203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_1_control_0-1</th>\n",
       "      <td>control</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_1_control</td>\n",
       "      <td>1</td>\n",
       "      <td>11336428.0</td>\n",
       "      <td>16.243532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_2_perturbated_0-1</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_2_perturbated</td>\n",
       "      <td>1</td>\n",
       "      <td>5791472.0</td>\n",
       "      <td>15.571897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_3_perturbated_0-1</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_3_perturbated</td>\n",
       "      <td>1</td>\n",
       "      <td>6216921.0</td>\n",
       "      <td>15.642785</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_4_control_0-1</th>\n",
       "      <td>control</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_4_control</td>\n",
       "      <td>1</td>\n",
       "      <td>22339083.0</td>\n",
       "      <td>16.921848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donor_plate_5_perturbated_0-1</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_5_perturbated</td>\n",
       "      <td>1</td>\n",
       "      <td>8803763.0</td>\n",
       "      <td>15.990690</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   condition    cell_type  \\\n",
       "donor_plate_0_control_0-0-0          control      B_cells   \n",
       "donor_plate_2_perturbated_0-0-0  perturbated      B_cells   \n",
       "donor_plate_3_perturbated_0-0-0  perturbated      B_cells   \n",
       "donor_plate_4_control_0-0-0          control      B_cells   \n",
       "donor_plate_5_perturbated_0-0-0  perturbated      B_cells   \n",
       "donor_plate_0_control_0-1-0          control     NK_cells   \n",
       "donor_plate_1_control_0-1-0          control     NK_cells   \n",
       "donor_plate_2_perturbated_0-1-0  perturbated     NK_cells   \n",
       "donor_plate_3_perturbated_0-1-0  perturbated     NK_cells   \n",
       "donor_plate_4_control_0-1-0          control     NK_cells   \n",
       "donor_plate_5_perturbated_0-1-0  perturbated     NK_cells   \n",
       "donor_plate_0_control_0-1            control  T_cells_CD4   \n",
       "donor_plate_1_control_0-1            control  T_cells_CD4   \n",
       "donor_plate_2_perturbated_0-1    perturbated  T_cells_CD4   \n",
       "donor_plate_3_perturbated_0-1    perturbated  T_cells_CD4   \n",
       "donor_plate_4_control_0-1            control  T_cells_CD4   \n",
       "donor_plate_5_perturbated_0-1    perturbated  T_cells_CD4   \n",
       "\n",
       "                                              sample batch    lib_size  \\\n",
       "donor_plate_0_control_0-0-0          plate_0_control     0    356966.0   \n",
       "donor_plate_2_perturbated_0-0-0  plate_2_perturbated     0    240979.0   \n",
       "donor_plate_3_perturbated_0-0-0  plate_3_perturbated     0    410903.0   \n",
       "donor_plate_4_control_0-0-0          plate_4_control     0    186590.0   \n",
       "donor_plate_5_perturbated_0-0-0  plate_5_perturbated     0    235308.0   \n",
       "donor_plate_0_control_0-1-0          plate_0_control     0   3020734.0   \n",
       "donor_plate_1_control_0-1-0          plate_1_control     0   2173727.0   \n",
       "donor_plate_2_perturbated_0-1-0  plate_2_perturbated     0    267785.0   \n",
       "donor_plate_3_perturbated_0-1-0  plate_3_perturbated     0    246949.0   \n",
       "donor_plate_4_control_0-1-0          plate_4_control     0   3421251.0   \n",
       "donor_plate_5_perturbated_0-1-0  plate_5_perturbated     0    396553.0   \n",
       "donor_plate_0_control_0-1            plate_0_control     1  41624636.0   \n",
       "donor_plate_1_control_0-1            plate_1_control     1  11336428.0   \n",
       "donor_plate_2_perturbated_0-1    plate_2_perturbated     1   5791472.0   \n",
       "donor_plate_3_perturbated_0-1    plate_3_perturbated     1   6216921.0   \n",
       "donor_plate_4_control_0-1            plate_4_control     1  22339083.0   \n",
       "donor_plate_5_perturbated_0-1    plate_5_perturbated     1   8803763.0   \n",
       "\n",
       "                                 log_lib_size  \n",
       "donor_plate_0_control_0-0-0         12.785396  \n",
       "donor_plate_2_perturbated_0-0-0     12.392465  \n",
       "donor_plate_3_perturbated_0-0-0     12.926112  \n",
       "donor_plate_4_control_0-0-0         12.136669  \n",
       "donor_plate_5_perturbated_0-0-0     12.368651  \n",
       "donor_plate_0_control_0-1-0         14.921010  \n",
       "donor_plate_1_control_0-1-0         14.591954  \n",
       "donor_plate_2_perturbated_0-1-0     12.497940  \n",
       "donor_plate_3_perturbated_0-1-0     12.416937  \n",
       "donor_plate_4_control_0-1-0         15.045517  \n",
       "donor_plate_5_perturbated_0-1-0     12.890565  \n",
       "donor_plate_0_control_0-1           17.544203  \n",
       "donor_plate_1_control_0-1           16.243532  \n",
       "donor_plate_2_perturbated_0-1       15.571897  \n",
       "donor_plate_3_perturbated_0-1       15.642785  \n",
       "donor_plate_4_control_0-1           16.921848  \n",
       "donor_plate_5_perturbated_0-1       15.990690  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_pb.obs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "0cb965d3-5627-4c1b-b746-cb41a318fee0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 727.8x2880 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sc.pl.pca(cdata_pb, color=cdata_pb.obs, ncols=1, size=200)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e38d207-4d38-4ac0-b9c6-0f0cb16dffd5",
   "metadata": {},
   "source": [
    "The PCA plots show the separation of reducted gene expression across cell types as well as the separation of reducted gene expression between perturbated and control cells. Although batch and library size appear to account for the major variations in PCA components, they are captured in the variable cell type, so we do not repetitively include any of them into the design matrix."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "326eaf35-762c-4cf7-a1fb-fac7ffd2d5f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "cdata_pb.X = cdata_pb.layers['counts'].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "1ca5bd77-3dbc-4dd0-9b04-54c55a7e0455",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[2.0, 4.0, 0.0, ..., 11.0, 3.0, 7.0],\n",
       "       [0.0, 2.0, 0.0, ..., 3.0, 1.0, 2.0],\n",
       "       [2.0, 6.0, 0.0, ..., 13.0, 9.0, 4.0],\n",
       "       ...,\n",
       "       [13.0, 24.0, 2.0, ..., 100.0, 194.0, 85.0],\n",
       "       [41.0, 99.0, 177.0, ..., 383.0, 871.0, 372.0],\n",
       "       [24.0, 40.0, 8.0, ..., 150.0, 308.0, 162.0]], dtype=object)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_pb.X"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3feb8dc0-412a-475c-9d14-b7678a58da57",
   "metadata": {},
   "source": [
    "# Pseudobulk-based DE analysis in CD4+ T cells"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "f37370a4-20d2-411d-8b42-b030bde65f5e",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%R\n",
    "fit_model <- function(adata_obj){\n",
    "    y <- DGEList(assay(adata_obj, \"X\"), group = colData(adata_obj)$condition)\n",
    "    cat(\"Dimensions before subsetting: \", paste(dim(y), collapse = \"x\"), \"\\n\")\n",
    "    keep <- filterByExpr(y)\n",
    "    # Filtering keeps genes that have count-per-million (CPM) above k in n samples, where k is determined by min.count and by the sample library sizes, and n is determined by the design matrix.\n",
    "    # n is essentially the smallest group sample size or, more generally, the minimum inverse leverage of any fitted value. \n",
    "    y <- y[keep, , keep.lib.sizes=FALSE] # The lib.size for each sample will be recalculated after filtering\n",
    "    cat(\"Dimensions after subsetting: \", paste(dim(y), collapse = \"x\"), \"\\n\")\n",
    "    y <- calcNormFactors(y)\n",
    "    # Generate a vector by combining the condition and cell type, for the subsequent use of creating contrasts.\n",
    "    group <- paste0(colData(adata_obj)$condition, \".\", colData(adata_obj)$cell_type)\n",
    "    if (length(colData(adata_obj)$replicate) != 0) {\n",
    "        replicate <- colData(adata_obj)$replicate\n",
    "        design_mat <- model.matrix(~ group + replicate) \n",
    "    } else {\n",
    "    design_mat <- model.matrix(~ group) }\n",
    "    # design_mat <- model.matrix(~ 0 + group + replicate) \n",
    "    # The design matrix will exclude a column of ones for the intercept.\n",
    "    # A model without an intercept term would only be recommended in cases where there is a strong biological reason why a zero covariate should be associated with a zero expression value, and such contexts are rare in gene expression modelling.\n",
    "    y <- estimateDisp(y, design = design_mat)\n",
    "    fit <- glmQLFit(y, design_mat)\n",
    "    return(list(\"fit\"=fit, \"design\"=design_mat, \"y\"=y))\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "1cf59b84-27a6-4174-a0cc-83aad0d82c86",
   "metadata": {},
   "outputs": [],
   "source": [
    "cdata_cd4 = cdata_pb[cdata_pb.obs['cell_type'] == 'T_cells_CD4']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "04cf9046-ed2a-4e78-bdf3-351e037268c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['donor_plate_0_control_0-1', 'donor_plate_1_control_0-1',\n",
       "       'donor_plate_2_perturbated_0-1', 'donor_plate_3_perturbated_0-1',\n",
       "       'donor_plate_4_control_0-1', 'donor_plate_5_perturbated_0-1'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_cd4.obs_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "4ab65753-a9e5-41ef-9e9d-9bf3eaa78bb7",
   "metadata": {},
   "outputs": [],
   "source": [
    "cdata_cd4.obs_names = [\n",
    "    i.split(\"_\")[2] + \"_\" + i.split(\"_\")[3] for i in cdata_cd4.obs_names\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "9db9cfee-eea8-4c2a-b794-ae0df010fe22",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['0_control', '1_control', '2_perturbated', '3_perturbated', '4_control',\n",
       "       '5_perturbated'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_cd4.obs_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "8cf887a8-ae44-4177-bdbd-35524fbb168b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>condition</th>\n",
       "      <th>cell_type</th>\n",
       "      <th>sample</th>\n",
       "      <th>batch</th>\n",
       "      <th>lib_size</th>\n",
       "      <th>log_lib_size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0_control</th>\n",
       "      <td>control</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_0_control</td>\n",
       "      <td>1</td>\n",
       "      <td>41624636.0</td>\n",
       "      <td>17.544203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1_control</th>\n",
       "      <td>control</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_1_control</td>\n",
       "      <td>1</td>\n",
       "      <td>11336428.0</td>\n",
       "      <td>16.243532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2_perturbated</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_2_perturbated</td>\n",
       "      <td>1</td>\n",
       "      <td>5791472.0</td>\n",
       "      <td>15.571897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3_perturbated</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_3_perturbated</td>\n",
       "      <td>1</td>\n",
       "      <td>6216921.0</td>\n",
       "      <td>15.642785</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4_control</th>\n",
       "      <td>control</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_4_control</td>\n",
       "      <td>1</td>\n",
       "      <td>22339083.0</td>\n",
       "      <td>16.921848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5_perturbated</th>\n",
       "      <td>perturbated</td>\n",
       "      <td>T_cells_CD4</td>\n",
       "      <td>plate_5_perturbated</td>\n",
       "      <td>1</td>\n",
       "      <td>8803763.0</td>\n",
       "      <td>15.990690</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 condition    cell_type               sample batch  \\\n",
       "0_control          control  T_cells_CD4      plate_0_control     1   \n",
       "1_control          control  T_cells_CD4      plate_1_control     1   \n",
       "2_perturbated  perturbated  T_cells_CD4  plate_2_perturbated     1   \n",
       "3_perturbated  perturbated  T_cells_CD4  plate_3_perturbated     1   \n",
       "4_control          control  T_cells_CD4      plate_4_control     1   \n",
       "5_perturbated  perturbated  T_cells_CD4  plate_5_perturbated     1   \n",
       "\n",
       "                 lib_size  log_lib_size  \n",
       "0_control      41624636.0     17.544203  \n",
       "1_control      11336428.0     16.243532  \n",
       "2_perturbated   5791472.0     15.571897  \n",
       "3_perturbated   6216921.0     15.642785  \n",
       "4_control      22339083.0     16.921848  \n",
       "5_perturbated   8803763.0     15.990690  "
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_cd4.obs"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b488b0ef-c58c-49bf-909a-802dd51c8391",
   "metadata": {},
   "source": [
    "## Two issues when using an anndata object in %%R"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ed1a01b-cff6-4ac6-a70f-67f214934d61",
   "metadata": {},
   "source": [
    "1. ValueError: Converting pandas 'Category' series to R factor is only possible when categories are strings\n",
    "2. [NotImplementedError: Conversion 'py2rpy' not defined for objects of type '<class 'numpy.ndarray'>'](https://github.com/theislab/anndata2ri/issues/50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "5bd9d806-1ace-4dff-b2f9-8d4816bfac2e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "category\n",
      "category\n",
      "category\n",
      "category\n",
      "object\n",
      "float64\n"
     ]
    }
   ],
   "source": [
    "# Ensure that all data types (dtypes) in the columns of a pandas DataFrame within the AnnData object are not of type \"object\"\n",
    "# Ensure that all categorical series are strings\n",
    "for column in cdata_cd4.obs.columns:\n",
    "    print(cdata_cd4.obs[column].dtype)\n",
    "    if cdata_cd4.obs[column].dtype == 'object':\n",
    "        cdata_cd4.obs[column] = cdata_cd4.obs[column].astype('category')\n",
    "    if cdata_cd4.obs[column].dtype == 'category':\n",
    "        cdata_cd4.obs[column] = cdata_cd4.obs[column].astype('str')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "645b0d6d-e482-4ad0-8b30-1066cbbafffb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AnnData object with n_obs × n_vars = 6 × 21231\n",
       "    obs: 'condition', 'cell_type', 'sample', 'batch', 'lib_size', 'log_lib_size'\n",
       "    uns: 'log1p', 'pca', 'condition_colors', 'cell_type_colors', 'sample_colors', 'batch_colors'\n",
       "    obsm: 'X_pca'\n",
       "    varm: 'PCs'\n",
       "    layers: 'counts'"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_cd4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "4dfd812c-f0f5-4330-b087-43916ac20399",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Ensure that all numpy.array are sparse.csc_matrix\n",
    "import numpy as np\n",
    "import scipy.sparse as sp\n",
    "\n",
    "def convert_to_sparse_float(array):\n",
    "    if not sp.issparse(array):\n",
    "        if array.dtype == np.dtype('O'):\n",
    "            array = array.astype(float)\n",
    "        array = sp.csc_matrix(array)\n",
    "    return array\n",
    "\n",
    "cdata_cd4.X = convert_to_sparse_float(cdata_cd4.X)\n",
    "cdata_cd4.layers['counts'] = convert_to_sparse_float(cdata_cd4.layers['counts'])\n",
    "cdata_cd4.obsm['X_pca'] = convert_to_sparse_float(cdata_cd4.obsm['X_pca'])\n",
    "cdata_cd4.varm['PCs'] = convert_to_sparse_float(cdata_cd4.varm['PCs'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "56dc33c0-f82c-4345-9a33-9f0c9efe9bb3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<6x21231 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 109464 stored elements in Compressed Sparse Column format>"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata_cd4.X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "ee51a273-b561-45bb-86b4-38dd7a5dac7a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dimensions before subsetting:  21231x6 \n",
      "Dimensions after subsetting:  12819x6 \n"
     ]
    }
   ],
   "source": [
    "%%R -i cdata_cd4\n",
    "fit_result <- fit_model(adata_obj = cdata_cd4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "59a9aa42-1675-4037-9ae7-3f2b6b460ac2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "An object of class \"DGEGLM\"\n",
      "$coefficients\n",
      "         (Intercept) groupperturbated.T_cells_CD4\n",
      "A1BG       -13.05474                   0.19826520\n",
      "A1BG-AS1   -12.36697                   0.23025545\n",
      "A2M        -12.23982                  -1.74638233\n",
      "A2M-AS1    -11.57138                  -1.20628927\n",
      "AAAS       -11.65340                   0.07192908\n",
      "12814 more rows ...\n",
      "\n",
      "$fitted.values\n",
      "         0_control 1_control 2_perturbated 3_perturbated 4_control\n",
      "A1BG      95.03083  24.05589     14.851850     15.158310  47.91356\n",
      "A1BG-AS1 189.37843  47.93884     30.549978     31.180361  95.48265\n",
      "A2M      215.09665  54.44909      4.769608      4.868026 108.44951\n",
      "A2M-AS1  420.02471 106.32413     16.073412     16.405079 211.77213\n",
      "AAAS     386.92355  97.94497     53.262115     54.361152 195.08286\n",
      "         5_perturbated\n",
      "A1BG          22.77496\n",
      "A1BG-AS1      46.84766\n",
      "A2M            7.31408\n",
      "A2M-AS1       24.64820\n",
      "AAAS          81.67619\n",
      "12814 more rows ...\n",
      "\n",
      "$deviance\n",
      "     A1BG  A1BG-AS1       A2M   A2M-AS1      AAAS \n",
      " 0.737650  3.340505 19.688368 13.198046  3.570182 \n",
      "12814 more elements ...\n",
      "\n",
      "$method\n",
      "[1] \"oneway\"\n",
      "\n",
      "$counts\n",
      "         0_control 1_control 2_perturbated 3_perturbated 4_control\n",
      "A1BG            99        27            16            13        41\n",
      "A1BG-AS1       183        48            43            24        99\n",
      "A2M             89        55             7             2       177\n",
      "A2M-AS1        261       112            18             9       288\n",
      "AAAS           368       123            47            62       163\n",
      "         5_perturbated\n",
      "A1BG                24\n",
      "A1BG-AS1            40\n",
      "A2M                  8\n",
      "A2M-AS1             31\n",
      "AAAS                80\n",
      "12814 more rows ...\n",
      "\n",
      "$unshrunk.coefficients\n",
      "         (Intercept) groupperturbated.T_cells_CD4\n",
      "A1BG       -13.05831                   0.19892357\n",
      "A1BG-AS1   -12.36876                   0.23061720\n",
      "A2M        -12.24142                  -1.75382288\n",
      "A2M-AS1    -11.57220                  -1.20814654\n",
      "AAAS       -11.65428                   0.07199868\n",
      "12814 more rows ...\n",
      "\n",
      "$df.residual\n",
      "[1] 4 4 4 4 4\n",
      "12814 more elements ...\n",
      "\n",
      "$design\n",
      "  (Intercept) groupperturbated.T_cells_CD4\n",
      "1           1                            0\n",
      "2           1                            0\n",
      "3           1                            1\n",
      "4           1                            1\n",
      "5           1                            0\n",
      "6           1                            1\n",
      "attr(,\"assign\")\n",
      "[1] 0 1\n",
      "attr(,\"contrasts\")\n",
      "attr(,\"contrasts\")$group\n",
      "[1] \"contr.treatment\"\n",
      "\n",
      "\n",
      "$offset\n",
      "         [,1]     [,2]     [,3]     [,4]     [,5]     [,6]\n",
      "[1,] 17.61251 16.23869 15.55751 15.57793 16.92771 15.98505\n",
      "attr(,\"class\")\n",
      "[1] \"CompressedMatrix\"\n",
      "attr(,\"Dims\")\n",
      "[1] 5 6\n",
      "attr(,\"repeat.row\")\n",
      "[1] TRUE\n",
      "attr(,\"repeat.col\")\n",
      "[1] FALSE\n",
      "12814 more rows ...\n",
      "\n",
      "$dispersion\n",
      "[1] 0.05369842 0.03635119 0.04451897 0.02902754 0.02419408\n",
      "12814 more elements ...\n",
      "\n",
      "$prior.count\n",
      "[1] 0.125\n",
      "\n",
      "$AveLogCPM\n",
      "[1] 1.266864 2.255592 1.790750 2.778343 3.181257\n",
      "12814 more elements ...\n",
      "\n",
      "$df.residual.zeros\n",
      "[1] 4 4 4 4 4\n",
      "12814 more elements ...\n",
      "\n",
      "$df.prior\n",
      "[1] 5.572563\n",
      "\n",
      "$var.post\n",
      "     A1BG  A1BG-AS1       A2M   A2M-AS1      AAAS \n",
      "0.4854327 0.7122836 2.4398013 1.7228313 0.7041830 \n",
      "12814 more elements ...\n",
      "\n",
      "$var.prior\n",
      "     A1BG  A1BG-AS1       A2M   A2M-AS1      AAAS \n",
      "0.7015058 0.6241069 0.6580066 0.5910863 0.5689760 \n",
      "12814 more elements ...\n",
      "\n",
      "$samples\n",
      "                    group lib.size norm.factors\n",
      "0_control         control 41473627    1.0745918\n",
      "1_control         control 11309106    0.9975719\n",
      "2_perturbated perturbated  5775575    0.9884277\n",
      "3_perturbated perturbated  6204297    0.9391129\n",
      "4_control         control 22283949    1.0083640\n",
      "5_perturbated perturbated  8783851    0.9966260\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%%R\n",
    "(model_fit <- fit_result$fit)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "493ea19a-4c48-4058-b683-cbc26fa16dc1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "An object of class \"DGEList\"\n",
      "$counts\n",
      "         0_control 1_control 2_perturbated 3_perturbated 4_control\n",
      "A1BG            99        27            16            13        41\n",
      "A1BG-AS1       183        48            43            24        99\n",
      "A2M             89        55             7             2       177\n",
      "A2M-AS1        261       112            18             9       288\n",
      "AAAS           368       123            47            62       163\n",
      "         5_perturbated\n",
      "A1BG                24\n",
      "A1BG-AS1            40\n",
      "A2M                  8\n",
      "A2M-AS1             31\n",
      "AAAS                80\n",
      "12814 more rows ...\n",
      "\n",
      "$samples\n",
      "                    group lib.size norm.factors\n",
      "0_control         control 41473627    1.0745918\n",
      "1_control         control 11309106    0.9975719\n",
      "2_perturbated perturbated  5775575    0.9884277\n",
      "3_perturbated perturbated  6204297    0.9391129\n",
      "4_control         control 22283949    1.0083640\n",
      "5_perturbated perturbated  8783851    0.9966260\n",
      "\n",
      "$design\n",
      "  (Intercept) groupperturbated.T_cells_CD4\n",
      "1           1                            0\n",
      "2           1                            0\n",
      "3           1                            1\n",
      "4           1                            1\n",
      "5           1                            0\n",
      "6           1                            1\n",
      "attr(,\"assign\")\n",
      "[1] 0 1\n",
      "attr(,\"contrasts\")\n",
      "attr(,\"contrasts\")$group\n",
      "[1] \"contr.treatment\"\n",
      "\n",
      "\n",
      "$common.dispersion\n",
      "[1] 0.02242221\n",
      "\n",
      "$trended.dispersion\n",
      "[1] 0.05369842 0.03635119 0.04451897 0.02902754 0.02419408\n",
      "12814 more elements ...\n",
      "\n",
      "$tagwise.dispersion\n",
      "[1] 0.02702210 0.02899591 0.15896762 0.07056275 0.02290690\n",
      "12814 more elements ...\n",
      "\n",
      "$AveLogCPM\n",
      "[1] 1.266864 2.255592 1.790750 2.778343 3.181257\n",
      "12814 more elements ...\n",
      "\n",
      "$trend.method\n",
      "[1] \"locfit\"\n",
      "\n",
      "$prior.df\n",
      "[1] 5.572563\n",
      "\n",
      "$prior.n\n",
      "[1] 1.393141\n",
      "\n",
      "$span\n",
      "[1] 0.2968402\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%%R\n",
    "(y <- fit_result$y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "a98917fd-41b7-4f57-ac6a-087f6bacc8f6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R\n",
    "plotMDS(y, col=ifelse(y$samples$group == 'perturbated', 'maroon', 'forestgreen'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "0d56fc8b-f011-423f-aa6b-c1e995378a63",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R\n",
    "plotBCV(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "788d0ece-36a5-48e0-b35a-093c7fc1b241",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1] \"(Intercept)\"                  \"groupperturbated.T_cells_CD4\"\n"
     ]
    }
   ],
   "source": [
    "%%R\n",
    "colnames(y$design)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7aff6d4-bbe4-4396-9f84-3e01cdb18f32",
   "metadata": {},
   "source": [
    "## Estimate all DE genes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "4c719429-b911-457a-a846-ff54aaa3ead9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  (Intercept) groupperturbated.T_cells_CD4\n",
      "1           1                            0\n",
      "2           1                            0\n",
      "3           1                            1\n",
      "4           1                            1\n",
      "5           1                            0\n",
      "6           1                            1\n",
      "attr(,\"assign\")\n",
      "[1] 0 1\n",
      "attr(,\"contrasts\")\n",
      "attr(,\"contrasts\")$group\n",
      "[1] \"contr.treatment\"\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%%R\n",
    "y$design"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27fe88fc-6fa2-4a60-b768-3d42735345dc",
   "metadata": {},
   "source": [
    "## Design matrix: ~ group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "984add3e-7dc8-4163-9ec1-211d048dc3f3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "In addition: Warning message:\n",
       "In makeContrasts(\"groupperturbated.T_cells_CD4\", levels = y$design) :\n",
       "  Renaming (Intercept) to Intercept\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R -o tt\n",
    "contrast <- makeContrasts('groupperturbated.T_cells_CD4', levels=y$design)\n",
    "qlf <- glmQLFTest(model_fit, contrast=contrast)\n",
    "tt <- topTags(qlf, n=Inf)\n",
    "tt <- tt$table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "decec64f-873c-4851-90e7-1cfb9637527c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>logFC</th>\n",
       "      <th>logCPM</th>\n",
       "      <th>F</th>\n",
       "      <th>PValue</th>\n",
       "      <th>FDR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>IFIT3</th>\n",
       "      <td>-3.092989</td>\n",
       "      <td>5.680483</td>\n",
       "      <td>4.218179e+02</td>\n",
       "      <td>3.078748e-09</td>\n",
       "      <td>0.000015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IFI44L</th>\n",
       "      <td>-3.263613</td>\n",
       "      <td>6.556056</td>\n",
       "      <td>4.155431e+02</td>\n",
       "      <td>3.302870e-09</td>\n",
       "      <td>0.000015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BEX5</th>\n",
       "      <td>2.092208</td>\n",
       "      <td>5.161734</td>\n",
       "      <td>4.122098e+02</td>\n",
       "      <td>3.429941e-09</td>\n",
       "      <td>0.000015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DNAJB1</th>\n",
       "      <td>-1.757186</td>\n",
       "      <td>6.495728</td>\n",
       "      <td>3.647760e+02</td>\n",
       "      <td>6.078582e-09</td>\n",
       "      <td>0.000016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ADI1</th>\n",
       "      <td>-3.057785</td>\n",
       "      <td>4.985321</td>\n",
       "      <td>3.611911e+02</td>\n",
       "      <td>6.365750e-09</td>\n",
       "      <td>0.000016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LINC02210</th>\n",
       "      <td>-0.000199</td>\n",
       "      <td>3.642759</td>\n",
       "      <td>6.980715e-07</td>\n",
       "      <td>9.993505e-01</td>\n",
       "      <td>0.999662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUP98</th>\n",
       "      <td>-0.000063</td>\n",
       "      <td>4.509983</td>\n",
       "      <td>3.018454e-07</td>\n",
       "      <td>9.995729e-01</td>\n",
       "      <td>0.999763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TMEM199</th>\n",
       "      <td>-0.016617</td>\n",
       "      <td>3.808436</td>\n",
       "      <td>2.140934e-02</td>\n",
       "      <td>9.996069e-01</td>\n",
       "      <td>0.999763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CCSER2</th>\n",
       "      <td>0.000037</td>\n",
       "      <td>6.192699</td>\n",
       "      <td>6.597543e-08</td>\n",
       "      <td>9.998003e-01</td>\n",
       "      <td>0.999823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>POMC</th>\n",
       "      <td>-0.000055</td>\n",
       "      <td>3.010311</td>\n",
       "      <td>5.176259e-08</td>\n",
       "      <td>9.998231e-01</td>\n",
       "      <td>0.999823</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12819 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              logFC    logCPM             F        PValue       FDR\n",
       "IFIT3     -3.092989  5.680483  4.218179e+02  3.078748e-09  0.000015\n",
       "IFI44L    -3.263613  6.556056  4.155431e+02  3.302870e-09  0.000015\n",
       "BEX5       2.092208  5.161734  4.122098e+02  3.429941e-09  0.000015\n",
       "DNAJB1    -1.757186  6.495728  3.647760e+02  6.078582e-09  0.000016\n",
       "ADI1      -3.057785  4.985321  3.611911e+02  6.365750e-09  0.000016\n",
       "...             ...       ...           ...           ...       ...\n",
       "LINC02210 -0.000199  3.642759  6.980715e-07  9.993505e-01  0.999662\n",
       "NUP98     -0.000063  4.509983  3.018454e-07  9.995729e-01  0.999763\n",
       "TMEM199   -0.016617  3.808436  2.140934e-02  9.996069e-01  0.999763\n",
       "CCSER2     0.000037  6.192699  6.597543e-08  9.998003e-01  0.999823\n",
       "POMC      -0.000055  3.010311  5.176259e-08  9.998231e-01  0.999823\n",
       "\n",
       "[12819 rows x 5 columns]"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe001ab0-e39d-4a66-8682-d63c14ccad67",
   "metadata": {},
   "source": [
    "## Design matrix: ~ 0 + group"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "29e0fb31-96b3-4560-b628-5fe5f069e286",
   "metadata": {},
   "source": [
    "[When to include or not the intercept in the design matrix](https://support.bioconductor.org/p/128516/)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9288606d-640d-4323-a5ba-3151ff3fa119",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%R -o tt\n",
    "contrast <- makeContrasts('groupperturbated.T_cells_CD4-groupcontrol.T_cells_CD4', levels=y$design)\n",
    "qlf <- glmQLFTest(model_fit, contrast=contrast)\n",
    "tt <- topTags(qlf, n=Inf)\n",
    "tt <- tt$table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f38272be-9ac2-4d1f-b8ec-e628daee2c2f",
   "metadata": {},
   "outputs": [],
   "source": [
    "tt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8dd1a282-5100-4cb1-b56c-d798583e4fd5",
   "metadata": {},
   "source": [
    "## Genewise statistical tests for a given coefficient or contrast relative to a specified fold-change threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "7c588e73-7340-4a38-a7e4-57c3b3edddd3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Coefficient:  1*groupperturbated.T_cells_CD4 \n",
      "            logFC unshrunk.logFC   logCPM       PValue          FDR\n",
      "IFI44L  -3.263613      -3.264177 6.556056 9.533981e-08 0.0007429515\n",
      "IFIT3   -3.092989      -3.093920 5.680483 1.159141e-07 0.0007429515\n",
      "ADI1    -3.057785      -3.059266 4.985321 2.571450e-07 0.0010987807\n",
      "METRN    2.749798       2.750406 5.847060 7.349532e-07 0.0023553413\n",
      "CTNNAL1  3.112246       3.115402 3.747237 1.423659e-06 0.0034630877\n",
      "NRGN     4.502046       4.528753 1.886567 1.620916e-06 0.0034630877\n",
      "RSAD2   -2.822134      -2.823467 4.899794 2.281681e-06 0.0035348597\n",
      "TRIP10   3.061221       3.068475 2.394709 2.372549e-06 0.0035348597\n",
      "HERC6   -3.291868      -3.296417 3.686144 2.820902e-06 0.0035348597\n",
      "AVEN    -3.496108      -3.512929 2.196805 3.055553e-06 0.0035348597\n"
     ]
    }
   ],
   "source": [
    "%%R\n",
    "qlt <- glmTreat(model_fit, contrast=contrast, lfc=1.5)\n",
    "topTags(qlt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "bbd5a31b-7ddb-4814-83e2-0a9dd2149c7b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R\n",
    "plotSmear(qlf, de.tags = rownames(tt)[which(tt$FDR<0.01)])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}