notebooks/melanoma_selected_cell_comparison.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "df5125dc",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Comparing selected cells for melanoma response status classification\n",
    "***\n",
    "Jan T. Schleicher, 2022\n",
    "\n",
    "In this notebook, we compare the selected cells for the distinction between respodner an non-responder melanoma lesions. Specifically, we compare DICSIT selected cells with cells that can be considered selected by scCODA, Wilcoxon rank-sum tests, or milo. This analysis is based on the data from [Sade-Feldman et al. 2018](https://doi.org/10.1016/j.cell.2018.10.038)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3a3bcbe9",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "import matplotlib as mpl\n",
    "import seaborn as sns\n",
    "import pickle\n",
    "import upsetplot as up\n",
    "\n",
    "%matplotlib inline\n",
    "%config Completer.use_jedi = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f2a1f352",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "plot_dir = \"../output/figures/melanoma_response_status\"\n",
    "os.makedirs(plot_dir, exist_ok=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83d4f0fe",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Load all the data\n",
    "\n",
    "We load the results of the different analyses: scCODA simulation results, cell scores based on milo neighborhood analysis, and DICSIT filter responses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8d334663",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "sccoda_results = pickle.load(open(\"../output/differential_abundance/melanoma_sccoda_results.pkl\", \"rb\"))\n",
    "sccoda_results.set_fdr(est_fdr=0.1)\n",
    "cell_scores_milo = pd.read_csv(\"../output/differential_abundance/melanoma_cell_scores_milo.csv\")\n",
    "differential_abundance = pd.read_csv(\"../output/differential_abundance/melanoma_differential_abundance_wilcoxon.csv\",\n",
    "                                     index_col=0)\n",
    "response_df = pd.read_csv(\"../output/filter_responses/melanoma_response_status_filter_response.csv\",\n",
    "                          index_col=0)\\\n",
    "    .reset_index().rename(columns={\"Row.names\": \"cell\"})\n",
    "meta_data_umap = pd.read_csv(\"../data/melanoma_umap.txt\", sep=\"\\t\", index_col=0)\\\n",
    "    .reset_index().rename(columns={\"index\": \"cell\"})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb5c7d68",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Determine selected cells for each method\n",
    "\n",
    "In contrast to DICSIT, the other methods do not have a predefined notion of selected cells. For scCODA and Wilcoxon rank-sum tests, which operate on the level of whole clusters, we define all cells from significantly different clusters as selected cells. For milo, we use scores for individual cells computed from neighborhood scores analogously to the DICSIT filter response and set the same threshold of $0.3 \\times (\\text{maximum filter response})$, after splitting the scores into scores for non-responder and responder lesions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9374979a",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "fdr = .1\n",
    "ratio_max_response = .3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "7c8173f0",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>cluster_index</th>\n",
       "      <th>cluster</th>\n",
       "      <th>log2fc</th>\n",
       "      <th>p</th>\n",
       "      <th>padj</th>\n",
       "      <th>selected_wilcoxon_NR</th>\n",
       "      <th>selected_wilcoxon_R</th>\n",
       "      <th>selected_sccoda_NR</th>\n",
       "      <th>selected_sccoda_R</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>G1</td>\n",
       "      <td>-2.158012</td>\n",
       "      <td>0.002407</td>\n",
       "      <td>0.013238</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>G2</td>\n",
       "      <td>0.578780</td>\n",
       "      <td>0.780688</td>\n",
       "      <td>0.780688</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>G3</td>\n",
       "      <td>2.065498</td>\n",
       "      <td>0.010422</td>\n",
       "      <td>0.028662</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>G4</td>\n",
       "      <td>1.278635</td>\n",
       "      <td>0.089436</td>\n",
       "      <td>0.157844</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>G5</td>\n",
       "      <td>-1.032665</td>\n",
       "      <td>0.100446</td>\n",
       "      <td>0.157844</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>G6</td>\n",
       "      <td>1.962359</td>\n",
       "      <td>0.005844</td>\n",
       "      <td>0.021429</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>G7</td>\n",
       "      <td>0.252103</td>\n",
       "      <td>0.290052</td>\n",
       "      <td>0.398822</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>G8</td>\n",
       "      <td>0.480612</td>\n",
       "      <td>0.372954</td>\n",
       "      <td>0.455833</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>G9</td>\n",
       "      <td>0.283632</td>\n",
       "      <td>0.635986</td>\n",
       "      <td>0.699584</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9</td>\n",
       "      <td>G10</td>\n",
       "      <td>-1.860139</td>\n",
       "      <td>0.042104</td>\n",
       "      <td>0.092629</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>10</td>\n",
       "      <td>G11</td>\n",
       "      <td>1.830245</td>\n",
       "      <td>0.000238</td>\n",
       "      <td>0.002614</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    cluster_index cluster    log2fc         p      padj  selected_wilcoxon_NR  \\\n",
       "0               0      G1 -2.158012  0.002407  0.013238                 False   \n",
       "1               1      G2  0.578780  0.780688  0.780688                 False   \n",
       "2               2      G3  2.065498  0.010422  0.028662                  True   \n",
       "3               3      G4  1.278635  0.089436  0.157844                 False   \n",
       "4               4      G5 -1.032665  0.100446  0.157844                 False   \n",
       "5               5      G6  1.962359  0.005844  0.021429                  True   \n",
       "6               6      G7  0.252103  0.290052  0.398822                 False   \n",
       "7               7      G8  0.480612  0.372954  0.455833                 False   \n",
       "8               8      G9  0.283632  0.635986  0.699584                 False   \n",
       "9               9     G10 -1.860139  0.042104  0.092629                 False   \n",
       "10             10     G11  1.830245  0.000238  0.002614                  True   \n",
       "\n",
       "    selected_wilcoxon_R  selected_sccoda_NR  selected_sccoda_R  \n",
       "0                  True               False               True  \n",
       "1                 False               False              False  \n",
       "2                 False               False              False  \n",
       "3                 False               False              False  \n",
       "4                 False               False              False  \n",
       "5                 False                True              False  \n",
       "6                 False               False              False  \n",
       "7                 False               False              False  \n",
       "8                 False               False              False  \n",
       "9                  True               False              False  \n",
       "10                False                True              False  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "differential_abundance[\"selected_wilcoxon_NR\"] = (differential_abundance.padj < fdr) & \\\n",
    "    (differential_abundance.log2fc > 0)\n",
    "differential_abundance[\"selected_wilcoxon_R\"] = (differential_abundance.padj < fdr) & \\\n",
    "    (differential_abundance.log2fc < 0)\n",
    "differential_abundance[\"selected_sccoda_NR\"] = (sccoda_results.credible_effects().values) & \\\n",
    "    (differential_abundance.log2fc > 0)\n",
    "differential_abundance[\"selected_sccoda_R\"] = (sccoda_results.credible_effects().values) & \\\n",
    "    (differential_abundance.log2fc < 0)\n",
    "differential_abundance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4122442f",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "cell_scores_milo[\"selected_milo_NR\"] = \\\n",
    "    (cell_scores_milo.score_NR > ratio_max_response * cell_scores_milo.score_NR.max())\n",
    "cell_scores_milo[\"selected_milo_R\"] = \\\n",
    "    (cell_scores_milo.score_R > ratio_max_response * cell_scores_milo.score_R.max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d1786ca2",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "response_df[\"selected_dicsit_R\"] = \\\n",
    "    (response_df.response_filter_0 > ratio_max_response * response_df.response_filter_0.max())\n",
    "response_df[\"selected_dicsit_NR\"] = \\\n",
    "    (response_df.response_filter_1 > ratio_max_response * response_df.response_filter_1.max())\n",
    "response_df[\"selected_dicsit_R_2\"] = \\\n",
    "    (response_df.response_filter_2 > ratio_max_response * response_df.response_filter_2.max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "db4f1a0c",
   "metadata": {
    "scrolled": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>cell</th>\n",
       "      <th>sample_name</th>\n",
       "      <th>patient</th>\n",
       "      <th>cluster</th>\n",
       "      <th>response_status</th>\n",
       "      <th>selected_wilcoxon_NR</th>\n",
       "      <th>selected_wilcoxon_R</th>\n",
       "      <th>selected_sccoda_NR</th>\n",
       "      <th>selected_sccoda_R</th>\n",
       "      <th>selected_milo_NR</th>\n",
       "      <th>selected_milo_R</th>\n",
       "      <th>selected_dicsit_NR</th>\n",
       "      <th>selected_dicsit_R</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A10_P1_M15</td>\n",
       "      <td>Post_P1</td>\n",
       "      <td>P1</td>\n",
       "      <td>G9</td>\n",
       "      <td>R</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A10_P4_M15</td>\n",
       "      <td>Post_P1</td>\n",
       "      <td>P1</td>\n",
       "      <td>G9</td>\n",
       "      <td>R</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A10_P5_M15</td>\n",
       "      <td>Post_P1</td>\n",
       "      <td>P1</td>\n",
       "      <td>G9</td>\n",
       "      <td>R</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>A11_P2_M15</td>\n",
       "      <td>Post_P1</td>\n",
       "      <td>P1</td>\n",
       "      <td>G9</td>\n",
       "      <td>R</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>A11_P6_M15</td>\n",
       "      <td>Post_P1</td>\n",
       "      <td>P1</td>\n",
       "      <td>G9</td>\n",
       "      <td>R</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14967</th>\n",
       "      <td>B7_P7_MMD7_L001_myeloid_enriched</td>\n",
       "      <td>Post_P19</td>\n",
       "      <td>P19</td>\n",
       "      <td>G4</td>\n",
       "      <td>R</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14968</th>\n",
       "      <td>D9_P7_MMD7_L001_myeloid_enriched</td>\n",
       "      <td>Post_P19</td>\n",
       "      <td>P19</td>\n",
       "      <td>G4</td>\n",
       "      <td>R</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14969</th>\n",
       "      <td>G12_P7_MMD7_L001_myeloid_enriched</td>\n",
       "      <td>Post_P19</td>\n",
       "      <td>P19</td>\n",
       "      <td>G4</td>\n",
       "      <td>R</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14970</th>\n",
       "      <td>G8_P7_MMD7_L001_myeloid_enriched</td>\n",
       "      <td>Post_P19</td>\n",
       "      <td>P19</td>\n",
       "      <td>G4</td>\n",
       "      <td>R</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14971</th>\n",
       "      <td>H7_P7_MMD7_L001_myeloid_enriched</td>\n",
       "      <td>Post_P19</td>\n",
       "      <td>P19</td>\n",
       "      <td>G4</td>\n",
       "      <td>R</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14972 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                    cell sample_name patient cluster  \\\n",
       "0                             A10_P1_M15     Post_P1      P1      G9   \n",
       "1                             A10_P4_M15     Post_P1      P1      G9   \n",
       "2                             A10_P5_M15     Post_P1      P1      G9   \n",
       "3                             A11_P2_M15     Post_P1      P1      G9   \n",
       "4                             A11_P6_M15     Post_P1      P1      G9   \n",
       "...                                  ...         ...     ...     ...   \n",
       "14967   B7_P7_MMD7_L001_myeloid_enriched    Post_P19     P19      G4   \n",
       "14968   D9_P7_MMD7_L001_myeloid_enriched    Post_P19     P19      G4   \n",
       "14969  G12_P7_MMD7_L001_myeloid_enriched    Post_P19     P19      G4   \n",
       "14970   G8_P7_MMD7_L001_myeloid_enriched    Post_P19     P19      G4   \n",
       "14971   H7_P7_MMD7_L001_myeloid_enriched    Post_P19     P19      G4   \n",
       "\n",
       "      response_status  selected_wilcoxon_NR  selected_wilcoxon_R  \\\n",
       "0                   R                 False                False   \n",
       "1                   R                 False                False   \n",
       "2                   R                 False                False   \n",
       "3                   R                 False                False   \n",
       "4                   R                 False                False   \n",
       "...               ...                   ...                  ...   \n",
       "14967               R                 False                False   \n",
       "14968               R                 False                False   \n",
       "14969               R                 False                False   \n",
       "14970               R                 False                False   \n",
       "14971               R                 False                False   \n",
       "\n",
       "       selected_sccoda_NR  selected_sccoda_R  selected_milo_NR  \\\n",
       "0                   False              False             False   \n",
       "1                   False              False             False   \n",
       "2                   False              False             False   \n",
       "3                   False              False             False   \n",
       "4                   False              False             False   \n",
       "...                   ...                ...               ...   \n",
       "14967               False              False             False   \n",
       "14968               False              False             False   \n",
       "14969               False              False             False   \n",
       "14970               False              False             False   \n",
       "14971               False              False             False   \n",
       "\n",
       "       selected_milo_R  selected_dicsit_NR  selected_dicsit_R  \n",
       "0                False               False              False  \n",
       "1                False               False              False  \n",
       "2                False               False              False  \n",
       "3                False               False              False  \n",
       "4                False               False              False  \n",
       "...                ...                 ...                ...  \n",
       "14967            False               False              False  \n",
       "14968            False               False              False  \n",
       "14969            False               False              False  \n",
       "14970            False               False              False  \n",
       "14971            False               False              False  \n",
       "\n",
       "[14972 rows x 13 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comp_df = meta_data_umap[[\"cell\", \"sample_name\", \"patient\", \"cluster\", \"response_status\"]]\\\n",
    "    .merge(differential_abundance[[\"cluster\", \"selected_wilcoxon_NR\", \"selected_wilcoxon_R\",\n",
    "                                   \"selected_sccoda_NR\", \"selected_sccoda_R\"]], on=\"cluster\")\\\n",
    "    .merge(cell_scores_milo[[\"cell\", \"selected_milo_NR\", \"selected_milo_R\"]], on=\"cell\")\\\n",
    "    .merge(response_df[[\"cell\", \"selected_dicsit_NR\", \"selected_dicsit_R\"]], on=\"cell\")\n",
    "comp_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5cb4cefe",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Upset plots\n",
    "\n",
    "To quantify the size of overlaps and unique sets of selected cells between the different methods, we generate upset plots."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8225d466",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "data_upset_R = up.from_indicators(indicators=comp_df.filter(regex=\"selected_.*_R\").columns.tolist(), data=comp_df)\n",
    "data_upset_NR = up.from_indicators(indicators=comp_df.filter(regex=\"selected_.*_NR\").columns.tolist(), data=comp_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "3376293a",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "sns.set_theme(style=\"ticks\", font_scale=1.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "803812c0",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 844.444x444.444 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "upset = up.UpSet(data_upset_R, min_degree=1, sort_categories_by=None, show_counts=True)\n",
    "subplots = upset.plot(fig=fig)\n",
    "subplots[\"intersections\"].grid(visible=False)\n",
    "subplots[\"totals\"].grid(visible=False)\n",
    "subplots[\"totals\"].set_yticklabels([\"Wilcoxon R\", \"scCODA R\", \"milo R\", \"DICSIT R\"])\n",
    "fig.savefig(os.path.join(plot_dir, \"melanoma_selected_cells_R_upset.svg\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "3c55df19",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x444.444 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "subplots = up.plot(data_upset_NR, min_degree=1, sort_categories_by=None, show_counts=True, fig=fig)\n",
    "subplots[\"intersections\"].grid(visible=False)\n",
    "subplots[\"totals\"].grid(visible=False)\n",
    "subplots[\"totals\"].set_yticklabels([\"Wilcoxon NR\", \"scCODA NR\", \"milo NR\", \"DICSIT NR\"])\n",
    "fig.savefig(os.path.join(plot_dir, \"melanoma_selected_cells_NR_upset.svg\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4d7c882f",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "upset",
   "language": "python",
   "name": "upset"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}