test_deconveil.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "31e57b4d-e596-462c-9790-b20eff06818c",
   "metadata": {},
   "source": [
    "## DeConveil test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "8972c233-8bd3-4152-bbe9-21edb9a0a92a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pickle as pkl\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "af99fd5f-819c-4676-ac36-6195c24df8ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pydeseq2\n",
    "from pydeseq2.dds import DeseqDataSet\n",
    "from pydeseq2.default_inference import DefaultInference\n",
    "from pydeseq2.ds import DeseqStats\n",
    "from pydeseq2.utils import build_design_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f1ae78b9-2dd9-4320-bc6e-f48e8e8ca960",
   "metadata": {},
   "outputs": [],
   "source": [
    "from deconveil.dds import deconveil_fit\n",
    "from deconveil.inference import Inference\n",
    "from deconveil.default_inference import DefInference\n",
    "from deconveil.utils_CNaware import *\n",
    "from deconveil import deconveil_fit\n",
    "from deconveil.ds import deconveil_stats"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fc521e2-d04a-430f-b3c0-52a2cd0c4bc6",
   "metadata": {},
   "source": [
    "### Functions to run PyDESeq2 (CN-naive) & DeConveil (CN-aware)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f7a1acfc-4abe-4be4-90f7-68c88c6bd13e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def run_pydeseq2(rna_counts, metadata, output_path, design_factors=\"condition\", alpha=0.05):\n",
    "    \"\"\"\n",
    "    Runs PyDESeq2 analysis and saves the results.\n",
    "\n",
    "    Parameters:\n",
    "        rna_counts (pd.DataFrame): Count matrix with genes as rows and samples as columns.\n",
    "        metadata (pd.DataFrame): Metadata for the samples with design factors.\n",
    "        output_path (str): Directory to save the results.\n",
    "        design_factors (str): Column in metadata to use for design.\n",
    "        alpha (float): Significance level for statistical tests.\n",
    "    \"\"\"\n",
    "    os.makedirs(output_path, exist_ok=True)\n",
    "    \n",
    "    # Initialize DESeq2 analysis\n",
    "    inference = DefaultInference(n_cpus=8)\n",
    "    dds = DeseqDataSet(\n",
    "        counts=rna_counts,\n",
    "        metadata=metadata,\n",
    "        design_factors=\"condition\",\n",
    "        refit_cooks=True,\n",
    "        inference=inference,\n",
    "    )\n",
    "\n",
    "    # Fit DESeq2 model\n",
    "    dds.fit_size_factors()\n",
    "    dds.fit_genewise_dispersions()\n",
    "    dds.fit_dispersion_trend()\n",
    "    dds.fit_dispersion_prior()\n",
    "    dds.fit_MAP_dispersions()\n",
    "    dds.fit_LFC()\n",
    "    dds.calculate_cooks()\n",
    "    \n",
    "    if dds.refit_cooks:\n",
    "        dds.refit()\n",
    "\n",
    "    # Perform statistical analysis\n",
    "    stat_res_pydeseq = DeseqStats(dds, alpha=alpha, cooks_filter=True, independent_filter=True)\n",
    "    stat_res_pydeseq.run_wald_test()\n",
    "\n",
    "    if stat_res_pydeseq.cooks_filter:\n",
    "        stat_res_pydeseq._cooks_filtering()\n",
    "    stat_res_pydeseq.p_values\n",
    "\n",
    "    if stat_res_pydeseq.independent_filter:\n",
    "        stat_res_pydeseq._independent_filtering()\n",
    "    else:\n",
    "        stat_res_pydeseq._p_value_adjustment()\n",
    "\n",
    "    # Log-fold change shrinkage\n",
    "    stat_res_pydeseq.lfc_shrink(coeff=\"condition_B_vs_A\")\n",
    "    stat_res_pydeseq.summary()\n",
    "\n",
    "    # Save results\n",
    "    results_path = os.path.join(output_path, \"res_CNnaive.csv\")\n",
    "    stat_res_pydeseq.results_df.to_csv(results_path)\n",
    "    return(stat_res_pydeseq.results_df)\n",
    "\n",
    "\n",
    "def run_deconveil(rna_counts, metadata, cnv, output_path, design_factors=\"condition\", alpha=0.05):\n",
    "    \"\"\"\n",
    "    Runs DeConveil analysis and saves the results.\n",
    "\n",
    "    Parameters:\n",
    "        rna_counts (pd.DataFrame): Count matrix with genes as rows and samples as columns.\n",
    "        metadata (pd.DataFrame): Metadata for the samples with design factors.\n",
    "        cnv (pd.DataFrame): Copy number variation (CNV) data matrix  with genes as rows and samples as columns.\n",
    "        output_path (str): Directory to save the results.\n",
    "        design_factors (str): Column in metadata to use for design.\n",
    "        alpha (float): Significance level for statistical tests.\n",
    "    \"\"\"\n",
    "    os.makedirs(output_path, exist_ok=True)\n",
    "    \n",
    "    # Initialize DeConveil inference\n",
    "    inference = DefInference(n_cpus=8)\n",
    "\n",
    "    # Fit DeConveil model\n",
    "    dds = deconveil_fit(\n",
    "        counts=rna_counts,\n",
    "        metadata=metadata,\n",
    "        cnv=cnv,\n",
    "        design_factors=design_factors,\n",
    "        inference=inference,\n",
    "        refit_cooks=True\n",
    "    )\n",
    "    dds.fit_size_factors()\n",
    "    dds.fit_genewise_dispersions()\n",
    "    dds.fit_dispersion_trend()\n",
    "    dds.fit_dispersion_prior()\n",
    "    dds.fit_MAP_dispersions()\n",
    "    dds.fit_LFC()\n",
    "    dds.calculate_cooks()\n",
    "\n",
    "    if dds.refit_cooks:\n",
    "        dds.refit()  # Replace outlier counts\n",
    "\n",
    "    # Statistical analysis\n",
    "    stat_res_deconveil = deconveil_stats(\n",
    "        dds, \n",
    "        alpha=alpha, \n",
    "        independent_filter=True, \n",
    "        cooks_filter=True\n",
    "    )\n",
    "    stat_res_deconveil.run_wald_test()\n",
    "\n",
    "    if stat_res_deconveil.independent_filter:\n",
    "        stat_res_deconveil._independent_filtering()\n",
    "    else:\n",
    "        stat_res_deconveil._p_value_adjustment()\n",
    "\n",
    "    # Log-fold change shrinkage\n",
    "    stat_res_deconveil.lfc_shrink(coeff=\"condition_B_vs_A\")\n",
    "    stat_res_deconveil.summary()\n",
    "\n",
    "    # Save results\n",
    "    results_path = os.path.join(output_path, \"res_CNaware.csv\")\n",
    "    stat_res_deconveil.results_df.to_csv(results_path)\n",
    "    return(stat_res_deconveil.results_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "289bd585-7214-4c47-ad23-b725b9174d09",
   "metadata": {},
   "source": [
    "### Examples of usage | Perform Inference and Statistical test\n",
    "\n",
    "We need to perform separate DGE testing using CN-naive method, e.g *PyDESeq2* and CN-aware (*DeConveil*).\n",
    "DeConveil requires the following input data: \n",
    "- matched matrixes of *mRNA read counts* (normal and tumor samples) and *CN data* (normal and tumor samples), where column names are genes and rownames are samples;\n",
    "- *metadata* containing sample-specific information\n",
    "\n",
    "Below, we provide an example using TCGA cancer data to demonstrate the application of this approach"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "61c71d01-e45e-4666-a772-259e47e2c1f5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Test simulated data\n",
    "#DATA_PATH = \"/Users/katsiarynadavydzenka/Documents/PhD_AI/DeConveil/datasets/\"\n",
    "#rna_counts = pd.read_csv(os.path.join(DATA_PATH, \"rna_counts.csv\"), index_col=0)\n",
    "#rna_counts = rna_counts.T\n",
    "#metadata = pd.read_csv(os.path.join(DATA_PATH, \"metadata.csv\"), index_col=0)\n",
    "#cnv = pd.read_csv(os.path.join(DATA_PATH, \"cnv.csv\"), index_col=0)\n",
    "#cnv = cnv.T"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2c8182da-1167-4771-8d79-cf1f88f4c411",
   "metadata": {},
   "source": [
    "Load TCGA cancer data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "aae22ec6-1039-40fd-936d-a17121202eb9",
   "metadata": {},
   "outputs": [],
   "source": [
    "DATA_PATH = \"/Users/katsiarynadavydzenka/Documents/PhD_AI/TCGA/BRCA/test\"\n",
    "rna_counts = pd.read_csv(os.path.join(DATA_PATH, \"rna.csv\"), index_col=0)\n",
    "rna_counts = rna_counts.T\n",
    "metadata = pd.read_csv(os.path.join(DATA_PATH, \"metadata.csv\"), index_col=0)\n",
    "cnv = pd.read_csv(os.path.join(DATA_PATH, \"cnv.csv\"), index_col=0)\n",
    "cnv = cnv.T\n",
    "cnv = (cnv * 2).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "2898a2f5-4257-44e2-ae6e-69f5144291bd",
   "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>A2ML1</th>\n",
       "      <th>A2MP1</th>\n",
       "      <th>A4GALT</th>\n",
       "      <th>AAAS</th>\n",
       "      <th>AACS</th>\n",
       "      <th>AADAC</th>\n",
       "      <th>...</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",
       "      <th>ZZZ3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>sample1</th>\n",
       "      <td>18</td>\n",
       "      <td>62</td>\n",
       "      <td>70194</td>\n",
       "      <td>133</td>\n",
       "      <td>35</td>\n",
       "      <td>29</td>\n",
       "      <td>1642</td>\n",
       "      <td>1910</td>\n",
       "      <td>2235</td>\n",
       "      <td>1041</td>\n",
       "      <td>...</td>\n",
       "      <td>398</td>\n",
       "      <td>266</td>\n",
       "      <td>353</td>\n",
       "      <td>898</td>\n",
       "      <td>1854</td>\n",
       "      <td>157</td>\n",
       "      <td>3238</td>\n",
       "      <td>10766</td>\n",
       "      <td>4396</td>\n",
       "      <td>2922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample2</th>\n",
       "      <td>42</td>\n",
       "      <td>89</td>\n",
       "      <td>66754</td>\n",
       "      <td>250</td>\n",
       "      <td>52</td>\n",
       "      <td>42</td>\n",
       "      <td>310</td>\n",
       "      <td>2116</td>\n",
       "      <td>3232</td>\n",
       "      <td>188</td>\n",
       "      <td>...</td>\n",
       "      <td>1007</td>\n",
       "      <td>474</td>\n",
       "      <td>697</td>\n",
       "      <td>2229</td>\n",
       "      <td>2851</td>\n",
       "      <td>107</td>\n",
       "      <td>6250</td>\n",
       "      <td>10970</td>\n",
       "      <td>6296</td>\n",
       "      <td>8211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample3</th>\n",
       "      <td>14</td>\n",
       "      <td>72</td>\n",
       "      <td>183713</td>\n",
       "      <td>193</td>\n",
       "      <td>60</td>\n",
       "      <td>148</td>\n",
       "      <td>1062</td>\n",
       "      <td>2122</td>\n",
       "      <td>2901</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>737</td>\n",
       "      <td>450</td>\n",
       "      <td>666</td>\n",
       "      <td>1777</td>\n",
       "      <td>1976</td>\n",
       "      <td>152</td>\n",
       "      <td>4618</td>\n",
       "      <td>13426</td>\n",
       "      <td>6621</td>\n",
       "      <td>5511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample4</th>\n",
       "      <td>4</td>\n",
       "      <td>44</td>\n",
       "      <td>138900</td>\n",
       "      <td>277</td>\n",
       "      <td>160</td>\n",
       "      <td>115</td>\n",
       "      <td>1341</td>\n",
       "      <td>1939</td>\n",
       "      <td>2746</td>\n",
       "      <td>221</td>\n",
       "      <td>...</td>\n",
       "      <td>986</td>\n",
       "      <td>308</td>\n",
       "      <td>630</td>\n",
       "      <td>1880</td>\n",
       "      <td>2334</td>\n",
       "      <td>226</td>\n",
       "      <td>7503</td>\n",
       "      <td>16279</td>\n",
       "      <td>6930</td>\n",
       "      <td>5255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample5</th>\n",
       "      <td>8</td>\n",
       "      <td>35</td>\n",
       "      <td>58746</td>\n",
       "      <td>112</td>\n",
       "      <td>54</td>\n",
       "      <td>19</td>\n",
       "      <td>1036</td>\n",
       "      <td>1556</td>\n",
       "      <td>3372</td>\n",
       "      <td>26</td>\n",
       "      <td>...</td>\n",
       "      <td>470</td>\n",
       "      <td>480</td>\n",
       "      <td>314</td>\n",
       "      <td>1001</td>\n",
       "      <td>2141</td>\n",
       "      <td>78</td>\n",
       "      <td>2772</td>\n",
       "      <td>13915</td>\n",
       "      <td>4423</td>\n",
       "      <td>2986</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22076 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         A1BG  A1BG-AS1     A2M  A2M-AS1  A2ML1  A2MP1  A4GALT  AAAS  AACS  \\\n",
       "sample1    18        62   70194      133     35     29    1642  1910  2235   \n",
       "sample2    42        89   66754      250     52     42     310  2116  3232   \n",
       "sample3    14        72  183713      193     60    148    1062  2122  2901   \n",
       "sample4     4        44  138900      277    160    115    1341  1939  2746   \n",
       "sample5     8        35   58746      112     54     19    1036  1556  3372   \n",
       "\n",
       "         AADAC  ...  ZWILCH  ZWINT  ZXDA  ZXDB  ZXDC  ZYG11A  ZYG11B    ZYX  \\\n",
       "sample1   1041  ...     398    266   353   898  1854     157    3238  10766   \n",
       "sample2    188  ...    1007    474   697  2229  2851     107    6250  10970   \n",
       "sample3      8  ...     737    450   666  1777  1976     152    4618  13426   \n",
       "sample4    221  ...     986    308   630  1880  2334     226    7503  16279   \n",
       "sample5     26  ...     470    480   314  1001  2141      78    2772  13915   \n",
       "\n",
       "         ZZEF1  ZZZ3  \n",
       "sample1   4396  2922  \n",
       "sample2   6296  8211  \n",
       "sample3   6621  5511  \n",
       "sample4   6930  5255  \n",
       "sample5   4423  2986  \n",
       "\n",
       "[5 rows x 22076 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rna_counts.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "295d03cc-f092-4114-aa05-babc61f4e746",
   "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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>sample1</th>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample2</th>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample3</th>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample4</th>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample5</th>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        condition\n",
       "sample1         A\n",
       "sample2         A\n",
       "sample3         A\n",
       "sample4         A\n",
       "sample5         A"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metadata.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "7d4efa13-0745-4f7d-85a1-a8e22dc93975",
   "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>A2ML1</th>\n",
       "      <th>A2MP1</th>\n",
       "      <th>A4GALT</th>\n",
       "      <th>AAAS</th>\n",
       "      <th>AACS</th>\n",
       "      <th>AADAC</th>\n",
       "      <th>...</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",
       "      <th>ZZZ3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>sample1</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample2</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample3</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample4</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sample5</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22076 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         A1BG  A1BG-AS1  A2M  A2M-AS1  A2ML1  A2MP1  A4GALT  AAAS  AACS  \\\n",
       "sample1     2         2    2        2      2      2       2     2     2   \n",
       "sample2     2         2    2        2      2      2       2     2     2   \n",
       "sample3     2         2    2        2      2      2       2     2     2   \n",
       "sample4     2         2    2        2      2      2       2     2     2   \n",
       "sample5     2         2    2        2      2      2       2     2     2   \n",
       "\n",
       "         AADAC  ...  ZWILCH  ZWINT  ZXDA  ZXDB  ZXDC  ZYG11A  ZYG11B  ZYX  \\\n",
       "sample1      2  ...       2      2     2     2     2       2       2    2   \n",
       "sample2      2  ...       2      2     2     2     2       2       2    2   \n",
       "sample3      2  ...       2      2     2     2     2       2       2    2   \n",
       "sample4      2  ...       2      2     2     2     2       2       2    2   \n",
       "sample5      2  ...       2      2     2     2     2       2       2    2   \n",
       "\n",
       "         ZZEF1  ZZZ3  \n",
       "sample1      2     2  \n",
       "sample2      2     2  \n",
       "sample3      2     2  \n",
       "sample4      2     2  \n",
       "sample5      2     2  \n",
       "\n",
       "[5 rows x 22076 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cnv.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f3ce39c-b796-404f-b438-e5e8b7456416",
   "metadata": {},
   "source": [
    "Run PyDESeq2 (CN-naive) and DeConveil (CN-aware) methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "6dca20f7-2921-487b-b0a3-7f7a12f08247",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Fitting size factors...\n",
      "... done in 0.17 seconds.\n",
      "\n",
      "Fitting dispersions...\n",
      "... done in 7.10 seconds.\n",
      "\n",
      "Fitting dispersion trend curve...\n",
      "... done in 2.15 seconds.\n",
      "\n",
      "Fitting MAP dispersions...\n",
      "... done in 8.26 seconds.\n",
      "\n",
      "Fitting LFCs...\n",
      "... done in 3.62 seconds.\n",
      "\n",
      "Calculating cook's distance...\n",
      "... done in 0.58 seconds.\n",
      "\n",
      "Replacing 2218 outlier genes.\n",
      "\n",
      "Fitting dispersions...\n",
      "... done in 0.79 seconds.\n",
      "\n",
      "Fitting MAP dispersions...\n",
      "... done in 0.77 seconds.\n",
      "\n",
      "Fitting LFCs...\n",
      "... done in 0.46 seconds.\n",
      "\n",
      "Running Wald tests...\n",
      "... done in 0.97 seconds.\n",
      "\n",
      "Fitting MAP LFCs...\n",
      "... done in 5.52 seconds.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Log2 fold change & Wald test p-value: condition B vs A\n",
      "              baseMean  log2FoldChange     lfcSE       stat        pvalue  \\\n",
      "A1BG         12.101259        0.216933  0.127349   1.737456  8.230666e-02   \n",
      "A1BG-AS1     77.956081        0.420832  0.097706   4.359761  1.302049e-05   \n",
      "A2M       60702.289378       -0.265263  0.098391 -11.738713  8.070482e-32   \n",
      "A2M-AS1     126.644961       -1.196387  0.109734 -11.021924  2.995857e-28   \n",
      "A2ML1       371.981713        3.397818  0.293877  10.548439  5.164780e-26   \n",
      "...                ...             ...       ...        ...           ...   \n",
      "ZYG11A      241.321578        1.658324  0.161510  10.455134  1.388031e-25   \n",
      "ZYG11B     3289.954445       -0.357462  0.060106  -3.333155  8.586712e-04   \n",
      "ZYX       10157.755495       -0.425009  0.082316  -4.406815  1.049018e-05   \n",
      "ZZEF1      4358.415548       -0.439077  0.065587  -6.128182  8.888876e-10   \n",
      "ZZZ3       3497.912776       -0.065804  0.060881  -0.948312  3.429706e-01   \n",
      "\n",
      "                  padj  \n",
      "A1BG      1.032387e-01  \n",
      "A1BG-AS1  2.449636e-05  \n",
      "A2M       7.328834e-31  \n",
      "A2M-AS1   2.282932e-27  \n",
      "A2ML1     3.475090e-25  \n",
      "...                ...  \n",
      "ZYG11A    9.065731e-25  \n",
      "ZYG11B    1.370348e-03  \n",
      "ZYX       1.987821e-05  \n",
      "ZZEF1     2.301828e-09  \n",
      "ZZZ3      3.858048e-01  \n",
      "\n",
      "[22076 rows x 6 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Fitting size factors...\n",
      "... done in 0.16 seconds.\n",
      "\n",
      "Fitting dispersions...\n",
      "... done in 5.70 seconds.\n",
      "\n",
      "Fitting dispersion trend curve...\n",
      "... done in 0.38 seconds.\n",
      "\n",
      "Fitting MAP dispersions...\n",
      "... done in 6.97 seconds.\n",
      "\n",
      "Fitting LFCs...\n",
      "... done in 4.11 seconds.\n",
      "\n",
      "Calculating cook's distance...\n",
      "... done in 0.53 seconds.\n",
      "\n",
      "Replacing 1970 outlier genes.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "replace_mask before filtering: (220, 1970)\n",
      "Number of True values in replace_mask: 3200\n",
      "replacement_counts_trimmed shape: (187, 1970)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Fitting dispersions...\n",
      "... done in 0.51 seconds.\n",
      "\n",
      "Fitting MAP dispersions...\n",
      "... done in 0.81 seconds.\n",
      "\n",
      "Fitting LFCs...\n",
      "... done in 0.30 seconds.\n",
      "\n",
      "Running Wald tests...\n",
      "... done in 0.97 seconds.\n",
      "\n",
      "Fitting MAP LFCs...\n",
      "... done in 6.32 seconds.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Log2 fold change & Wald test p-value: condition B vs A\n",
      "              baseMean  log2FoldChange     lfcSE       stat        pvalue  \\\n",
      "A1BG         12.101259       -0.248101  0.128128  -1.859431  6.296607e-02   \n",
      "A1BG-AS1     77.956081       -0.040320  0.097715  -0.326320  7.441820e-01   \n",
      "A2M       60702.289378        0.396316  0.099753 -13.259452  3.977901e-40   \n",
      "A2M-AS1     126.644961       -1.582486  0.109889 -14.425079  3.598561e-47   \n",
      "A2ML1       464.103817        2.806622  0.303290   9.488769  2.337791e-21   \n",
      "...                ...             ...       ...        ...           ...   \n",
      "ZYG11A      241.321578        1.391794  0.161342   8.857747  8.164764e-19   \n",
      "ZYG11B     3232.777025       -0.603744  0.025856   2.988412  2.804312e-03   \n",
      "ZYX       10157.755495       -0.680184  0.082936  -8.471871  2.414823e-17   \n",
      "ZZEF1      4358.415548       -0.316560  0.065445  -5.467741  4.558067e-08   \n",
      "ZZZ3       3497.912776       -0.331767  0.060655  -5.041385  4.621743e-07   \n",
      "\n",
      "                  padj  \n",
      "A1BG      7.863546e-02  \n",
      "A1BG-AS1  7.722366e-01  \n",
      "A2M       5.046905e-39  \n",
      "A2M-AS1   6.125044e-46  \n",
      "A2ML1     1.199653e-20  \n",
      "...                ...  \n",
      "ZYG11A    3.627397e-18  \n",
      "ZYG11B    4.101768e-03  \n",
      "ZYX       9.832096e-17  \n",
      "ZZEF1     9.981537e-08  \n",
      "ZZZ3      9.375997e-07  \n",
      "\n",
      "[22076 rows x 6 columns]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>baseMean</th>\n",
       "      <th>log2FoldChange</th>\n",
       "      <th>lfcSE</th>\n",
       "      <th>stat</th>\n",
       "      <th>pvalue</th>\n",
       "      <th>padj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>A1BG</th>\n",
       "      <td>12.101259</td>\n",
       "      <td>-0.248101</td>\n",
       "      <td>0.128128</td>\n",
       "      <td>-1.859431</td>\n",
       "      <td>6.296607e-02</td>\n",
       "      <td>7.863546e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1BG-AS1</th>\n",
       "      <td>77.956081</td>\n",
       "      <td>-0.040320</td>\n",
       "      <td>0.097715</td>\n",
       "      <td>-0.326320</td>\n",
       "      <td>7.441820e-01</td>\n",
       "      <td>7.722366e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2M</th>\n",
       "      <td>60702.289378</td>\n",
       "      <td>0.396316</td>\n",
       "      <td>0.099753</td>\n",
       "      <td>-13.259452</td>\n",
       "      <td>3.977901e-40</td>\n",
       "      <td>5.046905e-39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2M-AS1</th>\n",
       "      <td>126.644961</td>\n",
       "      <td>-1.582486</td>\n",
       "      <td>0.109889</td>\n",
       "      <td>-14.425079</td>\n",
       "      <td>3.598561e-47</td>\n",
       "      <td>6.125044e-46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2ML1</th>\n",
       "      <td>464.103817</td>\n",
       "      <td>2.806622</td>\n",
       "      <td>0.303290</td>\n",
       "      <td>9.488769</td>\n",
       "      <td>2.337791e-21</td>\n",
       "      <td>1.199653e-20</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZYG11A</th>\n",
       "      <td>241.321578</td>\n",
       "      <td>1.391794</td>\n",
       "      <td>0.161342</td>\n",
       "      <td>8.857747</td>\n",
       "      <td>8.164764e-19</td>\n",
       "      <td>3.627397e-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZYG11B</th>\n",
       "      <td>3232.777025</td>\n",
       "      <td>-0.603744</td>\n",
       "      <td>0.025856</td>\n",
       "      <td>2.988412</td>\n",
       "      <td>2.804312e-03</td>\n",
       "      <td>4.101768e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZYX</th>\n",
       "      <td>10157.755495</td>\n",
       "      <td>-0.680184</td>\n",
       "      <td>0.082936</td>\n",
       "      <td>-8.471871</td>\n",
       "      <td>2.414823e-17</td>\n",
       "      <td>9.832096e-17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZZEF1</th>\n",
       "      <td>4358.415548</td>\n",
       "      <td>-0.316560</td>\n",
       "      <td>0.065445</td>\n",
       "      <td>-5.467741</td>\n",
       "      <td>4.558067e-08</td>\n",
       "      <td>9.981537e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZZZ3</th>\n",
       "      <td>3497.912776</td>\n",
       "      <td>-0.331767</td>\n",
       "      <td>0.060655</td>\n",
       "      <td>-5.041385</td>\n",
       "      <td>4.621743e-07</td>\n",
       "      <td>9.375997e-07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>22076 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              baseMean  log2FoldChange     lfcSE       stat        pvalue  \\\n",
       "A1BG         12.101259       -0.248101  0.128128  -1.859431  6.296607e-02   \n",
       "A1BG-AS1     77.956081       -0.040320  0.097715  -0.326320  7.441820e-01   \n",
       "A2M       60702.289378        0.396316  0.099753 -13.259452  3.977901e-40   \n",
       "A2M-AS1     126.644961       -1.582486  0.109889 -14.425079  3.598561e-47   \n",
       "A2ML1       464.103817        2.806622  0.303290   9.488769  2.337791e-21   \n",
       "...                ...             ...       ...        ...           ...   \n",
       "ZYG11A      241.321578        1.391794  0.161342   8.857747  8.164764e-19   \n",
       "ZYG11B     3232.777025       -0.603744  0.025856   2.988412  2.804312e-03   \n",
       "ZYX       10157.755495       -0.680184  0.082936  -8.471871  2.414823e-17   \n",
       "ZZEF1      4358.415548       -0.316560  0.065445  -5.467741  4.558067e-08   \n",
       "ZZZ3       3497.912776       -0.331767  0.060655  -5.041385  4.621743e-07   \n",
       "\n",
       "                  padj  \n",
       "A1BG      7.863546e-02  \n",
       "A1BG-AS1  7.722366e-01  \n",
       "A2M       5.046905e-39  \n",
       "A2M-AS1   6.125044e-46  \n",
       "A2ML1     1.199653e-20  \n",
       "...                ...  \n",
       "ZYG11A    3.627397e-18  \n",
       "ZYG11B    4.101768e-03  \n",
       "ZYX       9.832096e-17  \n",
       "ZZEF1     9.981537e-08  \n",
       "ZZZ3      9.375997e-07  \n",
       "\n",
       "[22076 rows x 6 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# PyDESeq2\n",
    "pydeseq2_output_path = \"/Users/katsiarynadavydzenka/Documents/PhD_AI/deconveilCaseStudies/results/BRCA/\"\n",
    "run_pydeseq2(rna_counts, metadata, pydeseq2_output_path)\n",
    "\n",
    "# DeConveil\n",
    "deconveil_output_path = \"/Users/katsiarynadavydzenka/Documents/PhD_AI/deconveilCaseStudies/results/BRCA/\"\n",
    "run_deconveil(rna_counts, metadata, cnv, deconveil_output_path)    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72febdd5-f938-4191-94f4-12b303a696ae",
   "metadata": {},
   "source": [
    "### Define gene groups\n",
    "\n",
    "We are going to load previosly saved results of *PyDESeq2* and *DeConveil* methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "1e8641a0-4e49-4380-8475-a922dda9db86",
   "metadata": {},
   "outputs": [],
   "source": [
    "pydeseq2_output_path = \"/Users/katsiarynadavydzenka/Documents/PhD_AI/deconveilCaseStudies/results/BRCA/res_CNnaive.csv\"\n",
    "deconveil_output_path = \"/Users/katsiarynadavydzenka/Documents/PhD_AI/deconveilCaseStudies/results/BRCA/res_CNaware.csv\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f13c323f-ec5b-46d6-a52e-083b9e9bbcb2",
   "metadata": {},
   "source": [
    "Data preprocessing\n",
    "\n",
    "We are going to define DE genes based on the following thresholds *lfc_cut = 1.0 & pval_cut = 0.05*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "c12242fb-208f-436f-97c8-2ab48cda3785",
   "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>padj</th>\n",
       "      <th>isDE</th>\n",
       "      <th>DEtype</th>\n",
       "      <th>method</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>A1BG</th>\n",
       "      <td>0.216933</td>\n",
       "      <td>1.032387e-01</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1BG-AS1</th>\n",
       "      <td>0.420832</td>\n",
       "      <td>2.449636e-05</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2M</th>\n",
       "      <td>-0.265263</td>\n",
       "      <td>7.328834e-31</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2M-AS1</th>\n",
       "      <td>-1.196387</td>\n",
       "      <td>2.282932e-27</td>\n",
       "      <td>True</td>\n",
       "      <td>Down-reg</td>\n",
       "      <td>CN naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2ML1</th>\n",
       "      <td>3.397818</td>\n",
       "      <td>3.475090e-25</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>CN naive</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>ZYG11A</th>\n",
       "      <td>1.658324</td>\n",
       "      <td>9.065731e-25</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>CN naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZYG11B</th>\n",
       "      <td>-0.357462</td>\n",
       "      <td>1.370348e-03</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZYX</th>\n",
       "      <td>-0.425009</td>\n",
       "      <td>1.987821e-05</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZZEF1</th>\n",
       "      <td>-0.439077</td>\n",
       "      <td>2.301828e-09</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZZZ3</th>\n",
       "      <td>-0.065804</td>\n",
       "      <td>3.858048e-01</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>22076 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             logFC          padj   isDE    DEtype    method\n",
       "A1BG      0.216933  1.032387e-01  False      n.s.  CN naive\n",
       "A1BG-AS1  0.420832  2.449636e-05  False      n.s.  CN naive\n",
       "A2M      -0.265263  7.328834e-31  False      n.s.  CN naive\n",
       "A2M-AS1  -1.196387  2.282932e-27   True  Down-reg  CN naive\n",
       "A2ML1     3.397818  3.475090e-25   True    Up-reg  CN naive\n",
       "...            ...           ...    ...       ...       ...\n",
       "ZYG11A    1.658324  9.065731e-25   True    Up-reg  CN naive\n",
       "ZYG11B   -0.357462  1.370348e-03  False      n.s.  CN naive\n",
       "ZYX      -0.425009  1.987821e-05  False      n.s.  CN naive\n",
       "ZZEF1    -0.439077  2.301828e-09  False      n.s.  CN naive\n",
       "ZZZ3     -0.065804  3.858048e-01  False      n.s.  CN naive\n",
       "\n",
       "[22076 rows x 5 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Preprocess PyDESeq2 results\n",
    "res_pydeseq = process_results(pydeseq2_output_path, \"CN naive\", lfc_cut = 1.0, pval_cut = 0.05)\n",
    "res_pydeseq.columns = [\"logFC\", \"padj\", \"isDE\", \"DEtype\", \"method\"]\n",
    "res_pydeseq"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "8b3c2ab9-207a-47c0-a4ef-48594f0f0af0",
   "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>padj</th>\n",
       "      <th>isDE</th>\n",
       "      <th>DEtype</th>\n",
       "      <th>method</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>A1BG</th>\n",
       "      <td>-0.248101</td>\n",
       "      <td>7.863546e-02</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1BG-AS1</th>\n",
       "      <td>-0.040320</td>\n",
       "      <td>7.722366e-01</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2M</th>\n",
       "      <td>0.396316</td>\n",
       "      <td>5.046905e-39</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2M-AS1</th>\n",
       "      <td>-1.582486</td>\n",
       "      <td>6.125044e-46</td>\n",
       "      <td>True</td>\n",
       "      <td>Down-reg</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2ML1</th>\n",
       "      <td>2.806622</td>\n",
       "      <td>1.199653e-20</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>CN aware</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>ZYG11A</th>\n",
       "      <td>1.391794</td>\n",
       "      <td>3.627397e-18</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZYG11B</th>\n",
       "      <td>-0.603744</td>\n",
       "      <td>4.101768e-03</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZYX</th>\n",
       "      <td>-0.680184</td>\n",
       "      <td>9.832096e-17</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZZEF1</th>\n",
       "      <td>-0.316560</td>\n",
       "      <td>9.981537e-08</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZZZ3</th>\n",
       "      <td>-0.331767</td>\n",
       "      <td>9.375997e-07</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>22076 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             logFC          padj   isDE    DEtype    method\n",
       "A1BG     -0.248101  7.863546e-02  False      n.s.  CN aware\n",
       "A1BG-AS1 -0.040320  7.722366e-01  False      n.s.  CN aware\n",
       "A2M       0.396316  5.046905e-39  False      n.s.  CN aware\n",
       "A2M-AS1  -1.582486  6.125044e-46   True  Down-reg  CN aware\n",
       "A2ML1     2.806622  1.199653e-20   True    Up-reg  CN aware\n",
       "...            ...           ...    ...       ...       ...\n",
       "ZYG11A    1.391794  3.627397e-18   True    Up-reg  CN aware\n",
       "ZYG11B   -0.603744  4.101768e-03  False      n.s.  CN aware\n",
       "ZYX      -0.680184  9.832096e-17  False      n.s.  CN aware\n",
       "ZZEF1    -0.316560  9.981537e-08  False      n.s.  CN aware\n",
       "ZZZ3     -0.331767  9.375997e-07  False      n.s.  CN aware\n",
       "\n",
       "[22076 rows x 5 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Preprocess DeConveil results\n",
    "res_deconveil = process_results(deconveil_output_path, \"CN aware\", lfc_cut = 1.0, pval_cut = 0.05)\n",
    "res_deconveil.columns = [\"logFC\", \"padj\", \"isDE\", \"DEtype\", \"method\"]\n",
    "res_deconveil"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7ad5100-0f13-4e4b-90e4-5e24cbe0098f",
   "metadata": {},
   "source": [
    "Define gene groups based on the results of *CN-naive* and *CN-aware* methods: DSGs, DIGs, DCGs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "d516e337-7326-4af9-bb4f-c31ec263f60b",
   "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_naive</th>\n",
       "      <th>padj_naive</th>\n",
       "      <th>isDE_naive</th>\n",
       "      <th>DEtype_naive</th>\n",
       "      <th>method_naive</th>\n",
       "      <th>logFC_aware</th>\n",
       "      <th>padj_aware</th>\n",
       "      <th>isDE_aware</th>\n",
       "      <th>DEtype_aware</th>\n",
       "      <th>method_aware</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>A1BG</th>\n",
       "      <td>0.216933</td>\n",
       "      <td>1.032387e-01</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "      <td>-0.248101</td>\n",
       "      <td>7.863546e-02</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1BG-AS1</th>\n",
       "      <td>0.420832</td>\n",
       "      <td>2.449636e-05</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "      <td>-0.040320</td>\n",
       "      <td>7.722366e-01</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2M</th>\n",
       "      <td>-0.265263</td>\n",
       "      <td>7.328834e-31</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "      <td>0.396316</td>\n",
       "      <td>5.046905e-39</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2M-AS1</th>\n",
       "      <td>-1.196387</td>\n",
       "      <td>2.282932e-27</td>\n",
       "      <td>True</td>\n",
       "      <td>Down-reg</td>\n",
       "      <td>CN naive</td>\n",
       "      <td>-1.582486</td>\n",
       "      <td>6.125044e-46</td>\n",
       "      <td>True</td>\n",
       "      <td>Down-reg</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2ML1</th>\n",
       "      <td>3.397818</td>\n",
       "      <td>3.475090e-25</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>CN naive</td>\n",
       "      <td>2.806622</td>\n",
       "      <td>1.199653e-20</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>CN aware</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZYG11A</th>\n",
       "      <td>1.658324</td>\n",
       "      <td>9.065731e-25</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>CN naive</td>\n",
       "      <td>1.391794</td>\n",
       "      <td>3.627397e-18</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZYG11B</th>\n",
       "      <td>-0.357462</td>\n",
       "      <td>1.370348e-03</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "      <td>-0.603744</td>\n",
       "      <td>4.101768e-03</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZYX</th>\n",
       "      <td>-0.425009</td>\n",
       "      <td>1.987821e-05</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "      <td>-0.680184</td>\n",
       "      <td>9.832096e-17</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZZEF1</th>\n",
       "      <td>-0.439077</td>\n",
       "      <td>2.301828e-09</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "      <td>-0.316560</td>\n",
       "      <td>9.981537e-08</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZZZ3</th>\n",
       "      <td>-0.065804</td>\n",
       "      <td>3.858048e-01</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN naive</td>\n",
       "      <td>-0.331767</td>\n",
       "      <td>9.375997e-07</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>22076 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          logFC_naive    padj_naive  isDE_naive DEtype_naive method_naive  \\\n",
       "A1BG         0.216933  1.032387e-01       False         n.s.     CN naive   \n",
       "A1BG-AS1     0.420832  2.449636e-05       False         n.s.     CN naive   \n",
       "A2M         -0.265263  7.328834e-31       False         n.s.     CN naive   \n",
       "A2M-AS1     -1.196387  2.282932e-27        True     Down-reg     CN naive   \n",
       "A2ML1        3.397818  3.475090e-25        True       Up-reg     CN naive   \n",
       "...               ...           ...         ...          ...          ...   \n",
       "ZYG11A       1.658324  9.065731e-25        True       Up-reg     CN naive   \n",
       "ZYG11B      -0.357462  1.370348e-03       False         n.s.     CN naive   \n",
       "ZYX         -0.425009  1.987821e-05       False         n.s.     CN naive   \n",
       "ZZEF1       -0.439077  2.301828e-09       False         n.s.     CN naive   \n",
       "ZZZ3        -0.065804  3.858048e-01       False         n.s.     CN naive   \n",
       "\n",
       "          logFC_aware    padj_aware  isDE_aware DEtype_aware method_aware  \n",
       "A1BG        -0.248101  7.863546e-02       False         n.s.     CN aware  \n",
       "A1BG-AS1    -0.040320  7.722366e-01       False         n.s.     CN aware  \n",
       "A2M          0.396316  5.046905e-39       False         n.s.     CN aware  \n",
       "A2M-AS1     -1.582486  6.125044e-46        True     Down-reg     CN aware  \n",
       "A2ML1        2.806622  1.199653e-20        True       Up-reg     CN aware  \n",
       "...               ...           ...         ...          ...          ...  \n",
       "ZYG11A       1.391794  3.627397e-18        True       Up-reg     CN aware  \n",
       "ZYG11B      -0.603744  4.101768e-03       False         n.s.     CN aware  \n",
       "ZYX         -0.680184  9.832096e-17       False         n.s.     CN aware  \n",
       "ZZEF1       -0.316560  9.981537e-08       False         n.s.     CN aware  \n",
       "ZZZ3        -0.331767  9.375997e-07       False         n.s.     CN aware  \n",
       "\n",
       "[22076 rows x 10 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res_joint = pd.concat([\n",
    "    res_pydeseq.add_suffix(\"_naive\"),\n",
    "    res_deconveil.add_suffix(\"_aware\")\n",
    "], axis=1)\n",
    "res_joint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "45cdf116-fb65-4394-846e-4a59d7b47236",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'DSGs':            logFC_naive    padj_naive  isDE_naive DEtype_naive method_naive  \\\n",
       " AARD          1.976721  1.546922e-18        True       Up-reg     CN naive   \n",
       " ABAT          1.641054  6.843344e-24        True       Up-reg     CN naive   \n",
       " ABCA3         1.078939  3.786570e-18        True       Up-reg     CN naive   \n",
       " ABCB9         1.007015  3.142395e-20        True       Up-reg     CN naive   \n",
       " ABCC12        1.238684  4.221455e-04        True       Up-reg     CN naive   \n",
       " ...                ...           ...         ...          ...          ...   \n",
       " ZNF887P       1.350576  8.388363e-19        True       Up-reg     CN naive   \n",
       " ZNF92         1.038093  4.490923e-16        True       Up-reg     CN naive   \n",
       " ZNF93         1.151330  6.585788e-33        True       Up-reg     CN naive   \n",
       " ZSCAN1        1.056472  4.653589e-07        True       Up-reg     CN naive   \n",
       " ZSCAN12P1     1.171981  6.129064e-27        True       Up-reg     CN naive   \n",
       " \n",
       "            logFC_aware    padj_aware  isDE_aware DEtype_aware method_aware  \\\n",
       " AARD          0.876650  2.435353e-05       False         n.s.     CN aware   \n",
       " ABAT          0.858547  3.727558e-08       False         n.s.     CN aware   \n",
       " ABCA3         0.255437  9.447905e-03       False         n.s.     CN aware   \n",
       " ABCB9         0.622781  8.020406e-09       False         n.s.     CN aware   \n",
       " ABCC12        0.977844  3.936172e-03       False         n.s.     CN aware   \n",
       " ...                ...           ...         ...          ...          ...   \n",
       " ZNF887P       0.826794  5.369497e-08       False         n.s.     CN aware   \n",
       " ZNF92         0.674702  1.170686e-07       False         n.s.     CN aware   \n",
       " ZNF93         0.749968  4.168194e-15       False         n.s.     CN aware   \n",
       " ZSCAN1        0.461479  2.331138e-02       False         n.s.     CN aware   \n",
       " ZSCAN12P1     0.747862  4.964798e-12       False         n.s.     CN aware   \n",
       " \n",
       "           gene_category  \n",
       " AARD               DSGs  \n",
       " ABAT               DSGs  \n",
       " ABCA3              DSGs  \n",
       " ABCB9              DSGs  \n",
       " ABCC12             DSGs  \n",
       " ...                 ...  \n",
       " ZNF887P            DSGs  \n",
       " ZNF92              DSGs  \n",
       " ZNF93              DSGs  \n",
       " ZSCAN1             DSGs  \n",
       " ZSCAN12P1          DSGs  \n",
       " \n",
       " [1101 rows x 11 columns],\n",
       " 'DIGs':          logFC_naive    padj_naive  isDE_naive DEtype_naive method_naive  \\\n",
       " A2M-AS1    -1.196387  2.282932e-27        True     Down-reg     CN naive   \n",
       " A2ML1       3.397818  3.475090e-25        True       Up-reg     CN naive   \n",
       " AADAC      -3.931007  7.657307e-30        True     Down-reg     CN naive   \n",
       " AADACL3    -4.843213  3.611030e-27        True     Down-reg     CN naive   \n",
       " AADACP1    -2.329860  1.278256e-33        True     Down-reg     CN naive   \n",
       " ...              ...           ...         ...          ...          ...   \n",
       " ZPLD2P     -1.280341  5.457419e-10        True     Down-reg     CN naive   \n",
       " ZSCAN23    -1.173883  2.884302e-11        True     Down-reg     CN naive   \n",
       " ZWILCH      1.281154  3.958932e-65        True       Up-reg     CN naive   \n",
       " ZWINT       2.589727  2.491737e-99        True       Up-reg     CN naive   \n",
       " ZYG11A      1.658324  9.065731e-25        True       Up-reg     CN naive   \n",
       " \n",
       "          logFC_aware    padj_aware  isDE_aware DEtype_aware method_aware  \\\n",
       " A2M-AS1    -1.582486  6.125044e-46        True     Down-reg     CN aware   \n",
       " A2ML1       2.806622  1.199653e-20        True       Up-reg     CN aware   \n",
       " AADAC      -4.307937  6.787572e-35        True     Down-reg     CN aware   \n",
       " AADACL3    -4.816975  1.895547e-26        True     Down-reg     CN aware   \n",
       " AADACP1    -2.761676  1.132554e-43        True     Down-reg     CN aware   \n",
       " ...              ...           ...         ...          ...          ...   \n",
       " ZPLD2P     -1.516430  4.315149e-13        True     Down-reg     CN aware   \n",
       " ZSCAN23    -1.442373  5.214744e-16        True     Down-reg     CN aware   \n",
       " ZWILCH      1.003728  6.315242e-41        True       Up-reg     CN aware   \n",
       " ZWINT       2.197742  4.163823e-73        True       Up-reg     CN aware   \n",
       " ZYG11A      1.391794  3.627397e-18        True       Up-reg     CN aware   \n",
       " \n",
       "         gene_category  \n",
       " A2M-AS1          DIGs  \n",
       " A2ML1            DIGs  \n",
       " AADAC            DIGs  \n",
       " AADACL3          DIGs  \n",
       " AADACP1          DIGs  \n",
       " ...               ...  \n",
       " ZPLD2P           DIGs  \n",
       " ZSCAN23          DIGs  \n",
       " ZWILCH           DIGs  \n",
       " ZWINT            DIGs  \n",
       " ZYG11A           DIGs  \n",
       " \n",
       " [3962 rows x 11 columns],\n",
       " 'DCGs':              logFC_naive    padj_naive  isDE_naive DEtype_naive method_naive  \\\n",
       " A2MP1          -0.956441  4.707418e-18       False         n.s.     CN naive   \n",
       " AADACL2        -0.307907  1.490300e-03       False         n.s.     CN naive   \n",
       " AADACL2-AS1    -0.995814  2.216018e-06       False         n.s.     CN naive   \n",
       " AAMDC          -0.288829  1.167507e-06       False         n.s.     CN naive   \n",
       " ABCA13          1.736127  3.993827e-01       False         n.s.     CN naive   \n",
       " ...                  ...           ...         ...          ...          ...   \n",
       " ZNF727         -0.885087  1.000554e-12       False         n.s.     CN naive   \n",
       " ZNF843         -0.445272  1.381829e-07       False         n.s.     CN naive   \n",
       " ZNF853         -0.655175  7.079033e-10       False         n.s.     CN naive   \n",
       " ZNNT1           0.103425  4.581577e-01       False         n.s.     CN naive   \n",
       " ZSCAN16-AS1    -0.657664  9.256828e-11       False         n.s.     CN naive   \n",
       " \n",
       "              logFC_aware    padj_aware  isDE_aware DEtype_aware method_aware  \\\n",
       " A2MP1          -1.432029  8.211742e-16        True     Down-reg     CN aware   \n",
       " AADACL2        -1.023168  2.083816e-15        True     Down-reg     CN aware   \n",
       " AADACL2-AS1    -1.364874  2.424046e-10        True     Down-reg     CN aware   \n",
       " AAMDC          -1.020283  2.619830e-17        True     Down-reg     CN aware   \n",
       " ABCA13          1.201785  7.375775e-13        True       Up-reg     CN aware   \n",
       " ...                  ...           ...         ...          ...          ...   \n",
       " ZNF727         -1.459048  9.096181e-17        True     Down-reg     CN aware   \n",
       " ZNF843         -1.032809  2.056135e-33        True     Down-reg     CN aware   \n",
       " ZNF853         -1.097249  1.266383e-23        True     Down-reg     CN aware   \n",
       " ZNNT1          -1.063952  5.980050e-16        True     Down-reg     CN aware   \n",
       " ZSCAN16-AS1    -1.060131  2.936606e-25        True     Down-reg     CN aware   \n",
       " \n",
       "             gene_category  \n",
       " A2MP1                DCGs  \n",
       " AADACL2              DCGs  \n",
       " AADACL2-AS1          DCGs  \n",
       " AAMDC                DCGs  \n",
       " ABCA13               DCGs  \n",
       " ...                   ...  \n",
       " ZNF727               DCGs  \n",
       " ZNF843               DCGs  \n",
       " ZNF853               DCGs  \n",
       " ZNNT1                DCGs  \n",
       " ZSCAN16-AS1          DCGs  \n",
       " \n",
       " [1307 rows x 11 columns],\n",
       " 'non_DEGs':           logFC_naive    padj_naive  isDE_naive DEtype_naive method_naive  \\\n",
       " A1BG         0.216933  1.032387e-01       False         n.s.     CN naive   \n",
       " A1BG-AS1     0.420832  2.449636e-05       False         n.s.     CN naive   \n",
       " A2M         -0.265263  7.328834e-31       False         n.s.     CN naive   \n",
       " A4GALT      -0.362543  2.405339e-03       False         n.s.     CN naive   \n",
       " AAAS         0.150714  2.814126e-03       False         n.s.     CN naive   \n",
       " ...               ...           ...         ...          ...          ...   \n",
       " ZXDC         0.034641  5.447087e-01       False         n.s.     CN naive   \n",
       " ZYG11B      -0.357462  1.370348e-03       False         n.s.     CN naive   \n",
       " ZYX         -0.425009  1.987821e-05       False         n.s.     CN naive   \n",
       " ZZEF1       -0.439077  2.301828e-09       False         n.s.     CN naive   \n",
       " ZZZ3        -0.065804  3.858048e-01       False         n.s.     CN naive   \n",
       " \n",
       "           logFC_aware    padj_aware  isDE_aware DEtype_aware method_aware  \\\n",
       " A1BG        -0.248101  7.863546e-02       False         n.s.     CN aware   \n",
       " A1BG-AS1    -0.040320  7.722366e-01       False         n.s.     CN aware   \n",
       " A2M          0.396316  5.046905e-39       False         n.s.     CN aware   \n",
       " A4GALT      -0.435287  4.097918e-04       False         n.s.     CN aware   \n",
       " AAAS        -0.232639  8.505371e-06       False         n.s.     CN aware   \n",
       " ...               ...           ...         ...          ...          ...   \n",
       " ZXDC        -0.417872  1.157913e-16       False         n.s.     CN aware   \n",
       " ZYG11B      -0.603744  4.101768e-03       False         n.s.     CN aware   \n",
       " ZYX         -0.680184  9.832096e-17       False         n.s.     CN aware   \n",
       " ZZEF1       -0.316560  9.981537e-08       False         n.s.     CN aware   \n",
       " ZZZ3        -0.331767  9.375997e-07       False         n.s.     CN aware   \n",
       " \n",
       "          gene_category  \n",
       " A1BG          non-DEGs  \n",
       " A1BG-AS1      non-DEGs  \n",
       " A2M           non-DEGs  \n",
       " A4GALT        non-DEGs  \n",
       " AAAS          non-DEGs  \n",
       " ...                ...  \n",
       " ZXDC          non-DEGs  \n",
       " ZYG11B        non-DEGs  \n",
       " ZYX           non-DEGs  \n",
       " ZZEF1         non-DEGs  \n",
       " ZZZ3          non-DEGs  \n",
       " \n",
       " [15706 rows x 11 columns]}"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gene_groups = define_gene_groups(res_joint)\n",
    "gene_groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "ec86b6cf-d489-47ef-8673-1f51f8a4b30f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Extract each gene category\n",
    "\n",
    "# Dosage-sensitive genes\n",
    "d_sensitive = (\n",
    "    gene_groups['DSGs']\n",
    "    .rename(columns={\n",
    "        'logFC_aware': 'log2FC',\n",
    "        'padj_aware': 'padj',\n",
    "        'isDE_aware': 'isDE',\n",
    "        'DEtype_aware': 'DEtype',\n",
    "        'method_aware': 'method',\n",
    "        'gene_category': 'gene_group'\n",
    "    })[['log2FC', 'padj', 'isDE', 'DEtype', 'method', 'gene_group']]\n",
    ")\n",
    "\n",
    "# Dosage-insensitive genes\n",
    "d_insensitive = (\n",
    "    gene_groups['DIGs']\n",
    "    .rename(columns={\n",
    "        'logFC_aware': 'log2FC',\n",
    "        'padj_aware': 'padj',\n",
    "        'isDE_aware': 'isDE',\n",
    "        'DEtype_aware': 'DEtype',\n",
    "        'method_aware': 'method',\n",
    "        'gene_category': 'gene_group'\n",
    "    })[['log2FC', 'padj', 'isDE', 'DEtype', 'method', 'gene_group']]\n",
    ")\n",
    "\n",
    "# Dosage-compensated genes\n",
    "d_compensated = (\n",
    "    gene_groups['DCGs']\n",
    "    .rename(columns={\n",
    "        'logFC_aware': 'log2FC',\n",
    "        'padj_aware': 'padj',\n",
    "        'isDE_aware': 'isDE',\n",
    "        'DEtype_aware': 'DEtype',\n",
    "        'method_aware': 'method',\n",
    "        'gene_category': 'gene_group'\n",
    "    })[['log2FC', 'padj', 'isDE', 'DEtype', 'method', 'gene_group']]\n",
    ")\n",
    "\n",
    "# non-DEGs\n",
    "\n",
    "non_DEGs = (\n",
    "    gene_groups['non_DEGs']\n",
    "    .rename(columns={\n",
    "        'logFC_aware': 'log2FC',\n",
    "        'padj_aware': 'padj',\n",
    "        'isDE_aware': 'isDE',\n",
    "        'DEtype_aware': 'DEtype',\n",
    "        'method_aware': 'method',\n",
    "        'gene_category': 'gene_group'\n",
    "    })[['log2FC', 'padj', 'isDE', 'DEtype', 'method', 'gene_group']]\n",
    ")\n",
    "\n",
    "cn_aware = pd.concat([d_sensitive, d_insensitive, d_compensated], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "a9af4112-a34e-4d62-8aaa-ddd942968e55",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dosage-sensitive genes\n",
    "d_sensitive = (\n",
    "    gene_groups['DSGs']\n",
    "    .rename(columns={\n",
    "        'logFC_naive': 'log2FC',\n",
    "        'padj_naive': 'padj',\n",
    "        'isDE_naive': 'isDE',\n",
    "        'DEtype_naive': 'DEtype',\n",
    "        'method_naive': 'method',\n",
    "        'gene_category': 'gene_group'\n",
    "    })[['log2FC', 'padj', 'isDE', 'DEtype', 'method', 'gene_group']]\n",
    ")\n",
    "\n",
    "# Dosage-insensitive genes\n",
    "d_insensitive = (\n",
    "    gene_groups['DIGs']\n",
    "    .rename(columns={\n",
    "        'logFC_naive': 'log2FC',\n",
    "        'padj_naive': 'padj',\n",
    "        'isDE_naive': 'isDE',\n",
    "        'DEtype_naive': 'DEtype',\n",
    "        'method_naive': 'method',\n",
    "        'gene_category': 'gene_group'\n",
    "    })[['log2FC', 'padj', 'isDE', 'DEtype', 'method', 'gene_group']]\n",
    ")\n",
    "\n",
    "# Dosage-compensated genes\n",
    "d_compensated = (\n",
    "    gene_groups['DCGs']\n",
    "    .rename(columns={\n",
    "        'logFC_naive': 'log2FC',\n",
    "        'padj_naive': 'padj',\n",
    "        'isDE_naive': 'isDE',\n",
    "        'DEtype_naive': 'DEtype',\n",
    "        'method_naive': 'method',\n",
    "        'gene_category': 'gene_group'\n",
    "    })[['log2FC', 'padj', 'isDE', 'DEtype', 'method', 'gene_group']]\n",
    ")\n",
    "\n",
    "# non-DEGs\n",
    "\n",
    "non_DEGs = (\n",
    "    gene_groups['non_DEGs']\n",
    "    .rename(columns={\n",
    "        'logFC_naive': 'log2FC',\n",
    "        'padj_naive': 'padj',\n",
    "        'isDE_naive': 'isDE',\n",
    "        'DEtype_naive': 'DEtype',\n",
    "        'method_naive': 'method',\n",
    "        'gene_category': 'gene_group'\n",
    "    })[['log2FC', 'padj', 'isDE', 'DEtype', 'method', 'gene_group']]\n",
    ")\n",
    "\n",
    "cn_naive = pd.concat([d_sensitive, d_insensitive, d_compensated], axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc0b8f40-d789-4c6e-bf16-5690312b9b3f",
   "metadata": {},
   "source": [
    "### Data visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "477c0731-f3bb-4a36-a137-f43e09896d00",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.11/site-packages/pandas/core/arraylike.py:402: RuntimeWarning: divide by zero encountered in log10\n",
      "  result = getattr(ufunc, method)(*inputs, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "# Volcano plot\n",
    "plot_volcano = pd.concat([cn_naive, cn_aware], axis=0)\n",
    "plot_volcano['gene_group'] = plot_volcano['gene_group'].astype(\"category\")\n",
    "plot_volcano['-log10(padj)'] = -np.log10(plot_volcano['padj'])\n",
    "plot_volcano['tumor_type'] = \"BRCA\"\n",
    "plot_volcano['method'] = pd.Categorical(plot_volcano['method'], \n",
    "                                           categories=[\"CN naive\", \"CN aware\"], \n",
    "                                           ordered=True)\n",
    "plot_volcano['method'] = plot_volcano['method'].map({\"CN naive\": \"PyDESeq2\", \"CN aware\": \"DeConveil\"})\n",
    "\n",
    "lfc_cut = 1\n",
    "pval_cut = 0.05"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "17a0004c-6681-4f04-8a6a-3df4d631d61d",
   "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>log2FC</th>\n",
       "      <th>padj</th>\n",
       "      <th>isDE</th>\n",
       "      <th>DEtype</th>\n",
       "      <th>method</th>\n",
       "      <th>gene_group</th>\n",
       "      <th>-log10(padj)</th>\n",
       "      <th>tumor_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AARD</th>\n",
       "      <td>1.976721</td>\n",
       "      <td>1.546922e-18</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>PyDESeq2</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>17.810532</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABAT</th>\n",
       "      <td>1.641054</td>\n",
       "      <td>6.843344e-24</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>PyDESeq2</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>23.164732</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCA3</th>\n",
       "      <td>1.078939</td>\n",
       "      <td>3.786570e-18</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>PyDESeq2</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>17.421754</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCB9</th>\n",
       "      <td>1.007015</td>\n",
       "      <td>3.142395e-20</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>PyDESeq2</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>19.502739</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCC12</th>\n",
       "      <td>1.238684</td>\n",
       "      <td>4.221455e-04</td>\n",
       "      <td>True</td>\n",
       "      <td>Up-reg</td>\n",
       "      <td>PyDESeq2</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>3.374538</td>\n",
       "      <td>BRCA</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZNF727</th>\n",
       "      <td>-1.459048</td>\n",
       "      <td>9.096181e-17</td>\n",
       "      <td>True</td>\n",
       "      <td>Down-reg</td>\n",
       "      <td>DeConveil</td>\n",
       "      <td>DCGs</td>\n",
       "      <td>16.041141</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZNF843</th>\n",
       "      <td>-1.032809</td>\n",
       "      <td>2.056135e-33</td>\n",
       "      <td>True</td>\n",
       "      <td>Down-reg</td>\n",
       "      <td>DeConveil</td>\n",
       "      <td>DCGs</td>\n",
       "      <td>32.686948</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZNF853</th>\n",
       "      <td>-1.097249</td>\n",
       "      <td>1.266383e-23</td>\n",
       "      <td>True</td>\n",
       "      <td>Down-reg</td>\n",
       "      <td>DeConveil</td>\n",
       "      <td>DCGs</td>\n",
       "      <td>22.897435</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZNNT1</th>\n",
       "      <td>-1.063952</td>\n",
       "      <td>5.980050e-16</td>\n",
       "      <td>True</td>\n",
       "      <td>Down-reg</td>\n",
       "      <td>DeConveil</td>\n",
       "      <td>DCGs</td>\n",
       "      <td>15.223295</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZSCAN16-AS1</th>\n",
       "      <td>-1.060131</td>\n",
       "      <td>2.936606e-25</td>\n",
       "      <td>True</td>\n",
       "      <td>Down-reg</td>\n",
       "      <td>DeConveil</td>\n",
       "      <td>DCGs</td>\n",
       "      <td>24.532154</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12740 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               log2FC          padj  isDE    DEtype     method gene_group  \\\n",
       "AARD         1.976721  1.546922e-18  True    Up-reg   PyDESeq2       DSGs   \n",
       "ABAT         1.641054  6.843344e-24  True    Up-reg   PyDESeq2       DSGs   \n",
       "ABCA3        1.078939  3.786570e-18  True    Up-reg   PyDESeq2       DSGs   \n",
       "ABCB9        1.007015  3.142395e-20  True    Up-reg   PyDESeq2       DSGs   \n",
       "ABCC12       1.238684  4.221455e-04  True    Up-reg   PyDESeq2       DSGs   \n",
       "...               ...           ...   ...       ...        ...        ...   \n",
       "ZNF727      -1.459048  9.096181e-17  True  Down-reg  DeConveil       DCGs   \n",
       "ZNF843      -1.032809  2.056135e-33  True  Down-reg  DeConveil       DCGs   \n",
       "ZNF853      -1.097249  1.266383e-23  True  Down-reg  DeConveil       DCGs   \n",
       "ZNNT1       -1.063952  5.980050e-16  True  Down-reg  DeConveil       DCGs   \n",
       "ZSCAN16-AS1 -1.060131  2.936606e-25  True  Down-reg  DeConveil       DCGs   \n",
       "\n",
       "             -log10(padj) tumor_type  \n",
       "AARD            17.810532       BRCA  \n",
       "ABAT            23.164732       BRCA  \n",
       "ABCA3           17.421754       BRCA  \n",
       "ABCB9           19.502739       BRCA  \n",
       "ABCC12           3.374538       BRCA  \n",
       "...                   ...        ...  \n",
       "ZNF727          16.041141       BRCA  \n",
       "ZNF843          32.686948       BRCA  \n",
       "ZNF853          22.897435       BRCA  \n",
       "ZNNT1           15.223295       BRCA  \n",
       "ZSCAN16-AS1     24.532154       BRCA  \n",
       "\n",
       "[12740 rows x 8 columns]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_volcano"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "e808e966-2a6f-4509-83e5-98a0517f96bd",
   "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>log2FC</th>\n",
       "      <th>padj</th>\n",
       "      <th>isDE</th>\n",
       "      <th>DEtype</th>\n",
       "      <th>method</th>\n",
       "      <th>gene_group</th>\n",
       "      <th>tumor_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AARD</th>\n",
       "      <td>0.876650</td>\n",
       "      <td>2.435353e-05</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABAT</th>\n",
       "      <td>0.858547</td>\n",
       "      <td>3.727558e-08</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCA3</th>\n",
       "      <td>0.255437</td>\n",
       "      <td>9.447905e-03</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCB9</th>\n",
       "      <td>0.622781</td>\n",
       "      <td>8.020406e-09</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCC12</th>\n",
       "      <td>0.977844</td>\n",
       "      <td>3.936172e-03</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          log2FC          padj   isDE DEtype    method gene_group tumor_type\n",
       "AARD    0.876650  2.435353e-05  False   n.s.  CN aware       DSGs       BRCA\n",
       "ABAT    0.858547  3.727558e-08  False   n.s.  CN aware       DSGs       BRCA\n",
       "ABCA3   0.255437  9.447905e-03  False   n.s.  CN aware       DSGs       BRCA\n",
       "ABCB9   0.622781  8.020406e-09  False   n.s.  CN aware       DSGs       BRCA\n",
       "ABCC12  0.977844  3.936172e-03  False   n.s.  CN aware       DSGs       BRCA"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cn_aware['tumor_type'] = \"BRCA\"\n",
    "cn_aware.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "f945a64e-1786-4516-ad4d-6f0a7e9fe4b0",
   "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>cnv_mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>A1BG</th>\n",
       "      <td>3.119266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1BG-AS1</th>\n",
       "      <td>3.119266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2M</th>\n",
       "      <td>2.981651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2M-AS1</th>\n",
       "      <td>2.981651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2ML1</th>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          cnv_mean\n",
       "A1BG      3.119266\n",
       "A1BG-AS1  3.119266\n",
       "A2M       2.981651\n",
       "A2M-AS1   2.981651\n",
       "A2ML1     3.000000"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Histogram of CN distribution\n",
    "cnv = pd.read_csv(os.path.join(DATA_PATH, \"cnv.csv\"), index_col=0)\n",
    "cnv = cnv.iloc[:, 111:220]\n",
    "cnv['cnv_mean'] = cnv.mean(axis=1)\n",
    "cnv_mean = cnv['cnv_mean'] * 2\n",
    "cnv_mean = pd.DataFrame(cnv_mean)\n",
    "cnv_mean.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "3b31863a-1a91-46cf-919c-c0181ed12ac2",
   "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>log2FC</th>\n",
       "      <th>padj</th>\n",
       "      <th>isDE</th>\n",
       "      <th>DEtype</th>\n",
       "      <th>method</th>\n",
       "      <th>gene_group</th>\n",
       "      <th>tumor_type</th>\n",
       "      <th>cnv_mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AARD</th>\n",
       "      <td>0.876650</td>\n",
       "      <td>2.435353e-05</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABAT</th>\n",
       "      <td>0.858547</td>\n",
       "      <td>3.727558e-08</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "      <td>3.568807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCA3</th>\n",
       "      <td>0.255437</td>\n",
       "      <td>9.447905e-03</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "      <td>3.697248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCB9</th>\n",
       "      <td>0.622781</td>\n",
       "      <td>8.020406e-09</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "      <td>2.816514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCC12</th>\n",
       "      <td>0.977844</td>\n",
       "      <td>3.936172e-03</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "      <td>2.568807</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          log2FC          padj   isDE DEtype    method gene_group tumor_type  \\\n",
       "AARD    0.876650  2.435353e-05  False   n.s.  CN aware       DSGs       BRCA   \n",
       "ABAT    0.858547  3.727558e-08  False   n.s.  CN aware       DSGs       BRCA   \n",
       "ABCA3   0.255437  9.447905e-03  False   n.s.  CN aware       DSGs       BRCA   \n",
       "ABCB9   0.622781  8.020406e-09  False   n.s.  CN aware       DSGs       BRCA   \n",
       "ABCC12  0.977844  3.936172e-03  False   n.s.  CN aware       DSGs       BRCA   \n",
       "\n",
       "        cnv_mean  \n",
       "AARD    5.000000  \n",
       "ABAT    3.568807  \n",
       "ABCA3   3.697248  \n",
       "ABCB9   2.816514  \n",
       "ABCC12  2.568807  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combined_data = cn_aware.merge(cnv_mean, left_index=True, right_index=True, how=\"left\")\n",
    "combined_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "16345882-5ccb-4ed4-8703-2be7fe973899",
   "metadata": {},
   "outputs": [],
   "source": [
    "def assign_cnv_group(cnv_mean):\n",
    "    if 0.0 < cnv_mean <= 1.7:\n",
    "        return \"loss\"\n",
    "    elif 1.7 < cnv_mean <= 2.5:\n",
    "        return \"neutral\"\n",
    "    elif 2.5 < cnv_mean <= 4.0:\n",
    "        return \"gain\"\n",
    "    elif cnv_mean > 4.0:\n",
    "        return \"amplification\"\n",
    "    else:\n",
    "        return None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "5375ea85-1dfc-480b-acdd-6f6cef4b0a3a",
   "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>tumor_type</th>\n",
       "      <th>gene_group</th>\n",
       "      <th>Count</th>\n",
       "      <th>total</th>\n",
       "      <th>percentage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BRCA</td>\n",
       "      <td>DCGs</td>\n",
       "      <td>1307</td>\n",
       "      <td>6370</td>\n",
       "      <td>20.518053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BRCA</td>\n",
       "      <td>DIGs</td>\n",
       "      <td>3962</td>\n",
       "      <td>6370</td>\n",
       "      <td>62.197802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BRCA</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>1101</td>\n",
       "      <td>6370</td>\n",
       "      <td>17.284144</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  tumor_type gene_group  Count  total  percentage\n",
       "0       BRCA       DCGs   1307   6370   20.518053\n",
       "1       BRCA       DIGs   3962   6370   62.197802\n",
       "2       BRCA       DSGs   1101   6370   17.284144"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Group by 'tumor_type' and 'gene_group' and calculate counts, percentages\n",
    "combined_data['cnv_group'] = combined_data['cnv_mean'].apply(assign_cnv_group)\n",
    "barplot_data = combined_data.groupby(['tumor_type', 'gene_group']).size().reset_index(name='Count')\n",
    "barplot_data['total'] = barplot_data.groupby('tumor_type')['Count'].transform('sum')\n",
    "barplot_data['percentage'] = (barplot_data['Count'] / barplot_data['total']) * 100\n",
    "barplot_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "9d25be0b-5d07-43b3-89ba-6034d71affd4",
   "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>log2FC</th>\n",
       "      <th>padj</th>\n",
       "      <th>isDE</th>\n",
       "      <th>DEtype</th>\n",
       "      <th>method</th>\n",
       "      <th>gene_group</th>\n",
       "      <th>tumor_type</th>\n",
       "      <th>cnv_mean</th>\n",
       "      <th>cnv_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AARD</th>\n",
       "      <td>0.876650</td>\n",
       "      <td>2.435353e-05</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>amplification</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABAT</th>\n",
       "      <td>0.858547</td>\n",
       "      <td>3.727558e-08</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "      <td>3.568807</td>\n",
       "      <td>gain</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCA3</th>\n",
       "      <td>0.255437</td>\n",
       "      <td>9.447905e-03</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "      <td>3.697248</td>\n",
       "      <td>gain</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCB9</th>\n",
       "      <td>0.622781</td>\n",
       "      <td>8.020406e-09</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "      <td>2.816514</td>\n",
       "      <td>gain</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABCC12</th>\n",
       "      <td>0.977844</td>\n",
       "      <td>3.936172e-03</td>\n",
       "      <td>False</td>\n",
       "      <td>n.s.</td>\n",
       "      <td>CN aware</td>\n",
       "      <td>DSGs</td>\n",
       "      <td>BRCA</td>\n",
       "      <td>2.568807</td>\n",
       "      <td>gain</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          log2FC          padj   isDE DEtype    method gene_group tumor_type  \\\n",
       "AARD    0.876650  2.435353e-05  False   n.s.  CN aware       DSGs       BRCA   \n",
       "ABAT    0.858547  3.727558e-08  False   n.s.  CN aware       DSGs       BRCA   \n",
       "ABCA3   0.255437  9.447905e-03  False   n.s.  CN aware       DSGs       BRCA   \n",
       "ABCB9   0.622781  8.020406e-09  False   n.s.  CN aware       DSGs       BRCA   \n",
       "ABCC12  0.977844  3.936172e-03  False   n.s.  CN aware       DSGs       BRCA   \n",
       "\n",
       "        cnv_mean      cnv_group  \n",
       "AARD    5.000000  amplification  \n",
       "ABAT    3.568807           gain  \n",
       "ABCA3   3.697248           gain  \n",
       "ABCB9   2.816514           gain  \n",
       "ABCC12  2.568807           gain  "
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combined_data['gene_group'] = pd.Categorical(combined_data['gene_group'], categories=[\"DIGs\", \"DSGs\", \"DCGs\"], ordered=True)\n",
    "combined_data['tumor_type'] = pd.Categorical(combined_data['tumor_type'], categories=[\"LUSC\", \"BRCA\", \"LIHC\", \"KIRC\"], ordered=True)\n",
    "combined_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "3308b5f9-519c-4326-80b5-d1852d03b015",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 500x500 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n",
      "posx and posy should be finite values\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAHkCAYAAADvrlz5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABigUlEQVR4nO3deXxM5/4H8M9ksm8iEsQWa3ZL7EWlQm/tu1qTaqtSGrS1BVcltHXRqqJuY6tQtFQskYReEvsSoQQRhMSeCCL7OvP8/vDL1DQJkxHmJPm88/KSnPOcc75nTjKfOdtzZEIIASIiItIpPV0XQERERAxkIiIiSWAgExERSQADmYiISAIYyERERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAfq6LoCIdCs4OBizZs0qNlwmk8HS0hINGjTAe++9hw8++ACGhoaq8StWrMDKlStLna+hoSFsbW3RokULfPrpp3Byciq1bVZWFsLCwhAaGorExEQ8evQIZmZmcHFxQf/+/TFgwADo6b14/yE6OhqjR48GAHz11Veq74kqCgYyEQEAatSogU6dOql+ViqVSEtLw6VLl/Ddd9/h4MGD2Lhxo1ooA0D9+vXRqlUrtWFCCDx69AhxcXEIDw/HgQMHsG7dOnTo0KHYcqOjozFt2jQ8ePAAFhYWcHBwQIsWLfDgwQOcOnUKJ06cQHBwMAIDA2Fqalpq/Tt27IBMJoOBgQF+++03BjJVOAxkIgIANGnSBN99912x4VlZWRg/fjyio6OxceNGjBs3Tm1827Zt8Z///KfEeWZmZmLu3LkICwtDQEAAQkNDIZPJVOPPnTuHDz74AEqlEp9//jm8vb1hZmamGn/jxg1MnToVUVFR+PTTTxEUFKQ2/fM17tu3D66urrC1tUVkZCSio6PRtm1bbV8OojeO55CJ6IXMzMzw2WefAQAOHDhQpmnNzc2xYMECyOVy3LhxA7du3VKNy87OxrRp01BYWIiAgABMmDBBLYyBZx8S1q1bh2rVquH06dOIjIwscTlhYWHIzs7G22+/jb59+wIAfvvttzLVSqRrDGQieqlatWoBeLYnWlbm5uawsLAAADx9+lQ1/M8//8S9e/fg4uKCYcOGlTp9jRo1MHbsWHTq1AnZ2dkltgkODgYAeHh4oEePHjA3N8f+/fvx5MmTMtdLpCsMZCJ6qZiYGABAy5YtyzztuXPn8PTpU+jr66NRo0aq4eHh4QCA3r17l3gY+nkTJ07EL7/8otr7fV5CQgLOnTuHBg0aoFWrVjA2Nkbv3r2Rn5+vCmqiioCBTEQlUigUSE1Nxd69e7Fw4UJYWVnBx8dHo2kLCwuRkpKCPXv2YPLkyQCA0aNHo1q1aqo2N2/eBKBdyD9vx44dAIDBgwergn3w4MEAgG3btoGPfKeKghd1EREAICoqCo6OjiWOMzU1xcaNG1G/fv1i43bu3ImdO3eWOl99fX2MHTsW06dPVxuekpIC4NkhaW0pFArs3r0benp6GDhwoGq4u7s7GjdujJs3b+LEiRPo3Lmz1ssgelMYyEQEoPhtT0IIZGVl4fr167h79y4++ugjfP/99+jatavadM/f9qRUKhEXF4cbN25ALpdj4sSJGDVqFKytrYstTy6XA3gWqto6evQoHj58iC5dusDOzk5t3ODBg/Hdd9/ht99+YyBThcBAJiIApd/2BDzbC549ezZ8fX3x559/onbt2qpxJd329Ouvv2LBggVYt24d2rRpg7feeqvYPG1tbZGZmYnHjx9rXXPR4eqEhAR4eXmpjcvMzAQAREREIDk5WXVhGpFU8RwyEb3UoEGD4OHhgby8POzevful7ceMGYOxY8ciOzsbvr6+uH37drE2rq6uAP6+YOxFbt26hR9++AGnTp1SDXvy5InqNqh79+4hKipK7V9sbCyAZ+ezt2/frtF6EukSA5mINNKsWTMAwP379zVqP3XqVDRp0gSZmZmYOXNmsYurunfvDgDYv3//Sy+82rlzJ37++WcsWrRINWzPnj0oKChA3759cfXq1RL/LVu2DACwffv2Vzo0TvQmMJCJSCOJiYkAUOxcbWkMDQ3x9ddfQyaT4dy5c9i2bZva+B49esDe3h6XL19WHXouyZ07d7B161YAwKhRo1TDi6bp06dPqdN2794dlpaWSEpKwqFDhzSqm0hXGMhE9FJ//vknDhw4AAMDA/Tq1Uvj6Vq3bq3q9GPp0qVqHXUYGhpi3rx5kMvl+Oqrr7B27Vrk5OSoTR8XF4dPPvkET58+Rdu2bVW3M128eBHXrl2DlZUV3n777VKXb2hoqKq3KNSJpIoXdRERgGf9Rk+bNk1tWGFhIa5cuaLaO/bz84O9vX2Z5jt16lQcOHAAT548weLFi9UuAOvcuTNWrFiBqVOnYsmSJQgMDISLiwuqV6+O27dv4/LlywCA9u3bY+XKlaors4v2jv/1r3/BwMDghcsfOHAgfv/9dxw/fhx37twp8dYtIilgIBMRAODx48cICQlRG2ZkZIRatWqhb9++GDVqFNq0aVPm+VpZWWHGjBnw8/PDzp07MWTIELRr1041vnv37ggNDcXmzZtx4sQJxMTEIC8vD5aWlujSpQsGDhyIvn37qjr9yMvLQ2hoKACU2HPXP7Vu3RoNGzZEYmIitm3bhqlTp5Z5HYjeBJlgNzZEREQ6x3PIREREEsBAJiIikgAGMhERkQQwkImIiCSAgUxERCQBDGQiIiIJ4H3IEvDXX39BCPHSDg6IiKhiKSgogEwmg7u7+0vbcg9ZAoQQL+1cv6oQQiA/P5+vRyXGbVz5cRv/rSzv79xDloCiPePmzZvruBLdy87OxpUrV9C0aVOYmprquhx6DbiNKz9u479dvHhR47bcQyYiIpIABjIREZEEMJCJiIgkgIFMREQkAQxkIiIiCagwV1nn5eWhXbt2uHjxIq5fv46mTZuqjU9OTsaCBQuwb98+3Lt3D3Z2dhg2bBjmzp0Lc3PzYvNTKpXYsGEDVq1ahWvXrsHQ0BBdunTBV199hdatW5dYw/nz5xEQEIDTp08jLS0NDg4O8PHxgY+Pj+pZrURUeSgUChQUFOi6jAonLy9P9b+eXuXd7zMwMIBcLi+3+VWYQJ49e3apl48/ePAAnTp1QmJiItzc3NCnTx+cOXMGixcvxr59+3Ds2DFYWFioTTNx4kQEBgaievXq6NGjB5KTk7F7926EhYUhLCwMPXr0UGt/6NAh9OzZE/n5+ejatSusrKwQERGBCRMmICoqCuvXr39t605Eb5YQAklJSXj69KmuS6mQlEol9PX1cf/+/UodyABgZWWF2rVrl8tOWYUI5MjISPzwww+ljp80aRISExPh5+eHhQsXAgDy8/Ph5eWFbdu2wd/fH99//72q/d69exEYGAhXV1ccOnQINjY2AIAdO3Zg+PDhGDt2LOLj42FsbAzg2ae80aNHo6CgACEhIejTpw+AZx8EPD098csvv2DQoEHo16/f63oJiOgNKgrjmjVrwtTUlEfAykihUCAvLw9GRkblugcpJUIIZGdn4+HDhwAAOzu7cpmppKWmpor69esLBwcHUbt2bQFAXL9+XTU+Pj5e6Onpifr164v8/Hy1aZ8+fSosLCyEmZmZyM7OVg3v2rWrACDCw8OLLc/b21sAEEFBQaph69evFwDE8OHDi7WPiIgQAES3bt20XseYmBgRExOj9fSVSVZWloiOjhZZWVm6LoVeE6lv48LCQhEbGysePXqk61IqrMLCQpGZmSkKCwt1Xcpr9+jRIxEbG1vqupbl/V3yxxImTpyI+/fvY+PGjTAyMio2Pjw8HEqlEn369CnWF3S1atXg6emJrKwsHDp0CACQnp6OY8eOwdzcvNhhaQAYNGgQACA0NFQ1rOj7onHP8/DwQPXq1XHkyBFkZmZqvZ5EJA1F54yreg9TpJmi35PyuNZA0oG8detWbN26FbNmzUKHDh1KbHPp0iUAgJubW4njXVxcAPzdfVlsbCyUSiWcnJygr1/8iP0/279sGXp6enBycoJCoUBsbKymq0ZEEsfD1KSJ8vw9kWwg37lzBxMnToS7uzu++uqrUts9ePAAQOnH74uGJyUladVe22mIiIjKQpIXdQkh8MEHHyAnJwebNm164WMJiw4Tl3Z4ycTERK1dWdtrO01Zif+/QKCqy8nJUfufKh+pb+O8vDwolUooFAooFApdl1Mhif9/upEQotK/hgqFAkqlEjk5OVAqlcXGCyE03ouWZCAvXboUkZGR+O677+Dq6vrCtkVX8JW2ws//YmjS/p/TFU2jVCo1XoY2CgoKcOXKFa2nr2wSExN1XQK9ZlLexvr6+qp7aUl7VeE1zMvLQ2FhIW7evFlqG0NDQ43mJblAvnjxIubMmYOuXbviiy++eGn7ok4/Svu0nZubCwAwMzPTqH3R8KL2RdOkpqYiJyenxAvL/rkMbRgYGBTr7KQqysnJQWJiIho2bKg68kCVi9S3cV5eHu7fvw8jIyPVrY9UNkII1W1PVeFcvL6+Pho0aFBiPsTHx2s+n/IsqjzMmjVL1buLt7e32rhHjx4BAKZNmwZzc3PMmTMHderUAVD6+dt/nv8ta/uiaVJTU5GUlAQrKyuNpikrmUzGqzqfY2JiwtejkpPqNtbT04Oenh7kcrlO76EVQmDLli3YsmUL7ty5g1q1auH999/HuHHjMGvWLCQlJaFfv35YvXo17t27h8aNG2Pq1Knw8PBAUlISunXrBj8/P3zwwQeqeaanp6Nz586YMmUKxo0bp3EtO3fuxJo1a3Dnzh00atQIs2bNwscff4yvv/4agwcPRnBwMP79738jICAAP/74I/T09LBy5Uq4urpi//79WLt2LRISEmBqaoru3btj6tSpqFatGgBgxYoVWLlyJa5evaq2TEdHR/j6+mLSpEm4e/cuunfvjqVLlyIkJAQnT56ElZUVhg4dis8++0ynnY/I5XLo6enBxMSkxA9wZflAIrlALjoPW3SbUkl2794NABg3bhyaN28OAKVe4Xz58mUAULVzcXGBnp4e4uLioFQqi23If7Yv+v7y5cuIjY2Fk5OTWnulUom4uDjI5XLVFdqkPZlMBhMTkyrxqZroRZYuXYp169Zh7Nix6Ny5My5fvowffvgB+fn5AJ7d/fHw4UNMnjwZ5ubm+PHHHzF58mQcOXIEtWvXRocOHRAWFqYWyPv370dhYWGZOjHatWsX/Pz8MGzYMMyaNQsxMTGYOHFisXPDCoUCP//8M77++ms8evQIjRs3xs8//4zly5dj1KhR+OKLL3Dnzh38+OOPOH/+PLZt21bmIxD+/v7w8PDAihUrcPbsWaxatQpZWVnw8/Mr03wk61Vvin6T7O3ti3UMkpiYKACIRo0aFbsxu6hjEHNzc5Genq4a3q1bNwFAHDx4sNgyvLy8BADxyy+/qIYFBQUJAMLLy6tY+4MHDwoAwsPDQ+v10nXHIAqFQmfLrgj4+pQvqXcMkpOTI2JjY0VOTo7OakhLSxOurq7i22+/VRu+cOFC8eGHH4qZM2cKBwcHcevWLdW4qKgo4eDgIPbt2yeEECI4OFg4ODiIO3fuqNp4e3uLsWPHlqmWd955R/j4+KgNCwwMFA4ODmLHjh1CCCF27NghHBwcxLZt24QQzzoGuX//vnBzcxNz5sxRm/bMmTPCwcFBbN68WQghxPLly4WDg0Ox5To4OIjly5cLIYS4c+eOcHBwEN7e3mptvv76a+Hq6irS0tLKtE7l6WW/L2V5f5fcHnJZ2dvbo2/fvti7dy9mzJiB7777DjKZDPn5+fDx8UFGRgamTp2q1pf1Z599hsjISPj6+iIiIgK1a9cGAAQHB2PLli2oU6cORo4cqWo/ePBg+Pn5YcuWLRg4cCAGDx4M4Nlhb19fXwDPDqNXVHp6eggODkZKSoquS5EcW1tb1fYmelPOnz+PgoICvPvuu2rDi/YE/fz8YG1tjQYNGqjGFb2PFV0H869//QsBAQEICwvD+PHjkZKSgqioKFX3wpq4desW7t+/jylTpqgN79Onj1p3xEUcHBxU31+8eBH5+fnF9sbbtm2LunXr4vTp0xg1apTGtQBA//791X5+7733sHHjRpw/fx5du3Yt07ykqMIHMgCsXLkS586dw9KlSxEWFgY3NzdERUXh9u3baNOmDfz9/dXaDxkyBKNHj8bmzZvh6OgIT09PPHr0CMePH4ehoSE2b96sdnLe3Nwca9euxcCBAzF06FB07twZNjY2iIiIQHp6Onx8fNC3b983vNblKyUlhfdRE0lE0UMtrK2tS23zzwviik7zFN16Y2Zmhh49eqgCOTQ0FEZGRsVC/kWePHkCAKhRo4bacFtb2xLbP98uLS0NAFTPCniejY0NMjIyNK6jSM2aNdV+Lnp90tPTyzwvKZJsxyBlYW9vj6ioKIwbNw5paWkICQmBsbExZs+ejYiIiBIfvxgUFITly5fD3t4e+/btQ3x8PPr3749Tp07hnXfeKda+d+/eOHr0KHr16oVLly7hwIEDaNq0KdauXYtVq1a9gbUkoqrC0tISwN+BWOTBgwc4deqUxt00DhgwAFeuXEFiYiJCQ0PRo0ePMt0NUrTX/fjxY7Xh//y5JEUXbRVdjPu8lJQUVK9eHcDfHySePyedlZVV4jz/+fStojr++YGhoqpQe8gvum+xbt26WLNmjcbzksvlmDRpEiZNmqTxNB06dFDr45qI6HVo0aIFDAwMcPDgQbRt21Y1PCgoCDt37sTbb7+t0Xw6deoEW1tbbNq0CTExMZg8eXKZ6qhduzYaNGiA//3vfxg4cKBq+P79+186rZubGwwNDRESEqLW9XF0dDTu37+vusq7aIfpwYMHqFevHgDg3LlzJc4zIiJC7RD4/v37YWJigpYtW5ZpvaSqQgUyEVFVYG1tDW9vbwQFBcHQ0BAdO3bExYsX8euvv+LLL7/EtWvXNJqPXC5Hv379EBQUBFtbW3Tq1KlMdchkMkyePBnTpk3DvHnz8O677yIuLg4//fQTALzwdqNq1aph3LhxWLVqFQwMDNC9e3fcvXsXP/74I5o2baq6NsPDwwMLFy7E3Llz8cknnyApKQkrV64scU9+3759sLGxgYeHB6KiorB582Z88cUXkrx9ThsMZCIiCZo+fTpsbGywdetWrF+/HvXq1cPs2bMxatSoMt3mM2DAAKxfvx59+vTR6r7qfv36ITs7G+vWrcOOHTvQrFkzzJkzB3PmzHlpEPr6+qJmzZr49ddfsX37dlhZWaFnz574/PPPVefAGzVqhEWLFuG///0vxo8fjyZNmmDBggVYsGBBsflNmTIFUVFR+P3332FnZ4evvvpK7QLcik4mxCv090jloujJUs/f+/ymBQYG8qKuEtSuXRs+Pj66LqNSyc7OxpUrV+Ds7CzJPZvc3FwkJCSgUaNG7KkLwN69e+Hi4oLGjRurhh06dAg+Pj7YvXt3sb4ZgGfng3Nzc2FsbFwunasUdQyycOFCyd318LLfl7K8v3MPmYioihEaPvRBLpdjz549+OGHH/D555/Dzs4OiYmJWL58Odq3b19iGJP2GMhERFXMzp07MWvWrJe2W7hwIRYtWoTvv/8eS5YswZMnT2BjY4NevXqV+QIxejkGMhFRFdOtWzf88ccfL21Xr149VK9eHV9//fUbqOrFdfyzr+vKiIFMRFTFVK9eXXUfMElHpegYhIiIqKJjIBMREUkAA5mIiEgCGMhEREQSwEAmIiKSAAYyERGRBDCQiYg0pUHvVpVy2WVw+vRpuLi44P79+wAALy8vtb63V69ejY4dO6Jly5Y4cOAAHB0dERwcXC7LFkJg586dqscyBgcHw9HRsVzm/SbwPmQiIk3J5cDo0cCVK292uc7OwObNb3aZ5WTFihWq/qzT09OxdOlSjB8/HsOHD4e1tTWOHTsGCwuLclnWmTNn4Ofnh4MHDwJ49hx7TR9VKQUMZCKisrhyBfjrL11XUWFYWVmpvs/IyIAQAp06dULdunUBQPXUp/Lwz2clGRsbV6gHhPCQNRFRJXb9+nVMnDgRHTp0gJubG959910EBQUBeLb3OnbsWGzcuBFdunRBq1at8OWXXyIlJQUzZsyAu7s7PDw8sHPnTtX8vLy88O2332LGjBlo1aoVunbtitWrVxcLw+fb+/n54fTp0/D09AQAfPDBB6rv/3nIeu/evRgwYABatGiB7t2745dfftFoXU6fPg1vb28AQPfu3REcHFzskPXTp08REBAADw8PtGjRAiNHjkR0dLRq/IoVK+Dl5YU1a9aga9euaN68Oby9vXHz5s1X2gaaYiATEVVSOTk5+PDDD2FqaootW7YgNDQUvXr1wrfffosr/3/YPTo6GtHR0QgKCsKyZcuwf/9+9O3bF87OztixYwe6du2Kr776Cqmpqar5btmyBSYmJtixYwe++OIL/PTTT1izZs0La3F3d8f27dsBPAu+kvrS3rdvH6ZPn44+ffpgz549mDp1KpYtW4bt27e/dF3c3d2xYsUKAMD27dvRu3dvtXkrFAp89NFHiI6OxqJFi7Bz5044OTlh7NixqkckAsBff/2FM2fOYPXq1diwYQPu37+PgIAA7TZAGfGQNRFRJZWTkwNvb2+MGjUK5ubmAABfX18EBgaqHtagVCrx9ddfw9LSEk2aNIGzszMMDAzw4YcfAgDGjh2Lbdu24datW6r+rxs3bgx/f3/IZDI0adIEN27cwMaNG/HJJ5+UWouhoSGsra0BANWqVVN9/7wNGzagV69eGD9+PACgYcOGyMrKgomJyUvXxdnZGdWqVQMAWFtbFztUfezYMVy+fBkhISFwcHAAAHz11Ve4cOEC1q1bh2XLlgEACgsLsXjxYtWhdi8vLyxZsqRsL7yWGMhERJWUtbU1Ro0ahbCwMMTFxeHWrVuqPWOlUgkAqFGjBiwtLVXTmJiYwM7OTvWzkZERACAvL081rH379pDJZKqfW7VqhTVr1qjtRWvj6tWr6NWrl9qwYcOGqb5/2bq8yLVr12BhYaEKYwCQyWRo27Ytjh49qhpmY2Ojdt7bwsICBQUF2q5SmTCQiYgqqUePHuH9999H9erV0b17d7z11lto3rw5PDw8VG0MDAyKTaen9+Kzmfr66tFRdP646Gpqbenr66sF/fM0WZcXEUKUOG+lUqm2PoaGhtoVXw4YyERElVRISAiePn2K/fv3q4K36FB1aRdhaeL5c64AcO7cOdSrV091yFhbTZo0KTbvb7/9Fnfv3kW7du1eui6lhTnw7OKx9PR0XLt2TW0v+ezZs2jatOkr1V1eeFEXEVElVbt2beTk5CA8PBz379/HsWPH8OWXXwIA8vPztZ5vdHQ0li9fjoSEBPzxxx/YvHkzxo0b98r1jh8/HmFhYdi4cSNu376N0NBQ/Pbbb3j33Xc1WhdTU1MAQFxcHLKystTm3blzZzg6OmLq1Kk4ffo0bty4gYCAAFy7dg0ffPDBK9deHriHTERUFs7OFWaZPXv2xOXLl7Fo0SJkZmaibt26GDZsGA4ePIiYmBjUqVNHq/l2794d169fx4ABA1CzZk34+flh5MiRWs3reZ6enliwYAHWrFmDxYsXo27dupg9ezYGDRoEIcQL12XkyJFwcHCAh4cHPv/8c3z55Zdq54L19fXxyy+/YNGiRZg0aRLy8/Ph6uqKDRs2oFWrVq9ce3mQiVc5bkHlougQTfPmzXVWQ2BgIJKSknS2fKmqXbs2fHx8dF1GpZKdnY0rV67A2dlZtUcjJbm5uUhISECjRo2KdyqhUDzrrUsXdLns53h5eaFu3br4z3/+U2obhUKB3NxcGBsbv/J5Zal74e8Lyvb+zkPWRESa0mW4VPJgIwYyERGRJPAcMhERaWzTpk26LqHS4h4yERGRBDCQiYiIJICBTEREJAEMZCIiIglgIBMREUkAA5mIiEgCGMhEREQSwEAmItKQAooqt2xHR0cEBwfrZNlVDTsGISLSkBxyjMZoXMGVN7pcZzhjMza/0WUWOXbsGCwsLHSy7KqGgUxEVAZXcAV/4S9dl/HG2Nra6rqEKoOBTERUiT158gQLFizA0aNHIZfLMXToUFy8eBHt2rWDr68v1q1bhx07duDOnTswMjJC27Zt8e9//xv169cH8OyQ9cKFCzF48GD4+flBoVDAxsYGu3btQnZ2Njp37oyAgAAGdzngOWQiokpKqVTCx8cHt27dwpo1a7B+/XrExMQgKioKABAUFITAwEBMnz4d+/fvx6pVq5CQkPDCRyuGh4fj6dOn+PXXX7Fy5UqcPXsWP/zww5tapUqNe8gEgIelSsPXhSqyqKgoxMTEIDw8HI0bNwYALFu2DN26dQMANGjQAP/5z3/g6ekJAKhbty569eqF0NDQUudpbm6O+fPnw8DAAE2aNMGAAQNw+PDh178yVQADmaBUKjF48GBdlyFZSqUSeno8mEQVT2xsLKpVq6YKYwCoUaMGGjVqBADw9PTEhQsXsHz5cty6dQs3btzA9evXUatWrVLnaW9vDwMDA9XPFhYWKCgoeH0rUYUwkAl6enqYgzlIQIKuS5GcRmiEb/S+0XUZRFqRy+VQKpWljl+zZg1WrFiBwYMHo3379vDy8sLBgwdfuIdsaGj4OkolMJDp/4UjvEpdOaopd7jjGzCQqWJycnJCRkYGbty4gSZNmgAAnj59ilu3bgEA/vvf/8LX1xfjx49XTbNu3ToIIXRSb1XH43BERJVUhw4d0KpVK8yYMQPnz59HXFwcpk2bhpycHMhkMtjZ2eH48eOIj4/HzZs38cMPP+DPP/9Efn6+rkuvkriHTERUBs5wrlDLXL58OebPn4+xY8fCyMgIo0aNwo0bN2BgYIDFixdj/vz5GDJkCMzMzNCyZUsEBATA398fd+/eRb169cpxLehlGMhERBpSQKGzHrMUUEAOeZmmefLkCWJjY7Fs2TLVhVj5+fnYsGEDatWqBVdXV/z+++/FphsxYoTq+6tXr6q+L+l2qEmTJmHSpEllqotKxkAmItJQWQNR18vW19fHF198gREjRmDkyJEoKCjAunXrYGhoiK5du76GKulV8BwyEVElZWlpiZ9//hnnz5/HwIED8f777+PRo0fYuHEjrK2tdV0e/QP3kImIKrGOHTvit99+03UZpAHuIRMREUkAA5mIiEgCGMhEREQSwEAmIiKSAAYyERGRBDCQiYiIJICBTEREryQyMhLx8fGvNA8/Pz94eXmVU0UVEwOZiEhDL3qUYWVe9ovcu3cPn376KR4/fqzrUio8dgxCRKQhPT09BAcHIyUl5Y0u19bWFoMHD36jy9QUH9VYfhjIRERlkJKSgqSkJF2XoRFHR0csWLAA4eHhOHv2LKpVq4YxY8bAx8dH1SYyMhIrVqxAfHw8atWqhT59+mDixIkwNDRUzWPhwoVqHwg8PT0xaNAgDBo0CN27dwcAeHt7w9fXF+3bt4e3tzcmT56MoKAg1K1bF3/88QfOnz+PlStXIiYmBnl5eWjYsCE+/fRT9O3b982+KBLGQ9ZERJXY4sWLMXDgQOzevRtDhgzB0qVLER0dDQA4cuQIpkyZgmHDhmHv3r2YN28ewsPDMX36dI3mbWdnh+3btwMAVqxYgY8++kg17ujRo9i6dSu+/fZbPHr0CB999BGcnJwQHByM3bt3o3nz5pg1axYePXpU/itdQTGQiYgqsUGDBmHAgAFo1KgRPv/8c1SrVg1nz54FAPz8888YOnQoRo4ciQYNGqBLly4ICAjAvn37cPfu3ZfOWy6Xqx5SUa1aNZiZmanGeXt7o2HDhnB2dkZ+fj58fX0xffp02Nvbo0mTJvDx8UF+fj4SExNfy3pXRDxkTURUiTVp0kTtZ3NzcxQUFAAAYmNjERMTg507d6rGF50TvnHjBurVq6f1cuvXr6/2/ZAhQ/Drr78iPj4eiYmJuHLlCgBAoVBovYzKhoFMRFSJFZ0Lfl5R6CqVSowbNw6DBg0q1sbW1rZY+yJFgf4ixsbGqu9v3LiBkSNHwsXFBZ07d0b37t1RvXp1DBs2TOP1qAoYyEREVVSzZs1w8+ZN2Nvbq4ZFRUUhKCgI/v7+MDU1hYGBATIyMlTjMzMz8eTJE9XPMpnspcvZunUratSogQ0bNqiGRUREAOBV2s/jOWQioirqk08+wZ9//okVK1YgISEBJ0+exKxZs5Cenq7aQ3Z3d8fvv/+Oy5cv49q1a5gxYwb09f/elzM1NQUAXLt2TS24n1e7dm0kJSXh8OHDuHfvHv7880/4+/sDAPLz81/vSlYg3EMmIiqD5w/lVvRl9uzZEz/88AMCAwMRGBiIatWqoVu3bmpXWfv7+yMgIAAjRoyAtbU1PvzwQ2RnZ6vGV69eHUOGDMHixYtx69YtvPvuu8WW4+3tjZs3b2LGjBnIz89Hw4YN8eWXX2L58uWIiYlB165dX8v6VTQyweMFOnfx4kUAQPPmzXVWQ2u0xl/4S2fLlyp3uOMczum6jEolOzsbV65cgbOzs2rvSkpyc3ORkJCARo0aqZ0HBZ6dc9XT082BRV0uu6wUCgVyc3NhbGwMuVyu63Jeqxf9vgBle3+vGFuXiEgCdBmIFSWMSXvcwkRERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMRFQC3oBCmijP3xMGMhHRcwwMDABA7V5botIU/Z4U/d68Csl2DFJQUIDly5djw4YNuH79OszMzNC+fXt8/vnneO+994q1T05OxoIFC7Bv3z7cu3cPdnZ2GDZsGObOnQtzc/Ni7ZVKJTZs2IBVq1bh2rVrMDQ0RJcuXfDVV1+hdevWJdZ0/vx5BAQE4PTp00hLS4ODgwN8fHzg4+OjUfdxRCR9crkcVlZWePjwIYBnPVHx77tsFAoF8vLyAKDS3ocshEB2djYePnwIKyurcllPSQayEALDhw/Hzp07YWVlhR49eiA3NxcHDx7Evn378PXXX2POnDmq9g8ePECnTp2QmJgINzc39OnTB2fOnMHixYuxb98+HDt2DBYWFmrLmDhxIgIDA1G9enX06NEDycnJ2L17N8LCwhAWFoYePXqotT906BB69uyJ/Px8dO3aFVZWVoiIiMCECRMQFRWF9evXv5HXhohev9q1awOAKpSpbJRKJQoLC6Gvr1/p75+2srJS/b68KkkGcmBgIHbu3IlWrVrhwIEDqFGjBgAgJiYGb7/9NubOnYvBgwfD2dkZADBp0iQkJibCz88PCxcuBPCsf1QvLy9s27YN/v7++P7771Xz37t3LwIDA+Hq6opDhw7BxsYGALBjxw4MHz4cY8eORXx8vKrXlby8PIwePRoFBQUICQlBnz59ADz7IODp6YlffvkFgwYNQr9+/d7Ya0REr49MJoOdnR1q1qyp0ZONSF1OTg5u3ryJBg0awMTERNflvDYGBgblewRASFCnTp0EABEREVFs3IQJEwQA8f333wshhIiPjxd6enqifv36Ij8/X63t06dPhYWFhTAzMxPZ2dmq4V27dhUARHh4eLH5e3t7CwAiKChINWz9+vUCgBg+fHix9hEREQKA6Natm9brGxMTI2JiYrSevjy4C3cBfhX7chfuOt0ulVFWVpaIjo4WWVlZui6FXhNu47+V5f1dkscSDh48iPPnz5fY4XhmZiYAqJ42Eh4eDqVSiT59+hQ7qV6tWjV4enoiKysLhw4dAgCkp6fj2LFjMDc3L3ZYGoDquaChoaGqYUXfl/TMUA8PD1SvXh1HjhxR1UZERFRWkgxkY2NjtGzZstihgN27d2P79u0wNTXF4MGDAQCXLl0CALi5uZU4LxcXFwB/d/AdGxsLpVIJJycntUeIldb+ZcvQ09ODk5MTFAoFYmNjy7SeRERERSQZyM9LTU3FkCFD4OzsjIEDB8LW1hYhISGoV68egGfncQHAzs6uxOmLhiclJWnVXttpiIiIykKSF3U97+bNmwgODlb9LJPJcPnyZXh6egL4+xB2aY9xK7qgoKhdWdtrO01Zif+/hP5Nk8lklfqii/KSk5PDjiLKSU5Ojtr/VPlwG/9NCKHxbXOSD2RHR0c8efIEQgj873//w5QpUzB58mSkp6djzpw5qsPapa1w0Zto0f8va//P6YqmUSqVGi9DGwUFBbhy5YrW02vLxMREdZieSpeQkMA3l3KWmJio6xLoNeM2fsbQ0FCjdpIP5Oc79Rg+fDjq16+PLl26YOHChZgyZYpqfGlvlrm5uQAAMzMztfmV1r5oeFH7omlSU1ORk5MDIyOjly5DGwYGBmjatKnW02uLHR5oplGjRtxDLic5OTlITExEw4YNeXSmkuI2/lt8fLzGbSUfyP/UqVMnNGnSBPHx8bh27Rrq1KkDoPTzt/88/1vW9kXTpKamIikpCVZWVhpNU1YymazUQ+Kke1X9TeV1MDEx4e98JcdtXLadHsld1JWdnY2pU6dizJgxpe6RFO2lFhQUoHnz5gBQ6hXOly9fBgBVOxcXF+jp6SEuLg5KpfKl7Z//vqRlKJVKxMXFQS6X89AvERFpTXKBbGJigqCgIGzevBlHjhwpNj4hIQFXr16FkZER3Nzc0LNnTwDPet9SKBRqbdPS0hAZGQlzc3O8/fbbAJ5dmOXh4YG0tDTVvcnP27lzJwCgd+/eqmG9evUCAOzatatY+0OHDiE1NRVdunQp1j0nERGRpiQXyDKZDOPHjwcATJgwQXU4GADu3r2LESNGoLCwEBMmTICZmRns7e3Rt29fJCQkYMaMGaq96vz8fPj4+CAjIwM+Pj5qYfnZZ58BAHx9fdUOXQcHB2PLli2oU6cORo4cqRo+ePBg2NnZYcuWLWpXfCclJcHX1xcAMG3atNfwahARUZXxGnoKe2XZ2dmq7i3Nzc1Fz549haenpzAzMxMAxHvvvSdyc3NV7RMTE0WdOnUEAOHk5CSGDh0qGjRoIACINm3aiIyMjGLLGD16tAAgLC0txcCBA0WXLl2ETCYTRkZGIjIyslj70NBQYWBgIGQymejSpYsYOHCgsLS0FACEj4/PK60vu86U7he7zix/7Fax8uM2/luF7zrTxMQEBw4cwJIlS9C4cWNERkbi9OnTcHNzw88//4zQ0FC1q53t7e0RFRWFcePGIS0tDSEhITA2Nsbs2bMRERFR4uMXg4KCsHz5ctjb22Pfvn2Ij49H//79cerUKbzzzjvF2vfu3RtHjx5Fr169cOnSJRw4cABNmzbF2rVrsWrVqtf5chARURUgE4L3cuhaUTedz19I9qa1Rmv8hb90tnypcoc7zuGcrsuoVLKzs3HlyhU4OztX+StwKytu47+V5f1dknvIREREVQ0DmYiISAIYyERERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAQxkIiIiCWAgExERSQADmYiISAIYyERERBLAQCYiIpIABjIREZEEaB3IBQUFCAkJQWJiompYZGQkevfuDXd3d4wdOxZxcXHlUSMREVGlp1Ugp6enY9CgQZgxYwaio6MBALdu3cKkSZNw8+ZN5OTk4NSpUxgzZgzu3r1brgUTERFVRloF8rp16xAfH4+WLVvC2dkZAPDbb7+hsLAQAwYMQHR0NObOnYvMzEysXr26XAsmIiKqjLQK5IMHD8LGxgYbN26Eq6srAODAgQOQyWT49NNPYW5ujtGjR8PJyQnHjh0r14KJiIgqI60C+e7du2jVqhUMDQ0BPDtcfefOHdjZ2aFRo0aqdvb29khJSSmfSomIiCoxrQLZyMgIhYWFqp+PHj0KAOjUqZNau9TUVFVoExERUem0CuSGDRviwoULyMnJgRACe/bsgUwmQ7du3VRtEhIScO7cOTRr1qzciiUiIqqstArkXr16ITU1FYMHD8bIkSMRExODmjVromvXrgCAwMBAjB49GgqFAoMGDSrXgomIiCojfW0mGjt2LO7fv4+NGzcCAKpVq4YlS5bAwMAAALBjxw48efIE3t7eGD58ePlVS0REVElpFcgAMHv2bHz44YdISUmBg4MDjI2NVeM+//xzNG7cGE5OTuVSJBERUWWnVSCfOXMGNWrUQOPGjWFnZ1dsfO/evQEAFy5cwLVr1zBs2LBXq5KIiKiS0+ocspeXFwIDA1/abt26dfjPf/6jzSKIiIiqFI32kE+dOgUhhNqwlJQUnDx5stRpMjMzce7cuWLTERERUXEaBfKOHTuwd+9e1c8ymQwnT558YSADgBAC3bt3f7UKiYiIqgCNAnn69OlITk5W7e0+fw65JDKZDEZGRrC3t8fEiRPLr1oiIqJKSqNArlmzpuoWJwBwcnJC586dsXjx4tdWGBERUVWi1VXWBw8ehKmpaXnXQkREVGVpFch169YtNkypVJbaXk9Pq4u5iYiIqgytOwb5888/sXr1aly/fh35+fmltpPJZIiNjdV2MURERFWCVoEcGRmJKVOmaHRLE297IiIiejmtAnnt2rUQQuDDDz+Et7c3atasCblcXt61ERERVRlaBXJsbCycnZ0xc+bM8q6HiIioStLqait9fX3Uq1evvGshIiKqsrQK5JYtW+Ly5ctQKBTlXQ8REVGVpFUgT548GQ8fPsSiRYsYykREROVAq3PI58+fR5cuXbBp0yaEhobCzc0NlpaWkMlkxdrKZDIsWrTolQslIiKqzLQK5G+//RYymQxCCDx+/BiHDx8utS0DmYiI6OW0CuSFCxeWdx1ERERVmlaBPGjQoPKug4iIqEpjJ9NEREQSoNUe8sqVKzVuK5PJ8Nlnn2mzGCIioipD60Auuqjref+8yloIwUAmIiLSgFaB7OvrW+JwhUKBtLQ0nD9/HrGxsRg4cCB69uz5SgUSERFVBeUayM9bu3Ytli5diiFDhmizCCIioirltV3UNW7cONjb2+Onn356XYsgIiKqNF7rVdbNmjVDTEzM61wEERFRpfBaA/natWt8TjIREZEGtDqHfOfOnVLHKRQKPH78GJs2bcKtW7fw9ttva10cERFRVaFVIL/77rslPkjieUIIGBgYaHQBGBERUVWnVSDXqVOn1HF6enowNTWFk5MTvL294ebmpnVxREREVYVWgRwREVHedRAREVVp7MuaiIhIArTaQy6SkpKCrVu3IioqCikpKTA0NESNGjXQoUMHDBw4EHZ2duVVJxERUaWmdSAfPnwYU6dORVZWllqf1tevX8fp06exdu1aLFmyBJ6enuVSKBERUWWmVSDfuHEDU6ZMQV5eHgYOHIjevXujXr16EELgzp07CAsLw549ezB16lQEBwejUaNG5V03ERFRpaJVIAcGBiIvLw/ffPMNBg8erDaucePG8PDwQPv27TFnzhz88ssvmD9/frkUS0REVFlpdVHXyZMn4ejoWCyMnzdkyBA4OTnh2LFjWhdHRERUVWgVyKmpqRodhm7UqBEePXqkzSKIiIiqFK0CuXr16khISHhpu4SEBFSrVk2bRRAREVUpWgVyx44dcfXqVezatavUNrt27UJcXBw6dOigbW1ERERVhlYXdfn4+GD//v2YPXs2Tp8+jZ49e6JevXoAgLt372L//v3YtWsXDA0NMX78+HItmIiIqDLSKpCbNm2KpUuXYtq0adi5c2exPWUhBExMTLB48WI4ODiUR51ERESVmtYdg/To0QMHDhzAb7/9hujoaDx8+BBCCNSsWRPt2rXDsGHDUKtWrfKslYiIqNJ6pa4zbWxs+HhFIiKicqDVRV3p6eklDn/8+DEiIyOhUCheqSgiIqKqpkyBfPLkSQwZMgRLliwpcXxkZCQmTpyIXr164fjx4+VSIBERUVWgcSD/73//w/jx43H58mVcvny5xDbZ2dkwMjLC7du38cknn2Dv3r3lVigREVFlplEg3759G1OnTkVhYSF8fHywYcOGEtt5e3vj+PHj8PLyglKpxNy5c3H//v3yrJeIiKhS0iiQN27ciPz8fMycORNffPEFLC0tS21rZmaGOXPm4LPPPkNOTg42bdpUbsUSERFVVhoF8vHjx1GzZk14eXlpPGMfHx9Uq1YNR48e1bo4IiKiqkKjQL5//z4cHR0hl8s1nrGhoSHc3d1x9+5drYsjIiKqKjQKZAMDAxgYGJR55gYGBrwFioiISAMaBbKdnR1u375d5pnfunULNWrUKPN0REREVY1Ggdy2bVvcuHEDly5d0njGcXFxuHbtGlxcXLQqTKlUYvXq1XjrrbdgaWkJY2NjODo6YubMmXj69Gmx9snJyfD19UXTpk1hYmKCxo0bY+bMmcjMzCx1/uvXr0fbtm1haWkJGxsbDBw4EOfOnSu1pvPnz2PQoEGoU6cOzMzM4O7ujp9//hlCCK3WkYiIqIhGgTxs2DAIITBv3rxSA+55ubm5mD17NmQyGfr161fmopRKJYYOHQofHx9cuHABrVu3Ro8ePZCamorFixejXbt2SE5OVrV/8OABOnbsiJ9++gkmJibo06cPFAoFFi9ejM6dOyMjI6PYMiZOnIiPP/4YN2/eRI8ePeDo6Ijdu3ejY8eOOHDgQLH2hw4dQseOHbF79244ODjg3XffxY0bNzBhwgR8/PHHZV5HIiIiNUJD8+bNE46OjqJv374iIiJCKJXKYm0KCwtFZGSk6NOnj3B0dBTjxo3TdPZq1qxZIwCIZs2aiRs3bqiGp6eni379+gkA4v3331cNHzJkiAAg/Pz8VMPy8vLE+++/LwCIL7/8Um3+ISEhAoBwdXUVKSkpquF//PGHkMvlom7duiInJ0c1PDc3V9SpU0fo6emJvXv3qobfv39fODk5CQBiz549Wq2rEELExMSImJgYracvD+7CXYBfxb7chbtOt0tllJWVJaKjo0VWVpauS6HXhNv4b2V5f9c4kPPy8sT48eOFo6OjcHJyEm3atBHe3t7iyy+/FJMnTxZjxowRrVu3Fk5OTsLR0VGMGTNGZGdna7UCnTp1EgBESEhIsXEPHz4UMplMGBoaiuzsbBEfHy/09PRE/fr1RX5+vlrbp0+fCgsLC2FmZqZWS9euXQUAER4eXmz+3t7eAoAICgpSDVu/fr0AIIYPH16sfUREhAAgunXrptW6CsFAlvIXA7n88c268uM2/ltZ3t817jrT0NAQgYGBmDt3LurVq4fMzEycPn0aoaGh2L9/P86cOYOsrCw0btwY33zzDTZt2gQTExOt9tqtra3h5OSEt956q9g4W1tbVK9eHfn5+Xj06BHCw8OhVCrRp0+fYleCV6tWDZ6ensjKysKhQ4cAPHswxrFjx2Bubo4ePXoUm/+gQYMAAKGhoaphRd8XjXueh4cHqlevjiNHjmh0OJ+IiKgkZX784ujRozF69GhcunQJN2/eREpKCuRyOWrWrAlXV1fY29u/clEhISGljktISMCTJ09gaGgIW1tb1YVmbm5uJbZ3cXHB7t27cfHiRfTq1QuxsbFQKpVwcnKCvn7x1S+6CO3ixYuqYS9ahp6eHpycnHDy5EnExsaiffv2mq8oERHR/9P6echubm6lhuDrNGfOHABAnz59YGxsjAcPHgB4dmtWSYqGJyUlAUCZ22s7DRERUVloHci6sGLFCmzduhWmpqb49ttvAUB1mNjU1LTEaYoOmxe1K2t7bacpKyEEsrOztZ5eWzKZTOtTC1VJTk4Ob28rJzk5OWr/U+XDbfw3IQRkMplGbStMIC9fvhyff/45ZDIZ1q1bBycnJwBQdedZ2goXvYkW/f+y9v+crmgapVKp8TK0UVBQgCtXrmg9vbZMTEy0vle8KklISOCbSzlLTEzUdQn0mnEbP2NoaKhRO8kHshACfn5+WLx4MeRyOdatW4cRI0aoxpubmwMo/ZNYbm4ugGdPodKkfdHwovZF06SmpiInJwdGRkYvXYY2DAwM0LRpU62n15amn9yqukaNGnEPuZzk5OQgMTERDRs25NGZSorb+G/x8fEat5V0IOfk5GDMmDEIDg6GiYkJtm7digEDBqi1qVOnDoDSz9/+8/xvWdsXTZOamoqkpCRYWVlpNE1ZyWSyUg+Jk+5V9TeV18HExIS/85Uct3HZdno0vu3pTUtPT0f37t0RHByMmjVrIjIyslgYA0Dz5s0BALGxsSXO5/Lly2rtXFxcoKenh7i4OCiVype2f9kylEol4uLiIJfLeeiXiIi0JslALigoQJ8+fXDy5Ek0bdoUJ0+eRIcOHUps27NnTwDA3r17iz1ZKi0tDZGRkTA3N8fbb78N4NmFWR4eHkhLS1Pdm/y8nTt3AgB69+6tGtarVy8AwK5du4q1P3ToEFJTU9GlSxdYWFiUeV2JiIiAcgjkgoICxMTEYP/+/Th79iyAZ89PfhX+/v44duwYateujUOHDqFx48altrW3t0ffvn2RkJCAGTNmqM7z5efnw8fHBxkZGfDx8VELy88++wwA4Ovrq3boOjg4GFu2bEGdOnUwcuRI1fDBgwfDzs4OW7ZsQXBwsGp4UlISfH19AQDTpk17pXUmIqKqTetzyIWFhfjpp5+wefNm1cMb+vXrhzZt2mDmzJnIysrCDz/8UOaOQlJTU7Fs2TIAQK1atTBz5sxS237//feoVasWVq5ciXPnzmHp0qUICwuDm5sboqKicPv2bbRp0wb+/v5q0w0ZMgSjR4/G5s2b4ejoCE9PTzx69AjHjx+HoaEhNm/erHbxlrm5OdauXYuBAwdi6NCh6Ny5M2xsbBAREYH09HT4+Pigb9++ZVpPIiKi52kVyIWFhfDx8cGJEyegr68PNzc3tZ6tsrKyEBsbi9GjR2PXrl2wsbHReN6nT59W3Y974cIFXLhwodS2/v7+qFWrFuzt7REVFQV/f3+EhoYiJCQE9vb2mD17NmbOnKm6svp5QUFB6NChA9asWYN9+/bBysoK/fv3h7+/P1q1alWsfe/evXH06FHMnz8fJ06cQGFhIRwcHDBx4kR8+OGHGq8fERFRSbQK5M2bN+P48ePo0qULvv32W9SsWVN1XzAAbNmyBf7+/ti1axfWrVv3wr3cf+rZs6dWt5fUrVsXa9as0bi9XC7HpEmTMGnSJI2n6dChg1of10REROVFq3PIO3fuhJWVFX788UfUrFmz2HhjY2MsWLAANWvWxJEjR165SCIiospOq0BOTExE27ZtX9gRhoGBAVq0aPHKF3gRERFVBVoFsr6+PtLT01/a7unTpyU+UYmIiIjUaRXIzs7OuHjxIu7du1dqmzt37uDixYtq55aJiIioZFoF8siRI5GTk4OJEyeW2HvV1atXMWnSJOTn52Po0KGvXCQREVFlp9Xx5N69e+PEiRP4448/MGTIEFhYWEAmk+Ho0aN45513kJycDCEE+vTpU2J3l0RERKRO6xO8X3/9NVq1aoV169YhISEBwLNOPYBnD2P44IMP4O3tXT5VEhERVXKvdMXV0KFDMXToUKSkpODBgwdQKpWwtbVF3bp1y6s+IiKiKqFcLoG2tbWFra1tecyKiIioSnqlQI6Pj8e1a9eQm5tb4qMMi/DCLiIiohfTKpBzc3PxxRdflPj4wpIwkImIiF5Mq0AODAxEZGQk9PX10aZNG9SsWZMdgBAREb0CrVI0NDQUhoaG+O233+Di4lLeNREREVU5WnUM8uDBA7z11lsMYyIionKiVSDb2NigoKCgvGshIiKqsrQK5B49euCvv/7ik5yIiIjKiVaBPGXKFNSvXx/jx49HZGQknjx5AoVCAaVSWeI/IiIiejGtLury8vJCXl4ebt++jYkTJ76wrUwmK/EBFERERPQ3rQL5ypUrGrcVQmizCCIioipFq0COi4sr7zqIiIiqNK3OIRMREVH5euXutWJiYnD69GkkJSXByckJw4YNw6FDh9CiRQtYW1uXR41ERESVntaBfP/+fUyfPh3nzp1TDevXrx+GDRuG//73v4iLi8N3332Hd999t1wKJSIiqsy0OmSdmpqKMWPG4OzZs2jWrBk++ugjtYu37O3tkZeXhy+++ILnm4mIiDSgVSAHBgbi/v37mDBhAvbs2YPp06erjV+8eDH8/f1RWFiINWvWlEuhRERElZlWgXzgwAHY29tjypQppbYZMWIEmjVrhgsXLmhdHBERUVWhVSAnJyfDycnppe0aNWqEhw8farMIIiKiKkWrQLawsMC9e/de2u7u3buwsLDQZhFERERVilaB3LZtW8TGxiIqKqrUNidPnkRsbCzatGmjdXFERERVhVaB/Mknn0Amk2HChAnYsGGD6kpqhUKBO3fu4Ndff8XkyZOhp6eHDz/8sFwLJiIiqoy0ug+5efPm+Prrr/HVV19h0aJFAJ49RCIsLAxhYWEAAD09PcyZMwfu7u7lVy0REVElpXXHIIMGDUKLFi2wceNGVU9dCoUCtra2aN++Pby8vODq6lqetRIREVVar9R1ZpMmTRAQEFBetRAREVVZfLgEERGRBJR5D/nChQswNTVFs2bNAACrVq3C8ePHi7WrUaMGfvjhB8jl8levkoiIqJLTOJDT0tIwc+ZMHD58GF5eXpg9ezYAICEhAWfPni3WXiaTYe3atfDx8Sm/aomIiCopjQN54sSJOHv2LOrUqVPsYi2ZTIagoCDVz5cvX8aiRYuwdu1ajBkzBmZmZuVXMRERUSWkUSDv378fZ8+eRbt27bB69WqYmJgUa9O+fXu172NiYrBv3z7s3r0bo0aNKr+KiYiIKiGNLuoKDw+Hnp4eFixYUGIYl8TX1xdCCBw5cuSVCiQiIqoKNArkmJgYODg4oGHDhhrPuEmTJmjatCmuXr2qbW1ERERVhkaB/OTJEzRo0KDEcSYmJjA3Ny9xXMOGDfHkyRPtqyMiIqoiNApkAwMDKJXKEsfNnz8fZ86cKXFcYWGhxoe4iYiIqjKNArlWrVq4c+dOmWeekJCA2rVrl3k6IiKiqkajQHZ2dsa1a9dw9+5djWccFxeHW7du8fGLREREGtAokAcMGAAhBL799luNZiqEwDfffAOZTIb+/fu/UoFERERVgUaB3LlzZ7i4uCAyMhJTp05Fenp6qW0zMzMxZcoUREdHo1u3bmjZsmW5FUtERFRZadQxiEwmw7JlyzBq1CiEhYXh8OHD6NatG9q3b4+aNWsCAFJSUnDu3DkcOHAA6enpqFu3rsZ71ERERFWdxl1nNmjQADt27IC/vz8iIyMREhKCvXv3qrURQgAA3n33XXzzzTewtLQs32qJiIgqqTI97alWrVr473//izt37mDPnj24ePEikpKSoFAoULt2bTg7O6N///5o2rTp66qXiIioUirz4xcBoH79+vjss8/KuxYiIqIqS6OLuoiIiOj1YiATERFJAAOZiIhIAhjIREREEsBAJiKiciWTyWBiYgKZTKbrUioUra6yJiIiaVEqldDTk8Y+lomJCVxcXHRdRjFSeo1KwkAmIqoE9PT0EBwcjJSUFF2XIkm2trYYPHiwrst4IQYyEVElkZKSgqSkJF2XQVpiIBMRVRK2tra6LkGyKsJrw0AmIqoElEql5A/J6hrPIRMR0Wunp6eHOZiDBCTouhRJaoRG+EbvG12X8UIMZCKiSiIc4fgLf+m6DElyhzu+gbQDWbr77kRERFUIA5mIiEgCGMhEREQSwEAmIiKSAAYyERGRBDCQiYiIJICBTEREJAEMZCIiIglgIBMREUkAA5mIiEgCGMhEREQSwEAmojdKJpPBxMQEMplM16UQSQofLkFUBUjpsXMmJiZwcXHRdRlqpPT6UNVVYQJ506ZN8Pb2xv/+9z/06NGj2Pjk5GQsWLAA+/btw71792BnZ4dhw4Zh7ty5MDc3L9ZeqVRiw4YNWLVqFa5duwZDQ0N06dIFX331FVq3bl1iDefPn0dAQABOnz6NtLQ0ODg4wMfHBz4+Pvy0T5Kmp6eH4OBgpKSk6LoUybG1teVzhEkSKkQgnzlzBr6+vqWOf/DgATp16oTExES4ubmhT58+OHPmDBYvXox9+/bh2LFjsLCwUJtm4sSJCAwMRPXq1dGjRw8kJydj9+7dCAsLQ1hYWLHQP3ToEHr27In8/Hx07doVVlZWiIiIwIQJExAVFYX169e/lnUnKi8pKSlISkrSdRlEVArJB3JISAi8vb2Rnp5eaptJkyYhMTERfn5+WLhwIQAgPz8fXl5e2LZtG/z9/fH999+r2u/duxeBgYFwdXXFoUOHYGNjAwDYsWMHhg8fjrFjxyI+Ph7GxsYAgLy8PIwePRoFBQUICQlBnz59ADz7IODp6YlffvkFgwYNQr9+/V7Xy0D0ymxtbXVdgiTxdSHJEBJ19+5dMXbsWCGTyYSpqamoVauWACD+97//qbWLj48Xenp6on79+iI/P19t3NOnT4WFhYUwMzMT2dnZquFdu3YVAER4eHix5Xp7ewsAIigoSDVs/fr1AoAYPnx4sfYRERECgOjWrZvW6xoTEyNiYmK0nr48uAt3AX4V+3IX7jrdLuVFoVDougRJqyyvD/+OS//S1d9yWd7fJbuHPGfOHAQFBaFt27ZYv349Jk2ahOTk5GLtwsPDoVQq0adPHxgYGKiNq1atGjw9PbF7924cOnQIvXr1Qnp6Oo4dOwZzc/MSz0UPGjQIGzduRGhoKLy9vQEAoaGhqnH/5OHhgerVq+PIkSPIzMws8Xw1ka7p6elhDuYgAQm6LkVyGqERvtH7RtdlEEn3kLWTkxOCgoIwZsyYF179eOnSJQCAm5tbieNdXFywe/duXLx4Eb169UJsbCyUSiWcnJygr1989Yuu/rx48aJGy9DT04OTkxNOnjyJ2NhYtG/fXvOVJHqDwhGOv/CXrsuQHHe44xswkEn3JBvIfn5+GrV78OABAMDOzq7E8UXDiy5mKWt7bachIiIqC8kGsqYyMzMBAKampiWONzExUWtX1vbaTlNWQghkZ2drPb22ijppoBfLycmBEELXZWiF21gz3MZVw5vezkIIjW+LrfCBLJfLAaDUFS564Yv+f1n7f05XNI1SqdR4GdooKCjAlStXtJ5eW1LspEGKEhISkJOTo+sytMJtrBlu46pBF9vZ0NBQo3YVPpCLLqIq7QXOzc0FAJiZmWnUvmh4UfuiaVJTU5GTkwMjI6OXLkMbBgYGaNq0qdbTa4sdmmimUaNGFXrviV6O27hqeNPbOT4+XuO2FT6Q69SpA6D087f/PP9b1vZF06SmpiIpKQlWVlYaTVNWMpms1EPipHs8HFj5cRtXDW96O5flw1KF77y1efPmAIDY2NgSx1++fFmtnYuLC/T09BAXFwelUvnS9i9bhlKpRFxcHORyOQ8ZERGR1ip8IPfs2RPAs963FAqF2ri0tDRERkbC3Nwcb7/9NoBnF2Z5eHggLS0Nhw4dKja/nTt3AgB69+6tGtarVy8AwK5du4q1P3ToEFJTU9GlS5di3XMSERFpqsIHsr29Pfr27YuEhATMmDFDdW4gPz8fPj4+yMjIgI+Pj1pYfvbZZwAAX19ftUPXwcHB2LJlC+rUqYORI0eqhg8ePBh2dnbYsmULgoODVcOTkpJUfWxPmzbtta4nERFVbhX+HDIArFy5EufOncPSpUsRFhYGNzc3REVF4fbt22jTpg38/f3V2g8ZMgSjR4/G5s2b4ejoCE9PTzx69AjHjx+HoaEhNm/erHbxlrm5OdauXYuBAwdi6NCh6Ny5M2xsbBAREYH09HT4+Pigb9++b3itiYioMqnwe8jAs73kqKgojBs3DmlpaQgJCYGxsTFmz56NiIiIEruzDAoKwvLly2Fvb499+/YhPj4e/fv3x6lTp/DOO+8Ua9+7d28cPXoUvXr1wqVLl3DgwAE0bdoUa9euxapVq97AWhIRUWVWYfaQSzrf+7y6detizZo1Gs9PLpdj0qRJmDRpksbTdOjQQdWvNRERUXmqFHvIREREFR0DmYiISAIYyERERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAQxkIiIiCWAgExERSQADmYiISAIYyERERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAQxkIiIiCWAgExERSQADmYiISAIYyERERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAQxkIiIiCWAgExERSQADmYiISAIYyERERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAQxkIiIiCWAgExERSQADmYiISAIYyERERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAQxkIiIiCWAgExERSQADmYiISAIYyERERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAQxkIiIiCWAgExERSQADmYiISAIYyERERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAQxkIiIiCWAgExERSQADmYiISAIYyERERBLAQCYiIpIABjIREZEE6Ou6ACJ6M5zhrOsSJKkyvS6VaV3KW0V4bRjIBKBi/LLqQmV5XRRCgc2yzbouQ7IUQgG5TK7rMl4Jt/HLSX07M5CJf8gvIfU/Yk3IZXJgzhwgIUHXpUhPo0aQf/ONrqt4ZdzGL1EBtjMDmfiH/CIV4I9YY+HhwF9/6boK6XF3B7iNK78KsJ0ZyPQM/5BLVgH+iImocuBV1kRERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAQxkIiIiCWAgExERSQADmYiISAIYyERERBLAQNbC4cOH8a9//Qs1a9aEhYUFOnXqhO3bt+u6LCIiqsAYyGW0ZcsWdOvWDYcOHYK7uzu6du2Kc+fO4f3330dAQICuyyMiogqKgVwGDx8+xLhx42BqaooTJ05g//79CA0Nxfnz51GrVi0EBATgwoULui6TiIgqIAZyGfz000/IycmBr68v2rZtqxru5OSEhQsXQgiBH3/8UYcVEhFRRcVALoPQ0FAAwKBBg4qNGzBgAGQymaoNERFRWTCQy+Dy5csAAFdX12LjrK2tUbt2bTx8+BAPHz5806UREVEFx0DWUGpqKnJzc2FhYQFzc/MS29jZ2QEAkpKS3mRpRERUCejruoCKIjMzEwBgampaahsTExO1tpoqKCiAEAIxMTHaF/gKZDIZsGQJUFCgk+VLmoEBcPEihBC6ruSVcBu/ALdx1aCj7VxQUPBs22iAgawhuVwOAC98YYs2dFk3eNE8Nd1or0XNmrpbdgWg021TXriNX4jbuGp409tZJpMxkMtb0WHqnJycUtvk5uYCAMzMzMo0b3d3d+0LIyKiSoHnkDVUdO44LS2t1FB+8OABgL/PJRMREWmKgawhmUwGNzc3AMCVK1eKjX/y5AmSkpJga2uLWrVqvenyiIiogmMgl0GvXr0AALt27So2bteuXRBCoHfv3m+4KiIiqgxkoqJfWvgG3blzB05OTpDJZPjzzz/RqVMnAMDVq1fxzjvvIDk5GRcuXEDz5s11XCkREVU0DOQyWrNmDcaPHw+5XI5u3brByMgIBw8eRG5uLhYuXAg/Pz9dl0hERBUQA1kL+/btw6JFixAdHQ25XA5XV1dMnToVgwcP1nVpRERUQTGQiYiIJIAXdREREUkAA5mIiEgCGMhEREQSwEAmIiKSAAYyERGRBDCQiYiIJICBTEREJAEMZHpjeMs7EVHpGMj02igUCrWfK8UD4KkYpVJZbBg/fFUu/9ye3L6vh76uC6DKSaFQQC6XAwBCQkJw584dNGnSBN26dYOhoaGOq6Py8vx2vn//PlJSUtCyZUt++KpEnt/GAJCfn8+/4deEe8hU7pRKpeoPeMSIEZg0aRJmzJiBjRs34sGDB6p2/JRdsT3/Ru3j44MePXrA3d0dS5YsQXZ2NgBu44ru+W0cEBCAAQMGoGvXrrh27Rq37WvAPWQqd3p6zz7nffbZZzh9+jRWr14NJycnmJqawsbGBk+fPoWBgQHMzMygVCpV7aliKXqjHj16NE6ePInJkyfD2toabdq0gampKYC/T1MIIbjXXAEVbeOhQ4ciKioKb731FlxcXJCWlsbt+RowkKncKZVK3L9/HydPnkRAQAD+9a9/QaFQ4NSpUxg8eDCSkpJgbGyM8PBw1K1bV9flUhk9H64HDhzAyZMnsW7dOnTr1g0AcOvWLUyaNAlKpRKtW7fGBx98AH19vtVUVEuXLkV0dDRCQkLg6uoKfX19pKWlYcuWLbC0tISzszOaNGmi6zIrBe6a0CspOmz19OlTZGRkAHi2h2xgYIDs7Gzcvn0b+/btw8cff4y3334bWVlZaNeuHTIyMjBlyhQUFBTosnzSwvN7Rk+ePIFCoYCbmxsuXLgAf39/NG/eHFu3bsWWLVvw1Vdf4fTp0zqsll7V/fv30bFjR7Rs2RIPHjzAunXr4OzsjA8//BADBw7E/Pnz8fTpU12XWSnwYytpLSsrCwsWLMCpU6dw5coV2NjYYNGiRejbty+qV68OJycnLF++HI8fP0adOnUwf/58TJs2DcbGxhg6dChycnJgYGCg69Wgl8jNzcUff/yBU6dOISUlBe+88w4GDRqE2rVrw83NDcnJyXj33XeRlJSEp0+f4tNPP4Wvry9q1KiBWrVqISoqCp07d9b1atALFBQU4MqVK4iJiYG1tTXs7e3h6uoK4Nn2j4iIwLx58xAeHo6zZ89i8ODB+PTTT3H06FF8++23mDVrFqysrHS7EpUAA5m0kpGRgU6dOsHU1BRubm545513cOPGDdy8eRMAYGhoiK1bt2LHjh0wNjaGk5MT3NzcAABpaWnIy8tD3bp1UVhYCLlczvNREpWRkYGePXsiMzMThYWFMDIywvXr19GwYUP07NkTzs7O+OOPP7Bs2TK899576NSpEwYMGIDCwkI8ePAAjo6OqF27tq5Xg14gMzMTY8aMwZUrV5CYmIiCggIMHToU8+fPh5OTE/z8/HDt2jV8//33eOutt7B69Wp8/PHHAAAjIyMEBQWVeOsblZ1M8FI5KqPs7Gx4enrCwsICK1euhKOjI4BnQVutWjUAwJ9//omGDRvCwcEBABAXF4f4+HjI5XL88ccf2LlzJ44fPw5nZ2edrQe9WGZmJjp27IjatWsjICAArVu3homJCU6dOoWOHTsCAJKSklC7dm3k5ubC2NgYGRkZsLCwwLVr17Bs2TLs2bMHx48fh729vY7XhkqSmZmJdu3aoXbt2vjoo4/QvXt3XL9+HXfv3sXo0aPV2t66dUttOz558gSff/45Ll26hAMHDsDa2vpNl1/5CKIy+uabb0SXLl3EhQsXVMMKCwtV3wcEBAiZTCbef/99ERsbKxQKhejdu7eQy+XCzs5OtGzZUpw/f14XpZOGCgsLxUcffSQ8PT3FjRs3Smwzf/58IZPJxKFDh4QQQiQkJAg7OzthZ2cnnJ2dRf369cVff/31BqumssjLyxO9evUSPXr0EImJiSW2mT9/vvj4449VP585c0a88847wsfHR7z33nuiRo0a/FsuRzxkTWUWERGhdo5JCKF2r+KSJUvg4+ODjRs3QqFQYPny5fj9999x4sQJmJubo0mTJqhVq5YuV4FeIjU1FWfPnsUHH3yABg0aAIDaLWrz58/H119/jTp16qBPnz4IDw/H22+/jWXLluHo0aNo0qQJ+vfvj8aNG+tyNegFTp8+jbt37+I///mPahsXFhaqroj/5ptvMG/ePADPTkGtWrUKwLPzzefOnUOTJk1w9OhRHuUqT7r+REAVh0KhEPfv3xfVq1cX69evF0IIUVBQoBq/cuVKoaenJ3777TchhBBbt24VcrlczJ07Vyf1kvb27dsnZDKZSEhIKDZu+vTpwtTUVPzxxx/i6NGjomPHjsLAwEBcvnz5zRdKWps/f76ws7Mrcdy8efOEqamp+Pnnn8W8efOETCYTU6ZMUY3Pzs4WeXl5b6jSqoO3PZFGFAoF9PT0YGVlBUtLSxw7dgxCCLX7SxUKBUJCQvD+++8DAHr37g1TU1Pk5OToqmzSUtERj+TkZAB/396Wm5uLAwcO4KeffsKQIUPQpUsXTJkyBYWFhTh//ryuyiUtpKenw8TEBIWFhWq3Hx4+fBjz58/HunXr4OPjgy+++AJDhw7F+vXrcfHiRQCAiYkJu898DXjIml6oKEyL/nCNjIzg4OCA48eP49GjR7C1tVUd5po8ebLatFu2bIG1tTVGjRoFgL01VSSNGzeGnp4eIiIi0KFDB8hkMigUChgbGyMqKkrtg9jx48fh6OiI/v3767BiKitLS0vcvXsXT58+hY2NjeqUhIeHBy5duqQ6FF2tWjUkJyejXbt2aN68uY6rrty4h0ylUigU6NevH+zt7ZGVlQV9fX3o6elh9uzZSExMxNy5cwEA+vr6EEKgsLBQNW1SUhL279+PVq1aoVGjRgD4tKeKxNraGj179sRPP/2EgwcPAni216xQKNTC+OrVq7h48SJ69uwJIyMjXZVLWnjvvfdgbGyMmTNnqsI4Pz8fAODi4qL6ez169ChSUlIwYsQIAOyf/HXiHjKVSgiBYcOGIS4uDh06dMDp06dhZmYGBwcHfPTRR/j555+hp6eH77//HiYmJqo36uvXr2Px4sU4cuQIjh07xg4DJCw7Oxtr165FdHQ0CgoK0Lp1a0yfPh1WVlb45JNPMGLECPj7+0Mmk8HT01PtqT/JyclYsmQJbt68iXXr1rGTF4nKz8/H1atXcenSJZiZmaFx48Zwc3ODq6srunXrhu3bt8PZ2RlffvklDA0N1Z7m9PjxY6xduxYGBgbo1asXAH6wfq10eQKbpC83N1f8+uuvwsbGRri4uIiMjAwhhBCxsbFi3LhxQi6Xi/bt24u5c+eK8PBw4evrK3r27CmaNGnC2yEkLj09Xbi7u4vmzZuLTp06icaNG4shQ4aIS5cuqdr897//FXK5XDRt2lSsXLlS3Lp1S6Snp4u1a9eKESNGiFq1anE7S1h6erro27evaNq0qZDJZEImk4nevXurLtZLTk4WTk5OwsLCQkyePFlkZ2erpo2IiBBjx44V1tbWarc40uvDjkFITUFBAeLj45GQkABLS0u4uLjA2toa27dvx4QJE2Bra4szZ87A3Nwct27dwtmzZ7Fo0SLcuHEDBQUFsLe3x4ABA/Dhhx/ylhcJy87ORseOHWFra4sffvgBLVq0QHZ2Nu7evavqzKXIli1bsGrVKpw4cQJGRkZQKpVo0KABWrVqhfnz5/O2F4nKzMxE+/btUbduXXzyySfw9PTEmTNnYGZmhq5du6qu6UhOTsbo0aNx4sQJmJiYwNnZWdU3taWlJVavXq3qZY9eLwYyqRR1oRcbG6vqVatZs2bw8vLCrFmzsG3bNkyaNAnW1taqUAae3bv45MkTZGZmonHjxnykYgXwzTffYPfu3di0aRMcHR3VttmDBw9w9uxZ5OTkoHfv3jAzM8PDhw9x//59XL16FUIIdO3aFZaWlqrfAZIWhUIBHx8fJCQkYM2aNWjUqFGxQ83P33OcmZmJY8eOYd++fcjIyICxsTEGDBiAli1bss+AN4iBTACe3QLRtm1b1KxZE97e3nBwcMDdu3exbNkynDt3DiNHjsSqVauwf/9+TJw4ETVr1lR92i56My/6xC14NbXkDRs2DHp6evj9999Vw3JzczF16lSEhYXh1q1bAIBmzZph+/btaNGiha5KJS08ffoU//rXvzBy5EhMmTJF7e/zn/bs2cMr5CWCF3UR0tPT0bJlSzRq1Ajr169Hw4YNVeO6du2Kn376CUuWLAEAbN68GTKZDBMnTkT79u0RFRUFMzMzKBQK1QU/DGPpy8/PR2pqKoBnF+4cOXIE/v7+uHjxIlxcXPDzzz+joKAAq1evxqeffooTJ07ouGIqi4sXL+Ls2bNYuXKl6sjHP/8ulUoltmzZAm9vb5w8eRIdOnRQG88P1m8eA7mKy8zMROvWreHi4oI1a9agTp06AJ79scpkMjRo0ABffvklZDIZFi9eDGdnZ/z73/9Gfn4+/Pz80KxZM8THx8PU1FTHa0KaEkKgW7dumD17Nlq2bImCggLExcWhcePGmD17Nr744gvUqFEDwLNziAEBAbh79y7q1aun48qpLORyOfLy8gCgxNNIenp6qF+/PgAgPj6+WCAzjN88nuirwoQQGD16NG7evImPP/5YLYz19PRUf5C1atXCxIkT0blzZ2zatAkpKSkYNmwY5s6dC0tLSzx48ECXq0EvUfSs27CwMBw/fhxpaWnw9fXFnDlzYGZmBiMjI/j6+iIsLAxz585FjRo1oFAoADy7n9zAwED1FC+qGBwdHWFtbY1Vq1ahoKBAdci6SNH3Xbp0gb6+vqpHNtKxN39hN0nJ2bNnhaurq2jfvr04fPiwarhSqSzWdvfu3UImk4n9+/cLIYTIz88XaWlpb6xWKruMjAwxYMAA0axZMyGTyYS+vr5wcnISixcvVrVJSkpSm6Zo2ycnJ4thw4aJDz74QOTm5r7Ruklz+fn5IjY2VoSGhooTJ06Ie/fuCSGE8PLyEoaGhmL9+vWqp7EpFAq1affu3Svs7OzEsWPH3njdVBwDmcSFCxeEo6OjaNeunepRekL8/cZc9Ed87do1YWxsLIKCgnRSJ5VNWlqaaNasmejcubMIDAwUkZGRYtOmTaJNmzZCJpOJQYMGqQXt7du3Vd/Hx8eLcePGiRo1aojY2FhdlE8aKOkDl7Ozs1i6dKnIysoSjo6Ook6dOmLdunUiPz9fbdqUlBTh7e0tOnbsKJKTk3W0BvQ8BjIJIUoP5ec/Ue/cuVM0a9ZMXLx4URclUhmkpaWJhg0bim7duhV7YtOtW7fEjBkzhFwuF4MGDRJCPHtSV926dcXHH38svL29RadOnUS9evX4PGMJe9kHrgkTJojExETh5OQkTE1NxUcffSRiY2PF7du3xR9//CHGjBkjrK2tRUxMjK5Xhf4fA5lUSgtlIYR49OiR8PLyEj169BBPnjzRUYWkiYyMDNGkSRPRu3dv1eFLIZ59uCo66pGUlCRmzpwpZDKZmDdvnrh69arw8PAQzZo1Ey1bthRTpkwR169f19Uq0Eu86ANXYmKimDFjhiqUs7KyxIgRI0SNGjWETCYThoaGokGDBuKdd95R65WNdI+BTGpKCuX09HQxbtw4YWNjw2feSpxSqRT9+/cXMplM7NixQzX8n+cOhXi2p9ylSxfh6OgoMjIyRH5+vnj69KnIy8srsT1JQ1k/cH3//fdCiGenJH777TexYcMGER0dLR4/fqyT+ql0DGQq5vlQDg8PF59//rkwMTER586d03VppIFXuVCPpE2bD1wODg7izp07b7JM0hJve6JiWrRogW3btiEnJwcDBgzA6tWrcezYMbi7u+u6NNJA69atsWXLFqSlpWHatGk4fPgwAKh6UQOe3doGAM7OzjAyMkJSUpLO6iXNyWQyzJs3Dy4uLli0aBGOHDkCAMVuawKABg0aYPr06bh+/TquXr2qi3KpjBjIVKIWLVpg48aNaN68OaKiotC6dWtdl0RlUPShKj09HdOnT1cL5ec7ibh8+TLq16/P7VuBaPOB6969ezqrlzTHQKZSubu74/jx43B1ddV1KaSF0kK5KIwfP36M4OBg2Nvbo27duroslcqIH7gqJwYyvZCRkZGuS6BXUNobd0ZGBvz8/BAeHo4ff/wR1atX13GlVFb8wFX58GlPRFVATEwM3n//fVhaWmL+/PnYv38/AgMDcfz4cV4bUME9v22XLFkCDw8PZGRk4Msvv8SuXbtw+PBhuLi46LpM0gADmaiKiImJwejRo3Ht2jXo6+vj6NGjPJRZSfADV+XAQCaqQv766y988sknCAoK4rUBlQw/cFV8DGSiKiYvL4/XBlRS/MBVsTGQiYgqEX7gqrgYyERERBLA256IiIgkgIFMREQkAQxkIiIiCWAgExERSQADmYiISAIYyERERBLAQCYiIpIABjIREZEEMJCJiIgkgIFMREQkAf8HJgOk6OPfkNwAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 721.347x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_cnv_hist(cnv_mean, binwidth=0.2)\n",
    "plot_percentage_bar(barplot_data)\n",
    "plot_stacked_bar(combined_data)\n",
    "generate_volcano_plot(plot_volcano, xlim = (-10,10), ylim = (0, 200))"
   ]
  }
 ],
 "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}