4_Batch_score_calculation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f91a069f",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from sklearn.neighbors import NearestNeighbors\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import seaborn as sns\n",
    "\n",
    "from tqdm.auto import tqdm\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff131916",
   "metadata": {},
   "source": [
    "#### Paths of saved UMAP coordinates to calculate batch score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ffdba8dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "umap_csv_paths = sorted(list(Path('/path/to/umaps').glob('*umap*/*.csv')))\n",
    "print(umap_csv_paths)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f08bdc22",
   "metadata": {},
   "source": [
    "\n",
    "#### Count nearest neighbors\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7a06dafd",
   "metadata": {},
   "outputs": [],
   "source": [
    "def count_nearest_neighbor(df, batch_column='plate', k_nn=200):\n",
    "    X = df[['umap_x','umap_y']].to_numpy()\n",
    "    nbrs = NearestNeighbors(n_neighbors=k_nn, algorithm='ball_tree').fit(X)\n",
    "    _, indices = nbrs.kneighbors(X)\n",
    "\n",
    "    same_plate_nn_count = []\n",
    "    for idx in tqdm(range(df.shape[0])):\n",
    "        # convert indices to plate names\n",
    "        nearest_plates = df.loc[df.index[indices[idx]]][batch_column].to_numpy()\n",
    "        # first one is always self\n",
    "        selfplate = nearest_plates[0]\n",
    "        # others\n",
    "        nearest_plates = nearest_plates\n",
    "        # count how many cells from same plate\n",
    "        same_plate_nn_count.append((selfplate == nearest_plates).sum())\n",
    "        \n",
    "    df[f'same_plate_nn_{batch_column}'] = same_plate_nn_count\n",
    "    \n",
    "    return df\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9917500e",
   "metadata": {},
   "outputs": [],
   "source": [
    "k_nn = 200\n",
    "dfs = []\n",
    "for csv_path in umap_csv_paths:\n",
    "    df = pd.read_csv(csv_path, low_memory=False).reset_index(drop=True)\n",
    "    df['plate_well'] = df['plate'].astype('str') + '_' + df['well'].astype('str')\n",
    "    df['plate_well_fov'] = df['plate'].astype('str') + '_' + df['well'].astype('str') + '_' + df['fov'].astype('str')\n",
    "    # Calculate for different batch effects\n",
    "    df = count_nearest_neighbor(df, batch_column='plate')\n",
    "    df = count_nearest_neighbor(df, batch_column='plate_well')\n",
    "    df = count_nearest_neighbor(df, batch_column='plate_well_fov')\n",
    "    save_name = csv_path.parent / 'nn_count_1_200_platewell.csv'\n",
    "    # save\n",
    "    df.to_csv(save_name, index=False)\n",
    "    # add to list\n",
    "    dfs.append(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c452d2d8",
   "metadata": {},
   "source": [
    "#### Batch score calculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9b6f6c99",
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_column = 'plate'\n",
    "nn_count_column = f'same_plate_nn_{batch_column}'\n",
    "for df in dfs:\n",
    "    plate_percentage = {}\n",
    "    for grpn, grpdf in df.groupby(by=batch_column):\n",
    "        plate_percentage[grpn] = grpdf.shape[0] / df.shape[0]\n",
    "    df['plate_perc'] = df[batch_column].map(plate_percentage)\n",
    "       \n",
    "    df['same_plate_ratio'] = df[nn_count_column] / k_nn\n",
    "\n",
    "    df['same_plate_ratio'] = df['same_plate_ratio'] / df['plate_perc']\n",
    "    df['logp'] = np.log2(df['same_plate_ratio'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2170054e",
   "metadata": {},
   "source": [
    "Calculate mean batch score with confidence interval"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b3aedac2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy.stats as st \n",
    "def get_ci(df, col='same_plate_ratio', alpha=0.95):\n",
    "\n",
    "    meanv = df[col].mean()\n",
    "    stdv = df[col].std()\n",
    "\n",
    "    low, high = st.norm.interval(alpha=alpha,\n",
    "                     loc=df[col].mean(), \n",
    "                     scale=st.sem(df[col].to_numpy())) \n",
    "    return low, high, meanv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "92a43c6a",
   "metadata": {},
   "outputs": [],
   "source": [
    "stats = {}\n",
    "for df, csv_path in zip(dfs, umap_csv_paths):\n",
    "    low, high, meanv = get_ci(df, col='logp', alpha=0.95)\n",
    "    experiment_key = csv_path.parent.stem # Identifier between experiments\n",
    "    \n",
    "    stats[experiment_key] = [low, high, meanv] "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ab8d9742",
   "metadata": {},
   "source": [
    "Refactored dictionary {low_ci, high_ci, mean_score}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fce9b298",
   "metadata": {},
   "outputs": [],
   "source": [
    "stats = {\n",
    "    'No Correction' : [0.4000630751318601, 0.40246135396175764, 0.4012622145468089] ,\n",
    "    'LinearR, Plate' : [0.3683377601816099, 0.3705952979022169, 0.3694665290419134] ,\n",
    "    'LinearR, Plate+Well' : [0.36641501974092516, 0.3686653512623578, 0.36754018550164147] ,\n",
    "    'LinearR, Plate+Well+FOV' : [0.347603099393163, 0.3498209295966362, 0.3487120144948996] ,\n",
    "    'LinearR, Well' : [0.38915055758283285, 0.39145335820195387, 0.39030195789239336] ,\n",
    "    'LinearR, FOV' : [0.38218613291931197, 0.38449921063702613, 0.38334267177816905] ,\n",
    "    'LinearR, Well+FOV' : [0.375142960910258, 0.3774372131362851, 0.37629008702327155] ,\n",
    "    'PyComBat, Plate' : [0.20215757033009485, 0.2039106982775159, 0.20303413430380537] ,\n",
    "    'PyComBat, Plate+Well' : [0.16336055365998667, 0.16489708365335384, 0.16412881865667026] ,\n",
    "    'Harmony, Plate' : [0.15855122071425826, 0.16008917181439664, 0.15932019626432745] ,\n",
    "    'Harmony, Plate+Well' : [0.13770356569155595, 0.13912554180581116, 0.13841455374868356] ,\n",
    "    'No batch effect (shuffled data)' : [0.09625352242313197, 0.09745159170076954, 0.09685255706195076] ,\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5fe88e7c",
   "metadata": {},
   "source": [
    "Corrected percentage against 'No Correction'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b6ca1533",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.00 No Correction\n",
      "-7.92 LinearR, Plate\n",
      "-8.40 LinearR, Plate+Well\n",
      "-13.10 LinearR, Plate+Well+FOV\n",
      "-2.73 LinearR, Well\n",
      "-4.47 LinearR, FOV\n",
      "-6.22 LinearR, Well+FOV\n",
      "-49.40 PyComBat, Plate\n",
      "-59.10 PyComBat, Plate+Well\n",
      "-60.30 Harmony, Plate\n",
      "-65.51 Harmony, Plate+Well\n",
      "-75.86 No batch effect (shuffled data)\n"
     ]
    }
   ],
   "source": [
    "for key in stats1:\n",
    "    print(f\"{100*(stats1[key][2] - stats1['No Correction'][2])/stats1['No Correction'][2]:.2f}\", key)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d2b24aea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x864 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_style('whitegrid')\n",
    "\n",
    "def plot_confidence_intervaly(x, values, color='#f44336', horizontal_line_width=0.5):\n",
    "    top = values[0]\n",
    "    bottom = values[1]\n",
    "    meanv = values[2]\n",
    "\n",
    "    plt.plot(meanv, x, 'd', color='#2187bb', markersize=20)\n",
    "    # left = x - horizontal_line_width / 2\n",
    "    # right = x + horizontal_line_width / 2\n",
    "    # plt.plot([top, bottom], [x, x], color=color)\n",
    "    # plt.plot([top, top], [left, right], color=color)\n",
    "    # plt.plot([bottom, bottom], [left, right], color=color)\n",
    "\n",
    "xtick_pos = list(range(len(stats)))\n",
    "xtick_pos.reverse()\n",
    "xtick_names = list(stats.keys())\n",
    "\n",
    "plt.figure(figsize=(10,12))\n",
    "plt.yticks(xtick_pos, xtick_names, fontsize=20)\n",
    "plt.xticks(fontsize=20)\n",
    "plt.xlabel(\"Batch effect score\", fontsize=20)\n",
    "\n",
    "for xtick, key in zip(xtick_pos, stats):\n",
    "    plot_confidence_intervaly(xtick, stats[key])\n",
    "\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}