Pseudomonas/Pseudomonas_sample_lvl_sim.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pseudomonas sample level analysis\n",
    "\n",
    "Main notebook to run sample-level simulation experiment using *P. aeruginosa* gene expression data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "import os\n",
    "import sys\n",
    "import ast\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import random\n",
    "from plotnine import (ggplot,\n",
    "                      labs,  \n",
    "                      geom_line, \n",
    "                      geom_point,\n",
    "                      geom_errorbar,\n",
    "                      aes, \n",
    "                      ggsave, \n",
    "                      theme_bw,\n",
    "                      theme,\n",
    "                      facet_wrap,\n",
    "                      scale_color_manual,\n",
    "                      guides, \n",
    "                      guide_legend,\n",
    "                      element_blank,\n",
    "                      element_text,\n",
    "                      element_rect,\n",
    "                      element_line,\n",
    "                      coords)\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(action='ignore')\n",
    "\n",
    "from simulate_expression_compendia_modules import pipeline\n",
    "from ponyo import utils, train_vae_modules\n",
    "\n",
    "from numpy.random import seed\n",
    "randomState = 123\n",
    "seed(randomState)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Read in config variables\n",
    "base_dir = os.path.abspath(os.path.join(os.getcwd(),\"../\"))\n",
    "config_file = os.path.abspath(os.path.join(base_dir,\n",
    "                                           \"configs\", \n",
    "                                           \"config_Pa_sample_limma.tsv\"))\n",
    "params = utils.read_config(config_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load parameters\n",
    "local_dir = params[\"local_dir\"]\n",
    "dataset_name = params['dataset_name']\n",
    "analysis_name = params[\"simulation_type\"]\n",
    "correction_method = params[\"correction_method\"]\n",
    "lst_num_experiments = params[\"lst_num_experiments\"]\n",
    "train_architecture = params['NN_architecture']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Input files\n",
    "normalized_data_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"data\",\n",
    "    \"input\",\n",
    "    \"train_set_normalized.pcl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Output files\n",
    "normalized_processed_data_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"data\",\n",
    "    \"input\",\n",
    "    \"train_set_normalized_processed.txt.xz\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup directories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "utils.setup_dir(config_file)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Process data\n",
    "This pipeline is expecting data to be of the form sample x gene. The downloaded data is gene x sample."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipeline.transpose_data(normalized_data_file,\n",
    "                        normalized_processed_data_file)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train VAE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Directory containing log information from VAE training\n",
    "vae_log_dir = os.path.join(\n",
    "    base_dir, \n",
    "    dataset_name,\n",
    "    \"logs\",\n",
    "    train_architecture)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input dataset contains 950 samples and 5549 genes\n",
      "Train on 855 samples, validate on 95 samples\n",
      "Epoch 1/100\n",
      "855/855 [==============================] - 7s 8ms/step - loss: 3611.1525 - val_loss: 3816.1125\n",
      "Epoch 2/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3495.2920 - val_loss: 3551.1631\n",
      "Epoch 3/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3473.2418 - val_loss: 3482.1953\n",
      "Epoch 4/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3456.7678 - val_loss: 3456.3940\n",
      "Epoch 5/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3444.9030 - val_loss: 3459.7307\n",
      "Epoch 6/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3436.7637 - val_loss: 3416.9893\n",
      "Epoch 7/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3431.3799 - val_loss: 3434.4470\n",
      "Epoch 8/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3427.3698 - val_loss: 3411.0513\n",
      "Epoch 9/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3419.9824 - val_loss: 3410.5171\n",
      "Epoch 10/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3415.9700 - val_loss: 3404.3218\n",
      "Epoch 11/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3416.2802 - val_loss: 3403.4036\n",
      "Epoch 12/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3411.1866 - val_loss: 3401.8889\n",
      "Epoch 13/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3407.4503 - val_loss: 3393.6938\n",
      "Epoch 14/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3406.3521 - val_loss: 3394.1853\n",
      "Epoch 15/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3402.1354 - val_loss: 3398.5710\n",
      "Epoch 16/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3398.7678 - val_loss: 3387.9973\n",
      "Epoch 17/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3392.9006 - val_loss: 3390.2500\n",
      "Epoch 18/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3393.4382 - val_loss: 3394.5674\n",
      "Epoch 19/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3390.7812 - val_loss: 3390.8088\n",
      "Epoch 20/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3387.2417 - val_loss: 3386.5652\n",
      "Epoch 21/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3388.4475 - val_loss: 3382.1531\n",
      "Epoch 22/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3384.6557 - val_loss: 3392.9460\n",
      "Epoch 23/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3384.2207 - val_loss: 3371.8027\n",
      "Epoch 24/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3379.0954 - val_loss: 3374.3967\n",
      "Epoch 25/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3378.7492 - val_loss: 3382.1201\n",
      "Epoch 26/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3377.1324 - val_loss: 3371.3918\n",
      "Epoch 27/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3376.2099 - val_loss: 3366.2043\n",
      "Epoch 28/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3373.7145 - val_loss: 3373.8674\n",
      "Epoch 29/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3371.5058 - val_loss: 3363.1316\n",
      "Epoch 30/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3368.7515 - val_loss: 3365.1526\n",
      "Epoch 31/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3368.3223 - val_loss: 3358.7625\n",
      "Epoch 32/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3366.2654 - val_loss: 3359.6509\n",
      "Epoch 33/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3362.6108 - val_loss: 3359.0671\n",
      "Epoch 34/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3360.8647 - val_loss: 3356.6375\n",
      "Epoch 35/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3360.1895 - val_loss: 3358.7507\n",
      "Epoch 36/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3359.6459 - val_loss: 3352.0322\n",
      "Epoch 37/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3358.1640 - val_loss: 3358.1504\n",
      "Epoch 38/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3359.3125 - val_loss: 3361.4226\n",
      "Epoch 39/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3354.2152 - val_loss: 3361.4690\n",
      "Epoch 40/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3355.1488 - val_loss: 3352.0862\n",
      "Epoch 41/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3350.8853 - val_loss: 3352.0505\n",
      "Epoch 42/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3352.2767 - val_loss: 3354.8279\n",
      "Epoch 43/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3352.1681 - val_loss: 3348.0120\n",
      "Epoch 44/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3346.9439 - val_loss: 3345.2312\n",
      "Epoch 45/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3345.0559 - val_loss: 3343.7517\n",
      "Epoch 46/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3345.5782 - val_loss: 3341.2114\n",
      "Epoch 47/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3343.4898 - val_loss: 3339.5859\n",
      "Epoch 48/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3342.3118 - val_loss: 3340.1978\n",
      "Epoch 49/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3339.2598 - val_loss: 3335.8904\n",
      "Epoch 50/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3340.1402 - val_loss: 3339.4487\n",
      "Epoch 51/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3339.4786 - val_loss: 3334.0637\n",
      "Epoch 52/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3340.0821 - val_loss: 3335.1003\n",
      "Epoch 53/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3339.0124 - val_loss: 3335.1655\n",
      "Epoch 54/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3337.5843 - val_loss: 3332.3411\n",
      "Epoch 55/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3335.8808 - val_loss: 3329.6599\n",
      "Epoch 56/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3334.3341 - val_loss: 3328.8118\n",
      "Epoch 57/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3333.5022 - val_loss: 3326.2810\n",
      "Epoch 58/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3331.7643 - val_loss: 3321.8445\n",
      "Epoch 59/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3331.5002 - val_loss: 3328.5710\n",
      "Epoch 60/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3328.0937 - val_loss: 3322.7188\n",
      "Epoch 61/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3329.5366 - val_loss: 3319.4250\n",
      "Epoch 62/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3329.0135 - val_loss: 3322.3164\n",
      "Epoch 63/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3327.9080 - val_loss: 3319.1584\n",
      "Epoch 64/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3326.2012 - val_loss: 3323.1682\n",
      "Epoch 65/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3325.5615 - val_loss: 3317.9329\n",
      "Epoch 66/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3324.5384 - val_loss: 3318.2307\n",
      "Epoch 67/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3325.4764 - val_loss: 3315.2583\n",
      "Epoch 68/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3323.2634 - val_loss: 3316.4932\n",
      "Epoch 69/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3321.6376 - val_loss: 3316.7112\n",
      "Epoch 70/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3322.4927 - val_loss: 3316.6938\n",
      "Epoch 71/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3320.9720 - val_loss: 3318.5056\n",
      "Epoch 72/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3321.2227 - val_loss: 3314.7500\n",
      "Epoch 73/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3317.7966 - val_loss: 3313.7090\n",
      "Epoch 74/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3317.4775 - val_loss: 3315.4575\n",
      "Epoch 75/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3317.7560 - val_loss: 3316.9973\n",
      "Epoch 76/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "855/855 [==============================] - 5s 6ms/step - loss: 3317.6487 - val_loss: 3316.3596\n",
      "Epoch 77/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3317.1997 - val_loss: 3313.5674\n",
      "Epoch 78/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3314.9600 - val_loss: 3307.8259\n",
      "Epoch 79/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3314.9352 - val_loss: 3311.3474\n",
      "Epoch 80/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3314.3590 - val_loss: 3307.4714\n",
      "Epoch 81/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3312.7606 - val_loss: 3307.6812\n",
      "Epoch 82/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3312.9953 - val_loss: 3306.6299\n",
      "Epoch 83/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3312.4761 - val_loss: 3305.8276\n",
      "Epoch 84/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3312.7004 - val_loss: 3308.9504\n",
      "Epoch 85/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3311.1963 - val_loss: 3307.9629\n",
      "Epoch 86/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3310.8995 - val_loss: 3304.0220\n",
      "Epoch 87/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3310.9430 - val_loss: 3303.5293\n",
      "Epoch 88/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3308.6254 - val_loss: 3304.0569\n",
      "Epoch 89/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3309.3757 - val_loss: 3303.5723\n",
      "Epoch 90/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3308.6128 - val_loss: 3307.2759\n",
      "Epoch 91/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3307.1831 - val_loss: 3305.9128\n",
      "Epoch 92/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3306.1485 - val_loss: 3306.4150\n",
      "Epoch 93/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3305.2085 - val_loss: 3306.5774\n",
      "Epoch 94/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3305.6563 - val_loss: 3303.8933\n",
      "Epoch 95/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3305.7663 - val_loss: 3303.1541\n",
      "Epoch 96/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3305.7277 - val_loss: 3305.0710\n",
      "Epoch 97/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3303.9779 - val_loss: 3301.2756\n",
      "Epoch 98/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3303.6548 - val_loss: 3302.7622\n",
      "Epoch 99/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3302.6424 - val_loss: 3298.0505\n",
      "Epoch 100/100\n",
      "855/855 [==============================] - 5s 6ms/step - loss: 3301.7371 - val_loss: 3298.5310\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Train VAE\n",
    "if len(os.listdir(vae_log_dir)) == 0:\n",
    "    train_vae_modules.train_vae(config_file,\n",
    "                                normalized_processed_data_file)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run simulation experiment without noise correction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=5)]: Using backend LokyBackend with 5 concurrent workers.\n",
      "[Parallel(n_jobs=5)]: Done   1 tasks      | elapsed: 56.6min\n",
      "[Parallel(n_jobs=5)]: Done   2 out of   5 | elapsed: 56.7min remaining: 85.1min\n",
      "[Parallel(n_jobs=5)]: Done   3 out of   5 | elapsed: 56.8min remaining: 37.9min\n",
      "[Parallel(n_jobs=5)]: Done   5 out of   5 | elapsed: 57.0min remaining:    0.0s\n",
      "[Parallel(n_jobs=5)]: Done   5 out of   5 | elapsed: 57.0min finished\n",
      "                          score\n",
      "number of experiments          \n",
      "1                      0.999994\n",
      "2                      0.900765\n",
      "5                      0.607684\n",
      "10                     0.130417\n",
      "20                     0.042388\n",
      "50                     0.142795\n",
      "100                    0.237797\n",
      "500                    0.723494\n",
      "1000                   0.933498\n",
      "2000                   0.964786\n",
      "3000                   0.969494\n",
      "6000                   0.973936\n",
      "                              score\n",
      "number of experiments              \n",
      "1                      5.300363e-08\n",
      "2                      1.075690e-04\n",
      "5                      5.964708e-04\n",
      "10                     3.973912e-04\n",
      "20                     4.729727e-04\n",
      "50                     2.149832e-03\n",
      "100                    6.785386e-04\n",
      "500                    7.939839e-03\n",
      "1000                   1.320741e-03\n",
      "2000                   4.912390e-04\n",
      "3000                   8.520896e-04\n",
      "6000                   2.965269e-04\n",
      "                           ymin      ymax\n",
      "number of experiments                    \n",
      "1                      0.999994  0.999994\n",
      "2                      0.900555  0.900976\n",
      "5                      0.606515  0.608853\n",
      "10                     0.129638  0.131196\n",
      "20                     0.041461  0.043315\n",
      "50                     0.138581  0.147009\n",
      "100                    0.236467  0.239127\n",
      "500                    0.707932  0.739056\n",
      "1000                   0.930910  0.936087\n",
      "2000                   0.963823  0.965749\n",
      "3000                   0.967824  0.971164\n",
      "6000                   0.973354  0.974517\n"
     ]
    }
   ],
   "source": [
    "# Run simulation without correction \n",
    "pipeline.run_simulation(config_file,\n",
    "                        normalized_processed_data_file,\n",
    "                        corrected=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run simulation with correction applied"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=5)]: Using backend LokyBackend with 5 concurrent workers.\n",
      "[Parallel(n_jobs=5)]: Done   1 tasks      | elapsed: 119.7min\n",
      "[Parallel(n_jobs=5)]: Done   2 out of   5 | elapsed: 119.8min remaining: 179.6min\n",
      "[Parallel(n_jobs=5)]: Done   3 out of   5 | elapsed: 119.8min remaining: 79.9min\n",
      "[Parallel(n_jobs=5)]: Done   5 out of   5 | elapsed: 119.9min remaining:    0.0s\n",
      "[Parallel(n_jobs=5)]: Done   5 out of   5 | elapsed: 119.9min finished\n",
      "                          score\n",
      "number of experiments          \n",
      "1                      0.999994\n",
      "2                      0.999944\n",
      "5                      0.999698\n",
      "10                     0.999283\n",
      "20                     0.998476\n",
      "50                     0.995811\n",
      "100                    0.991450\n",
      "500                    0.956889\n",
      "1000                   0.913081\n",
      "2000                   0.814558\n",
      "3000                   0.690986\n",
      "6000                   0.011998\n",
      "                              score\n",
      "number of experiments              \n",
      "1                      4.885070e-08\n",
      "2                      9.122247e-06\n",
      "5                      2.951236e-05\n",
      "10                     1.881513e-05\n",
      "20                     1.909100e-05\n",
      "50                     9.588336e-05\n",
      "100                    1.194778e-04\n",
      "500                    4.641674e-04\n",
      "1000                   4.601200e-04\n",
      "2000                   4.821576e-04\n",
      "3000                   9.459170e-03\n",
      "6000                   3.245946e-03\n",
      "                           ymin      ymax\n",
      "number of experiments                    \n",
      "1                      0.999994  0.999994\n",
      "2                      0.999926  0.999962\n",
      "5                      0.999641  0.999756\n",
      "10                     0.999246  0.999320\n",
      "20                     0.998439  0.998514\n",
      "50                     0.995623  0.995999\n",
      "100                    0.991216  0.991684\n",
      "500                    0.955980  0.957799\n",
      "1000                   0.912179  0.913983\n",
      "2000                   0.813613  0.815503\n",
      "3000                   0.672446  0.709526\n",
      "6000                   0.005636  0.018360\n"
     ]
    }
   ],
   "source": [
    "# Run simulation without correction \n",
    "pipeline.run_simulation(config_file,\n",
    "                        normalized_processed_data_file,\n",
    "                        corrected=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Make figures"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "pca_ind = [0,1,2,-3,-2,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "# File directories\n",
    "similarity_uncorrected_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"results\",\n",
    "    \"saved_variables\",\n",
    "    dataset_name + \"_\" + analysis_name + \"_svcca_uncorrected_\" + correction_method + \".pickle\")\n",
    "\n",
    "ci_uncorrected_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"results\",\n",
    "    \"saved_variables\",\n",
    "    dataset_name + \"_\" + analysis_name + \"_ci_uncorrected_\" + correction_method + \".pickle\")\n",
    "\n",
    "similarity_corrected_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"results\",\n",
    "    \"saved_variables\",\n",
    "    dataset_name + \"_\" + analysis_name + \"_svcca_corrected_\" + correction_method + \".pickle\")\n",
    "\n",
    "ci_corrected_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"results\",\n",
    "    \"saved_variables\",\n",
    "    dataset_name + \"_\" + analysis_name + \"_ci_corrected_\" + correction_method + \".pickle\")\n",
    "\n",
    "permuted_score_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"results\",\n",
    "    \"saved_variables\",\n",
    "    dataset_name + \"_\" + analysis_name + \"_permuted.npy\")\n",
    "\n",
    "compendia_dir = os.path.join(\n",
    "    local_dir,\n",
    "    \"experiment_simulated\",\n",
    "    dataset_name + \"_\" + analysis_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Output files\n",
    "svcca_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"results\",\n",
    "    dataset_name +\"_\"+analysis_name+\"_svcca_\"+correction_method+\".svg\")\n",
    "\n",
    "svcca_png_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"results\",\n",
    "    dataset_name +\"_\"+analysis_name+\"_svcca_\"+correction_method+\".png\")\n",
    "\n",
    "pca_uncorrected_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"results\",\n",
    "    dataset_name +\"_\"+analysis_name+\"_pca_uncorrected_\"+correction_method+\".svg\")\n",
    "\n",
    "pca_corrected_file = os.path.join(\n",
    "    base_dir,\n",
    "    dataset_name,\n",
    "    \"results\",\n",
    "    dataset_name +\"_\"+analysis_name+\"_pca_corrected_\"+correction_method+\".svg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load pickled files\n",
    "uncorrected_svcca = pd.read_pickle(similarity_uncorrected_file)\n",
    "err_uncorrected_svcca = pd.read_pickle(ci_uncorrected_file)\n",
    "corrected_svcca = pd.read_pickle(similarity_corrected_file)\n",
    "err_corrected_svcca = pd.read_pickle(ci_corrected_file)\n",
    "\n",
    "permuted_score = np.load(permuted_score_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Concatenate error bars\n",
    "uncorrected_svcca_err = pd.concat([uncorrected_svcca, err_uncorrected_svcca], axis=1)\n",
    "corrected_svcca_err = pd.concat([corrected_svcca, err_corrected_svcca], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add group label\n",
    "uncorrected_svcca_err['Group'] = 'uncorrected'\n",
    "corrected_svcca_err['Group'] = 'corrected'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "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>score</th>\n",
       "      <th>ymin</th>\n",
       "      <th>ymax</th>\n",
       "      <th>Group</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number of experiments</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.999994</td>\n",
       "      <td>0.999994</td>\n",
       "      <td>0.999994</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.900765</td>\n",
       "      <td>0.900555</td>\n",
       "      <td>0.900976</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.607684</td>\n",
       "      <td>0.606515</td>\n",
       "      <td>0.608853</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.130417</td>\n",
       "      <td>0.129638</td>\n",
       "      <td>0.131196</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>0.042388</td>\n",
       "      <td>0.041461</td>\n",
       "      <td>0.043315</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>0.142795</td>\n",
       "      <td>0.138581</td>\n",
       "      <td>0.147009</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>0.237797</td>\n",
       "      <td>0.236467</td>\n",
       "      <td>0.239127</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>500</th>\n",
       "      <td>0.723494</td>\n",
       "      <td>0.707932</td>\n",
       "      <td>0.739056</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1000</th>\n",
       "      <td>0.933498</td>\n",
       "      <td>0.930910</td>\n",
       "      <td>0.936087</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>0.964786</td>\n",
       "      <td>0.963823</td>\n",
       "      <td>0.965749</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3000</th>\n",
       "      <td>0.969494</td>\n",
       "      <td>0.967824</td>\n",
       "      <td>0.971164</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6000</th>\n",
       "      <td>0.973936</td>\n",
       "      <td>0.973354</td>\n",
       "      <td>0.974517</td>\n",
       "      <td>uncorrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.999994</td>\n",
       "      <td>0.999994</td>\n",
       "      <td>0.999994</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.999944</td>\n",
       "      <td>0.999926</td>\n",
       "      <td>0.999962</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.999698</td>\n",
       "      <td>0.999641</td>\n",
       "      <td>0.999756</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.999283</td>\n",
       "      <td>0.999246</td>\n",
       "      <td>0.999320</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>0.998476</td>\n",
       "      <td>0.998439</td>\n",
       "      <td>0.998514</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>0.995811</td>\n",
       "      <td>0.995623</td>\n",
       "      <td>0.995999</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>0.991450</td>\n",
       "      <td>0.991216</td>\n",
       "      <td>0.991684</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>500</th>\n",
       "      <td>0.956889</td>\n",
       "      <td>0.955980</td>\n",
       "      <td>0.957799</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1000</th>\n",
       "      <td>0.913081</td>\n",
       "      <td>0.912179</td>\n",
       "      <td>0.913983</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>0.814558</td>\n",
       "      <td>0.813613</td>\n",
       "      <td>0.815503</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3000</th>\n",
       "      <td>0.690986</td>\n",
       "      <td>0.672446</td>\n",
       "      <td>0.709526</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6000</th>\n",
       "      <td>0.011998</td>\n",
       "      <td>0.005636</td>\n",
       "      <td>0.018360</td>\n",
       "      <td>corrected</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          score      ymin      ymax        Group\n",
       "number of experiments                                           \n",
       "1                      0.999994  0.999994  0.999994  uncorrected\n",
       "2                      0.900765  0.900555  0.900976  uncorrected\n",
       "5                      0.607684  0.606515  0.608853  uncorrected\n",
       "10                     0.130417  0.129638  0.131196  uncorrected\n",
       "20                     0.042388  0.041461  0.043315  uncorrected\n",
       "50                     0.142795  0.138581  0.147009  uncorrected\n",
       "100                    0.237797  0.236467  0.239127  uncorrected\n",
       "500                    0.723494  0.707932  0.739056  uncorrected\n",
       "1000                   0.933498  0.930910  0.936087  uncorrected\n",
       "2000                   0.964786  0.963823  0.965749  uncorrected\n",
       "3000                   0.969494  0.967824  0.971164  uncorrected\n",
       "6000                   0.973936  0.973354  0.974517  uncorrected\n",
       "1                      0.999994  0.999994  0.999994    corrected\n",
       "2                      0.999944  0.999926  0.999962    corrected\n",
       "5                      0.999698  0.999641  0.999756    corrected\n",
       "10                     0.999283  0.999246  0.999320    corrected\n",
       "20                     0.998476  0.998439  0.998514    corrected\n",
       "50                     0.995811  0.995623  0.995999    corrected\n",
       "100                    0.991450  0.991216  0.991684    corrected\n",
       "500                    0.956889  0.955980  0.957799    corrected\n",
       "1000                   0.913081  0.912179  0.913983    corrected\n",
       "2000                   0.814558  0.813613  0.815503    corrected\n",
       "3000                   0.690986  0.672446  0.709526    corrected\n",
       "6000                   0.011998  0.005636  0.018360    corrected"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Concatenate dataframes\n",
    "all_svcca = pd.concat([uncorrected_svcca_err, corrected_svcca_err])\n",
    "all_svcca"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SVCCA "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<ggplot: (-9223363244010200516)>\n"
     ]
    }
   ],
   "source": [
    "# Plot\n",
    "lst_num_partitions = list(all_svcca.index)\n",
    "\n",
    "threshold = pd.DataFrame(\n",
    "    pd.np.tile(\n",
    "        permuted_score,\n",
    "        (len(lst_num_partitions), 1)),\n",
    "    index=lst_num_partitions,\n",
    "    columns=['score'])\n",
    "\n",
    "panel_A = ggplot(all_svcca) \\\n",
    "    + geom_line(all_svcca,\n",
    "                aes(x=lst_num_partitions, y='score', color='Group'),\n",
    "                size=1.5) \\\n",
    "    + geom_point(aes(x=lst_num_partitions, y='score'), \n",
    "                 color ='darkgrey',\n",
    "                size=0.5) \\\n",
    "    + geom_errorbar(all_svcca,\n",
    "                  aes(x=lst_num_partitions, ymin='ymin', ymax='ymax'),\n",
    "                   color='darkgrey') \\\n",
    "    + geom_line(threshold, \n",
    "                aes(x=lst_num_partitions, y='score'), \n",
    "                linetype='dashed',\n",
    "                size=1,\n",
    "                color=\"darkgrey\",\n",
    "                show_legend=False) \\\n",
    "    + labs(x = \"Number of Partitions\", \n",
    "           y = \"Similarity score (SVCCA)\", \n",
    "           title = \"Similarity across varying numbers of partitions\") \\\n",
    "    + theme(\n",
    "            plot_background=element_rect(fill=\"white\"),\n",
    "            panel_background=element_rect(fill=\"white\"),\n",
    "            panel_grid_major_x=element_line(color=\"lightgrey\"),\n",
    "            panel_grid_major_y=element_line(color=\"lightgrey\"),\n",
    "            axis_line=element_line(color=\"grey\"),\n",
    "            legend_key=element_rect(fill='white', colour='white'),\n",
    "            legend_title=element_text(family='sans-serif', size=15),\n",
    "            legend_text=element_text(family='sans-serif', size=12),\n",
    "            plot_title=element_text(family='sans-serif', size=15),\n",
    "            axis_text=element_text(family='sans-serif', size=12),\n",
    "            axis_title=element_text(family='sans-serif', size=15)\n",
    "           ) \\\n",
    "    + scale_color_manual(['#1976d2', '#b3e5fc']) \\\n",
    "\n",
    "\n",
    "print(panel_A)\n",
    "ggsave(plot=panel_A, filename=svcca_file, device=\"svg\", dpi=300)\n",
    "ggsave(plot=panel_A, filename=svcca_png_file, device=\"svg\", dpi=300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Uncorrected PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plotting PCA of 1 experiment vs 1 experiments...\n",
      "[0.23503587 0.16136218]\n",
      "Plotting PCA of 1 experiment vs 2 experiments...\n",
      "[0.5229778  0.27045144]\n",
      "Plotting PCA of 1 experiment vs 5 experiments...\n",
      "[0.20313402 0.19038807]\n",
      "Plotting PCA of 1 experiment vs 2000 experiments...\n",
      "[0.03987891 0.02667372]\n",
      "Plotting PCA of 1 experiment vs 3000 experiments...\n",
      "[0.04240282 0.02977522]\n",
      "Plotting PCA of 1 experiment vs 6000 experiments...\n",
      "[0.04329961 0.03028643]\n"
     ]
    }
   ],
   "source": [
    "lst_num_experiments = [lst_num_experiments[i] for i in pca_ind]\n",
    "\n",
    "all_data_df = pd.DataFrame()\n",
    "\n",
    "# Get batch 1 data\n",
    "experiment_1_file = os.path.join(\n",
    "    compendia_dir,\n",
    "    \"Experiment_1_0.txt.xz\")\n",
    "\n",
    "experiment_1 = pd.read_table(\n",
    "    experiment_1_file,\n",
    "    header=0,\n",
    "    index_col=0,\n",
    "    sep='\\t')\n",
    "\n",
    "\n",
    "for i in lst_num_experiments:\n",
    "    print('Plotting PCA of 1 experiment vs {} experiments...'.format(i))\n",
    "    \n",
    "    # Simulated data with all samples in a single batch\n",
    "    original_data_df =  experiment_1.copy()\n",
    "    \n",
    "    # Add grouping column for plotting\n",
    "    original_data_df['num_experiments'] = '1'\n",
    "    \n",
    "    # downsample\n",
    "    original_data_df = original_data_df.sample(n=500)\n",
    "    \n",
    "    # Get data with additional batch effects added\n",
    "    experiment_other_file = os.path.join(\n",
    "        compendia_dir,\n",
    "        \"Experiment_\"+str(i)+\"_0.txt.xz\")\n",
    "\n",
    "    experiment_other = pd.read_table(\n",
    "        experiment_other_file,\n",
    "        header=0,\n",
    "        index_col=0,\n",
    "        sep='\\t')\n",
    "    \n",
    "    # Simulated data with i batch effects\n",
    "    experiment_data_df =  experiment_other\n",
    "    \n",
    "    # Add grouping column for plotting\n",
    "    experiment_data_df['num_experiments'] = 'multiple'\n",
    "    \n",
    "    # downsample\n",
    "    experiment_data_df = experiment_data_df.sample(n=500)\n",
    "    \n",
    "    # Concatenate datasets together\n",
    "    combined_data_df = pd.concat([original_data_df, experiment_data_df])\n",
    "\n",
    "    # PCA projection\n",
    "    pca = PCA(n_components=2)\n",
    "\n",
    "    # Encode expression data into 2D PCA space\n",
    "    combined_data_numeric_df = combined_data_df.drop(['num_experiments'], axis=1)\n",
    "    combined_data_PCAencoded = pca.fit_transform(combined_data_numeric_df)\n",
    "\n",
    "\n",
    "    combined_data_PCAencoded_df = pd.DataFrame(combined_data_PCAencoded,\n",
    "                                               index=combined_data_df.index,\n",
    "                                               columns=['PC1', 'PC2']\n",
    "                                              )\n",
    "                                              \n",
    "    # Variance explained\n",
    "    print(pca.explained_variance_ratio_)  \n",
    "    \n",
    "    # Add back in batch labels (i.e. labels = \"batch_\"<how many batch effects were added>)\n",
    "    combined_data_PCAencoded_df['num_experiments'] = combined_data_df['num_experiments']\n",
    "    \n",
    "    # Add column that designates which batch effect comparision (i.e. comparison of 1 batch vs 5 batches\n",
    "    # is represented by label = 5)\n",
    "    combined_data_PCAencoded_df['comparison'] = str(i)\n",
    "    \n",
    "    # Concatenate ALL comparisons\n",
    "    all_data_df = pd.concat([all_data_df, combined_data_PCAencoded_df])   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert 'num_experiments' into categories to preserve the ordering\n",
    "lst_num_experiments_str = [str(i) for i in lst_num_experiments]\n",
    "num_experiments_cat = pd.Categorical(all_data_df['num_experiments'], categories=['1', 'multiple'])\n",
    "\n",
    "# Convert 'comparison' into categories to preserve the ordering\n",
    "comparison_cat = pd.Categorical(all_data_df['comparison'], categories=lst_num_experiments_str)\n",
    "\n",
    "# Assign to a new column in the df\n",
    "all_data_df = all_data_df.assign(num_experiments_cat = num_experiments_cat)\n",
    "all_data_df = all_data_df.assign(comparison_cat = comparison_cat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_data_df.columns = ['PC1', 'PC2', 'num_experiments', 'comparison', 'No. of experiments', 'Comparison']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<ggplot: (-9223363244009375053)>\n"
     ]
    }
   ],
   "source": [
    "# Plot all comparisons in one figure\n",
    "panel_B = ggplot(all_data_df[all_data_df['Comparison'] != '1'],\n",
    "                 aes(x='PC1', y='PC2')) \\\n",
    "    + geom_point(aes(color='No. of experiments'), \n",
    "                 alpha=0.2) \\\n",
    "    + facet_wrap('~Comparison') \\\n",
    "    + labs(x = \"PC 1\", \n",
    "           y = \"PC 2\", \n",
    "           title = \"PCA of experiment 1 vs multiple experiments\") \\\n",
    "    + theme_bw() \\\n",
    "    + theme(\n",
    "        legend_title_align = \"center\",\n",
    "        plot_background=element_rect(fill='white'),\n",
    "        legend_key=element_rect(fill='white', colour='white'), \n",
    "        legend_title=element_text(family='sans-serif', size=15),\n",
    "        legend_text=element_text(family='sans-serif', size=12),\n",
    "        plot_title=element_text(family='sans-serif', size=15),\n",
    "        axis_text=element_text(family='sans-serif', size=12),\n",
    "        axis_title=element_text(family='sans-serif', size=15)\n",
    "    ) \\\n",
    "    + guides(colour=guide_legend(override_aes={'alpha': 1})) \\\n",
    "    + scale_color_manual(['#bdbdbd', '#b3e5fc']) \\\n",
    "    + geom_point(data=all_data_df[all_data_df['Comparison'] == '1'],\n",
    "                 alpha=0.1, \n",
    "                 color='#bdbdbd')\n",
    "\n",
    "print(panel_B)\n",
    "ggsave(plot=panel_B, filename=pca_uncorrected_file)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Corrected PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plotting PCA of 1 experiment vs 1 experiments...\n",
      "Plotting PCA of 1 experiment vs 2 experiments...\n",
      "Plotting PCA of 1 experiment vs 5 experiments...\n",
      "Plotting PCA of 1 experiment vs 2000 experiments...\n",
      "Plotting PCA of 1 experiment vs 3000 experiments...\n",
      "Plotting PCA of 1 experiment vs 6000 experiments...\n"
     ]
    }
   ],
   "source": [
    "lst_num_experiments = [lst_num_experiments[i] for i in pca_ind]\n",
    "\n",
    "all_corrected_data_df = pd.DataFrame()\n",
    "\n",
    "# Get batch 1 data\n",
    "experiment_1_file = os.path.join(\n",
    "    compendia_dir,\n",
    "    \"Experiment_corrected_1_0.txt.xz\")\n",
    "\n",
    "experiment_1 = pd.read_table(\n",
    "    experiment_1_file,\n",
    "    header=0,\n",
    "    index_col=0,\n",
    "    sep='\\t')\n",
    "\n",
    "# Transpose data to df: sample x gene\n",
    "experiment_1 = experiment_1.T\n",
    "\n",
    "for i in lst_num_experiments:\n",
    "    print('Plotting PCA of 1 experiment vs {} experiments...'.format(i))\n",
    "    \n",
    "     # Simulated data with all samples in a single batch\n",
    "    original_data_df =  experiment_1.copy()\n",
    "    \n",
    "    # Match format of column names in before and after df\n",
    "    original_data_df.columns = original_data_df.columns.astype(str)\n",
    "    \n",
    "    # Add grouping column for plotting\n",
    "    original_data_df['num_experiments'] = '1'\n",
    "    \n",
    "    # downsample\n",
    "    original_data_df = original_data_df.sample(n=500)\n",
    "    \n",
    "    # Get data with additional batch effects added and corrected\n",
    "    experiment_other_file = os.path.join(\n",
    "        compendia_dir,\n",
    "        \"Experiment_corrected_\"+str(i)+\"_0.txt.xz\")\n",
    "\n",
    "    experiment_other = pd.read_table(\n",
    "        experiment_other_file,\n",
    "        header=0,\n",
    "        index_col=0,\n",
    "        sep='\\t')\n",
    "    \n",
    "    # Transpose data to df: sample x gene\n",
    "    experiment_other = experiment_other.T\n",
    "    \n",
    "    # Simulated data with i batch effects that are corrected\n",
    "    experiment_data_df =  experiment_other\n",
    "    \n",
    "    # Match format of column names in before and after df\n",
    "    experiment_data_df.columns = experiment_data_df.columns.astype(str)\n",
    "    \n",
    "    # Add grouping column for plotting\n",
    "    experiment_data_df['num_experiments'] = 'multiple'\n",
    "    \n",
    "    # downsample\n",
    "    experiment_data_df = experiment_data_df.sample(n=500)\n",
    "        \n",
    "    # Concatenate datasets together\n",
    "    combined_data_df = pd.concat([original_data_df, experiment_data_df])\n",
    "    \n",
    "    # PCA projection\n",
    "    pca = PCA(n_components=2)\n",
    "\n",
    "    # Encode expression data into 2D PCA space    \n",
    "    combined_data_numeric_df = combined_data_df.drop(['num_experiments'], axis=1)    \n",
    "    combined_data_PCAencoded = pca.fit_transform(combined_data_numeric_df)\n",
    "\n",
    "    \n",
    "    combined_data_PCAencoded_df = pd.DataFrame(combined_data_PCAencoded,\n",
    "                                               index=combined_data_df.index,\n",
    "                                               columns=['PC1', 'PC2']\n",
    "                                              )\n",
    "    \n",
    "    # Add back in batch labels (i.e. labels = \"batch_\"<how many batch effects were added>)\n",
    "    combined_data_PCAencoded_df['num_experiments'] = combined_data_df['num_experiments']\n",
    "    \n",
    "    # Add column that designates which batch effect comparision (i.e. comparison of 1 batch vs 5 batches\n",
    "    # is represented by label = 5)\n",
    "    combined_data_PCAencoded_df['comparison'] = str(i)\n",
    "    \n",
    "    # Concatenate ALL comparisons\n",
    "    all_corrected_data_df = pd.concat([all_corrected_data_df, combined_data_PCAencoded_df])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert 'num_experiments' into categories to preserve the ordering\n",
    "lst_num_experiments_str = [str(i) for i in lst_num_experiments]\n",
    "num_experiments_cat = pd.Categorical(all_corrected_data_df['num_experiments'], categories=['1', 'multiple'])\n",
    "\n",
    "# Convert 'comparison' into categories to preserve the ordering\n",
    "comparison_cat = pd.Categorical(all_corrected_data_df['comparison'], categories=lst_num_experiments_str)\n",
    "\n",
    "# Assign to a new column in the df\n",
    "all_corrected_data_df = all_corrected_data_df.assign(num_experiments_cat = num_experiments_cat)\n",
    "all_corrected_data_df = all_corrected_data_df.assign(comparison_cat = comparison_cat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_corrected_data_df.columns = ['PC1', 'PC2', 'num_experiments', 'comparison', 'No. of experiments', 'Comparison']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<ggplot: (8792843720982)>\n"
     ]
    }
   ],
   "source": [
    "# Plot all comparisons in one figure\n",
    "panel_C = ggplot(all_corrected_data_df[all_corrected_data_df['Comparison'] != '1'],\n",
    "                 aes(x='PC1', y='PC2')) \\\n",
    "    + geom_point(aes(color='No. of experiments'), \n",
    "                 alpha=0.2) \\\n",
    "    + facet_wrap('~Comparison') \\\n",
    "    + labs(x = \"PC 1\",\n",
    "           y = \"PC 2\", \n",
    "           title = \"PCA of experiment 1 vs multiple experiments\") \\\n",
    "    + theme_bw() \\\n",
    "    + theme(\n",
    "        legend_title_align = \"center\",\n",
    "        plot_background=element_rect(fill='white'),\n",
    "        legend_key=element_rect(fill='white', colour='white'), \n",
    "        legend_title=element_text(family='sans-serif', size=15),\n",
    "        legend_text=element_text(family='sans-serif', size=12),\n",
    "        plot_title=element_text(family='sans-serif', size=15),\n",
    "        axis_text=element_text(family='sans-serif', size=12),\n",
    "        axis_title=element_text(family='sans-serif', size=15)\n",
    "    )\\\n",
    "    + guides(colour=guide_legend(override_aes={'alpha': 1})) \\\n",
    "    + scale_color_manual(['#bdbdbd', '#1976d2']) \\\n",
    "    + geom_point(data=all_corrected_data_df[all_corrected_data_df['Comparison'] == '1'],\n",
    "                 alpha=0.1, \n",
    "                 color='#bdbdbd')\n",
    "\n",
    "print(panel_C)\n",
    "ggsave(plot=panel_C, filename=pca_corrected_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:batch_effects]",
   "language": "python",
   "name": "conda-env-batch_effects-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}