examples/Example_simulate_by_random_sampling.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Template: simulating compendia by random sampling\n",
    "\n",
    "This template notebook simulates gene expression data by randomly sampling from the latent space distribution using the `simulate_by_random_sampling` function that is described [here](ponyo/simulate_expression_data.py)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/alexandra/Documents/Repos/ponyo/ponyo/helper_vae.py:21: The name tf.ConfigProto is deprecated. Please use tf.compat.v1.ConfigProto instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/alexandra/Documents/Repos/ponyo/ponyo/helper_vae.py:25: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/alexandra/Documents/Repos/ponyo/ponyo/helper_vae.py:25: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
      "\n"
     ]
    },
    {
     "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 pandas as pd\n",
    "import umap\n",
    "from plotnine import (ggplot,\n",
    "                      labs,  \n",
    "                      geom_point,\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",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(action='ignore')\n",
    "\n",
    "from ponyo import utils, train_vae_modules, simulate_expression_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/alexandra/Documents/Repos/ponyo/ponyo/train_vae_modules.py:56: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Set seeds to get reproducible VAE trained models\n",
    "train_vae_modules.set_all_seeds()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Read in config variables\n",
    "config_filename = \"config_simulate_by_random_sampling.tsv\"\n",
    "params = utils.read_config(config_filename)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load parameters\n",
    "local_dir = params[\"local_dir\"]\n",
    "raw_compendium_filename = params[\"raw_compendium_filename\"]\n",
    "normalized_compendium_filename = params[\"normalized_compendium_filename\"]\n",
    "num_simulated = params['num_simulated']\n",
    "training_stats_dir = params[\"training_stats_dir\"]\n",
    "vae_model_dir = params[\"vae_model_dir\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Input files\n",
    "assert os.path.exists(raw_compendium_filename)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup directories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "creating new directory: logs/\n",
      "creating new directory: models/\n",
      "creating new directory: ../example_results/\n"
     ]
    }
   ],
   "source": [
    "utils.setup_dir(config_filename)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check that directories were made correctly\n",
    "assert os.path.isdir(training_stats_dir)\n",
    "assert os.path.isdir(vae_model_dir)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pre-process data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input: dataset contains 273 samples and 5000 genes\n",
      "Output: normalized dataset contains 273 samples and 5000 genes\n"
     ]
    }
   ],
   "source": [
    "train_vae_modules.normalize_expression_data(config_filename,\n",
    "                                            raw_compendium_filename,\n",
    "                                            normalized_compendium_filename)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train VAE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input dataset contains 273 samples and 5000 genes\n",
      "WARNING:tensorflow:From /home/alexandra/anaconda3/envs/test_ponyo/lib/python3.7/site-packages/tensorflow_core/python/ops/resource_variable_ops.py:1630: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "If using Keras pass *_constraint arguments to layers.\n",
      "tracking <tf.Variable 'Variable:0' shape=() dtype=float32> beta\n",
      "WARNING:tensorflow:From /home/alexandra/anaconda3/envs/test_ponyo/lib/python3.7/site-packages/tensorflow_core/python/ops/nn_impl.py:183: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
      "WARNING:tensorflow:From /home/alexandra/anaconda3/envs/test_ponyo/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:422: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n",
      "\n",
      "Train on 246 samples, validate on 27 samples\n",
      "Epoch 1/10\n",
      "246/246 [==============================] - 21s 84ms/step - loss: 2340.0547 - kl_loss: 13.9695 - recons_loss: 2340.0549 - val_loss: 2576.0480 - val_kl_loss: 138.0642 - val_recons_loss: 2576.0481\n",
      "Epoch 2/10\n",
      "246/246 [==============================] - 20s 83ms/step - loss: 2167.1981 - kl_loss: 13.5955 - recons_loss: 2167.1982 - val_loss: 2336.0543 - val_kl_loss: 77.9998 - val_recons_loss: 2336.0542\n",
      "Epoch 3/10\n",
      "246/246 [==============================] - 21s 85ms/step - loss: 2154.3693 - kl_loss: 13.6479 - recons_loss: 2154.3691 - val_loss: 2209.4096 - val_kl_loss: 48.1001 - val_recons_loss: 2209.4097\n",
      "Epoch 4/10\n",
      "246/246 [==============================] - 20s 83ms/step - loss: 2145.3798 - kl_loss: 14.4935 - recons_loss: 2145.3799 - val_loss: 2148.8787 - val_kl_loss: 27.0799 - val_recons_loss: 2148.8787\n",
      "Epoch 5/10\n",
      "246/246 [==============================] - 21s 86ms/step - loss: 2140.5616 - kl_loss: 14.7198 - recons_loss: 2140.5618 - val_loss: 2125.0939 - val_kl_loss: 29.0089 - val_recons_loss: 2125.0940\n",
      "Epoch 6/10\n",
      "246/246 [==============================] - 21s 85ms/step - loss: 2135.2608 - kl_loss: 14.7896 - recons_loss: 2135.2607 - val_loss: 2106.7844 - val_kl_loss: 32.8426 - val_recons_loss: 2106.7847\n",
      "Epoch 7/10\n",
      "246/246 [==============================] - 21s 84ms/step - loss: 2123.8124 - kl_loss: 15.0116 - recons_loss: 2123.8123 - val_loss: 2084.1353 - val_kl_loss: 29.0683 - val_recons_loss: 2084.1355\n",
      "Epoch 8/10\n",
      "246/246 [==============================] - 21s 84ms/step - loss: 2123.9799 - kl_loss: 15.5945 - recons_loss: 2123.9795 - val_loss: 2089.8382 - val_kl_loss: 27.0921 - val_recons_loss: 2089.8381\n",
      "Epoch 9/10\n",
      "246/246 [==============================] - 20s 83ms/step - loss: 2090.6535 - kl_loss: 16.3353 - recons_loss: 2090.6541 - val_loss: 2126.9455 - val_kl_loss: 36.0171 - val_recons_loss: 2126.9456\n",
      "Epoch 10/10\n",
      "246/246 [==============================] - 20s 83ms/step - loss: 2101.7460 - kl_loss: 16.4985 - recons_loss: 2101.7461 - val_loss: 2079.0770 - val_kl_loss: 29.5822 - val_recons_loss: 2079.0769\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Train VAE\n",
    "train_vae_modules.train_vae(config_filename,\n",
    "                            normalized_compendium_filename)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check reproducibility of VAE training\n",
    "template_path = \"data/test_vae_log.tsv\"\n",
    "output_path = \"logs/tybalt_2layer_30latent_stats.tsv\"\n",
    "assert pd.read_csv(output_path, sep=\"\\t\")[\"val_loss\"].values[-1] < 2800, pd.read_csv(output_path, sep=\"\\t\")[\"val_loss\"].values[-1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Simulate data by random sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Normalized gene expression data contains 273 samples and 5000 genes\n",
      "Return: simulated gene expression data containing 100 samples and 5000 genes\n"
     ]
    }
   ],
   "source": [
    "# Run simulation\n",
    "simulated_data = simulate_expression_data.simulate_by_random_sampling(\n",
    "    normalized_compendium_filename,\n",
    "    num_simulated,\n",
    "    vae_model_dir,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>4990</th>\n",
       "      <th>4991</th>\n",
       "      <th>4992</th>\n",
       "      <th>4993</th>\n",
       "      <th>4994</th>\n",
       "      <th>4995</th>\n",
       "      <th>4996</th>\n",
       "      <th>4997</th>\n",
       "      <th>4998</th>\n",
       "      <th>4999</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.047630</td>\n",
       "      <td>0.036323</td>\n",
       "      <td>0.277738</td>\n",
       "      <td>0.119633</td>\n",
       "      <td>0.284286</td>\n",
       "      <td>0.346632</td>\n",
       "      <td>0.020977</td>\n",
       "      <td>0.096944</td>\n",
       "      <td>0.062515</td>\n",
       "      <td>0.418223</td>\n",
       "      <td>...</td>\n",
       "      <td>0.425015</td>\n",
       "      <td>0.361483</td>\n",
       "      <td>0.352412</td>\n",
       "      <td>0.020304</td>\n",
       "      <td>0.385193</td>\n",
       "      <td>0.157354</td>\n",
       "      <td>0.041844</td>\n",
       "      <td>0.049361</td>\n",
       "      <td>0.546658</td>\n",
       "      <td>0.421452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.105216</td>\n",
       "      <td>0.041911</td>\n",
       "      <td>0.326393</td>\n",
       "      <td>0.190275</td>\n",
       "      <td>0.330326</td>\n",
       "      <td>0.358407</td>\n",
       "      <td>0.049109</td>\n",
       "      <td>0.164491</td>\n",
       "      <td>0.118614</td>\n",
       "      <td>0.475839</td>\n",
       "      <td>...</td>\n",
       "      <td>0.401023</td>\n",
       "      <td>0.417316</td>\n",
       "      <td>0.386674</td>\n",
       "      <td>0.045546</td>\n",
       "      <td>0.385735</td>\n",
       "      <td>0.247026</td>\n",
       "      <td>0.101031</td>\n",
       "      <td>0.080914</td>\n",
       "      <td>0.595254</td>\n",
       "      <td>0.265012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.044584</td>\n",
       "      <td>0.052045</td>\n",
       "      <td>0.256216</td>\n",
       "      <td>0.125644</td>\n",
       "      <td>0.321328</td>\n",
       "      <td>0.339894</td>\n",
       "      <td>0.025672</td>\n",
       "      <td>0.089412</td>\n",
       "      <td>0.070641</td>\n",
       "      <td>0.389407</td>\n",
       "      <td>...</td>\n",
       "      <td>0.376761</td>\n",
       "      <td>0.370860</td>\n",
       "      <td>0.451509</td>\n",
       "      <td>0.024428</td>\n",
       "      <td>0.305814</td>\n",
       "      <td>0.152770</td>\n",
       "      <td>0.044391</td>\n",
       "      <td>0.052519</td>\n",
       "      <td>0.548577</td>\n",
       "      <td>0.277092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.213337</td>\n",
       "      <td>0.090136</td>\n",
       "      <td>0.331925</td>\n",
       "      <td>0.146207</td>\n",
       "      <td>0.348609</td>\n",
       "      <td>0.224303</td>\n",
       "      <td>0.078291</td>\n",
       "      <td>0.206755</td>\n",
       "      <td>0.176637</td>\n",
       "      <td>0.438614</td>\n",
       "      <td>...</td>\n",
       "      <td>0.411843</td>\n",
       "      <td>0.415836</td>\n",
       "      <td>0.401190</td>\n",
       "      <td>0.055254</td>\n",
       "      <td>0.388872</td>\n",
       "      <td>0.261685</td>\n",
       "      <td>0.179475</td>\n",
       "      <td>0.102899</td>\n",
       "      <td>0.560416</td>\n",
       "      <td>0.235998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.210298</td>\n",
       "      <td>0.179365</td>\n",
       "      <td>0.256450</td>\n",
       "      <td>0.219491</td>\n",
       "      <td>0.295060</td>\n",
       "      <td>0.208946</td>\n",
       "      <td>0.125192</td>\n",
       "      <td>0.166322</td>\n",
       "      <td>0.183197</td>\n",
       "      <td>0.408472</td>\n",
       "      <td>...</td>\n",
       "      <td>0.346551</td>\n",
       "      <td>0.352284</td>\n",
       "      <td>0.299130</td>\n",
       "      <td>0.072374</td>\n",
       "      <td>0.329897</td>\n",
       "      <td>0.240977</td>\n",
       "      <td>0.141765</td>\n",
       "      <td>0.070050</td>\n",
       "      <td>0.497483</td>\n",
       "      <td>0.220044</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 5000 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       0         1         2         3         4         5         6     \\\n",
       "0  0.047630  0.036323  0.277738  0.119633  0.284286  0.346632  0.020977   \n",
       "1  0.105216  0.041911  0.326393  0.190275  0.330326  0.358407  0.049109   \n",
       "2  0.044584  0.052045  0.256216  0.125644  0.321328  0.339894  0.025672   \n",
       "3  0.213337  0.090136  0.331925  0.146207  0.348609  0.224303  0.078291   \n",
       "4  0.210298  0.179365  0.256450  0.219491  0.295060  0.208946  0.125192   \n",
       "\n",
       "       7         8         9     ...      4990      4991      4992      4993  \\\n",
       "0  0.096944  0.062515  0.418223  ...  0.425015  0.361483  0.352412  0.020304   \n",
       "1  0.164491  0.118614  0.475839  ...  0.401023  0.417316  0.386674  0.045546   \n",
       "2  0.089412  0.070641  0.389407  ...  0.376761  0.370860  0.451509  0.024428   \n",
       "3  0.206755  0.176637  0.438614  ...  0.411843  0.415836  0.401190  0.055254   \n",
       "4  0.166322  0.183197  0.408472  ...  0.346551  0.352284  0.299130  0.072374   \n",
       "\n",
       "       4994      4995      4996      4997      4998      4999  \n",
       "0  0.385193  0.157354  0.041844  0.049361  0.546658  0.421452  \n",
       "1  0.385735  0.247026  0.101031  0.080914  0.595254  0.265012  \n",
       "2  0.305814  0.152770  0.044391  0.052519  0.548577  0.277092  \n",
       "3  0.388872  0.261685  0.179475  0.102899  0.560416  0.235998  \n",
       "4  0.329897  0.240977  0.141765  0.070050  0.497483  0.220044  \n",
       "\n",
       "[5 rows x 5000 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "simulated_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize random sampled compendium"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Read data\n",
    "normalized_data = pd.read_csv(normalized_compendium_filename, header=0, sep=\"\\t\", index_col=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>ENSG00000000003.14</th>\n",
       "      <th>ENSG00000000005.5</th>\n",
       "      <th>ENSG00000000419.12</th>\n",
       "      <th>ENSG00000000457.13</th>\n",
       "      <th>ENSG00000000460.16</th>\n",
       "      <th>ENSG00000000938.12</th>\n",
       "      <th>ENSG00000000971.15</th>\n",
       "      <th>ENSG00000001036.13</th>\n",
       "      <th>ENSG00000001084.10</th>\n",
       "      <th>ENSG00000001167.14</th>\n",
       "      <th>...</th>\n",
       "      <th>ENSG00000119953.12</th>\n",
       "      <th>ENSG00000119965.12</th>\n",
       "      <th>ENSG00000119969.14</th>\n",
       "      <th>ENSG00000119973.5</th>\n",
       "      <th>ENSG00000119977.20</th>\n",
       "      <th>ENSG00000119979.16</th>\n",
       "      <th>ENSG00000119986.6</th>\n",
       "      <th>ENSG00000120008.15</th>\n",
       "      <th>ENSG00000120029.12</th>\n",
       "      <th>ENSG00000120049.18</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>SRR493816</th>\n",
       "      <td>0.192519</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.214496</td>\n",
       "      <td>0.030213</td>\n",
       "      <td>0.185058</td>\n",
       "      <td>0.002518</td>\n",
       "      <td>0.037827</td>\n",
       "      <td>0.132201</td>\n",
       "      <td>0.344856</td>\n",
       "      <td>0.222818</td>\n",
       "      <td>...</td>\n",
       "      <td>0.098992</td>\n",
       "      <td>0.294837</td>\n",
       "      <td>0.061220</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.302188</td>\n",
       "      <td>0.119891</td>\n",
       "      <td>0.279795</td>\n",
       "      <td>0.024098</td>\n",
       "      <td>0.650383</td>\n",
       "      <td>0.022837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SRR493817</th>\n",
       "      <td>0.127613</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.191400</td>\n",
       "      <td>0.022873</td>\n",
       "      <td>0.148562</td>\n",
       "      <td>0.003878</td>\n",
       "      <td>0.001635</td>\n",
       "      <td>0.103004</td>\n",
       "      <td>0.406601</td>\n",
       "      <td>0.187205</td>\n",
       "      <td>...</td>\n",
       "      <td>0.132431</td>\n",
       "      <td>0.312500</td>\n",
       "      <td>0.089187</td>\n",
       "      <td>0.003215</td>\n",
       "      <td>0.275055</td>\n",
       "      <td>0.139419</td>\n",
       "      <td>0.919693</td>\n",
       "      <td>0.026674</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.039663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SRR547975</th>\n",
       "      <td>0.122846</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.483292</td>\n",
       "      <td>0.030554</td>\n",
       "      <td>0.324603</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.064989</td>\n",
       "      <td>0.391202</td>\n",
       "      <td>0.035818</td>\n",
       "      <td>0.801450</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.576087</td>\n",
       "      <td>0.506717</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.509190</td>\n",
       "      <td>0.282016</td>\n",
       "      <td>0.124300</td>\n",
       "      <td>0.018504</td>\n",
       "      <td>0.598180</td>\n",
       "      <td>0.126202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SRR547973</th>\n",
       "      <td>0.407407</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.316953</td>\n",
       "      <td>0.067338</td>\n",
       "      <td>0.308716</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.106436</td>\n",
       "      <td>0.353394</td>\n",
       "      <td>0.209581</td>\n",
       "      <td>...</td>\n",
       "      <td>0.201964</td>\n",
       "      <td>0.387228</td>\n",
       "      <td>0.085444</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.144639</td>\n",
       "      <td>0.187557</td>\n",
       "      <td>0.044633</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.430077</td>\n",
       "      <td>0.006010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SRR547968</th>\n",
       "      <td>0.078108</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.173464</td>\n",
       "      <td>0.038576</td>\n",
       "      <td>0.200086</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.463940</td>\n",
       "      <td>0.239900</td>\n",
       "      <td>0.570123</td>\n",
       "      <td>...</td>\n",
       "      <td>0.419321</td>\n",
       "      <td>0.585598</td>\n",
       "      <td>0.641268</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.341962</td>\n",
       "      <td>0.157735</td>\n",
       "      <td>0.023676</td>\n",
       "      <td>0.687739</td>\n",
       "      <td>0.080529</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 5000 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           ENSG00000000003.14  ENSG00000000005.5  ENSG00000000419.12  \\\n",
       "SRR493816            0.192519                0.0            0.214496   \n",
       "SRR493817            0.127613                0.0            0.191400   \n",
       "SRR547975            0.122846                0.0            0.483292   \n",
       "SRR547973            0.407407                0.0            0.316953   \n",
       "SRR547968            0.078108                0.0            0.173464   \n",
       "\n",
       "           ENSG00000000457.13  ENSG00000000460.16  ENSG00000000938.12  \\\n",
       "SRR493816            0.030213            0.185058            0.002518   \n",
       "SRR493817            0.022873            0.148562            0.003878   \n",
       "SRR547975            0.030554            0.324603            0.000000   \n",
       "SRR547973            0.067338            0.308716            0.000000   \n",
       "SRR547968            0.038576            0.200086            0.000000   \n",
       "\n",
       "           ENSG00000000971.15  ENSG00000001036.13  ENSG00000001084.10  \\\n",
       "SRR493816            0.037827            0.132201            0.344856   \n",
       "SRR493817            0.001635            0.103004            0.406601   \n",
       "SRR547975            0.064989            0.391202            0.035818   \n",
       "SRR547973            0.000000            0.106436            0.353394   \n",
       "SRR547968            0.000000            0.463940            0.239900   \n",
       "\n",
       "           ENSG00000001167.14  ...  ENSG00000119953.12  ENSG00000119965.12  \\\n",
       "SRR493816            0.222818  ...            0.098992            0.294837   \n",
       "SRR493817            0.187205  ...            0.132431            0.312500   \n",
       "SRR547975            0.801450  ...            1.000000            0.576087   \n",
       "SRR547973            0.209581  ...            0.201964            0.387228   \n",
       "SRR547968            0.570123  ...            0.419321            0.585598   \n",
       "\n",
       "           ENSG00000119969.14  ENSG00000119973.5  ENSG00000119977.20  \\\n",
       "SRR493816            0.061220           0.000000            0.302188   \n",
       "SRR493817            0.089187           0.003215            0.275055   \n",
       "SRR547975            0.506717           0.000000            0.509190   \n",
       "SRR547973            0.085444           0.000000            0.144639   \n",
       "SRR547968            0.641268           0.000000            1.000000   \n",
       "\n",
       "           ENSG00000119979.16  ENSG00000119986.6  ENSG00000120008.15  \\\n",
       "SRR493816            0.119891           0.279795            0.024098   \n",
       "SRR493817            0.139419           0.919693            0.026674   \n",
       "SRR547975            0.282016           0.124300            0.018504   \n",
       "SRR547973            0.187557           0.044633            1.000000   \n",
       "SRR547968            0.341962           0.157735            0.023676   \n",
       "\n",
       "           ENSG00000120029.12  ENSG00000120049.18  \n",
       "SRR493816            0.650383            0.022837  \n",
       "SRR493817            1.000000            0.039663  \n",
       "SRR547975            0.598180            0.126202  \n",
       "SRR547973            0.430077            0.006010  \n",
       "SRR547968            0.687739            0.080529  \n",
       "\n",
       "[5 rows x 5000 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "normalized_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = umap.UMAP(random_state=123).fit(normalized_data)\n",
    "\n",
    "input_data_UMAPencoded = model.transform(normalized_data)\n",
    "input_data_UMAPencoded_df = pd.DataFrame(data=input_data_UMAPencoded,\n",
    "                                         index=normalized_data.index,\n",
    "                                         columns=['1','2'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# UMAP embedding of simulated data\n",
    "\n",
    "simulated_data_UMAPencoded = model.transform(simulated_data)\n",
    "simulated_data_UMAPencoded_df = pd.DataFrame(data=simulated_data_UMAPencoded,\n",
    "                                         index=simulated_data.index,\n",
    "                                         columns=['1','2'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": false
   },
   "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: (8773546736685)>\n"
     ]
    }
   ],
   "source": [
    "# Overlay original input vs simulated data\n",
    "\n",
    "# Add label for input or simulated dataset\n",
    "input_data_UMAPencoded_df['dataset'] = 'original'\n",
    "simulated_data_UMAPencoded_df['dataset'] = 'simulated'\n",
    "\n",
    "# Concatenate input and simulated dataframes together\n",
    "combined_data_df = pd.concat([input_data_UMAPencoded_df, simulated_data_UMAPencoded_df])\n",
    "\n",
    "# Plot\n",
    "g_input_sim = ggplot(combined_data_df, aes(x='1', y='2'))\n",
    "g_input_sim += geom_point(aes(color='dataset'),alpha=0.3)\n",
    "g_input_sim += labs(x = \"UMAP 1\",\n",
    "                    y = \"UMAP 2\", \n",
    "                    title = \"Original and simulated data\")\n",
    "g_input_sim += theme_bw()\n",
    "g_input_sim += 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",
    ")\n",
    "\n",
    "print(g_input_sim)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}