notebooks/GNPS_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "%pylab inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analysis of User-defined Metabolite Sets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook demonstrates how PALS can be used to analyse user-defined metabolite sets. Here PALS is used to analyse Molecular Families from GNPS, as well as Mass2Motifs from MS2LDA for differentially expressed metabolite sets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import sys\n",
    "import pathlib\n",
    "import pickle\n",
    "\n",
    "sys.path.append('..')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import zipfile\n",
    "import seaborn as sns\n",
    "import matplotlib.patches as mpatches\n",
    "from IPython.display import display, HTML\n",
    "from loguru import logger"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-01-07 17:14:29.113 | INFO     | pals.reactome:get_neo4j_driver:24 - Created graph database driver for bolt://localhost:7687 (neo4j)\n"
     ]
    }
   ],
   "source": [
    "from pals.feature_extraction import DataSource\n",
    "from pals.loader import GNPSLoader\n",
    "from pals.PLAGE import PLAGE\n",
    "from pals.common import *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GNPS Molecular Family Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load GNPS results using the loader"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The loader is used to retrieve molecular networking results from GNPS and extract the necessary data for PALS analysis. \n",
    "\n",
    "Input:\n",
    "- URL to FBMN GNPS results\n",
    "- A metadata CSV"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Provide the link to your FBMN GNPS experiment results below"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "database_name = DATABASE_GNPS_MOLECULAR_FAMILY\n",
    "gnps_url = 'https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=0a8432b5891a48d7ad8459ba4a89969f'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Provide a metadata CSV file describing the sample groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>sample</th>\n",
       "      <th>group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11337_P4_RE11_01_484.mzML</td>\n",
       "      <td>Less than 10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>13313_P4_RE2_01_474.mzML</td>\n",
       "      <td>Less than 10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>22510_P3_RG10_01_390.mzML</td>\n",
       "      <td>Less than 10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>27840_P1_RG3_01_118.mzML</td>\n",
       "      <td>Less than 10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>28742_P6_RH3_01_699.mzML</td>\n",
       "      <td>Less than 10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>41935_P1_RG11_01_126.mzML</td>\n",
       "      <td>More than 30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82</th>\n",
       "      <td>42739_P3_RH1_01_395.mzML</td>\n",
       "      <td>More than 30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>43160_P3_RE11_01_361.mzML</td>\n",
       "      <td>More than 30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84</th>\n",
       "      <td>49025_P6_RA4_01_676.mzML</td>\n",
       "      <td>More than 30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85</th>\n",
       "      <td>5990_P2_RB6_01_175.mzML</td>\n",
       "      <td>More than 30</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>86 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       sample         group\n",
       "0   11337_P4_RE11_01_484.mzML  Less than 10\n",
       "1    13313_P4_RE2_01_474.mzML  Less than 10\n",
       "2   22510_P3_RG10_01_390.mzML  Less than 10\n",
       "3    27840_P1_RG3_01_118.mzML  Less than 10\n",
       "4    28742_P6_RH3_01_699.mzML  Less than 10\n",
       "..                        ...           ...\n",
       "81  41935_P1_RG11_01_126.mzML  More than 30\n",
       "82   42739_P3_RH1_01_395.mzML  More than 30\n",
       "83  43160_P3_RE11_01_361.mzML  More than 30\n",
       "84   49025_P6_RA4_01_676.mzML  More than 30\n",
       "85    5990_P2_RB6_01_175.mzML  More than 30\n",
       "\n",
       "[86 rows x 2 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metadata_file = os.path.join('test_data', 'AGP', 'AG_Plants_extremes_metadata_df.csv')\n",
    "metadata_df = pd.read_csv(metadata_file)\n",
    "metadata_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Define case and control groups, and create a comparison."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'case': 'More than 30',\n",
       "  'control': 'Less than 10',\n",
       "  'name': 'more_plants/no_plants'}]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "case = 'More than 30'\n",
    "control = 'Less than 10'\n",
    "comp_name = 'more_plants/no_plants'\n",
    "comparisons = [{'case': case, 'control': control, 'name': comp_name },]\n",
    "comparisons"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Instantiate the loader object, and load the data into a `database` object."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-01-07 17:14:32.686 | INFO     | pals.loader:load_data:166 - Retrieving clustering and quantification information from GNPS\n",
      "2021-01-07 17:14:32.688 | DEBUG    | pals.loader:load_data:167 - https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=0a8432b5891a48d7ad8459ba4a89969f\n",
      "2021-01-07 17:14:32.688 | INFO     | pals.loader:_download_gnps:374 - Found GNPS task 0a8432b5891a48d7ad8459ba4a89969f\n",
      "39.8MiB [00:16, 2.87MiB/s]2021-01-07 17:14:50.007 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tDB_result/8d5e3e402de7496ab290b1ed0963bba8.tsv\n",
      "2021-01-07 17:14:50.009 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tFEATURE-BASED-MOLECULAR-NETWORKING-0a8432b5-download_cytoscape_data-main.graphml\n",
      "2021-01-07 17:14:50.011 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tclusterinfo_summary/cffab31fda574b1f961b8b336a3fc9bd.tsv\n",
      "2021-01-07 17:14:50.012 | DEBUG    | pals.loader:_parse_gnps_molfam:433 - Found cluster info: clusterinfo_summary/cffab31fda574b1f961b8b336a3fc9bd.tsv\n",
      "2021-01-07 17:14:50.013 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tgnps_molecular_network_graphml/54687b8ba4644b50bad04fd355a13254.graphml\n",
      "2021-01-07 17:14:50.014 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tnetworkedges_selfloop/e4fdc54c7eb644c7b4e7cef6f23077d7..selfloop\n",
      "2021-01-07 17:14:50.015 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tnetworking_pairs_results_file_filtered/79d9ba858c8b4f4e8f85939ad4c38b0c.tsv\n",
      "2021-01-07 17:14:50.016 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tparams.xml\n",
      "2021-01-07 17:14:50.017 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tqiime2_output/qiime2_emperor.qzv\n",
      "2021-01-07 17:14:50.018 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tqiime2_output/qiime2_manifest.tsv\n",
      "2021-01-07 17:14:50.018 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tqiime2_output/qiime2_metadata.tsv\n",
      "2021-01-07 17:14:50.019 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tqiime2_output/qiime2_table.qza\n",
      "2021-01-07 17:14:50.019 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tquantification_table/quantification_table-00000.csv\n",
      "2021-01-07 17:14:50.020 | DEBUG    | pals.loader:_parse_gnps_molfam:437 - Found quantification table: quantification_table/quantification_table-00000.csv\n",
      "2021-01-07 17:14:50.020 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tspectra/specs_ms.mgf\n",
      "2021-01-07 17:14:50.021 | DEBUG    | pals.loader:_parse_gnps_molfam:443 - Loading clusterinfo clusterinfo_summary/cffab31fda574b1f961b8b336a3fc9bd.tsv\n",
      "2021-01-07 17:14:50.109 | DEBUG    | pals.loader:_parse_gnps_molfam:446 - Loading quantification table quantification_table/quantification_table-00000.csv\n",
      "40.0MiB [00:16, 2.36MiB/s]\n"
     ]
    }
   ],
   "source": [
    "loader = GNPSLoader(database_name, gnps_url, metadata_df, comparisons)\n",
    "database = loader.load_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create data source. For the GNPS database, we get *measurement_df*, *annotation_df* and *experimental_design* parameters from *database* before passing them to the *DataSource* constructor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "measurement_df = database.extra_data['measurement_df']\n",
    "annotation_df = database.extra_data['annotation_df']\n",
    "experimental_design = database.extra_data['experimental_design']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>11337_P4_RE11_01_484.mzML</th>\n",
       "      <th>13313_P4_RE2_01_474.mzML</th>\n",
       "      <th>22510_P3_RG10_01_390.mzML</th>\n",
       "      <th>27840_P1_RG3_01_118.mzML</th>\n",
       "      <th>28742_P6_RH3_01_699.mzML</th>\n",
       "      <th>30483_P1_GD1_01_145.mzML</th>\n",
       "      <th>37543_P6_RG2_01_695.mzML</th>\n",
       "      <th>37937_P6_RD2_01_686.mzML</th>\n",
       "      <th>38057_P5_RE1_01_602.mzML</th>\n",
       "      <th>38093_P5_RE5_01_606.mzML</th>\n",
       "      <th>...</th>\n",
       "      <th>41323_P1_RG12_01_127.mzML</th>\n",
       "      <th>41341_P2_RE8_01_224.mzML</th>\n",
       "      <th>41534_P6_RB1_01_677.mzML</th>\n",
       "      <th>41698_P3_RC9_01_329.mzML</th>\n",
       "      <th>41744_P6_RE1_01_688.mzML</th>\n",
       "      <th>41935_P1_RG11_01_126.mzML</th>\n",
       "      <th>42739_P3_RH1_01_395.mzML</th>\n",
       "      <th>43160_P3_RE11_01_361.mzML</th>\n",
       "      <th>49025_P6_RA4_01_676.mzML</th>\n",
       "      <th>5990_P2_RB6_01_175.mzML</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>peak_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>16621.9880</td>\n",
       "      <td>7.116694e+04</td>\n",
       "      <td>6.055347e+05</td>\n",
       "      <td>1095.7485</td>\n",
       "      <td>1.140599e+06</td>\n",
       "      <td>14605.3060</td>\n",
       "      <td>4.579952e+04</td>\n",
       "      <td>1.098218e+06</td>\n",
       "      <td>9.364128e+03</td>\n",
       "      <td>1.653913e+04</td>\n",
       "      <td>...</td>\n",
       "      <td>1715.1885</td>\n",
       "      <td>1.936919e+06</td>\n",
       "      <td>1.551468e+05</td>\n",
       "      <td>8492.3235</td>\n",
       "      <td>2.391372e+05</td>\n",
       "      <td>94469.0770</td>\n",
       "      <td>1.985232e+06</td>\n",
       "      <td>3.165443e+05</td>\n",
       "      <td>291132.561</td>\n",
       "      <td>57110.3940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>226.5455</td>\n",
       "      <td>1.566927e+04</td>\n",
       "      <td>5.025617e+04</td>\n",
       "      <td>395148.9380</td>\n",
       "      <td>1.934524e+05</td>\n",
       "      <td>142826.9005</td>\n",
       "      <td>4.360585e+02</td>\n",
       "      <td>1.532759e+05</td>\n",
       "      <td>2.788818e+04</td>\n",
       "      <td>1.289885e+04</td>\n",
       "      <td>...</td>\n",
       "      <td>145.8310</td>\n",
       "      <td>4.268199e+05</td>\n",
       "      <td>6.492751e+03</td>\n",
       "      <td>7720.6630</td>\n",
       "      <td>1.263886e+04</td>\n",
       "      <td>480.2375</td>\n",
       "      <td>5.066707e+05</td>\n",
       "      <td>1.663545e+04</td>\n",
       "      <td>20276.257</td>\n",
       "      <td>168.3725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>992235.6470</td>\n",
       "      <td>1.293665e+06</td>\n",
       "      <td>1.181147e+06</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>9.435760e+05</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1.147643e+06</td>\n",
       "      <td>9.106171e+05</td>\n",
       "      <td>1.458229e+06</td>\n",
       "      <td>1.350834e+06</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.673888e+06</td>\n",
       "      <td>14355.9335</td>\n",
       "      <td>1.304436e+06</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1.380754e+06</td>\n",
       "      <td>1.227284e+06</td>\n",
       "      <td>1444913.288</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>931484.0325</td>\n",
       "      <td>1.214219e+06</td>\n",
       "      <td>9.357495e+05</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>7.189252e+05</td>\n",
       "      <td>145.0350</td>\n",
       "      <td>7.851636e+05</td>\n",
       "      <td>6.281846e+05</td>\n",
       "      <td>1.044920e+06</td>\n",
       "      <td>9.298224e+05</td>\n",
       "      <td>...</td>\n",
       "      <td>269.6995</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.086346e+06</td>\n",
       "      <td>7135.2980</td>\n",
       "      <td>9.033790e+05</td>\n",
       "      <td>694.5275</td>\n",
       "      <td>1.056104e+06</td>\n",
       "      <td>9.832828e+05</td>\n",
       "      <td>1043496.384</td>\n",
       "      <td>433.5000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>810685.3930</td>\n",
       "      <td>8.649341e+05</td>\n",
       "      <td>4.130054e+05</td>\n",
       "      <td>2470.5085</td>\n",
       "      <td>7.716864e+05</td>\n",
       "      <td>2717.1375</td>\n",
       "      <td>6.570987e+05</td>\n",
       "      <td>4.635221e+05</td>\n",
       "      <td>8.705470e+02</td>\n",
       "      <td>1.174299e+03</td>\n",
       "      <td>...</td>\n",
       "      <td>2350.6205</td>\n",
       "      <td>1.245302e+03</td>\n",
       "      <td>1.519240e+06</td>\n",
       "      <td>709596.0825</td>\n",
       "      <td>1.246027e+06</td>\n",
       "      <td>2860.2950</td>\n",
       "      <td>2.444535e+06</td>\n",
       "      <td>4.618985e+05</td>\n",
       "      <td>1872717.354</td>\n",
       "      <td>3621.1830</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 84 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         11337_P4_RE11_01_484.mzML  13313_P4_RE2_01_474.mzML  \\\n",
       "peak_id                                                        \n",
       "1                       16621.9880              7.116694e+04   \n",
       "2                         226.5455              1.566927e+04   \n",
       "3                      992235.6470              1.293665e+06   \n",
       "4                      931484.0325              1.214219e+06   \n",
       "5                      810685.3930              8.649341e+05   \n",
       "\n",
       "         22510_P3_RG10_01_390.mzML  27840_P1_RG3_01_118.mzML  \\\n",
       "peak_id                                                        \n",
       "1                     6.055347e+05                 1095.7485   \n",
       "2                     5.025617e+04               395148.9380   \n",
       "3                     1.181147e+06                    0.0000   \n",
       "4                     9.357495e+05                    0.0000   \n",
       "5                     4.130054e+05                 2470.5085   \n",
       "\n",
       "         28742_P6_RH3_01_699.mzML  30483_P1_GD1_01_145.mzML  \\\n",
       "peak_id                                                       \n",
       "1                    1.140599e+06                14605.3060   \n",
       "2                    1.934524e+05               142826.9005   \n",
       "3                    9.435760e+05                    0.0000   \n",
       "4                    7.189252e+05                  145.0350   \n",
       "5                    7.716864e+05                 2717.1375   \n",
       "\n",
       "         37543_P6_RG2_01_695.mzML  37937_P6_RD2_01_686.mzML  \\\n",
       "peak_id                                                       \n",
       "1                    4.579952e+04              1.098218e+06   \n",
       "2                    4.360585e+02              1.532759e+05   \n",
       "3                    1.147643e+06              9.106171e+05   \n",
       "4                    7.851636e+05              6.281846e+05   \n",
       "5                    6.570987e+05              4.635221e+05   \n",
       "\n",
       "         38057_P5_RE1_01_602.mzML  38093_P5_RE5_01_606.mzML  ...  \\\n",
       "peak_id                                                      ...   \n",
       "1                    9.364128e+03              1.653913e+04  ...   \n",
       "2                    2.788818e+04              1.289885e+04  ...   \n",
       "3                    1.458229e+06              1.350834e+06  ...   \n",
       "4                    1.044920e+06              9.298224e+05  ...   \n",
       "5                    8.705470e+02              1.174299e+03  ...   \n",
       "\n",
       "         41323_P1_RG12_01_127.mzML  41341_P2_RE8_01_224.mzML  \\\n",
       "peak_id                                                        \n",
       "1                        1715.1885              1.936919e+06   \n",
       "2                         145.8310              4.268199e+05   \n",
       "3                           0.0000              0.000000e+00   \n",
       "4                         269.6995              0.000000e+00   \n",
       "5                        2350.6205              1.245302e+03   \n",
       "\n",
       "         41534_P6_RB1_01_677.mzML  41698_P3_RC9_01_329.mzML  \\\n",
       "peak_id                                                       \n",
       "1                    1.551468e+05                 8492.3235   \n",
       "2                    6.492751e+03                 7720.6630   \n",
       "3                    1.673888e+06                14355.9335   \n",
       "4                    1.086346e+06                 7135.2980   \n",
       "5                    1.519240e+06               709596.0825   \n",
       "\n",
       "         41744_P6_RE1_01_688.mzML  41935_P1_RG11_01_126.mzML  \\\n",
       "peak_id                                                        \n",
       "1                    2.391372e+05                 94469.0770   \n",
       "2                    1.263886e+04                   480.2375   \n",
       "3                    1.304436e+06                     0.0000   \n",
       "4                    9.033790e+05                   694.5275   \n",
       "5                    1.246027e+06                  2860.2950   \n",
       "\n",
       "         42739_P3_RH1_01_395.mzML  43160_P3_RE11_01_361.mzML  \\\n",
       "peak_id                                                        \n",
       "1                    1.985232e+06               3.165443e+05   \n",
       "2                    5.066707e+05               1.663545e+04   \n",
       "3                    1.380754e+06               1.227284e+06   \n",
       "4                    1.056104e+06               9.832828e+05   \n",
       "5                    2.444535e+06               4.618985e+05   \n",
       "\n",
       "         49025_P6_RA4_01_676.mzML  5990_P2_RB6_01_175.mzML  \n",
       "peak_id                                                     \n",
       "1                      291132.561               57110.3940  \n",
       "2                       20276.257                 168.3725  \n",
       "3                     1444913.288                   0.0000  \n",
       "4                     1043496.384                 433.5000  \n",
       "5                     1872717.354                3621.1830  \n",
       "\n",
       "[5 rows x 84 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "measurement_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>entity_id</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>peak_id</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        entity_id\n",
       "peak_id          \n",
       "2               2\n",
       "4               4\n",
       "5               5\n",
       "6               6\n",
       "7               7"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "annotation_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create data source and run PLAGE analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create a PALS data source and run PLAGE analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-01-07 17:15:15.080 | DEBUG    | pals.feature_extraction:__init__:48 - Using user-provided database\n",
      "2021-01-07 17:15:15.080 | DEBUG    | pals.feature_extraction:__init__:56 - Mapping pathway to unique ids\n",
      "2021-01-07 17:15:15.085 | DEBUG    | pals.feature_extraction:__init__:70 - Creating dataset to pathway mapping\n",
      "2021-01-07 17:15:15.489 | DEBUG    | pals.feature_extraction:__init__:98 - Computing unique id counts\n"
     ]
    }
   ],
   "source": [
    "gnps_ds = DataSource(measurement_df, annotation_df, experimental_design, None, database=database, min_replace=SMALL)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-01-07 17:15:17.767 | DEBUG    | pals.PLAGE:__init__:29 - PLAGE initialised\n",
      "2021-01-07 17:15:17.954 | DEBUG    | pals.preprocessing:process:20 - Performing min-value imputation\n",
      "2021-01-07 17:15:17.988 | DEBUG    | pals.preprocessing:process:36 - Performing row average imputation\n",
      "2021-01-07 17:15:18.016 | DEBUG    | pals.preprocessing:process:46 - Applying log normalisation\n",
      "2021-01-07 17:15:18.024 | DEBUG    | pals.preprocessing:process:53 - Scaling the data across the sample: zero mean and unit variance\n",
      "2021-01-07 17:15:18.164 | DEBUG    | pals.PLAGE:get_plage_activity_df:84 - Mean values of the rows in the DF is [-0.  0. -0. ...  0.  0. -0.]\n",
      "2021-01-07 17:15:18.164 | DEBUG    | pals.PLAGE:get_plage_activity_df:85 - Variance in the rows of the DF is [1. 1. 1. ... 1. 1. 1.]\n",
      "2021-01-07 17:15:18.700 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:96 - Calculating plage p-values with resampling\n",
      "2021-01-07 17:15:18.701 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:103 - Comparison more_plants/no_plants\n",
      "2021-01-07 17:15:18.701 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 0/1000\n",
      "2021-01-07 17:15:18.821 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 100/1000\n",
      "2021-01-07 17:15:18.938 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 200/1000\n",
      "2021-01-07 17:15:19.058 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 300/1000\n",
      "2021-01-07 17:15:19.177 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 400/1000\n",
      "2021-01-07 17:15:19.302 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 500/1000\n",
      "2021-01-07 17:15:19.458 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 600/1000\n",
      "2021-01-07 17:15:19.583 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 700/1000\n",
      "2021-01-07 17:15:19.705 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 800/1000\n",
      "2021-01-07 17:15:19.830 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 900/1000\n",
      "2021-01-07 17:15:20.188 | DEBUG    | pals.feature_extraction:_calculate_coverage_df:329 - Calculating dataset formula coverage\n",
      "c:\\users\\joewa\\.virtualenvs\\pals-4ls44jcw\\lib\\site-packages\\scipy\\stats\\_discrete_distns.py:463: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  g1 = (m - n)*(M-2*N) / (M-2.0) * sqrt((M-1.0) / (m*n*N*(M-N)))\n",
      "c:\\users\\joewa\\.virtualenvs\\pals-4ls44jcw\\lib\\site-packages\\scipy\\stats\\_discrete_distns.py:468: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  g2 /= n * N * (M-N) * m * (M-2.) * (M-3.)\n"
     ]
    }
   ],
   "source": [
    "plage = PLAGE(gnps_ds)\n",
    "df = plage.get_results()"
   ]
  },
  {
   "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>pw_name</th>\n",
       "      <th>more_plants/no_plants p-value</th>\n",
       "      <th>unq_pw_F</th>\n",
       "      <th>tot_ds_F</th>\n",
       "      <th>F_coverage</th>\n",
       "      <th>sf</th>\n",
       "      <th>exp_F</th>\n",
       "      <th>Ex_Cov</th>\n",
       "      <th>more_plants/no_plants comb_p</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>65</th>\n",
       "      <td>Molecular Family #65</td>\n",
       "      <td>0.0</td>\n",
       "      <td>81</td>\n",
       "      <td>81</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>81.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Molecular Family #21</td>\n",
       "      <td>0.0</td>\n",
       "      <td>64</td>\n",
       "      <td>64</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>223</th>\n",
       "      <td>Molecular Family #223</td>\n",
       "      <td>0.0</td>\n",
       "      <td>46</td>\n",
       "      <td>46</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>590</th>\n",
       "      <td>Molecular Family #590</td>\n",
       "      <td>0.0</td>\n",
       "      <td>33</td>\n",
       "      <td>33</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>781</th>\n",
       "      <td>Molecular Family #781</td>\n",
       "      <td>0.0</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1319</th>\n",
       "      <td>Molecular Family #1319</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2776</th>\n",
       "      <td>Molecular Family #2776</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1063</th>\n",
       "      <td>Molecular Family #1063</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3804</th>\n",
       "      <td>Molecular Family #3804</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1668</th>\n",
       "      <td>Molecular Family #1668</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>983 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     pw_name  more_plants/no_plants p-value  unq_pw_F  \\\n",
       "65      Molecular Family #65                            0.0        81   \n",
       "21      Molecular Family #21                            0.0        64   \n",
       "223    Molecular Family #223                            0.0        46   \n",
       "590    Molecular Family #590                            0.0        33   \n",
       "781    Molecular Family #781                            0.0        27   \n",
       "...                      ...                            ...       ...   \n",
       "1319  Molecular Family #1319                            1.0         2   \n",
       "2776  Molecular Family #2776                            1.0         2   \n",
       "1063  Molecular Family #1063                            1.0         2   \n",
       "3804  Molecular Family #3804                            1.0         2   \n",
       "1668  Molecular Family #1668                            1.0         2   \n",
       "\n",
       "      tot_ds_F  F_coverage   sf  exp_F  Ex_Cov  more_plants/no_plants comb_p  \n",
       "65          81       100.0  1.0   81.0   100.0                           NaN  \n",
       "21          64       100.0  1.0   64.0   100.0                           NaN  \n",
       "223         46       100.0  1.0   46.0   100.0                           NaN  \n",
       "590         33       100.0  1.0   33.0   100.0                           NaN  \n",
       "781         27       100.0  1.0   27.0   100.0                           NaN  \n",
       "...        ...         ...  ...    ...     ...                           ...  \n",
       "1319         2       100.0  1.0    2.0   100.0                           NaN  \n",
       "2776         2       100.0  1.0    2.0   100.0                           NaN  \n",
       "1063         2       100.0  1.0    2.0   100.0                           NaN  \n",
       "3804         2       100.0  1.0    2.0   100.0                           NaN  \n",
       "1668         2       100.0  1.0    2.0   100.0                           NaN  \n",
       "\n",
       "[983 rows x 9 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p_value_col = '%s p-value' % comp_name\n",
    "count_col = 'unq_pw_F'\n",
    "df.sort_values([p_value_col, count_col], ascending=[True, False], inplace=True)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Checking results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.set()\n",
    "pd.set_option('display.max_colwidth', None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# cmap = 'RdBu_r'\n",
    "# cmap = 'jet'\n",
    "cmap = 'vlag'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Filter significant molecular families by p-value. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>pw_name</th>\n",
       "      <th>more_plants/no_plants p-value</th>\n",
       "      <th>unq_pw_F</th>\n",
       "      <th>tot_ds_F</th>\n",
       "      <th>F_coverage</th>\n",
       "      <th>sf</th>\n",
       "      <th>exp_F</th>\n",
       "      <th>Ex_Cov</th>\n",
       "      <th>more_plants/no_plants comb_p</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>65</th>\n",
       "      <td>Molecular Family #65</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>81</td>\n",
       "      <td>81</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>81.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Molecular Family #21</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>64</td>\n",
       "      <td>64</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>223</th>\n",
       "      <td>Molecular Family #223</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>46</td>\n",
       "      <td>46</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>590</th>\n",
       "      <td>Molecular Family #590</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>33</td>\n",
       "      <td>33</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>781</th>\n",
       "      <td>Molecular Family #781</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91</th>\n",
       "      <td>Molecular Family #91</td>\n",
       "      <td>0.030027</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116</th>\n",
       "      <td>Molecular Family #116</td>\n",
       "      <td>0.033380</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2784</th>\n",
       "      <td>Molecular Family #2784</td>\n",
       "      <td>0.034841</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2067</th>\n",
       "      <td>Molecular Family #2067</td>\n",
       "      <td>0.038444</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>539</th>\n",
       "      <td>Molecular Family #539</td>\n",
       "      <td>0.038952</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>165 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     pw_name  more_plants/no_plants p-value  unq_pw_F  \\\n",
       "65      Molecular Family #65                       0.000000        81   \n",
       "21      Molecular Family #21                       0.000000        64   \n",
       "223    Molecular Family #223                       0.000000        46   \n",
       "590    Molecular Family #590                       0.000000        33   \n",
       "781    Molecular Family #781                       0.000000        27   \n",
       "...                      ...                            ...       ...   \n",
       "91      Molecular Family #91                       0.030027         3   \n",
       "116    Molecular Family #116                       0.033380         9   \n",
       "2784  Molecular Family #2784                       0.034841         2   \n",
       "2067  Molecular Family #2067                       0.038444         2   \n",
       "539    Molecular Family #539                       0.038952         4   \n",
       "\n",
       "      tot_ds_F  F_coverage   sf  exp_F  Ex_Cov  more_plants/no_plants comb_p  \n",
       "65          81       100.0  1.0   81.0   100.0                           NaN  \n",
       "21          64       100.0  1.0   64.0   100.0                           NaN  \n",
       "223         46       100.0  1.0   46.0   100.0                           NaN  \n",
       "590         33       100.0  1.0   33.0   100.0                           NaN  \n",
       "781         27       100.0  1.0   27.0   100.0                           NaN  \n",
       "...        ...         ...  ...    ...     ...                           ...  \n",
       "91           3       100.0  1.0    3.0   100.0                           NaN  \n",
       "116          9       100.0  1.0    9.0   100.0                           NaN  \n",
       "2784         2       100.0  1.0    2.0   100.0                           NaN  \n",
       "2067         2       100.0  1.0    2.0   100.0                           NaN  \n",
       "539          4       100.0  1.0    4.0   100.0                           NaN  \n",
       "\n",
       "[165 rows x 9 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pval_threshold = 0.05\n",
    "df = df[df[p_value_col] < pval_threshold]\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Count how many significant molecular families having at least 10 members."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(34, 9)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "min_members = 10\n",
    "df[df[count_col] >= min_members].shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot significant molecular families"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_samples = []\n",
    "all_groups = []\n",
    "for group in experimental_design['groups']:\n",
    "    samples = experimental_design['groups'][group]\n",
    "    all_samples.extend(samples)\n",
    "    all_groups.extend([group] * len(samples))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-01-07 17:16:02.903 | DEBUG    | pals.feature_extraction:change_zero_peak_ints:307 - Setting the zero intensity values in the dataframe\n",
      "2021-01-07 17:16:02.909 | DEBUG    | pals.feature_extraction:change_zero_peak_ints:309 - 0\n",
      "2021-01-07 17:16:02.964 | DEBUG    | pals.feature_extraction:standardize_intensity_df:276 - Scaling the data across the sample: zero mean and unit variance\n"
     ]
    }
   ],
   "source": [
    "entity_dict = gnps_ds.entity_dict\n",
    "intensities_df = gnps_ds.standardize_intensity_df()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Molecular Family #65\n",
      "Molecular Family #21\n",
      "Molecular Family #223\n",
      "Molecular Family #590\n",
      "Molecular Family #781\n",
      "Molecular Family #148\n",
      "pw_name                          Molecular Family #148\n",
      "more_plants/no_plants p-value                        0\n",
      "unq_pw_F                                            23\n",
      "tot_ds_F                                            23\n",
      "F_coverage                                         100\n",
      "sf                                                   1\n",
      "exp_F                                               23\n",
      "Ex_Cov                                             100\n",
      "more_plants/no_plants comb_p                       NaN\n",
      "Name: 148, dtype: object\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>LibraryID</th>\n",
       "      <th>Precursor m/z</th>\n",
       "      <th>RTConsensus</th>\n",
       "      <th>PrecursorInt</th>\n",
       "      <th>no_spectra</th>\n",
       "      <th>link</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1187</th>\n",
       "      <td>NaN</td>\n",
       "      <td>357.2056</td>\n",
       "      <td>3.0498</td>\n",
       "      <td>0.037737</td>\n",
       "      <td>61</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1189</th>\n",
       "      <td>NaN</td>\n",
       "      <td>385.2352</td>\n",
       "      <td>3.4066</td>\n",
       "      <td>0.005774</td>\n",
       "      <td>85</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1197</th>\n",
       "      <td>NaN</td>\n",
       "      <td>343.2264</td>\n",
       "      <td>3.6391</td>\n",
       "      <td>0.018462</td>\n",
       "      <td>28</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1216</th>\n",
       "      <td>NaN</td>\n",
       "      <td>367.2269</td>\n",
       "      <td>3.3133</td>\n",
       "      <td>0.003620</td>\n",
       "      <td>86</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1217</th>\n",
       "      <td>NaN</td>\n",
       "      <td>339.1964</td>\n",
       "      <td>3.0423</td>\n",
       "      <td>0.003786</td>\n",
       "      <td>109</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3498</th>\n",
       "      <td>NaN</td>\n",
       "      <td>769.4667</td>\n",
       "      <td>3.3526</td>\n",
       "      <td>0.001045</td>\n",
       "      <td>174</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3678</th>\n",
       "      <td>NaN</td>\n",
       "      <td>399.3253</td>\n",
       "      <td>5.7595</td>\n",
       "      <td>0.000823</td>\n",
       "      <td>159</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7085</th>\n",
       "      <td>NaN</td>\n",
       "      <td>713.4052</td>\n",
       "      <td>3.0489</td>\n",
       "      <td>0.003077</td>\n",
       "      <td>176</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7091</th>\n",
       "      <td>Spectral Match to Mestranol from NIST14</td>\n",
       "      <td>311.2008</td>\n",
       "      <td>3.0540</td>\n",
       "      <td>0.001190</td>\n",
       "      <td>24</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7136</th>\n",
       "      <td>NaN</td>\n",
       "      <td>343.2255</td>\n",
       "      <td>3.2039</td>\n",
       "      <td>0.000161</td>\n",
       "      <td>25</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7892</th>\n",
       "      <td>NaN</td>\n",
       "      <td>385.2367</td>\n",
       "      <td>3.1743</td>\n",
       "      <td>0.013286</td>\n",
       "      <td>68</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8428</th>\n",
       "      <td>adrenosterone</td>\n",
       "      <td>301.1801</td>\n",
       "      <td>2.7956</td>\n",
       "      <td>0.003410</td>\n",
       "      <td>98</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8496</th>\n",
       "      <td>NaN</td>\n",
       "      <td>383.3314</td>\n",
       "      <td>8.3489</td>\n",
       "      <td>0.007551</td>\n",
       "      <td>196</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9249</th>\n",
       "      <td>NaN</td>\n",
       "      <td>343.2264</td>\n",
       "      <td>3.7812</td>\n",
       "      <td>0.003326</td>\n",
       "      <td>25</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9263</th>\n",
       "      <td>NaN</td>\n",
       "      <td>369.2410</td>\n",
       "      <td>4.2581</td>\n",
       "      <td>0.000547</td>\n",
       "      <td>56</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9612</th>\n",
       "      <td>NaN</td>\n",
       "      <td>371.2578</td>\n",
       "      <td>4.3831</td>\n",
       "      <td>0.008217</td>\n",
       "      <td>90</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10364</th>\n",
       "      <td>Spectral Match to Boldione from NIST14</td>\n",
       "      <td>285.1853</td>\n",
       "      <td>3.1896</td>\n",
       "      <td>0.000941</td>\n",
       "      <td>52</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10369</th>\n",
       "      <td>NaN</td>\n",
       "      <td>413.3041</td>\n",
       "      <td>5.2675</td>\n",
       "      <td>0.000591</td>\n",
       "      <td>16</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10375</th>\n",
       "      <td>NaN</td>\n",
       "      <td>313.2165</td>\n",
       "      <td>3.7372</td>\n",
       "      <td>0.000360</td>\n",
       "      <td>28</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10649</th>\n",
       "      <td>NaN</td>\n",
       "      <td>685.4481</td>\n",
       "      <td>3.7336</td>\n",
       "      <td>0.001139</td>\n",
       "      <td>124</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10650</th>\n",
       "      <td>NaN</td>\n",
       "      <td>369.2419</td>\n",
       "      <td>4.1228</td>\n",
       "      <td>0.001922</td>\n",
       "      <td>30</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10653</th>\n",
       "      <td>NaN</td>\n",
       "      <td>341.2097</td>\n",
       "      <td>3.6374</td>\n",
       "      <td>0.000229</td>\n",
       "      <td>19</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10654</th>\n",
       "      <td>NaN</td>\n",
       "      <td>325.2157</td>\n",
       "      <td>3.6832</td>\n",
       "      <td>0.000355</td>\n",
       "      <td>17</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=148&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     LibraryID  Precursor m/z  RTConsensus  \\\n",
       "id                                                                           \n",
       "1187                                       NaN       357.2056       3.0498   \n",
       "1189                                       NaN       385.2352       3.4066   \n",
       "1197                                       NaN       343.2264       3.6391   \n",
       "1216                                       NaN       367.2269       3.3133   \n",
       "1217                                       NaN       339.1964       3.0423   \n",
       "3498                                       NaN       769.4667       3.3526   \n",
       "3678                                       NaN       399.3253       5.7595   \n",
       "7085                                       NaN       713.4052       3.0489   \n",
       "7091   Spectral Match to Mestranol from NIST14       311.2008       3.0540   \n",
       "7136                                       NaN       343.2255       3.2039   \n",
       "7892                                       NaN       385.2367       3.1743   \n",
       "8428                             adrenosterone       301.1801       2.7956   \n",
       "8496                                       NaN       383.3314       8.3489   \n",
       "9249                                       NaN       343.2264       3.7812   \n",
       "9263                                       NaN       369.2410       4.2581   \n",
       "9612                                       NaN       371.2578       4.3831   \n",
       "10364   Spectral Match to Boldione from NIST14       285.1853       3.1896   \n",
       "10369                                      NaN       413.3041       5.2675   \n",
       "10375                                      NaN       313.2165       3.7372   \n",
       "10649                                      NaN       685.4481       3.7336   \n",
       "10650                                      NaN       369.2419       4.1228   \n",
       "10653                                      NaN       341.2097       3.6374   \n",
       "10654                                      NaN       325.2157       3.6832   \n",
       "\n",
       "       PrecursorInt  no_spectra  \\\n",
       "id                                \n",
       "1187       0.037737          61   \n",
       "1189       0.005774          85   \n",
       "1197       0.018462          28   \n",
       "1216       0.003620          86   \n",
       "1217       0.003786         109   \n",
       "3498       0.001045         174   \n",
       "3678       0.000823         159   \n",
       "7085       0.003077         176   \n",
       "7091       0.001190          24   \n",
       "7136       0.000161          25   \n",
       "7892       0.013286          68   \n",
       "8428       0.003410          98   \n",
       "8496       0.007551         196   \n",
       "9249       0.003326          25   \n",
       "9263       0.000547          56   \n",
       "9612       0.008217          90   \n",
       "10364      0.000941          52   \n",
       "10369      0.000591          16   \n",
       "10375      0.000360          28   \n",
       "10649      0.001139         124   \n",
       "10650      0.001922          30   \n",
       "10653      0.000229          19   \n",
       "10654      0.000355          17   \n",
       "\n",
       "                                                                                                                              link  \n",
       "id                                                                                                                                  \n",
       "1187   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "1189   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "1197   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "1216   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "1217   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "3498   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "3678   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "7085   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "7091   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "7136   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "7892   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "8428   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "8496   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "9249   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "9263   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "9612   https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "10364  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "10369  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "10375  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "10649  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "10650  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "10653  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "10654  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=148&task=0a8432b5891a48d7ad8459ba4a89969f  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Molecular Family #54\n",
      "Molecular Family #95\n",
      "Molecular Family #528\n",
      "Molecular Family #215\n",
      "Molecular Family #279\n",
      "Molecular Family #192\n",
      "Molecular Family #817\n",
      "Molecular Family #321\n",
      "Molecular Family #234\n",
      "Molecular Family #107\n",
      "Molecular Family #27\n",
      "Molecular Family #522\n",
      "Molecular Family #1443\n",
      "Molecular Family #1923\n",
      "Molecular Family #48\n",
      "Molecular Family #283\n",
      "Molecular Family #806\n",
      "Molecular Family #643\n",
      "Molecular Family #303\n",
      "Molecular Family #59\n",
      "Molecular Family #832\n",
      "Molecular Family #715\n",
      "Molecular Family #67\n",
      "Molecular Family #720\n",
      "Molecular Family #152\n",
      "Molecular Family #117\n",
      "Molecular Family #418\n",
      "Molecular Family #454\n"
     ]
    }
   ],
   "source": [
    "for idx, row in df.iterrows():\n",
    "    members = gnps_ds.dataset_pathways_to_row_ids[idx]\n",
    "    if len(members) < min_members:\n",
    "        continue\n",
    "        \n",
    "    pw_name = row['pw_name']\n",
    "    p_value = row[p_value_col]\n",
    "    print(pw_name)\n",
    "    \n",
    "    # plotting for manuscript\n",
    "    to_plot = ['Molecular Family #148']\n",
    "    if pw_name not in to_plot:\n",
    "        continue\n",
    "    else:\n",
    "        print(row)\n",
    "                \n",
    "    # get group intensities\n",
    "    group_intensities = intensities_df.loc[members][all_samples]    \n",
    "    \n",
    "    # get group info\n",
    "    # print('%s p-value=%.4f' % (pw_name, p_value))\n",
    "    data = []\n",
    "    for member in members:\n",
    "        member_info = entity_dict[member]\n",
    "        unique_id = member_info['unique_id']\n",
    "        library_id = member_info['LibraryID']\n",
    "        gnps_linkout_network = member_info['GNPSLinkout_Network']\n",
    "        no_spectra = member_info['number of spectra']\n",
    "        rt = member_info['RT']\n",
    "        mz = member_info['mass']    \n",
    "        intensity = member_info['SumPeakIntensity']\n",
    "        row = [unique_id, library_id, mz, rt, intensity, no_spectra, gnps_linkout_network]\n",
    "        data.append(row)\n",
    "    member_df = pd.DataFrame(data, columns=['id', 'LibraryID', 'Precursor m/z', 'RTConsensus', 'PrecursorInt', 'no_spectra', 'link']).set_index('id')    \n",
    "    \n",
    "    # Create a categorical palette to identify the networks\n",
    "    used_groups = list(set(all_groups))\n",
    "    group_pal = sns.husl_palette(len(used_groups), s=.45)\n",
    "    group_lut = dict(zip(map(str, used_groups), group_pal))\n",
    "\n",
    "    # Convert the palette to vectors that will be drawn on the side of the matrix\n",
    "    group_colours = pd.Series(all_groups, index=group_intensities.columns).map(group_lut)    \n",
    "    group_colours.name = 'groups'\n",
    "    \n",
    "    # plot heatmap\n",
    "    g = sns.clustermap(group_intensities, center=0, cmap=cmap, col_colors=group_colours, \n",
    "                   col_cluster=False, linewidths=0.75, figsize=(10, 10), cbar_pos=(1.0, 0.3, 0.05, 0.5))\n",
    "    \n",
    "    title = pw_name\n",
    "    if pw_name == 'Molecular Family #127':\n",
    "        title = 'Cinnamic Acid-related Molecular Family'\n",
    "    elif pw_name == 'Molecular Family #148':\n",
    "        title = 'Steroid-related Molecular Family'\n",
    "    plt.suptitle('%s' % (title), fontsize=24, y=0.9)\n",
    "\n",
    "    # draw group legend\n",
    "    for group in used_groups:\n",
    "        g.ax_col_dendrogram.bar(0, 0, color=group_lut[group], label=group, linewidth=0)\n",
    "    g.ax_col_dendrogram.legend(loc=\"right\")\n",
    "    \n",
    "    # make the annotated peaks to have labels in bold\n",
    "    annotated_df = member_df[member_df['LibraryID'].notnull()]\n",
    "    annotated_peaks = annotated_df.index.values\n",
    "    for label in g.ax_heatmap.get_yticklabels():\n",
    "        if label.get_text() in annotated_peaks:\n",
    "            label.set_weight(\"bold\")\n",
    "            label.set_color(\"green\")   \n",
    "    plt.setp(g.ax_heatmap.get_yticklabels(), rotation=0)        \n",
    "\n",
    "    out_file = '%s.pdf' % pw_name\n",
    "    plt.savefig(out_file, dpi=300)\n",
    "    plt.show()\n",
    "    \n",
    "    out_file = '%s.csv' % pw_name\n",
    "    display(member_df)\n",
    "    member_df.drop('link', axis=1).round(4).to_csv(out_file, index=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GNPS-MS2LDA Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similar to above, but here we provide a link to GNPS-MS2LDA analysis to retrieve Mass2Motifs, and pass it to the loader."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "gnps_ms2lda_url = 'https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=7c34badae00e43bc87b195a706cf1f43'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-01-07 17:16:11.708 | INFO     | pals.loader:load_data:166 - Retrieving clustering and quantification information from GNPS\n",
      "2021-01-07 17:16:11.709 | DEBUG    | pals.loader:load_data:167 - https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=0a8432b5891a48d7ad8459ba4a89969f\n",
      "2021-01-07 17:16:11.709 | INFO     | pals.loader:_download_gnps:374 - Found GNPS task 0a8432b5891a48d7ad8459ba4a89969f\n",
      "39.9MiB [00:17, 4.69MiB/s]2021-01-07 17:16:30.002 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tDB_result/8d5e3e402de7496ab290b1ed0963bba8.tsv\n",
      "2021-01-07 17:16:30.003 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tFEATURE-BASED-MOLECULAR-NETWORKING-0a8432b5-download_cytoscape_data-main.graphml\n",
      "2021-01-07 17:16:30.004 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tclusterinfo_summary/cffab31fda574b1f961b8b336a3fc9bd.tsv\n",
      "2021-01-07 17:16:30.004 | DEBUG    | pals.loader:_parse_gnps_molfam:433 - Found cluster info: clusterinfo_summary/cffab31fda574b1f961b8b336a3fc9bd.tsv\n",
      "2021-01-07 17:16:30.004 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tgnps_molecular_network_graphml/54687b8ba4644b50bad04fd355a13254.graphml\n",
      "2021-01-07 17:16:30.005 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tnetworkedges_selfloop/e4fdc54c7eb644c7b4e7cef6f23077d7..selfloop\n",
      "2021-01-07 17:16:30.006 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tnetworking_pairs_results_file_filtered/79d9ba858c8b4f4e8f85939ad4c38b0c.tsv\n",
      "2021-01-07 17:16:30.006 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tparams.xml\n",
      "2021-01-07 17:16:30.007 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tqiime2_output/qiime2_emperor.qzv\n",
      "2021-01-07 17:16:30.008 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tqiime2_output/qiime2_manifest.tsv\n",
      "2021-01-07 17:16:30.008 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tqiime2_output/qiime2_metadata.tsv\n",
      "2021-01-07 17:16:30.009 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tqiime2_output/qiime2_table.qza\n",
      "2021-01-07 17:16:30.009 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tquantification_table/quantification_table-00000.csv\n",
      "2021-01-07 17:16:30.010 | DEBUG    | pals.loader:_parse_gnps_molfam:437 - Found quantification table: quantification_table/quantification_table-00000.csv\n",
      "2021-01-07 17:16:30.010 | DEBUG    | pals.loader:_parse_gnps_molfam:429 - \tspectra/specs_ms.mgf\n",
      "2021-01-07 17:16:30.011 | DEBUG    | pals.loader:_parse_gnps_molfam:443 - Loading clusterinfo clusterinfo_summary/cffab31fda574b1f961b8b336a3fc9bd.tsv\n",
      "2021-01-07 17:16:30.093 | DEBUG    | pals.loader:_parse_gnps_molfam:446 - Loading quantification table quantification_table/quantification_table-00000.csv\n",
      "40.0MiB [00:18, 2.21MiB/s]\n",
      "2021-01-07 17:16:30.534 | INFO     | pals.loader:load_data:231 - Retrieving motif information from GNPS\n",
      "2021-01-07 17:16:30.535 | DEBUG    | pals.loader:load_data:232 - https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=7c34badae00e43bc87b195a706cf1f43\n",
      "2021-01-07 17:16:30.535 | INFO     | pals.loader:_download_gnps:374 - Found GNPS task 7c34badae00e43bc87b195a706cf1f43\n",
      "140kiB [00:01, 111kiB/s]  2021-01-07 17:16:32.495 | DEBUG    | pals.loader:_parse_ms2lda_motifs:455 - Found motif table: MS2LDA_MOTIFDB-7c34bada-view_all_motifs-main.tsv\n",
      "163kiB [00:01, 149kiB/s]\n"
     ]
    }
   ],
   "source": [
    "database_name = DATABASE_GNPS_MS2LDA\n",
    "loader = GNPSLoader(database_name, gnps_url, metadata_df, comparisons, gnps_ms2lda_url=gnps_ms2lda_url)\n",
    "database = loader.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-01-07 17:16:33.580 | DEBUG    | pals.feature_extraction:__init__:48 - Using user-provided database\n",
      "2021-01-07 17:16:33.581 | DEBUG    | pals.feature_extraction:__init__:56 - Mapping pathway to unique ids\n",
      "2021-01-07 17:16:33.584 | DEBUG    | pals.feature_extraction:__init__:70 - Creating dataset to pathway mapping\n",
      "2021-01-07 17:16:33.959 | DEBUG    | pals.feature_extraction:__init__:98 - Computing unique id counts\n"
     ]
    }
   ],
   "source": [
    "measurement_df = database.extra_data['measurement_df']\n",
    "annotation_df = database.extra_data['annotation_df']\n",
    "experimental_design = database.extra_data['experimental_design']\n",
    "gnps_ds = DataSource(measurement_df, annotation_df, experimental_design, None, database=database, min_replace=SMALL)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-01-07 17:16:34.050 | DEBUG    | pals.PLAGE:__init__:29 - PLAGE initialised\n",
      "2021-01-07 17:16:34.255 | DEBUG    | pals.preprocessing:process:20 - Performing min-value imputation\n",
      "2021-01-07 17:16:34.288 | DEBUG    | pals.preprocessing:process:36 - Performing row average imputation\n",
      "2021-01-07 17:16:34.313 | DEBUG    | pals.preprocessing:process:46 - Applying log normalisation\n",
      "2021-01-07 17:16:34.320 | DEBUG    | pals.preprocessing:process:53 - Scaling the data across the sample: zero mean and unit variance\n",
      "2021-01-07 17:16:34.429 | DEBUG    | pals.PLAGE:get_plage_activity_df:84 - Mean values of the rows in the DF is [-0.  0. -0. ...  0.  0. -0.]\n",
      "2021-01-07 17:16:34.429 | DEBUG    | pals.PLAGE:get_plage_activity_df:85 - Variance in the rows of the DF is [1. 1. 1. ... 1. 1. 1.]\n",
      "2021-01-07 17:16:34.785 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:96 - Calculating plage p-values with resampling\n",
      "2021-01-07 17:16:34.786 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:103 - Comparison more_plants/no_plants\n",
      "2021-01-07 17:16:34.787 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 0/1000\n",
      "2021-01-07 17:16:34.878 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 100/1000\n",
      "2021-01-07 17:16:34.968 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 200/1000\n",
      "2021-01-07 17:16:35.055 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 300/1000\n",
      "2021-01-07 17:16:35.142 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 400/1000\n",
      "2021-01-07 17:16:35.233 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 500/1000\n",
      "2021-01-07 17:16:35.321 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 600/1000\n",
      "2021-01-07 17:16:35.406 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 700/1000\n",
      "2021-01-07 17:16:35.495 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 800/1000\n",
      "2021-01-07 17:16:35.588 | DEBUG    | pals.PLAGE:set_up_resample_plage_p_df:111 - Resampling 900/1000\n",
      "2021-01-07 17:16:35.770 | DEBUG    | pals.feature_extraction:_calculate_coverage_df:329 - Calculating dataset formula coverage\n"
     ]
    }
   ],
   "source": [
    "plage = PLAGE(gnps_ds)\n",
    "df = plage.get_results()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>pw_name</th>\n",
       "      <th>more_plants/no_plants p-value</th>\n",
       "      <th>unq_pw_F</th>\n",
       "      <th>tot_ds_F</th>\n",
       "      <th>F_coverage</th>\n",
       "      <th>sf</th>\n",
       "      <th>exp_F</th>\n",
       "      <th>Ex_Cov</th>\n",
       "      <th>more_plants/no_plants comb_p</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>motif_122</th>\n",
       "      <td>motif_122</td>\n",
       "      <td>0.0</td>\n",
       "      <td>91</td>\n",
       "      <td>77</td>\n",
       "      <td>84.62</td>\n",
       "      <td>0.027142</td>\n",
       "      <td>68.97</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_188</th>\n",
       "      <td>motif_188</td>\n",
       "      <td>0.0</td>\n",
       "      <td>78</td>\n",
       "      <td>63</td>\n",
       "      <td>80.77</td>\n",
       "      <td>0.184479</td>\n",
       "      <td>59.12</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_112</th>\n",
       "      <td>motif_112</td>\n",
       "      <td>0.0</td>\n",
       "      <td>67</td>\n",
       "      <td>51</td>\n",
       "      <td>76.12</td>\n",
       "      <td>0.541361</td>\n",
       "      <td>50.78</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_149</th>\n",
       "      <td>motif_149</td>\n",
       "      <td>0.0</td>\n",
       "      <td>61</td>\n",
       "      <td>54</td>\n",
       "      <td>88.52</td>\n",
       "      <td>0.010195</td>\n",
       "      <td>46.23</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_185</th>\n",
       "      <td>motif_185</td>\n",
       "      <td>0.0</td>\n",
       "      <td>61</td>\n",
       "      <td>47</td>\n",
       "      <td>77.05</td>\n",
       "      <td>0.477871</td>\n",
       "      <td>46.23</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_193</th>\n",
       "      <td>motif_193</td>\n",
       "      <td>1.0</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>100.00</td>\n",
       "      <td>0.020497</td>\n",
       "      <td>10.61</td>\n",
       "      <td>75.79</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_205</th>\n",
       "      <td>motif_205</td>\n",
       "      <td>1.0</td>\n",
       "      <td>13</td>\n",
       "      <td>11</td>\n",
       "      <td>84.62</td>\n",
       "      <td>0.356600</td>\n",
       "      <td>9.85</td>\n",
       "      <td>75.77</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_172</th>\n",
       "      <td>motif_172</td>\n",
       "      <td>1.0</td>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "      <td>77.78</td>\n",
       "      <td>0.622791</td>\n",
       "      <td>6.82</td>\n",
       "      <td>75.78</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_154</th>\n",
       "      <td>motif_154</td>\n",
       "      <td>1.0</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>100.00</td>\n",
       "      <td>0.249871</td>\n",
       "      <td>3.79</td>\n",
       "      <td>75.80</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gnps_motif_19.m2m</th>\n",
       "      <td>gnps_motif_19.m2m [(5-Hydroxy-2 2-dimethyl-4-oxo-3 4-dihydro-2H-chromen-7-yl)oxy substructure]</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>100.00</td>\n",
       "      <td>0.574351</td>\n",
       "      <td>1.52</td>\n",
       "      <td>76.00</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>102 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                          pw_name  \\\n",
       "motif_122                                                                                               motif_122   \n",
       "motif_188                                                                                               motif_188   \n",
       "motif_112                                                                                               motif_112   \n",
       "motif_149                                                                                               motif_149   \n",
       "motif_185                                                                                               motif_185   \n",
       "...                                                                                                           ...   \n",
       "motif_193                                                                                               motif_193   \n",
       "motif_205                                                                                               motif_205   \n",
       "motif_172                                                                                               motif_172   \n",
       "motif_154                                                                                               motif_154   \n",
       "gnps_motif_19.m2m  gnps_motif_19.m2m [(5-Hydroxy-2 2-dimethyl-4-oxo-3 4-dihydro-2H-chromen-7-yl)oxy substructure]   \n",
       "\n",
       "                   more_plants/no_plants p-value  unq_pw_F  tot_ds_F  \\\n",
       "motif_122                                    0.0        91        77   \n",
       "motif_188                                    0.0        78        63   \n",
       "motif_112                                    0.0        67        51   \n",
       "motif_149                                    0.0        61        54   \n",
       "motif_185                                    0.0        61        47   \n",
       "...                                          ...       ...       ...   \n",
       "motif_193                                    1.0        14        14   \n",
       "motif_205                                    1.0        13        11   \n",
       "motif_172                                    1.0         9         7   \n",
       "motif_154                                    1.0         5         5   \n",
       "gnps_motif_19.m2m                            1.0         2         2   \n",
       "\n",
       "                   F_coverage        sf  exp_F  Ex_Cov  \\\n",
       "motif_122               84.62  0.027142  68.97   75.79   \n",
       "motif_188               80.77  0.184479  59.12   75.79   \n",
       "motif_112               76.12  0.541361  50.78   75.79   \n",
       "motif_149               88.52  0.010195  46.23   75.79   \n",
       "motif_185               77.05  0.477871  46.23   75.79   \n",
       "...                       ...       ...    ...     ...   \n",
       "motif_193              100.00  0.020497  10.61   75.79   \n",
       "motif_205               84.62  0.356600   9.85   75.77   \n",
       "motif_172               77.78  0.622791   6.82   75.78   \n",
       "motif_154              100.00  0.249871   3.79   75.80   \n",
       "gnps_motif_19.m2m      100.00  0.574351   1.52   76.00   \n",
       "\n",
       "                   more_plants/no_plants comb_p  \n",
       "motif_122                                   0.0  \n",
       "motif_188                                   0.0  \n",
       "motif_112                                   0.0  \n",
       "motif_149                                   0.0  \n",
       "motif_185                                   0.0  \n",
       "...                                         ...  \n",
       "motif_193                                   1.0  \n",
       "motif_205                                   1.0  \n",
       "motif_172                                   1.0  \n",
       "motif_154                                   1.0  \n",
       "gnps_motif_19.m2m                           1.0  \n",
       "\n",
       "[102 rows x 9 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p_value_col = '%s p-value' % comp_name\n",
    "count_col = 'unq_pw_F'\n",
    "df.sort_values([p_value_col, count_col], ascending=[True, False], inplace=True)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>pw_name</th>\n",
       "      <th>more_plants/no_plants p-value</th>\n",
       "      <th>unq_pw_F</th>\n",
       "      <th>tot_ds_F</th>\n",
       "      <th>F_coverage</th>\n",
       "      <th>sf</th>\n",
       "      <th>exp_F</th>\n",
       "      <th>Ex_Cov</th>\n",
       "      <th>more_plants/no_plants comb_p</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>motif_122</th>\n",
       "      <td>motif_122</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>91</td>\n",
       "      <td>77</td>\n",
       "      <td>84.62</td>\n",
       "      <td>0.027142</td>\n",
       "      <td>68.97</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_188</th>\n",
       "      <td>motif_188</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>78</td>\n",
       "      <td>63</td>\n",
       "      <td>80.77</td>\n",
       "      <td>0.184479</td>\n",
       "      <td>59.12</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_112</th>\n",
       "      <td>motif_112</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>67</td>\n",
       "      <td>51</td>\n",
       "      <td>76.12</td>\n",
       "      <td>0.541361</td>\n",
       "      <td>50.78</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_149</th>\n",
       "      <td>motif_149</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>61</td>\n",
       "      <td>54</td>\n",
       "      <td>88.52</td>\n",
       "      <td>0.010195</td>\n",
       "      <td>46.23</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_185</th>\n",
       "      <td>motif_185</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>61</td>\n",
       "      <td>47</td>\n",
       "      <td>77.05</td>\n",
       "      <td>0.477871</td>\n",
       "      <td>46.23</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_127</th>\n",
       "      <td>motif_127</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>43</td>\n",
       "      <td>43</td>\n",
       "      <td>100.00</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>32.59</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_198</th>\n",
       "      <td>motif_198</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>39</td>\n",
       "      <td>32</td>\n",
       "      <td>82.05</td>\n",
       "      <td>0.237857</td>\n",
       "      <td>29.56</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_146</th>\n",
       "      <td>motif_146</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>15</td>\n",
       "      <td>14</td>\n",
       "      <td>93.33</td>\n",
       "      <td>0.090188</td>\n",
       "      <td>11.37</td>\n",
       "      <td>75.80</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gnps_motif_54.m2m</th>\n",
       "      <td>gnps_motif_54.m2m [Fragments indicative for ferulic acid based substructure (MzCloud)]</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>12</td>\n",
       "      <td>100.00</td>\n",
       "      <td>0.035748</td>\n",
       "      <td>9.09</td>\n",
       "      <td>75.75</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gnps_motif_170.m2m</th>\n",
       "      <td>gnps_motif_170.m2m [methoxy-methylaminophenyl substructure]</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>11</td>\n",
       "      <td>91.67</td>\n",
       "      <td>0.173221</td>\n",
       "      <td>9.09</td>\n",
       "      <td>75.75</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gnps_motif_37.m2m</th>\n",
       "      <td>gnps_motif_37.m2m [Fragments indicative for cinnamic/hydroxycinnamic acid substructure]</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>100.00</td>\n",
       "      <td>0.435232</td>\n",
       "      <td>2.27</td>\n",
       "      <td>75.67</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_215</th>\n",
       "      <td>motif_215</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>50.00</td>\n",
       "      <td>0.966504</td>\n",
       "      <td>4.55</td>\n",
       "      <td>75.83</td>\n",
       "      <td>0.000005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gnps_motif_6.m2m</th>\n",
       "      <td>gnps_motif_6.m2m [Diphenyl substructure [ClassyFire; Relevant terms - Substituents: Diphenylmethane - Taxa: Diphenylmethanes]]</td>\n",
       "      <td>0.000265</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>75.00</td>\n",
       "      <td>0.751565</td>\n",
       "      <td>3.03</td>\n",
       "      <td>75.75</td>\n",
       "      <td>0.000265</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_226</th>\n",
       "      <td>motif_226</td>\n",
       "      <td>0.011940</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>50.00</td>\n",
       "      <td>0.941421</td>\n",
       "      <td>1.52</td>\n",
       "      <td>76.00</td>\n",
       "      <td>0.011940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_240</th>\n",
       "      <td>motif_240</td>\n",
       "      <td>0.017400</td>\n",
       "      <td>19</td>\n",
       "      <td>13</td>\n",
       "      <td>68.42</td>\n",
       "      <td>0.846401</td>\n",
       "      <td>14.40</td>\n",
       "      <td>75.79</td>\n",
       "      <td>0.017400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motif_175</th>\n",
       "      <td>motif_175</td>\n",
       "      <td>0.034619</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>100.00</td>\n",
       "      <td>0.108638</td>\n",
       "      <td>6.06</td>\n",
       "      <td>75.75</td>\n",
       "      <td>0.034619</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                           pw_name  \\\n",
       "motif_122                                                                                                                                motif_122   \n",
       "motif_188                                                                                                                                motif_188   \n",
       "motif_112                                                                                                                                motif_112   \n",
       "motif_149                                                                                                                                motif_149   \n",
       "motif_185                                                                                                                                motif_185   \n",
       "motif_127                                                                                                                                motif_127   \n",
       "motif_198                                                                                                                                motif_198   \n",
       "motif_146                                                                                                                                motif_146   \n",
       "gnps_motif_54.m2m                                           gnps_motif_54.m2m [Fragments indicative for ferulic acid based substructure (MzCloud)]   \n",
       "gnps_motif_170.m2m                                                                     gnps_motif_170.m2m [methoxy-methylaminophenyl substructure]   \n",
       "gnps_motif_37.m2m                                          gnps_motif_37.m2m [Fragments indicative for cinnamic/hydroxycinnamic acid substructure]   \n",
       "motif_215                                                                                                                                motif_215   \n",
       "gnps_motif_6.m2m    gnps_motif_6.m2m [Diphenyl substructure [ClassyFire; Relevant terms - Substituents: Diphenylmethane - Taxa: Diphenylmethanes]]   \n",
       "motif_226                                                                                                                                motif_226   \n",
       "motif_240                                                                                                                                motif_240   \n",
       "motif_175                                                                                                                                motif_175   \n",
       "\n",
       "                    more_plants/no_plants p-value  unq_pw_F  tot_ds_F  \\\n",
       "motif_122                                0.000000        91        77   \n",
       "motif_188                                0.000000        78        63   \n",
       "motif_112                                0.000000        67        51   \n",
       "motif_149                                0.000000        61        54   \n",
       "motif_185                                0.000000        61        47   \n",
       "motif_127                                0.000000        43        43   \n",
       "motif_198                                0.000000        39        32   \n",
       "motif_146                                0.000000        15        14   \n",
       "gnps_motif_54.m2m                        0.000000        12        12   \n",
       "gnps_motif_170.m2m                       0.000000        12        11   \n",
       "gnps_motif_37.m2m                        0.000000         3         3   \n",
       "motif_215                                0.000005         6         3   \n",
       "gnps_motif_6.m2m                         0.000265         4         3   \n",
       "motif_226                                0.011940         2         1   \n",
       "motif_240                                0.017400        19        13   \n",
       "motif_175                                0.034619         8         8   \n",
       "\n",
       "                    F_coverage        sf  exp_F  Ex_Cov  \\\n",
       "motif_122                84.62  0.027142  68.97   75.79   \n",
       "motif_188                80.77  0.184479  59.12   75.79   \n",
       "motif_112                76.12  0.541361  50.78   75.79   \n",
       "motif_149                88.52  0.010195  46.23   75.79   \n",
       "motif_185                77.05  0.477871  46.23   75.79   \n",
       "motif_127               100.00  0.000006  32.59   75.79   \n",
       "motif_198                82.05  0.237857  29.56   75.79   \n",
       "motif_146                93.33  0.090188  11.37   75.80   \n",
       "gnps_motif_54.m2m       100.00  0.035748   9.09   75.75   \n",
       "gnps_motif_170.m2m       91.67  0.173221   9.09   75.75   \n",
       "gnps_motif_37.m2m       100.00  0.435232   2.27   75.67   \n",
       "motif_215                50.00  0.966504   4.55   75.83   \n",
       "gnps_motif_6.m2m         75.00  0.751565   3.03   75.75   \n",
       "motif_226                50.00  0.941421   1.52   76.00   \n",
       "motif_240                68.42  0.846401  14.40   75.79   \n",
       "motif_175               100.00  0.108638   6.06   75.75   \n",
       "\n",
       "                    more_plants/no_plants comb_p  \n",
       "motif_122                               0.000000  \n",
       "motif_188                               0.000000  \n",
       "motif_112                               0.000000  \n",
       "motif_149                               0.000000  \n",
       "motif_185                               0.000000  \n",
       "motif_127                               0.000000  \n",
       "motif_198                               0.000000  \n",
       "motif_146                               0.000000  \n",
       "gnps_motif_54.m2m                       0.000000  \n",
       "gnps_motif_170.m2m                      0.000000  \n",
       "gnps_motif_37.m2m                       0.000000  \n",
       "motif_215                               0.000005  \n",
       "gnps_motif_6.m2m                        0.000265  \n",
       "motif_226                               0.011940  \n",
       "motif_240                               0.017400  \n",
       "motif_175                               0.034619  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df[df[p_value_col] < 0.05]\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-01-07 17:16:36.160 | DEBUG    | pals.feature_extraction:change_zero_peak_ints:307 - Setting the zero intensity values in the dataframe\n",
      "2021-01-07 17:16:36.165 | DEBUG    | pals.feature_extraction:change_zero_peak_ints:309 - 0\n",
      "2021-01-07 17:16:36.216 | DEBUG    | pals.feature_extraction:standardize_intensity_df:276 - Scaling the data across the sample: zero mean and unit variance\n"
     ]
    }
   ],
   "source": [
    "entity_dict = gnps_ds.entity_dict\n",
    "intensities_df = gnps_ds.standardize_intensity_df()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gnps_motif_54.m2m [Fragments indicative for ferulic acid based substructure (MzCloud)]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>LibraryID</th>\n",
       "      <th>Precursor m/z</th>\n",
       "      <th>RTConsensus</th>\n",
       "      <th>PrecursorInt</th>\n",
       "      <th>no_spectra</th>\n",
       "      <th>link</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1326</th>\n",
       "      <td>Spectral Match to Curcumin from NIST14</td>\n",
       "      <td>369.1335</td>\n",
       "      <td>4.0253</td>\n",
       "      <td>0.032184</td>\n",
       "      <td>145</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3633</th>\n",
       "      <td>Spectral Match to Curcumin from NIST14</td>\n",
       "      <td>369.1336</td>\n",
       "      <td>3.1547</td>\n",
       "      <td>0.002445</td>\n",
       "      <td>97</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7879</th>\n",
       "      <td>Spectral Match to 3-Hydroxy-4-methoxycinnamic acid from NIST14</td>\n",
       "      <td>195.0655</td>\n",
       "      <td>2.4750</td>\n",
       "      <td>0.001998</td>\n",
       "      <td>192</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8837</th>\n",
       "      <td>Spectral Match to 3-Hydroxy-4-methoxycinnamic acid from NIST14</td>\n",
       "      <td>177.0538</td>\n",
       "      <td>4.0199</td>\n",
       "      <td>0.000481</td>\n",
       "      <td>127</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8838</th>\n",
       "      <td>NaN</td>\n",
       "      <td>371.1490</td>\n",
       "      <td>3.8366</td>\n",
       "      <td>0.004556</td>\n",
       "      <td>49</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8907</th>\n",
       "      <td>MoNA:3697220 Feruloyltyramine</td>\n",
       "      <td>265.1549</td>\n",
       "      <td>0.8890</td>\n",
       "      <td>0.001524</td>\n",
       "      <td>177</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9019</th>\n",
       "      <td>NCGC00095321-06!(1E,4Z,6E)-5-hydroxy-1,7-bis(4-hydroxy-3-methoxyphenyl)hepta-1,4,6-trien-3-one</td>\n",
       "      <td>369.1332</td>\n",
       "      <td>4.2750</td>\n",
       "      <td>0.001551</td>\n",
       "      <td>85</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9024</th>\n",
       "      <td>(1R,3R,4S,5R)-1,3,4-trihydroxy-5-[(E)-3-(4-hydroxyphenyl)prop-2-enoyl]oxycyclohexane-1-carboxylic acid</td>\n",
       "      <td>339.1230</td>\n",
       "      <td>3.1470</td>\n",
       "      <td>0.000508</td>\n",
       "      <td>103</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9040</th>\n",
       "      <td>Spectral Match to Curcumin from NIST14</td>\n",
       "      <td>369.1331</td>\n",
       "      <td>3.8770</td>\n",
       "      <td>0.003096</td>\n",
       "      <td>128</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9048</th>\n",
       "      <td>Spectral Match to Curcumin from NIST14</td>\n",
       "      <td>369.1331</td>\n",
       "      <td>4.4907</td>\n",
       "      <td>0.000681</td>\n",
       "      <td>84</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9056</th>\n",
       "      <td>NCGC00168971-02_C17H20O9_(1R,3R,4S,5R)-1,3,4-Trihydroxy-5-{[(2E)-3-(4-hydroxy-3-methoxyphenyl)-2-propenoyl]oxy}cyclohexanecarboxylic acid</td>\n",
       "      <td>369.1335</td>\n",
       "      <td>3.3287</td>\n",
       "      <td>0.000295</td>\n",
       "      <td>105</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9309</th>\n",
       "      <td>NaN</td>\n",
       "      <td>177.0552</td>\n",
       "      <td>2.4861</td>\n",
       "      <td>0.001077</td>\n",
       "      <td>177</td>\n",
       "      <td>https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&amp;componentindex=127&amp;task=0a8432b5891a48d7ad8459ba4a89969f</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                      LibraryID  \\\n",
       "id                                                                                                                                                \n",
       "1326                                                                                                     Spectral Match to Curcumin from NIST14   \n",
       "3633                                                                                                     Spectral Match to Curcumin from NIST14   \n",
       "7879                                                                             Spectral Match to 3-Hydroxy-4-methoxycinnamic acid from NIST14   \n",
       "8837                                                                             Spectral Match to 3-Hydroxy-4-methoxycinnamic acid from NIST14   \n",
       "8838                                                                                                                                        NaN   \n",
       "8907                                                                                                              MoNA:3697220 Feruloyltyramine   \n",
       "9019                                             NCGC00095321-06!(1E,4Z,6E)-5-hydroxy-1,7-bis(4-hydroxy-3-methoxyphenyl)hepta-1,4,6-trien-3-one   \n",
       "9024                                     (1R,3R,4S,5R)-1,3,4-trihydroxy-5-[(E)-3-(4-hydroxyphenyl)prop-2-enoyl]oxycyclohexane-1-carboxylic acid   \n",
       "9040                                                                                                     Spectral Match to Curcumin from NIST14   \n",
       "9048                                                                                                     Spectral Match to Curcumin from NIST14   \n",
       "9056  NCGC00168971-02_C17H20O9_(1R,3R,4S,5R)-1,3,4-Trihydroxy-5-{[(2E)-3-(4-hydroxy-3-methoxyphenyl)-2-propenoyl]oxy}cyclohexanecarboxylic acid   \n",
       "9309                                                                                                                                        NaN   \n",
       "\n",
       "      Precursor m/z  RTConsensus  PrecursorInt  no_spectra  \\\n",
       "id                                                           \n",
       "1326       369.1335       4.0253      0.032184         145   \n",
       "3633       369.1336       3.1547      0.002445          97   \n",
       "7879       195.0655       2.4750      0.001998         192   \n",
       "8837       177.0538       4.0199      0.000481         127   \n",
       "8838       371.1490       3.8366      0.004556          49   \n",
       "8907       265.1549       0.8890      0.001524         177   \n",
       "9019       369.1332       4.2750      0.001551          85   \n",
       "9024       339.1230       3.1470      0.000508         103   \n",
       "9040       369.1331       3.8770      0.003096         128   \n",
       "9048       369.1331       4.4907      0.000681          84   \n",
       "9056       369.1335       3.3287      0.000295         105   \n",
       "9309       177.0552       2.4861      0.001077         177   \n",
       "\n",
       "                                                                                                                             link  \n",
       "id                                                                                                                                 \n",
       "1326  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "3633  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "7879  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "8837  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "8838  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "8907  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "9019  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "9024  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "9040  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "9048  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "9056  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  \n",
       "9309  https://gnps.ucsd.edu/ProteoSAFe/result.jsp?view=network_displayer&componentindex=127&task=0a8432b5891a48d7ad8459ba4a89969f  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for idx, row in df.iterrows():\n",
    "    members = gnps_ds.dataset_pathways_to_row_ids[idx]\n",
    "    if len(members) < min_members:\n",
    "        continue\n",
    "        \n",
    "    pw_name = row['pw_name']\n",
    "    p_value = row[p_value_col]\n",
    "    \n",
    "    if 'gnps_motif_54.m2m' not in pw_name:\n",
    "        continue\n",
    "    else:\n",
    "        print(pw_name)\n",
    "                \n",
    "    # get group intensities\n",
    "    group_intensities = intensities_df.loc[members][all_samples]    \n",
    "    \n",
    "    # get group info\n",
    "    # print('%s p-value=%.4f' % (pw_name, p_value))\n",
    "    data = []\n",
    "    for member in members:\n",
    "        member_info = entity_dict[member]\n",
    "        unique_id = member_info['unique_id']\n",
    "        library_id = member_info['LibraryID']\n",
    "        gnps_linkout_network = member_info['GNPSLinkout_Network']\n",
    "        no_spectra = member_info['number of spectra']\n",
    "        rt = member_info['RT']\n",
    "        mz = member_info['mass']    \n",
    "        intensity = member_info['SumPeakIntensity']\n",
    "        row = [unique_id, library_id, mz, rt, intensity, no_spectra, gnps_linkout_network]\n",
    "        data.append(row)\n",
    "    member_df = pd.DataFrame(data, columns=['id', 'LibraryID', 'Precursor m/z', 'RTConsensus', 'PrecursorInt', 'no_spectra', 'link']).set_index('id')    \n",
    "    \n",
    "    # Create a categorical palette to identify the networks\n",
    "    used_groups = list(set(all_groups))\n",
    "    group_pal = sns.husl_palette(len(used_groups), s=.45)\n",
    "    group_lut = dict(zip(map(str, used_groups), group_pal))\n",
    "\n",
    "    # Convert the palette to vectors that will be drawn on the side of the matrix\n",
    "    group_colours = pd.Series(all_groups, index=group_intensities.columns).map(group_lut)    \n",
    "    group_colours.name = 'groups'\n",
    "    \n",
    "    # plot heatmap\n",
    "    g = sns.clustermap(group_intensities, center=0, cmap=cmap, col_colors=group_colours, \n",
    "                   col_cluster=False, linewidths=0.75, figsize=(10, 10), cbar_pos=(1.0, 0.3, 0.05, 0.5))\n",
    "\n",
    "    title = pw_name\n",
    "    if 'gnps_motif_54.m2m' in pw_name:\n",
    "        title = 'Ferulic-acid-related Mass2Motif'\n",
    "    plt.suptitle('%s' % (title), fontsize=24, y=0.9)\n",
    "\n",
    "    # draw group legend\n",
    "    for group in used_groups:\n",
    "        g.ax_col_dendrogram.bar(0, 0, color=group_lut[group], label=group, linewidth=0)\n",
    "    g.ax_col_dendrogram.legend(loc=\"right\")\n",
    "    \n",
    "    # make the annotated peaks to have labels in bold\n",
    "    annotated_df = member_df[member_df['LibraryID'].notnull()]\n",
    "    annotated_peaks = annotated_df.index.values\n",
    "    for label in g.ax_heatmap.get_yticklabels():\n",
    "        if label.get_text() in annotated_peaks:\n",
    "            label.set_weight(\"bold\")\n",
    "            label.set_color(\"green\")   \n",
    "    plt.setp(g.ax_heatmap.get_yticklabels(), rotation=0)        \n",
    "\n",
    "    out_file = '%s.pdf' % pw_name\n",
    "    plt.savefig(out_file, dpi=300)\n",
    "    plt.show()\n",
    "    \n",
    "    out_file = '%s.csv' % pw_name\n",
    "    display(member_df)\n",
    "    member_df.drop('link', axis=1).round(4).to_csv(out_file, index=True)"
   ]
  }
 ],
 "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}