ipynb/tutorials/IBD-Tutorial-standalone-API.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Tutorial Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Repeat measure experimental designs (e.g. time series) are a valid and powerful method to control for inter-individual variation. However, conventional dimensionality reduction methods can not account for the high-correlation of each subject to itself at a later time point. This inherent correlation structure can cause subject grouping to confound or even outweigh important phenotype groupings. To address this we will use Compositional Tensor Factorization (CTF) which we provide in the software package [gemelli](https://github.com/biocore/gemelli). CTF can account for repeated measures, compositionality, and sparsity in microbiome data.\n",
    "\n",
    "In this tutorial we use _gemelli_ to perform CTF on a time series dataset comparing Crohn's and control subjects over a period of 25 weeks published in [Vázquez-Baeza et al](https://gut.bmj.com/content/67/9/1743). First we will download the processed data originally from [here](https://qiita.ucsd.edu/study/description/2538#). This data can be downloaded with the following links:\n",
    "\n",
    "* **Table** (table.biom) | [download](https://github.com/biocore/gemelli/tree/master/ipynb/tutorials/IBD-2538/data/table.biom)\n",
    "* **Rarefied Table** (rarefied-table.biom) | [download](https://github.com/biocore/gemelli/tree/master/ipynb/tutorials/IBD-2538/data/rarefied-table.biom)\n",
    "* **Sample Metadata** (metadata.tsv) | [download](https://github.com/biocore/gemelli/tree/master/ipynb/tutorials/IBD-2538/data/metadata.tsv)\n",
    "* **Feature Metadata** (taxonomy.tsv) | [download](https://github.com/biocore/gemelli/tree/master/ipynb/tutorials/IBD-2538/data/taxonomy.tsv)\n",
    "* **Tree** (sepp-insertion-tree.nwk) | [download](https://github.com/biocore/gemelli/tree/master/ipynb/tutorials/IBD-2538/data/sepp-insertion-tree.nwk)\n",
    "\n",
    "**Note**: This tutorial assumes you have installed [QIIME2](https://qiime2.org/) using one of the procedures in the [install documents](https://docs.qiime2.org/2020.2/install/). This tutorial also assumed you have installed, [Qurro](https://github.com/biocore/qurro) and [gemelli](https://github.com/biocore/gemelli).\n",
    "\n",
    "First, we will make a tutorial directory and download the data above and move the files to the `IBD-2538/data` directory:\n",
    "\n",
    "```bash\n",
    "mkdir IBD-2538\n",
    "```\n",
    "```bash\n",
    "# move downloaded data here\n",
    "mkdir IBD-2538/data\n",
    "```\n",
    "\n",
    "\n",
    "First we will import our data with python:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import warnings\n",
    "import pandas as pd\n",
    "from biom import Table, load_table\n",
    "from skbio import TreeNode, read\n",
    "# hide pandas Future/Deprecation Warning(s) for tutorial\n",
    "warnings.filterwarnings(\"ignore\", category=DeprecationWarning) \n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
    "\n",
    "# import table(s)\n",
    "table = load_table('IBD-2538/data/table.biom')\n",
    "# import metadata\n",
    "metadata = pd.read_csv('IBD-2538/data/metadata.tsv', sep='\\t',\n",
    "                       dtype={'sample_name':'str'}).set_index('sample_name')\n",
    "# import taxonomy\n",
    "taxonomy = pd.read_csv('IBD-2538/data/taxonomy.tsv', sep='\\t',\n",
    "                       index_col=0, dtype={'sample_name':'str'})\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we will demonstrate the issues with using conventional dimensionality reduction methods on time series data. To do this we will perform PCoA dimensionality reduction on weighted and unweighted UniFrac $\\beta$-diversity distances. We will also run Aitchison Robust PCA which is built on the same framework as CTF but does not account for repeated measures.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/cmartino/miniconda3/envs/qiime2-2021.2.songbird.mmvec/lib/python3.6/site-packages/skbio/stats/ordination/_principal_coordinate_analysis.py:152: RuntimeWarning: The result contains negative eigenvalues. Please compare their magnitude with the magnitude of some of the largest positive eigenvalues. If the negative ones are smaller, it's probably safe to ignore them, but if they are large in magnitude, the results won't be useful. See the Notes section for more details. The smallest eigenvalue is -0.9329142170862077 and the largest is 24.591847406777045.\n",
      "  RuntimeWarning\n",
      "/Users/cmartino/miniconda3/envs/qiime2-2021.2.songbird.mmvec/lib/python3.6/site-packages/skbio/stats/ordination/_principal_coordinate_analysis.py:152: RuntimeWarning: The result contains negative eigenvalues. Please compare their magnitude with the magnitude of some of the largest positive eigenvalues. If the negative ones are smaller, it's probably safe to ignore them, but if they are large in magnitude, the results won't be useful. See the Notes section for more details. The smallest eigenvalue is -0.057899443060764624 and the largest is 12.71711105317202.\n",
      "  RuntimeWarning\n"
     ]
    }
   ],
   "source": [
    "import unifrac as qsu\n",
    "from emperor import Emperor\n",
    "from gemelli.rpca import rpca\n",
    "from skbio.stats.ordination import pcoa\n",
    "\n",
    "# save variables\n",
    "rarefied_table, phylogeny = ('IBD-2538/data/rarefied-table.biom',\n",
    "                             'IBD-2538/data/sepp-insertion-tree.nwk')\n",
    "variance_adjusted, bypass_tips = False, False\n",
    "\n",
    "# generate distances\n",
    "unweighted_unifrac_distance = qsu.ssu(str(rarefied_table), str(phylogeny), 'unweighted',\n",
    "                                          variance_adjusted, 1.0, bypass_tips, 1.0) \n",
    "weighted_unifrac_distancea_dist = qsu.ssu(str(rarefied_table), str(phylogeny), 'weighted_unnormalized',\n",
    "                                          variance_adjusted, 1.0, bypass_tips, 1.0)\n",
    "\n",
    "# perform PCoA\n",
    "unweighted_unifrac_pcoa = pcoa(weighted_unifrac_distancea_dist)\n",
    "weighted_unifrac_pcoa = pcoa(unweighted_unifrac_distance)\n",
    "\n",
    "# run RPCA and plot with emperor\n",
    "rpca_biplot, rpca_distance = rpca(table.copy())\n",
    "\n",
    "# use emperor to plot\n",
    "os.makedirs('IBD-2538/core-metric-output', exist_ok=True)\n",
    "# unweighted plot\n",
    "emp = Emperor(unweighted_unifrac_pcoa, metadata, remote='.')\n",
    "output_folder = 'IBD-2538/core-metric-output/unweighted_unifrac_emperor_plot' \n",
    "os.makedirs(output_folder, exist_ok=True)\n",
    "with open(os.path.join(output_folder, 'index.html'), 'w') as f:\n",
    "    f.write(emp.make_emperor(standalone=True))\n",
    "    emp.copy_support_files(output_folder)\n",
    "# weighted plot\n",
    "emp = Emperor(weighted_unifrac_pcoa, metadata, remote='.')\n",
    "output_folder = 'IBD-2538/core-metric-output/weighted_emperor_unifrac_plot' \n",
    "os.makedirs(output_folder, exist_ok=True)\n",
    "with open(os.path.join(output_folder, 'index.html'), 'w') as f:\n",
    "    f.write(emp.make_emperor(standalone=True))\n",
    "    emp.copy_support_files(output_folder)\n",
    "# RPCA plot   \n",
    "emp = Emperor(rpca_biplot, metadata, remote='.')\n",
    "output_folder = 'IBD-2538/core-metric-output/rpca_emperor_plot' \n",
    "os.makedirs(output_folder, exist_ok=True)\n",
    "with open(os.path.join(output_folder, 'index.html'), 'w') as f:\n",
    "    f.write(emp.make_emperor(standalone=True))\n",
    "    emp.copy_support_files(output_folder)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "Now we can visualize the sample groupings by host subject ID and IBD with [Emperor](https://biocore.github.io/emperor/). From this we can see the PCoA samples clearly separate by host subject ID which in some cases (e.g. UniFrac) can overwhelm the control (blue) v. Crohn's disease (orange) sample groupings. Even in the case where the IBD grouping is not completely lost (e.g. RPCA) we can still see confounding groupings in the control (blue) groups by subject ID. This can complicate the interpretation of these analysis.    \n",
    " \n",
    "![image.png](etc/subjectidsgroups.png)\n",
    "\n",
    "This confounding effect can also be observed in the statistics performed on pairwise $\\beta$-diversity distances (e.g. PERMANOVA). For the purpose of exploring distance matrices, [q2-longitudinal](https://msystems.asm.org/content/3/6/e00219-18) has many excellent methods to account for repeated measures data. You can find the q2-longitudinal tutorial [here](https://docs.qiime2.org/2020.2/tutorials/longitudinal/).\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Compositional Tensor Factorization (CTF) Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "In order to account for the correlation among samples from the same subject we will employ compositional tensor factorization (CTF). CTF builds on the ability to account for compositionality and sparsity using the robust center log-ratio transform covered in the RPCA tutorial (found [here](https://forum.qiime2.org/t/robust-aitchison-pca-beta-diversity-with-deicode)) but restructures and factors the data as a tensor. Here we will run CTF through _gemelli_ and explore/interpret the different results. \n",
    "\n",
    "\n",
    "To run CTF we only need to run one command (gemelli ctf). The required input requirements are:\n",
    "\n",
    "1. table\n",
    "    - The table is of type `FeatureTable[Frequency]` which is a table where the rows are features (e.g. ASVs/microbes), the columns are samples, and the entries are the number of sequences for each sample-feature pair.\n",
    "2. sample-metadata\n",
    "    - This is a QIIME2 formatted [metadata](https://docs.qiime2.org/2020.2/tutorials/metadata/) (e.g. tsv format) where the rows are samples matched to the (1) table and the columns are different sample data (e.g. time point).  \n",
    "3. individual-id-column\n",
    "    - This is the name of the column in the (2) metadata that indicates the individual subject/site (e.g. subject ID) that was sampled repeatedly.\n",
    "4. state-column\n",
    "    - This is the name of the column in the (2) metadata that indicates the numeric repeated measure (e.g., Time in months/days) or non-numeric category (i.e. decade/body-site). \n",
    "5. output-dir\n",
    "    - The desired location of the output. We will cover each output independently below.  \n",
    "\n",
    "There are also optional input parameters:\n",
    "\n",
    "* ( _Optional_ ) feature-metadata-file\n",
    "    - This is a metadata file (e.g. tsv, or `FeatureTable[Taxonomy]` .qza) where the rows are matched to the table features and the columns are feature metadata such as taxonomy, gene pathway, etc... \n",
    "\n",
    "In this tutorial our individual-id-column is `host_subject_id` and our state-column is different time points denoted as `timepoint` in the sample metadata. Now we are ready to run CTF:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from gemelli.ctf import ctf\n",
    "\n",
    "# run CTF\n",
    "ctf_results = ctf(table.copy(), metadata.reindex(table.ids()),\n",
    "                  'host_subject_id',\n",
    "                  'timepoint',\n",
    "                  feature_metadata=taxonomy)\n",
    "# expand the results\n",
    "subject_biplot = ctf_results[0]\n",
    "state_biplot = ctf_results[1]\n",
    "distance_matrix = ctf_results[2]\n",
    "state_subject_ordination = ctf_results[3]\n",
    "state_feature_ordination = ctf_results[4]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "We will now cover the output files:\n",
    "* subject_biplot\n",
    "* state_biplot\n",
    "* distance_matrix\n",
    "* state_subject_ordination\n",
    "* state_feature_ordination\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, we will explore the `subject_biplot` which is an ordination where dots represent _subjects_ not _samples_ and arrows represent features (e.g. ASVs). First, we will need to aggregate the metadata by subject (i.e. collapsing the metadata of all samples from a given subject). This can be done by hand or using DataFrames in python (with pandas) or R like so:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>ibd</th>\n",
       "      <th>active_disease</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>#SampleID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>s1000100</th>\n",
       "      <td>Control</td>\n",
       "      <td>quiescent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>s1000200</th>\n",
       "      <td>Control</td>\n",
       "      <td>quiescent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>s1000300</th>\n",
       "      <td>Control</td>\n",
       "      <td>quiescent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>s1000500</th>\n",
       "      <td>Control</td>\n",
       "      <td>quiescent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>s1000600</th>\n",
       "      <td>Control</td>\n",
       "      <td>quiescent</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               ibd active_disease\n",
       "#SampleID                        \n",
       "s1000100   Control      quiescent\n",
       "s1000200   Control      quiescent\n",
       "s1000300   Control      quiescent\n",
       "s1000500   Control      quiescent\n",
       "s1000600   Control      quiescent"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from qiime2 import Metadata\n",
    "\n",
    "# first we import the metdata into pandas\n",
    "mf = pd.read_csv('IBD-2538/data/metadata.tsv', sep='\\t',index_col=0)\n",
    "# next we aggregate by subjects (i.e. 'host_subject_id') \n",
    "# and keep the first instance of 'diagnosis_full' by subject.\n",
    "mf = mf.groupby('host_subject_id').agg({'ibd':'first','active_disease':'first'})\n",
    "# now we save the metadata in QIIME2 format.\n",
    "mf.index.name = '#SampleID'\n",
    "mf.to_csv('IBD-2538/data/subject-metadata.tsv', sep='\\t')\n",
    "mf.head(5)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "Now with the collapsed `subject-metadata.tsv` table we are ready to plot with emperor: \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# RPCA plot   \n",
    "emp = Emperor(subject_biplot, mf, remote='.',\n",
    "              feature_mapping_file=taxonomy)\n",
    "output_folder = 'IBD-2538/ctf_emperor_plot' \n",
    "os.makedirs(output_folder, exist_ok=True)\n",
    "with open(os.path.join(output_folder, 'index.html'), 'w') as f:\n",
    "    f.write(emp.make_emperor(standalone=True))\n",
    "    emp.copy_support_files(output_folder)\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From this visualization we can see that the Crohn's subjects clearly separate from the healthy controls.\n",
    "\n",
    "![image.png](etc/per_subject_biplot_2.png)\n",
    "\n",
    "\n",
    "We can also see that the IBD grouping is separated entirely along the first PC (axis 2). We can now use [Qurro](https://github.com/biocore/qurro) to explore the feature loading partitions (arrows) in this biplot as a log-ratio of the original table counts. This allows us to relate these low-dimensional representations back to our original data. Additionally, log-ratios provide a nice set of data points for additional analysis such as LME models. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "\n",
    "# make a directory to store results\n",
    "os.makedirs('qurro-results', exist_ok=True)\n",
    "# save input for Qurro\n",
    "subject_biplot.write('qurro-results/subject_biplot.txt')\n",
    "# run Qurro (capture will hide warning outputs)\n",
    "!qurro \\\n",
    "    --ranks qurro-results/subject_biplot.txt\\\n",
    "    --table IBD-2538/data/table.biom\\\n",
    "    --sample-metadata IBD-2538/data/metadata.tsv\\\n",
    "    --feature-metadata IBD-2538/data/taxonomy.tsv\\\n",
    "    --output-dir IBD-2538/qurro-results/qurro-visualization\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "From the Qurro output `qurro.qzv` we will simply choose the PC2 loadings above and below zero as the numerator (red ranks) and denominator (blue ranks) to create a log-ratio that differentiates the samples by IBD status. Log-ratios can also be chosen by taxonomy or sequence identifiers (see the Qurro tutorials [here](https://github.com/biocore/qurro#tutorials) for more information). We can plot this log-ratio in Qurro with the x-axis as time and the color as IBD, which clearly shows nice separation between phenotypes. \n",
    "\n",
    "![image.png](etc/qurro-plot.png)\n",
    "\n",
    "We can further explore these phenotype differences by exporting the `sample_plot_data.tsv` from Qurro (marked in a orange box above) which will provide the selected log-ratio values for each sample. We can then merge this `sample_plot_data` with our sample metadata in python or R. \n",
    "\n",
    "**Note:** Qurro will have an option to export all of the metadata or only the log-ratio data soon."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>Current_Natural_Log_Ratio</th>\n",
       "      <th>active_disease</th>\n",
       "      <th>age</th>\n",
       "      <th>age_unit</th>\n",
       "      <th>alcohol_frequency</th>\n",
       "      <th>anonymized_name</th>\n",
       "      <th>anti_tnf</th>\n",
       "      <th>antibiotic_select</th>\n",
       "      <th>appendix_removed</th>\n",
       "      <th>asa_5_asa</th>\n",
       "      <th>...</th>\n",
       "      <th>tanning_beds</th>\n",
       "      <th>tanning_sprays</th>\n",
       "      <th>taxon_id</th>\n",
       "      <th>teethbrushing_frequency</th>\n",
       "      <th>title</th>\n",
       "      <th>tonsils_removed</th>\n",
       "      <th>tot_mass</th>\n",
       "      <th>weeks</th>\n",
       "      <th>weight_change</th>\n",
       "      <th>timepoint</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>#SampleID</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>2538.10001</th>\n",
       "      <td>9.806334</td>\n",
       "      <td>quiescent</td>\n",
       "      <td>56</td>\n",
       "      <td>years</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>102</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>...</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>408170.0</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Guiding longitudinal sampling in inflammatory ...</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>129.3</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2538.10001</th>\n",
       "      <td>10.005570</td>\n",
       "      <td>quiescent</td>\n",
       "      <td>56</td>\n",
       "      <td>years</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>104</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>...</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>408170.0</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Guiding longitudinal sampling in inflammatory ...</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>129.3</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>Missing: Not provided</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 128 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Current_Natural_Log_Ratio active_disease  age age_unit  \\\n",
       "#SampleID                                                            \n",
       "2538.10001                   9.806334      quiescent   56    years   \n",
       "2538.10001                  10.005570      quiescent   56    years   \n",
       "\n",
       "                alcohol_frequency  anonymized_name               anti_tnf  \\\n",
       "#SampleID                                                                   \n",
       "2538.10001  Missing: Not provided              102  Missing: Not provided   \n",
       "2538.10001  Missing: Not provided              104  Missing: Not provided   \n",
       "\n",
       "                antibiotic_select       appendix_removed  \\\n",
       "#SampleID                                                  \n",
       "2538.10001  Missing: Not provided  Missing: Not provided   \n",
       "2538.10001  Missing: Not provided  Missing: Not provided   \n",
       "\n",
       "                        asa_5_asa  ...           tanning_beds  \\\n",
       "#SampleID                          ...                          \n",
       "2538.10001  Missing: Not provided  ...  Missing: Not provided   \n",
       "2538.10001  Missing: Not provided  ...  Missing: Not provided   \n",
       "\n",
       "                   tanning_sprays  taxon_id teethbrushing_frequency  \\\n",
       "#SampleID                                                             \n",
       "2538.10001  Missing: Not provided  408170.0   Missing: Not provided   \n",
       "2538.10001  Missing: Not provided  408170.0   Missing: Not provided   \n",
       "\n",
       "                                                        title  \\\n",
       "#SampleID                                                       \n",
       "2538.10001  Guiding longitudinal sampling in inflammatory ...   \n",
       "2538.10001  Guiding longitudinal sampling in inflammatory ...   \n",
       "\n",
       "                  tonsils_removed tot_mass                  weeks  \\\n",
       "#SampleID                                                           \n",
       "2538.10001  Missing: Not provided    129.3  Missing: Not provided   \n",
       "2538.10001  Missing: Not provided    129.3  Missing: Not provided   \n",
       "\n",
       "                    weight_change timepoint  \n",
       "#SampleID                                    \n",
       "2538.10001  Missing: Not provided        25  \n",
       "2538.10001  Missing: Not provided         6  \n",
       "\n",
       "[2 rows x 128 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# import log-ratio data\n",
    "metadata_one = pd.read_csv('IBD-2538/data/metadata.tsv',\n",
    "                           sep='\\t', index_col=0)\n",
    "# import rest of the metadata\n",
    "metadata_two = pd.read_csv('IBD-2538/qurro-results/sample_plot_data.tsv',\n",
    "                           sep='\\t', index_col=0)[['Current_Natural_Log_Ratio']]\n",
    "# merge the data\n",
    "log_ratio_metdata = pd.concat([metadata_two, metadata_one], axis=1)\n",
    "# ensure no duplicate columns\n",
    "log_ratio_metdata = log_ratio_metdata.dropna(subset=['Current_Natural_Log_Ratio'])\n",
    "# export in QIIME2 format\n",
    "log_ratio_metdata.index.name = '#SampleID'\n",
    "log_ratio_metdata.to_csv('IBD-2538/qurro-results/merged_sample_plot_data.tsv', sep='\\t')\n",
    "log_ratio_metdata.head(2)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see above the metadata now has the added column of `Current_Natural_Log_Ratio` from Qurro. So now we will continue to explore this log-ratio by first plotting it explicitly over time.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plotting\n",
    "import matplotlib\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.style.use('ggplot')\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['xtick.labelsize'] = 18\n",
    "plt.rcParams['ytick.labelsize'] = 18\n",
    "\n",
    "fig, ax = plt.subplots(1, 1, figsize=(16, 5))\n",
    "\n",
    "pal = {'Control':'#377eb8','Crohns':'#ff7f00'}\n",
    "\n",
    "sns.lineplot(y='Current_Natural_Log_Ratio',x='timepoint', hue='ibd', ci=68,\n",
    "             data=log_ratio_metdata,\n",
    "             ax=ax, palette=pal)\n",
    "\n",
    "ax.set_ylabel('$\\\\log(\\\\frac{Control\\,Associated\\,ASVs}'\n",
    "              '{Crohns\\,Associated\\,ASVs})$',\n",
    "              fontsize=25,\n",
    "              color='black')\n",
    "ax.set_xlabel('time (weeks)',\n",
    "              fontsize=25,\n",
    "              color='black')\n",
    "\n",
    "# fix spine colors\n",
    "ax.set_facecolor('white')\n",
    "ax.set_axisbelow(True)\n",
    "ax.spines['right'].set_visible(False)\n",
    "ax.spines['left'].set_visible(True)\n",
    "ax.spines['bottom'].set_visible(True)\n",
    "ax.spines['top'].set_visible(False)\n",
    "ax.spines['bottom'].set_bounds(1, 25)\n",
    "for child in ax.get_children():\n",
    "    if isinstance(child, matplotlib.spines.Spine):\n",
    "        child.set_color('grey')\n",
    "ax.tick_params(axis='y', colors='black')\n",
    "ax.tick_params(axis='x', colors='black')\n",
    "\n",
    "# generate legend from last iterations     \n",
    "handles, labels = ax.get_legend_handles_labels()\n",
    "legend = ax.legend(handles[1:], labels[1:], loc=2, \n",
    "                         bbox_to_anchor=(0.0, 1.1),\n",
    "                         prop={'size':30},\n",
    "                         fancybox=True, framealpha=0.5, \n",
    "                         ncol=2, markerscale=3,\n",
    "                         facecolor=\"white\")\n",
    "# increase the line width in the legend \n",
    "for line in legend.get_lines()[:]:\n",
    "    line.set_linewidth(8.0)\n",
    "for line in legend.get_lines()[:]:\n",
    "    line.set_linewidth(4.0)\n",
    "\n",
    "fig.patch.set_facecolor('white')\n",
    "fig.patch.set_alpha(0.0)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "This demonstrates that we can recreate the separation by IBD that we saw in both the `subject_biplot` & `state_subject_ordination`, allowing us to associate specific taxa (in the numerator or denominator) with a particular phenotype.\n",
    "\n",
    "We can test the statistical power of this log-ratio to differentiate samples by IBD status using a linear mixed effects (LME). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 Mixed Linear Model Regression Results\n",
      "=======================================================================\n",
      "Model:            MixedLM Dependent Variable: Current_Natural_Log_Ratio\n",
      "No. Observations: 279     Method:             REML                     \n",
      "No. Groups:       24      Scale:              0.9849                   \n",
      "Min. group size:  3       Log-Likelihood:     -409.1360                \n",
      "Max. group size:  16      Converged:          Yes                      \n",
      "Mean group size:  11.6                                                 \n",
      "------------------------------------------------------------------------\n",
      "                         Coef.   Std.Err.    z     P>|z|  [0.025  0.975]\n",
      "------------------------------------------------------------------------\n",
      "Intercept                 9.798     0.175  55.844  0.000   9.454  10.142\n",
      "ibd[T.Crohns]            -2.271     0.269  -8.437  0.000  -2.798  -1.743\n",
      "timepoint                -0.006     0.011  -0.539  0.590  -0.027   0.015\n",
      "timepoint:ibd[T.Crohns]  -0.007     0.018  -0.404  0.686  -0.043   0.028\n",
      "Group Var                 0.084     0.062                               \n",
      "=======================================================================\n",
      "\n"
     ]
    }
   ],
   "source": [
    "\n",
    "import statsmodels.formula.api as smf\n",
    "\n",
    "\n",
    "md = smf.mixedlm(\"Current_Natural_Log_Ratio ~ timepoint * ibd\",\n",
    "                 log_ratio_metdata,\n",
    "                 groups=log_ratio_metdata[\"host_subject_id\"])\n",
    "mdf = md.fit()\n",
    "print(mdf.summary())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "From this LME model we can see that indeed the IBD grouping is significant across time. "
   ]
  }
 ],
 "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": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}