1_3DMM_pre-processing_KatharoSeq.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os.path\n",
    "import glob\n",
    "import pandas as pd\n",
    "from biom import Table, load_table\n",
    "\n",
    "import warnings\n",
    "\n",
    "from qiime2.plugins.feature_table.methods import (rarefy, filter_samples, filter_features,\n",
    "                                                  filter_features_conditionally, merge)\n",
    "from qiime2 import Artifact\n",
    "from qiime2 import Metadata\n",
    "\n",
    "\n",
    "from skbio import OrdinationResults, TreeNode, DistanceMatrix\n",
    "from qiime2.plugins.gemelli.actions import (rpca, phylogenetic_rpca_without_taxonomy,\n",
    "                                            phylogenetic_rpca_with_taxonomy)\n",
    "from qiime2.plugins.emperor.actions import plot, biplot\n",
    "from qiime2.plugins.diversity.actions import (beta,\n",
    "                                              beta_phylogenetic,\n",
    "                                              pcoa)\n",
    "from qiime2.plugins.diversity.pipelines import (alpha,\n",
    "                                                alpha_phylogenetic)\n",
    "from qiime2.plugins.diversity.visualizers import (alpha_rarefaction,adonis,beta_group_significance)\n",
    "from qiime2.plugins.diversity.methods import pcoa\n",
    "from qiime2.plugins import taxa\n",
    "\n",
    "from qiime2.plugins import emperor\n",
    "\n",
    "from skbio.stats.distance import permanova, bioenv\n",
    "from skbio import DistanceMatrix\n",
    "import statsmodels.stats.multitest as multi \n",
    "from scipy.optimize import curve_fit\n",
    "from scipy import stats\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import re"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# classifier\n",
    "from scipy import interp\n",
    "from sklearn import svm, datasets\n",
    "from sklearn.metrics import (roc_curve, auc, precision_recall_curve)\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn import preprocessing\n",
    "from sklearn.preprocessing import label_binarize\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "# stats\n",
    "from statsmodels.stats.contingency_tables import mcnemar"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Importing files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Downloaded from qiita (older greengenes [gg-13-8])\n",
    "## gg-13-8\n",
    "collapsed_table = Artifact.load('./data/151594_collapsed_table.qza')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Downloaded from qiita (older greengenes [gg-13-8])\n",
    "##reference hit table\n",
    "deblur_table = Artifact.load('./data/151557_feature-table.qza')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Downloaded from qiita (older greengenes [gg-13-8])\n",
    "## gg-13-8\n",
    "taxonomy = Artifact.load('./data/151580_classification.qza')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "taxonomy_df = taxonomy.view(pd.DataFrame)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "##Updated. Need to update metadata in Qiita study\n",
    "metadata = Metadata.load('./metadata/2023_03_21_3DMM_metadata_to_update_qiita.txt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_metadata = Metadata.load('./metadata/gg-13-8_taxonomy.tsv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "metadata_df = metadata.to_dataframe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## KatharoSeq Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define the allosteric sigmoid equation\n",
    "def allosteric_sigmoid(x, h, k_prime):\n",
    "    y = x ** h / (k_prime + x ** h)\n",
    "    return y\n",
    "\n",
    "#Define curve fitting function\n",
    "def katharoseq_fit_curve(input_table_df,filter_out=None):\n",
    "    katharo_table_df = input_table_df[input_table_df.index.str.contains('KATHARO')].copy()\n",
    "    # Calculate the total number of reads per sample\n",
    "    katharo_table_df['deblur_reads'] = katharo_table_df.sum(axis=1)\n",
    "    \n",
    "    # calculate the number of reads aligning to the mock community input genera\n",
    "    katharo_table_df['kath_reads'] = katharo_table_df[['k__Bacteria;p__Proteobacteria;c__Betaproteobacteria;o__Burkholderiales;f__Comamonadaceae;g__Variovorax']].sum(1)\n",
    "    # calculate the percent correctly assigned\n",
    "    katharo_table_df['correct_assign'] = katharo_table_df['kath_reads'] / katharo_table_df['deblur_reads']\n",
    "    # subset out KatharoSeq columns and add in logarithim of reads for plotting\n",
    "    katharo = katharo_table_df[['correct_assign','deblur_reads','kath_reads']].copy()\n",
    "    katharo['log_deblur_reads'] = np.log10(katharo['deblur_reads'])\n",
    "    if filter_out != None:\n",
    "        # Filter out list of problematic samples\n",
    "        katharo = katharo[~katharo.index.isin(filter_out)]\n",
    "    #fit allosteric sigmoid\n",
    "    popt, pcov = curve_fit(allosteric_sigmoid, katharo['log_deblur_reads'], katharo['correct_assign'], method='dogbox')\n",
    "    \n",
    "    return katharo, popt, pcov\n",
    "\n",
    "#Define function that regresses a min read_threshold from a proportion of fidelity (from control samples)\n",
    "def katharo_read_threshold(popt,proportion):\n",
    "    # Determine the number of reads at which the specified proportion of reads aligning to the control reference\n",
    "\n",
    "    # assign variables and solve for X (number of reads to pass filter)\n",
    "    h = popt[0]  # first value printed above graph\n",
    "    k = popt[1]   # second value printed above graph\n",
    "    y = proportion ## what you want to solve for\n",
    "\n",
    "    min_log_reads = np.power((k/((1/y)-1)),(1/h))\n",
    "    min_freq = np.power(10, min_log_reads).astype(int)\n",
    "    return min_freq\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Katharoseq filtering on min_reads for 0.80 alignment to target. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First performs KatharoSeq analysis on a per_sequence_run basis, filters per_sequence_run feature_tables based on a read threshold that matches a 0.80 fidelity to the expected taxon from control samples, as modeled by the allosteric sigmoid fit.\n",
    "\n",
    "Then performs a second KatharoSeq analysis on a per_extraction_plate basis, filters per_extraction_plate feature_tables based on a read threshold that matches a 0.80 fidelity to the expected taxon from control samples, as modeled by the allosteric sigmoid fit.\n",
    "\n",
    "Lastly plots the results with matplotlib subplots. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "##setup subplots, 12 curves so 4 columns by 3 rows for a landscape view\n",
    "index=0\n",
    "rows=3\n",
    "columns=4\n",
    "f, axes = plt.subplots(nrows=rows, ncols=columns, figsize=(20,10),sharey=True,sharex=True)\n",
    "plt.subplots_adjust(hspace=0.3)\n",
    "\n",
    "#per_sequence_run loop\n",
    "for seq_run in metadata_df['run_prefix'].unique():\n",
    "    f_table = filter_samples(collapsed_table,\n",
    "                                     metadata = metadata,\n",
    "                                     where= f\"run_prefix == '{seq_run}'\")\n",
    "\n",
    "    filtered_table_df = f_table.filtered_table.view(pd.DataFrame)\n",
    "    katharo, popt, pcov = katharoseq_fit_curve(filtered_table_df)\n",
    "    \n",
    "    # plot fit curve\n",
    "    x = np.linspace(0, 6, 50)\n",
    "    y = allosteric_sigmoid(x, *popt)\n",
    "\n",
    "    # multi-plot the fit\n",
    "    axes[index//columns,index%columns].plot(katharo['log_deblur_reads'], katharo['correct_assign'], 'o', label='data')\n",
    "    axes[index//columns,index%columns].plot(x,y, label='fit')\n",
    "    axes[index//columns,index%columns].set_ylim([0,1.05])\n",
    "    #axes[index//columns,index%columns].legend(loc='best')\n",
    "    axes[index//columns,index%columns].set_title(seq_run.replace('_S1_L001',\"\"))\n",
    "    \n",
    "    \n",
    "    axes[index//columns,index%columns].axvline(np.log10(katharo_read_threshold(popt,0.80)),ls='--')\n",
    "    axes[index//columns,index%columns].axhline(0.80,ls='--')\n",
    "    index +=1\n",
    "    \n",
    "#per_extraction_plate loop\n",
    "for sample_plate in metadata_df['sample_plate'].unique():\n",
    "    f_table = filter_samples(collapsed_table,\n",
    "                                     metadata = metadata,\n",
    "                                     where= f\"sample_plate == '{sample_plate}'\")\n",
    "    plate_name = sample_plate.replace('3DMM_20220614_14542_','').replace('3DMM_20220623_14542_','')\n",
    "#     f_table.filtered_table.save('./data/per_plate/' + '150353_collapsed_table_' + plate_name+'.qza')\n",
    "    filtered_table_df = f_table.filtered_table.view(pd.DataFrame)\n",
    "    katharo, popt, pcov = katharoseq_fit_curve(filtered_table_df)\n",
    "    \n",
    "    # plot fit curve\n",
    "    x = np.linspace(0, 6, 50)\n",
    "    y = allosteric_sigmoid(x, *popt)\n",
    "\n",
    "    # plot the fit\n",
    "    axes[index//columns,index%columns].plot(katharo['log_deblur_reads'], katharo['correct_assign'], 'o', label='data')\n",
    "    axes[index//columns,index%columns].plot(x,y, label='fit')\n",
    "    axes[index//columns,index%columns].set_ylim([0,1.05])\n",
    "    #axes[index//columns,index%columns].legend(loc='best')\n",
    "    axes[index//columns,index%columns].set_title(plate_name + ' read_threshold: '\n",
    "                                                 + str(katharo_read_threshold(popt,0.80)))\n",
    "    \n",
    "    axes[index//columns,index%columns].axvline(np.log10(katharo_read_threshold(popt,0.80)),ls='--')\n",
    "    axes[index//columns,index%columns].axhline(0.80,ls='--')\n",
    "    index +=1\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The KatharoSeq analysis was rerun after fitlering outliers from Plate_2 and Plate_4 to improve allosteric sigmoidal curve fit. Rerunning will overwrite the files written on the cells above. \n",
    "\n",
    "Since the curve fit of sequencing run '3DMM_Plates_1-5' was poor (didn't generalize well, likely due to differential sequencing coverage across extraction_plates), a per_extraction_plate KatharoSeq analysis was chosen. \n",
    "\n",
    "The KatharoSeq analysis was run on per_extraction_plate basis, with a feature_table collapsed at the genera level (collapsed_table),but we want to keep the higher resolution sOTUs featues from deblur instead, so we filter deblur_tables (instead of collapsed_table) with the KatharoSeq minimum sequencing depth threshold and saved to file (lines 66-69). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "problematic_samples = ['150297.14542.KATHARO.P4.2','150297.14542.KATHARO.P2.4']\n",
    "##setup subplots, 12 curves so 4 columns by 3 rows for a landscape view\n",
    "index=0\n",
    "rows=3\n",
    "columns=4\n",
    "f, axes = plt.subplots(nrows=rows, ncols=columns, figsize=(12,10),sharey=True,sharex=True)\n",
    "plt.subplots_adjust(hspace=0.3)\n",
    "\n",
    "#per_sequence_run loop\n",
    "for seq_run in metadata_df['run_prefix'].unique():\n",
    "    f_table = filter_samples(collapsed_table,\n",
    "                                     metadata = metadata,\n",
    "                                     where= f\"run_prefix == '{seq_run}'\")\n",
    "    filtered_table_df = f_table.filtered_table.view(pd.DataFrame)\n",
    "    katharo, popt, pcov = katharoseq_fit_curve(filtered_table_df)\n",
    "    \n",
    "    # plot fit curve\n",
    "    x = np.linspace(0, 6, 50)\n",
    "    y = allosteric_sigmoid(x, *popt)\n",
    "\n",
    "    # plot the fit\n",
    "    axes[index//columns,index%columns].plot(katharo['log_deblur_reads'], katharo['correct_assign'], 'o', label='data')\n",
    "    axes[index//columns,index%columns].plot(x,y, label='fit')\n",
    "    axes[index//columns,index%columns].set_ylim([0,1.05])\n",
    "    axes[index//columns,index%columns].set_title(seq_run.replace('_S1_L001',\"\"))\n",
    "    axes[index//columns,index%columns].set_ylabel('Proportion of Reads on Target')\n",
    "    axes[index//columns,index%columns].set_xlabel('Log10 (Read Counts)')\n",
    "    \n",
    "    #Plot thresholding lines\n",
    "    axes[index//columns,index%columns].axvline(np.log10(katharo_read_threshold(popt,0.80)),ls='--')\n",
    "    axes[index//columns,index%columns].axhline(0.80,ls='--')\n",
    "    \n",
    "    \n",
    "    index +=1\n",
    "    \n",
    "#per_extraction_plate loop\n",
    "for sample_plate in metadata_df['sample_plate'].unique():\n",
    "    f_table = filter_samples(collapsed_table,\n",
    "                                     metadata = metadata,\n",
    "                                     where= f\"sample_plate == '{sample_plate}'\")\n",
    "    f_deblur = filter_samples(deblur_table,\n",
    "                                     metadata = metadata,\n",
    "                                     where= f\"sample_plate == '{sample_plate}'\")\n",
    "    plate_name = sample_plate.replace('3DMM_20220614_14542_','').replace('3DMM_20220623_14542_','')\n",
    "    filtered_table_df = f_table.filtered_table.view(pd.DataFrame)\n",
    "    katharo, popt, pcov = katharoseq_fit_curve(filtered_table_df,filter_out = problematic_samples)\n",
    "    \n",
    "    # plot fit curve\n",
    "    x = np.linspace(0, 6, 50)\n",
    "    y = allosteric_sigmoid(x, *popt)\n",
    "\n",
    "    # plot the fit\n",
    "    axes[index//columns,index%columns].plot(katharo['log_deblur_reads'], katharo['correct_assign'], 'o', label='data')\n",
    "    axes[index//columns,index%columns].plot(x,y, label='fit')\n",
    "    axes[index//columns,index%columns].set_ylim([0,1.05])\n",
    "    katharo_min_read = katharo_read_threshold(popt,0.80)\n",
    "    axes[index//columns,index%columns].set_title(plate_name + ' read_threshold: '\n",
    "                                                 + str(katharo_min_read))\n",
    "    axes[index//columns,index%columns].set_ylabel('Proportion of Reads on Target')\n",
    "    axes[index//columns,index%columns].set_xlabel('Log10 (Read Counts)')\n",
    "    \n",
    "    #plot thresholding lines\n",
    "    axes[index//columns,index%columns].axvline(np.log10(katharo_read_threshold(popt,0.80)),ls='--')\n",
    "    axes[index//columns,index%columns].axhline(0.80,ls='--')\n",
    "    \n",
    "    # apply katharoseq min_read_threshold per plate to deblur table. \n",
    "    katharo_filtered_plate = filter_samples(f_deblur.filtered_table,min_frequency=katharo_min_read)\n",
    "    katharo_filtered_plate.filtered_table.save('./data/katharo_filtered_deblur/151557_feature-table_' +\n",
    "                                               plate_name + '.qza')\n",
    "    index +=1\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In response to reviewer comments, we also plotted the KatharoSeq curves with a paired histogram displaying the distribution of samples and controls across the read depth (Log10[Read Counts])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "on_target_proportion = 0.80\n",
    "problematic_samples = ['150297.14542.KATHARO.P4.2','150297.14542.KATHARO.P2.4']\n",
    "\n",
    "for sample_plate in metadata_df['sample_plate'].unique():\n",
    "    ##setup subplots, 12 curves so 4 columns by 3 rows for a landscape view\n",
    "    index=0\n",
    "    rows=2\n",
    "    columns=1\n",
    "    f, (ax,ax2) = plt.subplots(nrows=rows, ncols=columns, figsize=(6,6),sharex=True,gridspec_kw={'height_ratios': [3, 1]})\n",
    "    f_table = filter_samples(collapsed_table,\n",
    "                                     metadata = metadata,\n",
    "                                     where= f\"sample_plate == '{sample_plate}'\")\n",
    "\n",
    "    plate_name = sample_plate.replace('3DMM_20220614_14542_','').replace('3DMM_20220623_14542_','')\n",
    "    #f_table.filtered_table.save('./data/per_plate/' + '150353_collapsed_table_' + plate_name+'.qza')\n",
    "    filtered_table_df = f_table.filtered_table.view(pd.DataFrame)\n",
    "    katharo, popt, pcov = katharoseq_fit_curve(filtered_table_df,filter_out = problematic_samples)\n",
    "    \n",
    "    # plot fit curve\n",
    "    x = np.linspace(0, 6, 50)\n",
    "    y = allosteric_sigmoid(x, *popt)\n",
    "\n",
    "    # plot the fit\n",
    "    ax.plot(katharo['log_deblur_reads'], katharo['correct_assign'], 'o', label='data')\n",
    "    ax.plot(x,y, label='fit')\n",
    "    ax.set_ylim([-.05,1.05])\n",
    "    katharo_min_read = katharo_read_threshold(popt,on_target_proportion)\n",
    "\n",
    "    ax.set_title(plate_name + ' read_threshold: '+ str(katharo_min_read))\n",
    "    ax.set_ylabel('Proportion of Reads on Target (Positive Control)')\n",
    "    ax.set_xlabel('Log10 (Read Counts)')\n",
    "    \n",
    "    #Plot thresholding lines\n",
    "    ax.axvline(np.log10(katharo_min_read),ls='--')\n",
    "    ax.axhline(on_target_proportion,ls='--')\n",
    "    \n",
    "    table_w_metadata =filtered_table_df.merge(metadata_df,left_index=True,right_index=True,how='left')\n",
    "    table_w_metadata = table_w_metadata.merge(pd.Series(filtered_table_df.sum(axis=1),name='read counts'),\n",
    "                                             left_index=True,right_index=True)\n",
    "    table_w_metadata['log_reads'] =np.log10(table_w_metadata['read counts'])\n",
    "    table_w_metadata.sort_values(by='sample_type',inplace=True)\n",
    "    #histogra,\n",
    "    sns.histplot(x='log_reads',hue='sample_type',data=table_w_metadata,ax=ax2)\n",
    "    ax2.axvline(np.log10(katharo_min_read),ls='--',color='red')\n",
    "    ax2.set_xlabel('Log10 (Read Counts)')\n",
    "    ax.spines[['right', 'top']].set_visible(False)\n",
    "    ax2.spines[['right', 'top']].set_visible(False)\n",
    "    plt.tight_layout()\n",
    "    plt.savefig(f'./figs/{sample_plate}.svg')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Merging per plate katharoseq filtered tables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "i = 0\n",
    "for file in glob.glob('./data/katharo_filtered_deblur/*.qza'):\n",
    "    if i > 0:\n",
    "        #Load new tables and merge them with growing_table\n",
    "        new_table = Artifact.load(file)\n",
    "        growing_table = merge([growing_table, new_table]).merged_table\n",
    "    else:\n",
    "        #initialize the growing table by loading the first .qza artifact\n",
    "        growing_table = Artifact.load(file)\n",
    "    i += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "katharo_deblur_gg = growing_table\n",
    "katharo_deblur_df = katharo_deblur_gg.view(pd.DataFrame)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "BLANK_katharo_deblur_df = katharo_deblur_df[katharo_deblur_df.index.str.contains('BLANK')].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TACAGAGGATGCAAGCGTTATCCGGAATGATTGGGCGTAAAGCGTCTGTAGGTGGCTTTTTAAGTCCGCCGTCAAATCCCAGGGCTCAACCCTGGACAGGCGGTGGAAACTACCAAGCTGGAGTACGGTAGGGGCAGAGGGAATTTCCGG    989870.0\n",
       "GACGGGGGGGGCAAGTGTTCTTCGGAATGACTGGGCGTAAAGGGCACGTAGGCGGTGAATCGGGTTGAAAGTGAAAGTCGCCAAAAACTGGTGGAATGCTCTCGAAACCAATTCACTTGAGTGAGACAGAGGAGAGTGGAATTTCGTGTG     84846.0\n",
       "TACGAAGGGGGCTAGCGTTGTTCGGAATTACTGGGCGTAAAGCGCACGTAGGCGGATATTTAAGTCAGGGGTGAAATCCCAGAGCTCAACTCTGGAACTGCCTTTGATACTGGGTATCTTGAGTATGGAAGAGGTAAGTGGAATTCCGAG      7825.0\n",
       "TACAGAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGCGCGTAGGTGGTTTGTTAAGTTGGATGTGAAAGCCCCGGGCTCAACCTGGGAACTGCATCCAAAACTGGCAAGCTAGAGTATGGTAGAGGGTGGTGGAATTTCCTG      4757.0\n",
       "TACGGAGGGTGCAAGCGTTATCCGGATTCACTGGGTTTAAAGGGTGCGTAGGCGGGTAGGTAAGTCAGAGGTGAAATCCTGGAGCTTAACTCCAGAACTGCCTTTGATACTATCTATCTTGAATATGGTGGAGGTAAGCGGAATATGTCA      3104.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "BLANK_katharo_deblur_df.sum(axis=0).sort_values(ascending=False).head(5)\n",
    "\n",
    "##Taxonomic assignment for sOTUS with highest read counts across all blanks(older greengenes [gg-13-8])\n",
    "## 1 k__Bacteria; p__Cyanobacteria; c__Chloroplast; o__Streptophyta; f__; g__; s__\n",
    "## 2 k__Bacteria; p__Proteobacteria; c__Alphaproteobacteria; o__Rickettsiales; f__mitochondria\n",
    "## 3 k__Bacteria; p__Proteobacteria; c__Alphaproteobacteria; o__Rhizobiales; f__Rhizobiaceae; g__Agrobacterium; s__\n",
    "## 4 k__Bacteria; p__Proteobacteria; c__Gammaproteobacteria; o__Pseudomonadales; f__Pseudomonadaceae; g__Pseudomonas\n",
    "## 5 k__Bacteria; p__Bacteroidetes; c__[Saprospirae]; o__[Saprospirales]; f__Chitinophagaceae; g__Sediminibacterium; s__\n",
    "\n",
    "##Taxonomic assignment for sOTUS with highest read counts across all blanks (greengenes2)\n",
    "## you can \"BLAST\" these sOTUs in the following website https://greengenes2.ucsd.edu\n",
    "\n",
    "## 1 d__Bacteria; p__Cyanobacteria; c__Cyanobacteriia; o__Cyanobacteriales; f__Coleofasciculaceae_23353; g__SIO2C1; s__SIO2C1 sp010672925\n",
    "## 2 d__Bacteria; p__Proteobacteria; c__Alphaproteobacteria; o__Sphingomonadales; f__Kordiimonadaceae_482971; g__Kordiimonas_482971; s__Kordiimonas sp002401685\n",
    "## 3 d__Bacteria; p__Proteobacteria; c__Alphaproteobacteria; o__Rhizobiales_A_500471; f__Rhizobiaceae_A_500471; g__Agrobacterium; s__Agrobacterium radiobacter\n",
    "## 4 d__Bacteria; p__Proteobacteria; c__Gammaproteobacteria; o__Pseudomonadales_650611; f__Pseudomonadaceae; g__Pseudomonas_E_647464; s__Pseudomonas_E_647464 cremoricolorata\n",
    "## 5 d__Bacteria; p__Bacteroidota; c__Bacteroidia; o__Chitinophagales; f__Chitinophagaceae_966727; g__Sediminibacterium; s__Sediminibacterium magnilacihabitans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "counts_table = pd.DataFrame(metadata_df['module'].value_counts())\n",
    "\n",
    "counts_table['katharoseq']= metadata_df.loc[metadata_df.index.isin(katharo_deblur_df.index),\n",
    "                                            'module'].value_counts()\n",
    "\n",
    "counts_table['% Retained'] = counts_table['katharoseq']/counts_table['module']\n",
    "\n",
    "#reordering the table to match other figures\n",
    "counts_table = counts_table.loc[['Airlock', 'Columbus', 'JLP', 'JPM', 'US Lab', 'Node 1', 'Node 2',\n",
    "       'Node 3', 'PMM']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "module_colors= {'Airlock':'tab:red',\n",
    "               'Columbus':'tab:blue',\n",
    "               'JLP':'tab:orange',\n",
    "               'JPM':'tab:green',\n",
    "               'US Lab':'cadetblue',\n",
    "               'Node 1':'tab:purple',\n",
    "               'Node 2':'yellow',\n",
    "               'Node 3':'tab:cyan',\n",
    "               'PMM':'tab:pink'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.barplot(x=counts_table.index,y='% Retained',data=counts_table,palette=module_colors)\n",
    "labels = ax.get_xticks()\n",
    "plt.xticks(rotation = 60,ha='center')\n",
    "plt.ylabel('Proportion of Samples Retained')\n",
    "plt.axhline(counts_table['% Retained'].mean(),linestyle='--',color='black')\n",
    "plt.savefig('./figs/katharo_retained.svg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = ['Airlock',\n",
    "         'Columbus',\n",
    "         'JLP',\n",
    "          'JPM',\n",
    "         'US Lab',\n",
    "         'Node 1',\n",
    "         'Node 2',\n",
    "         'Node 3',\n",
    "         'PMM']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Figure 1B"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "counts_table['filtered'] = counts_table['module']- counts_table['katharoseq']\n",
    "plt.bar(range(counts_table.shape[0]), counts_table['katharoseq'],color=module_colors.values())\n",
    "plt.bar(range(counts_table.shape[0]), counts_table['filtered'], bottom=counts_table['katharoseq']\n",
    "        , align='center',color='gray')\n",
    "\n",
    "\n",
    "plt.xticks(ticks=range(len(labels)),labels=labels,rotation=90)\n",
    "plt.tight_layout()\n",
    "plt.savefig('./figs/katharo_retained_fig_1B.svg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "803"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts_table['module'].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>module</th>\n",
       "      <th>katharoseq</th>\n",
       "      <th>% Retained</th>\n",
       "      <th>filtered</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Airlock</th>\n",
       "      <td>22</td>\n",
       "      <td>5</td>\n",
       "      <td>0.227273</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Columbus</th>\n",
       "      <td>70</td>\n",
       "      <td>23</td>\n",
       "      <td>0.328571</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JLP</th>\n",
       "      <td>38</td>\n",
       "      <td>21</td>\n",
       "      <td>0.552632</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JPM</th>\n",
       "      <td>102</td>\n",
       "      <td>30</td>\n",
       "      <td>0.294118</td>\n",
       "      <td>72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>US Lab</th>\n",
       "      <td>102</td>\n",
       "      <td>36</td>\n",
       "      <td>0.352941</td>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Node 1</th>\n",
       "      <td>130</td>\n",
       "      <td>66</td>\n",
       "      <td>0.507692</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Node 2</th>\n",
       "      <td>106</td>\n",
       "      <td>50</td>\n",
       "      <td>0.471698</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Node 3</th>\n",
       "      <td>163</td>\n",
       "      <td>113</td>\n",
       "      <td>0.693252</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PMM</th>\n",
       "      <td>70</td>\n",
       "      <td>49</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          module  katharoseq  % Retained  filtered\n",
       "Airlock       22           5    0.227273        17\n",
       "Columbus      70          23    0.328571        47\n",
       "JLP           38          21    0.552632        17\n",
       "JPM          102          30    0.294118        72\n",
       "US Lab       102          36    0.352941        66\n",
       "Node 1       130          66    0.507692        64\n",
       "Node 2       106          50    0.471698        56\n",
       "Node 3       163         113    0.693252        50\n",
       "PMM           70          49    0.700000        21"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "393"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Number of surface samples passing katharoseq filters\n",
    "counts_table['katharoseq'].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "counts_table_plate = pd.DataFrame(metadata_df['sample_plate'].value_counts())\n",
    "counts_table_plate['katharoseq']= metadata_df.loc[metadata_df.index.isin(katharo_deblur_df.index),'sample_plate'].value_counts()\n",
    "counts_table_plate['% Retained'] = counts_table_plate['katharoseq']/counts_table_plate['sample_plate']\n",
    "counts_table_plate.sort_values(by='% Retained',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 1.0)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors_ = sns.color_palette(palette='dark', n_colors=counts_table_plate.shape[0])\n",
    "sns.barplot(x=counts_table_plate.index,y='% Retained',data=counts_table_plate,palette='dark')\n",
    "ticks, labels = plt.xticks()\n",
    "plt.xticks(ticks=ticks, labels = ['Plate 2','Plate 8','Plate 6',\n",
    "                     'Plate 7','Plate 10','Plate 5',\n",
    "                     'Plate 9','Plate 4','Plate 3',\n",
    "                     'Plate 1'],rotation = 60,ha='right')\n",
    "plt.xlabel('Extraction Plate')\n",
    "plt.ylabel('Proportion of Samples Retained')\n",
    "plt.ylim(0,1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plate 1 gets fully filtered out. Upon further investigation, this was tracked to issues with automation infrastructure during nucleic acid extraction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Filtering feature table against Greengenes2 feature space. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Greengenes2 was released during the analysis of this study, so we had to bridge the analysis to have the most up to date feature space. In order to do that, a second feature table was downloaded from qiita (gg2_feature-table) and only the samples that passed KatharoSeq with greengenes (gg-13-8) were kept for further analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16026 x 960 <class 'biom.table.Table'> with 148219 nonzero entries (0% dense)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "##hit table greengenes2 filtered\n",
    "gg2_table = Artifact.load('./data/gg2_feature-table.qza')\n",
    "gg2_table.view(Table)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "26909 x 960 <class 'biom.table.Table'> with 162495 nonzero entries (0% dense)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "deblur_table.view(Table)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10883"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "deblur_table.view(Table).shape[0]-gg2_table.view(Table).shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.995139712162615"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#greengenes2 feature table is a nearly complete subset of the original gg-13-8 feature_table. \n",
    "gg2_table.view(pd.DataFrame).sum().sum()/deblur_table.view(pd.DataFrame).sum().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "there is a loss of rare features with low counts (possibly artifacts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "#deblur_table (greengenes [gg-13-8])\n",
    "deblur_features = set(deblur_table.view(pd.DataFrame).columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "gg2_features = set(gg2_table.view(pd.DataFrame).columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "#features exclusive to deblur \n",
    "deblur_exclusive = deblur_features - gg2_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10883"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(deblur_exclusive)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.004860287837384976"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#deblur_table (greengenes [gg-13-8])\n",
    "exlusive_deblur_gg_counts = deblur_table.view(pd.DataFrame)[deblur_exclusive].sum().sum()\n",
    "full_deblur_gg_counts = deblur_table.view(pd.DataFrame).sum().sum()\n",
    "\n",
    "exlusive_deblur_gg_counts/full_deblur_gg_counts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Less than 1% of the feature counts were lost to the features that are not represetned in greengenes2. We deemed this reasonable. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "we filtered the greengenes2 feature_table to keep the sample that passed katharoseq analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "keep_df = katharo_deblur_gg.view(pd.DataFrame)\n",
    "keep_df.index.names = ['#SampleID']\n",
    "# keep_df.to_csv('./metadata/2023_kataroseq_keep.tsv',\n",
    "#                                         sep='\\t', columns=[])\n",
    "keep = Metadata(keep_df)\n",
    "# filter the gg2_ft to only include katharoseq passing samples\n",
    "gg2_table = filter_samples(gg2_table,metadata=keep).filtered_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'./data/katharo_deblur_gg2_feature_table.qza'"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gg2_table.save('./data/katharo_deblur_gg2_feature_table.qza')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10892 x 464 <class 'biom.table.Table'> with 86054 nonzero entries (1% dense)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gg2_table.view(Table)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}