work/2_plasmodium_rebalancing/20191104_plasmodium_rebalancing_run9.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "READS = '../../../data/phylo_ampl_dada2/run9/dada2/output/stats.tsv'\n",
    "HAPS = '../../../data/phylo_ampl_dada2/run9/dada2/output/haplotypes.tsv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>s_Sample</th>\n",
       "      <th>target</th>\n",
       "      <th>input</th>\n",
       "      <th>filtered</th>\n",
       "      <th>denoisedF</th>\n",
       "      <th>denoisedR</th>\n",
       "      <th>merged</th>\n",
       "      <th>nonchim</th>\n",
       "      <th>final</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>21128</th>\n",
       "      <td>16_9d7</td>\n",
       "      <td>16</td>\n",
       "      <td>182</td>\n",
       "      <td>182</td>\n",
       "      <td>182</td>\n",
       "      <td>182</td>\n",
       "      <td>182</td>\n",
       "      <td>182</td>\n",
       "      <td>182</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      s_Sample target  input  filtered  denoisedF  denoisedR  merged  nonchim  \\\n",
       "21128   16_9d7     16    182       182        182        182     182      182   \n",
       "\n",
       "       final  \n",
       "21128    182  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reads = pd.read_csv(READS, sep='\\t')\n",
    "reads.sample()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(51392, 9)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(51200, 9)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# remove original mosquito samples\n",
    "display(reads.shape)\n",
    "reads = reads[~reads.s_Sample.str.startswith('A')]\n",
    "display(reads.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>s_Sample</th>\n",
       "      <th>input</th>\n",
       "      <th>filtered</th>\n",
       "      <th>denoisedF</th>\n",
       "      <th>denoisedR</th>\n",
       "      <th>merged</th>\n",
       "      <th>nonchim</th>\n",
       "      <th>final</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>154</th>\n",
       "      <td>13_0d5</td>\n",
       "      <td>79.983871</td>\n",
       "      <td>79.370968</td>\n",
       "      <td>79.209677</td>\n",
       "      <td>79.274194</td>\n",
       "      <td>78.354839</td>\n",
       "      <td>78.354839</td>\n",
       "      <td>78.048387</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    s_Sample      input   filtered  denoisedF  denoisedR     merged  \\\n",
       "154   13_0d5  79.983871  79.370968  79.209677  79.274194  78.354839   \n",
       "\n",
       "       nonchim      final target  \n",
       "154  78.354839  78.048387      M  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# mean read counts across mosquito targets\n",
    "mr = reads[~reads.target.str.startswith('P')].groupby('s_Sample').mean().reset_index()\n",
    "mr['target'] = 'M'\n",
    "mr.sample()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(51200, 9)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(1600, 9)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# subset to plasmodium targets only\n",
    "display(reads.shape)\n",
    "reads = reads[reads.target.str.startswith('P')]\n",
    "display(reads.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1600, 9)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(2400, 9)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# add averaged mosquito targets\n",
    "display(reads.shape)\n",
    "reads = pd.concat([reads, mr], sort=True)\n",
    "display(reads.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>denoisedF</th>\n",
       "      <th>denoisedR</th>\n",
       "      <th>filtered</th>\n",
       "      <th>final</th>\n",
       "      <th>input</th>\n",
       "      <th>merged</th>\n",
       "      <th>nonchim</th>\n",
       "      <th>s_Sample</th>\n",
       "      <th>target</th>\n",
       "      <th>pool</th>\n",
       "      <th>original_sample</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6654</th>\n",
       "      <td>41.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>12_0d4</td>\n",
       "      <td>P1</td>\n",
       "      <td>12</td>\n",
       "      <td>0d4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      denoisedF  denoisedR  filtered  final  input  merged  nonchim s_Sample  \\\n",
       "6654       41.0       41.0      41.0   41.0   41.0    41.0     41.0   12_0d4   \n",
       "\n",
       "     target pool original_sample  \n",
       "6654     P1   12             0d4  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# separate pool and original sample\n",
    "p = reads.s_Sample.str.split('_', n=1, expand=True)\n",
    "reads['pool'] = p[0]\n",
    "reads['original_sample'] = p[1]\n",
    "reads.sample()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert pool to two-digit string\n",
    "reads['pool'] = reads.pool.apply(lambda x: x.zfill(2))"
   ]
  },
  {
   "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>denoisedF</th>\n",
       "      <th>denoisedR</th>\n",
       "      <th>filtered</th>\n",
       "      <th>final</th>\n",
       "      <th>input</th>\n",
       "      <th>merged</th>\n",
       "      <th>nonchim</th>\n",
       "      <th>s_Sample</th>\n",
       "      <th>target</th>\n",
       "      <th>pool</th>\n",
       "      <th>original_sample</th>\n",
       "      <th>pool_target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>487</th>\n",
       "      <td>260.177419</td>\n",
       "      <td>260.177419</td>\n",
       "      <td>260.193548</td>\n",
       "      <td>259.419355</td>\n",
       "      <td>263.903226</td>\n",
       "      <td>259.467742</td>\n",
       "      <td>259.467742</td>\n",
       "      <td>3_ps10e4</td>\n",
       "      <td>M</td>\n",
       "      <td>03</td>\n",
       "      <td>ps10e4</td>\n",
       "      <td>03_M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      denoisedF   denoisedR    filtered       final       input      merged  \\\n",
       "487  260.177419  260.177419  260.193548  259.419355  263.903226  259.467742   \n",
       "\n",
       "        nonchim  s_Sample target pool original_sample pool_target  \n",
       "487  259.467742  3_ps10e4      M   03          ps10e4        03_M  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create pool-target combination\n",
    "reads['pool_target'] = reads['pool'] + '_' + reads['target']\n",
    "reads.sample()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# heatmap of final reads\n",
    "d = reads.pivot(index='pool_target', columns='original_sample', values='final').replace(0, np.nan)\n",
    "d = np.log10(d)\n",
    "fig, ax = plt.subplots(1,1, figsize=(12,12))\n",
    "sns.heatmap(d, ax=ax, cmap='coolwarm_r', vmin=1)\n",
    "ax.collections[0].colorbar.set_ticks(range(5));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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
}