Figure_4D.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Per-genome rarefaction curves drawn using \n",
    "```\n",
    "scripts/190308_gorg_rarefaction_tails.py\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import pandas as pd\n",
    "import os.path as op\n",
    "import os\n",
    "from collections import Counter\n",
    "import random\n",
    "import numpy as np\n",
    "\n",
    "import seaborn as sns\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import sklearn.metrics\n",
    "import statistics\n",
    "import math\n",
    "\n",
    "def fit(df, x, y):\n",
    "    \n",
    "    def _check_fit(p, x, y):\n",
    "        predicted = [p(i) for i in x]\n",
    "        observed = y\n",
    "        residuals = [x-y for x, y in zip(predicted, observed)]\n",
    "        r2 = sklearn.metrics.r2_score(observed, predicted)\n",
    "        stdev = np.std(residuals)\n",
    "        return r2, stdev\n",
    "\n",
    "    x = [float(i) for i in df[x].values]\n",
    "    y = [float(i) for i in df[y].values]\n",
    "\n",
    "    z = np.polyfit(x, y, 1)\n",
    "    p = np.poly1d(z)\n",
    "    r2_order1, sd_order1 = _check_fit(p, x, y)\n",
    "    return p, r2_order1, sd_order1\n",
    "\n",
    "\n",
    "wd = \"/mnt/scgc/simon/simonsproject/gorg-clustering/\"\n",
    "os.chdir(wd)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"analyses/190308_tail_rarefaction_calculations.csv\")\n",
    "df1s = df[df['sag_start'] == 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "means = []\n",
    "sds = []\n",
    "groups = []\n",
    "for g in df1s['group'].unique():\n",
    "    slopes = []\n",
    "    for i in df1s[df1s['group'] == g]['iteration'].unique():\n",
    "\n",
    "        idf = df1s[(df1s['group'] == g) & (df1s['iteration'] == i)]\n",
    "        \n",
    "        last_10 = idf[-10:].copy()\n",
    "        last_10['count'] = np.arange(0, 10, 1)\n",
    "        f = fit(last_10, 'count','clusters')\n",
    "        slope = f[0][1]\n",
    "        slopes.append(slope)\n",
    "            \n",
    "    mean_slopes = statistics.mean(slopes)\n",
    "    sd_slopes = statistics.stdev(slopes)\n",
    "    groups.append(g)\n",
    "    means.append(mean_slopes)\n",
    "    sds.append(sd_slopes)\n",
    "\n",
    "rdf = pd.DataFrame(data={'group':groups, 'mean':means, 'stdev':sds})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "genome_count = df.groupby(['group'])['sag_number'].max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>group</th>\n",
       "      <th>sag_number</th>\n",
       "      <th>mean</th>\n",
       "      <th>stdev</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AEGEAN-169</td>\n",
       "      <td>413</td>\n",
       "      <td>47.594545</td>\n",
       "      <td>9.331591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ActinomarinaOM1</td>\n",
       "      <td>208</td>\n",
       "      <td>41.023030</td>\n",
       "      <td>5.179071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Arctic97B-4</td>\n",
       "      <td>13</td>\n",
       "      <td>740.872727</td>\n",
       "      <td>123.955019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Ascidiaceihabitans</td>\n",
       "      <td>12</td>\n",
       "      <td>707.263636</td>\n",
       "      <td>177.203825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Ca._Nitrosopumilales</td>\n",
       "      <td>15</td>\n",
       "      <td>176.323636</td>\n",
       "      <td>25.951460</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  group  sag_number        mean       stdev\n",
       "0            AEGEAN-169         413   47.594545    9.331591\n",
       "1       ActinomarinaOM1         208   41.023030    5.179071\n",
       "2           Arctic97B-4          13  740.872727  123.955019\n",
       "3    Ascidiaceihabitans          12  707.263636  177.203825\n",
       "4  Ca._Nitrosopumilales          15  176.323636   25.951460"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pdf = genome_count.reset_index().merge(rdf, on='group', how='right')\n",
    "pdf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "groups = pd.read_excel('/mnt/scgc/simon/simonsproject/info/GORG_16S_basicinfo_20190123.xlsx')\n",
    "mean_est_group_genome_size = groups.dropna(subset=['Estimated_genome_size']).groupby('Group_short')['Estimated_genome_size'].mean().reset_index().copy()\n",
    "mean_est_group_genome_size.rename(columns={'Group_short':'group'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pdf = pdf.merge(mean_est_group_genome_size, on='group', how='left').sort_values(by='sag_number', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sns.set_style(\"whitegrid\")\n",
    "gcols = {'AEGEAN-169': '#9b4ec2',\n",
    " 'ActinomarinaOM1': '#4cacdb',\n",
    " 'NS4': '#469057',\n",
    " 'NS5': '#5c6918',\n",
    " 'Prochlorococcus': '#5753c0',\n",
    " 'SAR116': '#d07ce1',\n",
    " 'SAR86': '#d044ab',\n",
    " 'Surface1_SAR11': '#a95f24',\n",
    " 'Surface4_SAR11': '#b13a33',\n",
    " 'Unclass_Rhodobacteraceae': '#4168a7',\n",
    " 'Unclass_SAR11': '#da3956',\n",
    "  'Other':'#A9A9A9'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pdf['log_sag_number'] = [math.log(i, 10) for i in pdf['sag_number']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2aaaf253c8d0>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnEAAAGGCAYAAAAU8tcEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEC5IkjghhBBCCBckSZwQQgghhAv6H+hGu5C4sRi8AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2aaae877eb00>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib.patches import Patch\n",
    "\n",
    "\n",
    "sizes = pdf['Estimated_genome_size'] * 100\n",
    "pdf['color'] = [gcols.get(i, '#A9A9A9') for i in pdf['group']]\n",
    "\n",
    "#for i, l in pdf.iterrows():\n",
    "#    plt.errorbar(l['sag_number'], l['mean'], yerr = l['stdev'], fmt=\",\", color = l['color'], alpha=1)\n",
    "\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.scatter('sag_number','mean', s=sizes, data=pdf, color = pdf['color'], alpha = 0.9)\n",
    "plt.xlim(10,)\n",
    "plt.ylim(0,)\n",
    "plt.xlabel('# of SAGs in Group')\n",
    "plt.ylabel('Clusters per SAG added\\nby Group')\n",
    "plt.xscale('log')\n",
    "\n",
    "pdf['Estimated_genome_size']\n",
    "size_legend = [1, 2, 3, 4] \n",
    "sizes = [i*100 for i in size_legend]\n",
    "labels = ['1 Mbp','2 Mbp','3 Mbp','4 Mbp']\n",
    "\n",
    "markers = []\n",
    "\n",
    "for s in sizes:\n",
    "    marker = plt.scatter([],[], s=s, color='#A9A9A9', alpha=0.9)\n",
    "    markers.append(marker)\n",
    "\n",
    "legend1 = plt.legend(handles=markers, labels=labels, loc='right',\n",
    "    scatterpoints=1, title = 'Average Genome Size')\n",
    "\n",
    "ax = plt.gca().add_artist(legend1)\n",
    "\n",
    "legend_elements = [Patch(facecolor=gcols[i], edgecolor=gcols[i], label=i) for i in gcols]\n",
    "plt.legend(handles=legend_elements, loc=0,\n",
    "          fancybox=False, shadow=False, frameon=False, ncol=2)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}