Figure_3A_GORG-TARA_clustering.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import pandas as pd\n",
    "import os\n",
    "import os.path as op\n",
    "import math\n",
    "\n",
    "import matplotlib as m\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from collections import Counter\n",
    "\n",
    "cdir = '/mnt/scgc/simon/simonsproject/gorg-tara_clustering'\n",
    "os.chdir(cdir)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data prep:\n",
    "\n",
    "* Downloaded all tropics and subtropics coding sequences used in [this study](http://ocean-microbiome.embl.de/companion.html) that are between 40S and 40N from EBI.  \n",
    "\n",
    "* Translated all sequences from nt to aa sequences + add prefix to sequence names to indicate they came from TARA.  \n",
    "\n",
    "* Called proteins from all GORG SAGs using prodigal with the following parameters:\n",
    "```\n",
    "prodigal -i {input.nfa} -q -p meta \n",
    "```\n",
    "\n",
    "* add prefix to GORG sequence names to indicate they came from either the BATS248 (BATS248) sample, or Global Surface SAGs (GSS)\n",
    "\n",
    "* Combined TARA cds sequences and GORG sequences into one large fasta file (called gorg_tara_tropics.faa).\n",
    "\n",
    "* Run mmseq2 as such:\n",
    "\n",
    "```\n",
    "cd /mnt/scgc/simon/simonsproject/gorg-tara_clustering\n",
    "\n",
    "name=gorg_tara_tropics\n",
    "qual=80minid_m80\n",
    "\n",
    "faa=data/${name}.faa\n",
    "db=analyses/${name}.db\n",
    "db_clustered=analyses/${name}_${qual}.db_clustered\n",
    "rep_seqs=analyses/${name}_${qual}.reps\n",
    "out_fasta=analyses/${name}_${qual}.faa\n",
    "out_tsv=analyses/${name}_${qual}.tsv\n",
    "tmp_dir=data/tmp2/\n",
    "\n",
    "mkdir $tmp_dir\n",
    "\n",
    "threads=50\n",
    "\n",
    "mmseqs createdb $faa $db\n",
    "mmseqs linclust --threads $threads -v 1 --min-seq-id 0.6 --kmer-per-seq 80 $db $db_clustered $tmp_dir\n",
    "mmseqs result2repseq $db $db_clustered $rep_seqs  --threads $threads\n",
    "mmseqs result2flat $db $db $rep_seqs $out_fasta --use-fasta-header\n",
    "mmseqs createtsv $db $db $db_clustered $out_tsv --threads $threads -v 1\n",
    "```\n",
    "\n",
    "Which creates a number of output files, the one of interest is ```analyses/gorg_tara_tropics_80minid_m80.tsv```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How many clusters contain both TARA and GORG sequences, and what are their sizes?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "tsv_file = \"analyses/gorg_tara_tropics_80minid_m80.tsv\"\n",
    "\n",
    "# In output tsv file, the first column (c1) indicates the seed sequence for a given cluster.  \n",
    "# This will also serve as the cluster name.\n",
    "# The second column (c2) is a sequence that is a member of the cluster as indicated by c1\n",
    "tsv = pd.read_csv(tsv_file, sep=\"\\t\", names = ['c1','c2'])\n",
    "\n",
    "# extract source of sequence (TARA, GSS or BATS248)\n",
    "tsv['source'] = [i.split(\".\")[0].split(\"_\")[0] for i in tsv['c2']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calculate the number of sequences from each source in each cluster (source_count):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "source_count = tsv.groupby(['c1','source'])['c2'].count().reset_index().rename(columns={'c2':'seqs_included'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And pivot to count membership by cluster:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "scp = source_count.pivot(columns='source', index='c1', values='seqs_included')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Add count of all GORG sequences in cluster column and total sequences in each cluster."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "scp['gorg'] = scp['bats248'] + scp['gorgtropics']\n",
    "scp['member_count'] = scp['TARA'] + scp['gorg']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "scp = scp.fillna(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Print out some summary statistics:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 34855553 total clusters.\n",
      "There are 34351307 clusters with sequences from TARA metagenomes, that is 99 % of total clusters.\n",
      "There are 221401 clusters with sequences from GORG dataset, that is 1 % of total clusters.\n",
      "There are 22411436 singleton sequences from TARA metagenomes, that is 65 % of tara clusters and 64 % of all clusters.\n",
      "There are 0 singleton sequences from GORG dataset, that is 0 % of gorg clusters and 0 % of all clusters.\n",
      "There are 11733402 multi-sequence clusters only in TARA metagenomes, that is 34 % of tara clusters and 34 % of all clusters.\n",
      "There are 14932 multi-sequence clusters only in GORG dataset, that is 7 % of gorg clusters and 0.04 % of all clusters.\n",
      "There are 206469 clusters shared between TARA and GORG, that is 1 % of all clusters\n"
     ]
    }
   ],
   "source": [
    "total_tara_clusts = len(scp[scp['TARA'] > 0])\n",
    "total_gorg_clusts = len(scp[scp['gorg'] > 0])\n",
    "total_clusts = len(scp)\n",
    "\n",
    "print(\"There are\", total_clusts, \"total clusters.\")\n",
    "                        \n",
    "print(\"There are\", total_tara_clusts, \"clusters with sequences from TARA metagenomes,\", \n",
    "      \"that is\", round(total_tara_clusts / total_clusts * 100), \"% of total clusters.\")\n",
    "\n",
    "print(\"There are\", total_gorg_clusts, \"clusters with sequences from GORG dataset,\", \n",
    "      \"that is\", round(total_gorg_clusts / total_clusts * 100), \"% of total clusters.\")\n",
    "\n",
    "tara_singletons = len(scp[(scp['TARA'] == 1) & (scp['gorg'] == 0)])\n",
    "gorg_singletons = len(scp[(scp['gorg'] == 1) & (scp['TARA'] == 0)])\n",
    "\n",
    "print(\"There are\", tara_singletons, \"singleton sequences from TARA metagenomes,\", \"that is\", \n",
    "      round(tara_singletons / total_tara_clusts * 100), \"% of tara clusters and\",\n",
    "      round(tara_singletons / total_clusts * 100), \"% of all clusters.\")\n",
    "\n",
    "print(\"There are\", gorg_singletons, \"singleton sequences from GORG dataset,\",\n",
    "      \"that is\",\n",
    "     round(gorg_singletons / total_gorg_clusts * 100), \"% of gorg clusters and\",\n",
    "     round(gorg_singletons / total_clusts * 100), '% of all clusters.')\n",
    "\n",
    "tara_multi = len(scp[(scp['TARA'] > 1) & (scp['gorg'] == 0)])\n",
    "gorg_multi = len(scp[(scp['gorg'] > 1) & (scp['TARA'] == 0)])\n",
    "\n",
    "print(\"There are\", tara_multi, \"multi-sequence clusters only in TARA metagenomes,\",\n",
    "      \"that is\", \n",
    "      round(tara_multi / total_tara_clusts * 100), \"% of tara clusters and\",\n",
    "      round(tara_multi / total_clusts * 100), \"% of all clusters.\")\n",
    "\n",
    "print(\"There are\", gorg_multi, \"multi-sequence clusters only in GORG dataset,\",\"that is\",\n",
    "     round(gorg_multi / total_gorg_clusts * 100), \"% of gorg clusters and\",\n",
    "     round(gorg_multi / total_clusts * 100, 2), '% of all clusters.')\n",
    "\n",
    "shared_clusts = len(scp[(scp['TARA'] > 0) & (scp['gorg'] > 0)])\n",
    "\n",
    "print(\"There are\", shared_clusts, \"clusters shared between TARA and GORG,\",\n",
    "      \"that is\", round(shared_clusts / total_clusts * 100), \"% of all clusters\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "scp.to_csv(\"gorg_tara_tropics_80minid_kmer80_counts_by_source.csv\", index=True)"
   ]
  },
  {
   "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>source</th>\n",
       "      <th>TARA</th>\n",
       "      <th>bats248</th>\n",
       "      <th>gorgtropics</th>\n",
       "      <th>gorg</th>\n",
       "      <th>member_count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c1</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>TARA_004_DCM_0.22-1.6_C22365543_1_gene422493</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TARA_004_DCM_0.22-1.6_C22365545_1_gene422494</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TARA_004_DCM_0.22-1.6_C22365547_1_gene422497</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TARA_004_DCM_0.22-1.6_C22365549_1_gene422498</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TARA_004_DCM_0.22-1.6_C22365551_1_gene422500</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "source                                        TARA  bats248  gorgtropics  \\\n",
       "c1                                                                         \n",
       "TARA_004_DCM_0.22-1.6_C22365543_1_gene422493   1.0      0.0          0.0   \n",
       "TARA_004_DCM_0.22-1.6_C22365545_1_gene422494   1.0      0.0          0.0   \n",
       "TARA_004_DCM_0.22-1.6_C22365547_1_gene422497   1.0      0.0          0.0   \n",
       "TARA_004_DCM_0.22-1.6_C22365549_1_gene422498   1.0      0.0          0.0   \n",
       "TARA_004_DCM_0.22-1.6_C22365551_1_gene422500   1.0      0.0          0.0   \n",
       "\n",
       "source                                        gorg  member_count  \n",
       "c1                                                                \n",
       "TARA_004_DCM_0.22-1.6_C22365543_1_gene422493   0.0           0.0  \n",
       "TARA_004_DCM_0.22-1.6_C22365545_1_gene422494   0.0           0.0  \n",
       "TARA_004_DCM_0.22-1.6_C22365547_1_gene422497   0.0           0.0  \n",
       "TARA_004_DCM_0.22-1.6_C22365549_1_gene422498   0.0           0.0  \n",
       "TARA_004_DCM_0.22-1.6_C22365551_1_gene422500   0.0           0.0  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scp.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The above takes a while and requires a good amount of RAM.  I ran it, and then closed my notebook session for the day + reloaded the saved dataframe.\n",
    "\n",
    "Next Question: \n",
    "How are sequences from GORG and TARA distributed amongst clusters?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"/mnt/scgc/simon/simonsproject/gorg-tara_clustering/analyses/gorg_tara_tropics_80minid_kmer80_counts_by_source.csv\")[['c1','TARA','gorg']]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df['total'] = df[['TARA','gorg']].sum(axis=1)\n",
    "df['contains_gorg'] = [1 if i > 0 else 0 for i in df['gorg']]\n",
    "df['contains_tara'] = [1 if i > 0 else 0 for i in df['TARA']]\n",
    "\n",
    "binsplits = [0,1, 10, 100, 1000, 5000, 12000]\n",
    "df['bin'] = pd.cut(df['total'], bins=binsplits)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "tp = df.groupby(['bin','contains_gorg','contains_tara'])['total'].count().reset_index().rename(columns={'total':'bin_count'})\n",
    "ncd = dict(zip(list(tp['bin'].drop_duplicates()), ['1', '2-10','11-100','101-1000', '1001-5000','5001-12000']))\n",
    "tp['bin_label'] = [ncd[i] for i in tp['bin']]\n",
    "tp['gorg-presence'] = [\"GORG\" if i == 1 else \"\" for i in tp['contains_gorg']]\n",
    "tp['tara-presence'] = [\"Tara\" if i == 1 else \"\" for i in tp['contains_tara']]\n",
    "tp['membership'] = tp['gorg-presence'] + \" and \" + tp['tara-presence'] + ' containing'\n",
    "tp['membership'] = [i if i == 'GORG and Tara containing' else i.replace(\" and \",\"\") + \" only\" for i in tp['membership']]\n",
    "bin_sums = tp.groupby('bin')['bin_count'].sum().reset_index().rename(columns={'bin_count':'total_bin_count'})\n",
    "tp = tp.merge(bin_sums, on='bin', how='outer')\n",
    "tp['log_bin_count'] = [math.log(i, 10) for i in tp['bin_count']]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>bin</th>\n",
       "      <th>contains_gorg</th>\n",
       "      <th>contains_tara</th>\n",
       "      <th>bin_count</th>\n",
       "      <th>bin_label</th>\n",
       "      <th>gorg-presence</th>\n",
       "      <th>tara-presence</th>\n",
       "      <th>membership</th>\n",
       "      <th>total_bin_count</th>\n",
       "      <th>log_bin_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>(0, 1]</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>22263710</td>\n",
       "      <td>1</td>\n",
       "      <td></td>\n",
       "      <td>Tara</td>\n",
       "      <td>Tara containing only</td>\n",
       "      <td>22715792</td>\n",
       "      <td>7.347598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>(0, 1]</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>452082</td>\n",
       "      <td>1</td>\n",
       "      <td>GORG</td>\n",
       "      <td></td>\n",
       "      <td>GORG containing only</td>\n",
       "      <td>22715792</td>\n",
       "      <td>5.655217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(1, 10]</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10382961</td>\n",
       "      <td>2-10</td>\n",
       "      <td></td>\n",
       "      <td>Tara</td>\n",
       "      <td>Tara containing only</td>\n",
       "      <td>11021815</td>\n",
       "      <td>7.016321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>(1, 10]</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>51304</td>\n",
       "      <td>2-10</td>\n",
       "      <td>GORG</td>\n",
       "      <td></td>\n",
       "      <td>GORG containing only</td>\n",
       "      <td>11021815</td>\n",
       "      <td>4.710151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>(1, 10]</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>587550</td>\n",
       "      <td>2-10</td>\n",
       "      <td>GORG</td>\n",
       "      <td>Tara</td>\n",
       "      <td>GORG and Tara containing</td>\n",
       "      <td>11021815</td>\n",
       "      <td>5.769045</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       bin  contains_gorg  contains_tara  bin_count bin_label gorg-presence  \\\n",
       "0   (0, 1]              0              1   22263710         1                 \n",
       "1   (0, 1]              1              0     452082         1          GORG   \n",
       "2  (1, 10]              0              1   10382961      2-10                 \n",
       "3  (1, 10]              1              0      51304      2-10          GORG   \n",
       "4  (1, 10]              1              1     587550      2-10          GORG   \n",
       "\n",
       "  tara-presence                membership  total_bin_count  log_bin_count  \n",
       "0          Tara      Tara containing only         22715792       7.347598  \n",
       "1                    GORG containing only         22715792       5.655217  \n",
       "2          Tara      Tara containing only         11021815       7.016321  \n",
       "3                    GORG containing only         11021815       4.710151  \n",
       "4          Tara  GORG and Tara containing         11021815       5.769045  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/mnt/scgc_nfs/opt/common/anaconda3/4.0.0/lib/python3.5/site-packages/seaborn-0.7.1-py3.5.egg/seaborn/categorical.py:1468: FutureWarning: remove_na is deprecated and is a private function. Do not use.\n",
      "  stat_data = remove_na(group_data[hue_mask])\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2ab0f28cf860>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAekAAAFSCAYAAADWwychAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2ab0f27f5f60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_style('whitegrid')\n",
    "fig = plt.figure(figsize=(8,6))\n",
    "gs = m.gridspec.GridSpec(2,1,height_ratios=[10,1])\n",
    "ax = plt.subplot(gs[0])\n",
    "\n",
    "colors = [\"#44A8DE\",\n",
    "\"#8e5db0\",\n",
    "\"#729b57\"\n",
    "]\n",
    "\n",
    "sns.barplot(x = 'bin_label',\n",
    "            y = 'log_bin_count',\n",
    "            hue='membership', \n",
    "            data=tp, \n",
    "            hue_order=['GORG containing only', 'GORG and Tara containing','Tara containing only'],\n",
    "           palette = colors)\n",
    "\n",
    "\n",
    "plt.xlabel('Cluster Size')\n",
    "plt.ylabel('Log Total Number of Sequences')\n",
    "plt.title('Sequence Distribution by Cluster Size')\n",
    "plt.legend(loc=1)\n",
    "\n",
    "#plt.savefig(op.join(plotdir, \"gorg_tara_seq_clust_count_80id_80k.png\"), dpi=300)\n",
    "#plt.savefig(op.join(plotdir, \"gorg_tara_seq_clust_count_80id_80k.svg\"))"
   ]
  },
  {
   "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
}