notebooks/LCOResultExploration.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def load_test_prediction(model_name, class_name):\n",
    "    path = '../data/evaluation/lco-neg-10k/{}/{}.seed0.test.csv'.format(model_name, class_name)\n",
    "    domains = pd.read_csv(path, low_memory=False)\n",
    "    domains = domains.drop(['bitscore','evalue','domain_start','domain_end','gene_strand', 'locus_tag'], axis=1)\n",
    "    return domains\n",
    "\n",
    "def load_train_class(class_name):\n",
    "    path = '../data/evaluation/lco-neg-10k/splits/{}.seed0.train.csv'.format(class_name)\n",
    "    domains = pd.read_csv(path, low_memory=False)\n",
    "    domains = domains.drop(['bitscore','evalue','domain_start','domain_end','gene_strand', 'locus_tag'], axis=1)\n",
    "    return domains"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>locus_tag</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>gene_start</th>\n",
       "      <th>gene_end</th>\n",
       "      <th>gene_strand</th>\n",
       "      <th>pfam_id</th>\n",
       "      <th>domain_start</th>\n",
       "      <th>domain_end</th>\n",
       "      <th>evalue</th>\n",
       "      <th>bitscore</th>\n",
       "      <th>in_cluster</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BGC0000001.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>AEK75490.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1083</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13649</td>\n",
       "      <td>0</td>\n",
       "      <td>73</td>\n",
       "      <td>5.000000e-08</td>\n",
       "      <td>33.4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BGC0000001.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>AEK75490.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1083</td>\n",
       "      <td>1</td>\n",
       "      <td>PF08241</td>\n",
       "      <td>0</td>\n",
       "      <td>51</td>\n",
       "      <td>2.300000e-03</td>\n",
       "      <td>18.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BGC0000001.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>AEK75490.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1083</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13847</td>\n",
       "      <td>3</td>\n",
       "      <td>63</td>\n",
       "      <td>1.100000e-09</td>\n",
       "      <td>38.2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BGC0000001.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>AEK75490.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1083</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13489</td>\n",
       "      <td>12</td>\n",
       "      <td>62</td>\n",
       "      <td>5.100000e-04</td>\n",
       "      <td>19.8</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BGC0000001.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>AEK75490.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1083</td>\n",
       "      <td>1</td>\n",
       "      <td>PF02353</td>\n",
       "      <td>35</td>\n",
       "      <td>109</td>\n",
       "      <td>9.000000e-03</td>\n",
       "      <td>15.2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      contig_id          locus_tag  protein_id  gene_start  gene_end  \\\n",
       "0  BGC0000001.1  unknown_locus_tag  AEK75490.1           0      1083   \n",
       "1  BGC0000001.1  unknown_locus_tag  AEK75490.1           0      1083   \n",
       "2  BGC0000001.1  unknown_locus_tag  AEK75490.1           0      1083   \n",
       "3  BGC0000001.1  unknown_locus_tag  AEK75490.1           0      1083   \n",
       "4  BGC0000001.1  unknown_locus_tag  AEK75490.1           0      1083   \n",
       "\n",
       "   gene_strand  pfam_id  domain_start  domain_end        evalue  bitscore  \\\n",
       "0            1  PF13649             0          73  5.000000e-08      33.4   \n",
       "1            1  PF08241             0          51  2.300000e-03      18.5   \n",
       "2            1  PF13847             3          63  1.100000e-09      38.2   \n",
       "3            1  PF13489            12          62  5.100000e-04      19.8   \n",
       "4            1  PF02353            35         109  9.000000e-03      15.2   \n",
       "\n",
       "   in_cluster  \n",
       "0           1  \n",
       "1           1  \n",
       "2           1  \n",
       "3           1  \n",
       "4           1  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "domains = pd.read_csv('../data/training/positive/mibig_bgcs_all.csv')\n",
    "domains = domains[domains['evalue'] <= 0.01]\n",
    "domains.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "neg = pd.read_csv('../data/training/negative/geneswap_negatives.csv')\n",
    "neg_pfams = set(neg['pfam_id'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>BGC_ID</th>\n",
       "      <th>num_proteins</th>\n",
       "      <th>classes</th>\n",
       "      <th>source</th>\n",
       "      <th>species</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>contig_id</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>BGC0000001.1</th>\n",
       "      <td>BGC0000001</td>\n",
       "      <td>28</td>\n",
       "      <td>Polyketide</td>\n",
       "      <td>Verrucosispora maris AB-18-032</td>\n",
       "      <td>Verrucosispora</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000002.1</th>\n",
       "      <td>BGC0000002</td>\n",
       "      <td>34</td>\n",
       "      <td>Polyketide</td>\n",
       "      <td>Kutzneria albida DSM 43870</td>\n",
       "      <td>Kutzneria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000003.1</th>\n",
       "      <td>BGC0000003</td>\n",
       "      <td>6</td>\n",
       "      <td>Polyketide</td>\n",
       "      <td>Alternaria alternata</td>\n",
       "      <td>Alternaria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000004.1</th>\n",
       "      <td>BGC0000004</td>\n",
       "      <td>23</td>\n",
       "      <td>Polyketide</td>\n",
       "      <td>Aspergillus oryzae</td>\n",
       "      <td>Aspergillus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000005.1</th>\n",
       "      <td>BGC0000005</td>\n",
       "      <td>1</td>\n",
       "      <td>Polyketide</td>\n",
       "      <td>Aspergillus parasiticus</td>\n",
       "      <td>Aspergillus</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  BGC_ID  num_proteins     classes  \\\n",
       "contig_id                                            \n",
       "BGC0000001.1  BGC0000001            28  Polyketide   \n",
       "BGC0000002.1  BGC0000002            34  Polyketide   \n",
       "BGC0000003.1  BGC0000003             6  Polyketide   \n",
       "BGC0000004.1  BGC0000004            23  Polyketide   \n",
       "BGC0000005.1  BGC0000005             1  Polyketide   \n",
       "\n",
       "                                      source         species  \n",
       "contig_id                                                     \n",
       "BGC0000001.1  Verrucosispora maris AB-18-032  Verrucosispora  \n",
       "BGC0000002.1      Kutzneria albida DSM 43870       Kutzneria  \n",
       "BGC0000003.1            Alternaria alternata      Alternaria  \n",
       "BGC0000004.1              Aspergillus oryzae     Aspergillus  \n",
       "BGC0000005.1         Aspergillus parasiticus     Aspergillus  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "properties = pd.read_csv('../data/mibig/mibig_properties.csv').set_index('contig_id')\n",
    "properties.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "206"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ripp_ids = set(properties[properties['classes'] == 'RiPP'].index)\n",
    "len(ripp_ids)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Polyketide    434\n",
      "NRP           247\n",
      "Saccharide    128\n",
      "Terpene        91\n",
      "Alkaloid       28\n",
      "Name: classes, dtype: int64\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "928"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "other_classes = ['Alkaloid','NRP','Polyketide','Saccharide','Terpene'] \n",
    "other_bgcs = properties[properties['classes'].apply(lambda c: c in other_classes)]\n",
    "other_ids = set(other_bgcs.index)\n",
    "print(other_bgcs['classes'].value_counts())\n",
    "len(other_ids)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "ripp_index = domains['contig_id'].apply(lambda contig_id: contig_id in ripp_ids)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>locus_tag</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>gene_start</th>\n",
       "      <th>gene_end</th>\n",
       "      <th>gene_strand</th>\n",
       "      <th>pfam_id</th>\n",
       "      <th>domain_start</th>\n",
       "      <th>domain_end</th>\n",
       "      <th>evalue</th>\n",
       "      <th>bitscore</th>\n",
       "      <th>in_cluster</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>32238</th>\n",
       "      <td>BGC0000468.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>CCM09440.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1338</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF07690</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "      <td>2.000000e-13</td>\n",
       "      <td>49.9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32239</th>\n",
       "      <td>BGC0000468.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>CCM09440.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1338</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF13347</td>\n",
       "      <td>9</td>\n",
       "      <td>410</td>\n",
       "      <td>1.300000e-05</td>\n",
       "      <td>23.9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32240</th>\n",
       "      <td>BGC0000468.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>CCM09440.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1338</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF05977</td>\n",
       "      <td>19</td>\n",
       "      <td>401</td>\n",
       "      <td>6.000000e-14</td>\n",
       "      <td>51.2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32241</th>\n",
       "      <td>BGC0000468.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>CCM09440.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1338</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF00083</td>\n",
       "      <td>47</td>\n",
       "      <td>166</td>\n",
       "      <td>5.800000e-06</td>\n",
       "      <td>25.3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32242</th>\n",
       "      <td>BGC0000468.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>CCM09441.1</td>\n",
       "      <td>1791</td>\n",
       "      <td>2631</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF04072</td>\n",
       "      <td>4</td>\n",
       "      <td>186</td>\n",
       "      <td>3.100000e-20</td>\n",
       "      <td>72.8</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          contig_id          locus_tag  protein_id  gene_start  gene_end  \\\n",
       "32238  BGC0000468.1  unknown_locus_tag  CCM09440.1           0      1338   \n",
       "32239  BGC0000468.1  unknown_locus_tag  CCM09440.1           0      1338   \n",
       "32240  BGC0000468.1  unknown_locus_tag  CCM09440.1           0      1338   \n",
       "32241  BGC0000468.1  unknown_locus_tag  CCM09440.1           0      1338   \n",
       "32242  BGC0000468.1  unknown_locus_tag  CCM09441.1        1791      2631   \n",
       "\n",
       "       gene_strand  pfam_id  domain_start  domain_end        evalue  bitscore  \\\n",
       "32238           -1  PF07690             1         221  2.000000e-13      49.9   \n",
       "32239           -1  PF13347             9         410  1.300000e-05      23.9   \n",
       "32240           -1  PF05977            19         401  6.000000e-14      51.2   \n",
       "32241           -1  PF00083            47         166  5.800000e-06      25.3   \n",
       "32242           -1  PF04072             4         186  3.100000e-20      72.8   \n",
       "\n",
       "       in_cluster  \n",
       "32238           1  \n",
       "32239           1  \n",
       "32240           1  \n",
       "32241           1  \n",
       "32242           1  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ripp_domains = domains[ripp_index]\n",
    "ripp_domains.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>locus_tag</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>gene_start</th>\n",
       "      <th>gene_end</th>\n",
       "      <th>gene_strand</th>\n",
       "      <th>pfam_id</th>\n",
       "      <th>domain_start</th>\n",
       "      <th>domain_end</th>\n",
       "      <th>evalue</th>\n",
       "      <th>bitscore</th>\n",
       "      <th>in_cluster</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BGC0000001.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>AEK75490.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1083</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13649</td>\n",
       "      <td>0</td>\n",
       "      <td>73</td>\n",
       "      <td>5.000000e-08</td>\n",
       "      <td>33.4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BGC0000001.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>AEK75490.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1083</td>\n",
       "      <td>1</td>\n",
       "      <td>PF08241</td>\n",
       "      <td>0</td>\n",
       "      <td>51</td>\n",
       "      <td>2.300000e-03</td>\n",
       "      <td>18.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BGC0000001.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>AEK75490.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1083</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13847</td>\n",
       "      <td>3</td>\n",
       "      <td>63</td>\n",
       "      <td>1.100000e-09</td>\n",
       "      <td>38.2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BGC0000001.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>AEK75490.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1083</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13489</td>\n",
       "      <td>12</td>\n",
       "      <td>62</td>\n",
       "      <td>5.100000e-04</td>\n",
       "      <td>19.8</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BGC0000001.1</td>\n",
       "      <td>unknown_locus_tag</td>\n",
       "      <td>AEK75490.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1083</td>\n",
       "      <td>1</td>\n",
       "      <td>PF02353</td>\n",
       "      <td>35</td>\n",
       "      <td>109</td>\n",
       "      <td>9.000000e-03</td>\n",
       "      <td>15.2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      contig_id          locus_tag  protein_id  gene_start  gene_end  \\\n",
       "0  BGC0000001.1  unknown_locus_tag  AEK75490.1           0      1083   \n",
       "1  BGC0000001.1  unknown_locus_tag  AEK75490.1           0      1083   \n",
       "2  BGC0000001.1  unknown_locus_tag  AEK75490.1           0      1083   \n",
       "3  BGC0000001.1  unknown_locus_tag  AEK75490.1           0      1083   \n",
       "4  BGC0000001.1  unknown_locus_tag  AEK75490.1           0      1083   \n",
       "\n",
       "   gene_strand  pfam_id  domain_start  domain_end        evalue  bitscore  \\\n",
       "0            1  PF13649             0          73  5.000000e-08      33.4   \n",
       "1            1  PF08241             0          51  2.300000e-03      18.5   \n",
       "2            1  PF13847             3          63  1.100000e-09      38.2   \n",
       "3            1  PF13489            12          62  5.100000e-04      19.8   \n",
       "4            1  PF02353            35         109  9.000000e-03      15.2   \n",
       "\n",
       "   in_cluster  \n",
       "0           1  \n",
       "1           1  \n",
       "2           1  \n",
       "3           1  \n",
       "4           1  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nonripp_domains = domains[domains['contig_id'].apply(lambda contig_id: contig_id in other_ids)]\n",
    "nonripp_domains.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2634"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nonripp_pfam_counts = nonripp_domains['pfam_id'].value_counts()\n",
    "nonripp_pfam_set = set(nonripp_pfam_counts[nonripp_pfam_counts > 0].index)\n",
    "len(nonripp_pfam_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "ripp_samples = ripp_domains.groupby('contig_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "199"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Not all ripps have known pfam_IDs\n",
    "len(ripp_samples)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Comparing RiPPs to other BGCs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>domains_common</th>\n",
       "      <th>domains_total</th>\n",
       "      <th>domains_common_percent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BGC0000468.1</td>\n",
       "      <td>16</td>\n",
       "      <td>17</td>\n",
       "      <td>94.117647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BGC0000469.1</td>\n",
       "      <td>41</td>\n",
       "      <td>52</td>\n",
       "      <td>78.846154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BGC0000470.1</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BGC0000471.1</td>\n",
       "      <td>14</td>\n",
       "      <td>15</td>\n",
       "      <td>93.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BGC0000472.1</td>\n",
       "      <td>11</td>\n",
       "      <td>16</td>\n",
       "      <td>68.750000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      contig_id  domains_common  domains_total  domains_common_percent\n",
       "0  BGC0000468.1              16             17               94.117647\n",
       "1  BGC0000469.1              41             52               78.846154\n",
       "2  BGC0000470.1              11             11              100.000000\n",
       "3  BGC0000471.1              14             15               93.333333\n",
       "4  BGC0000472.1              11             16               68.750000"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "similarity = []\n",
    "for contig_id, sample_domains in ripp_samples:\n",
    "    sample_pfam_set = set(sample_domains['pfam_id'].unique())\n",
    "    similarity.append({\n",
    "        'contig_id': contig_id,\n",
    "        'domains_total': len(sample_pfam_set),\n",
    "        'domains_common': len(sample_pfam_set.intersection(nonripp_pfam_set))\n",
    "    })\n",
    "similarity = pd.DataFrame(similarity)\n",
    "similarity['domains_common_percent'] = similarity['domains_common'] / similarity['domains_total'] * 100\n",
    "similarity.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>domains_common</th>\n",
       "      <th>domains_total</th>\n",
       "      <th>domains_common_percent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>144</th>\n",
       "      <td>BGC0000618.1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>BGC0000485.1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>106</th>\n",
       "      <td>BGC0000580.1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>50.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>187</th>\n",
       "      <td>BGC0001398.15</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196</th>\n",
       "      <td>BGC0001398.7</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>BGC0001398.4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>192</th>\n",
       "      <td>BGC0001398.2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>190</th>\n",
       "      <td>BGC0001398.18</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>189</th>\n",
       "      <td>BGC0001398.17</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>188</th>\n",
       "      <td>BGC0001398.16</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         contig_id  domains_common  domains_total  domains_common_percent\n",
       "144   BGC0000618.1               0              3                     0.0\n",
       "14    BGC0000485.1               0              2                     0.0\n",
       "106   BGC0000580.1               1              2                    50.0\n",
       "187  BGC0001398.15               1              1                   100.0\n",
       "196   BGC0001398.7               1              1                   100.0\n",
       "194   BGC0001398.4               1              1                   100.0\n",
       "192   BGC0001398.2               1              1                   100.0\n",
       "190  BGC0001398.18               1              1                   100.0\n",
       "189  BGC0001398.17               1              1                   100.0\n",
       "188  BGC0001398.16               1              1                   100.0"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "different = similarity[similarity['domains_common'] <= 1].sort_values(by='domains_total', ascending=False)\n",
    "different.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# RiPP LCO"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>gene_end</th>\n",
       "      <th>gene_start</th>\n",
       "      <th>in_cluster</th>\n",
       "      <th>pfam_id</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2</td>\n",
       "      <td>1286</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>PF07690</td>\n",
       "      <td>CP002046_1414</td>\n",
       "      <td>0.102327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2</td>\n",
       "      <td>2392</td>\n",
       "      <td>1286</td>\n",
       "      <td>0</td>\n",
       "      <td>PF00817</td>\n",
       "      <td>CP002046_558</td>\n",
       "      <td>0.047054</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2</td>\n",
       "      <td>2392</td>\n",
       "      <td>1286</td>\n",
       "      <td>0</td>\n",
       "      <td>PF11798</td>\n",
       "      <td>CP002046_558</td>\n",
       "      <td>0.036830</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2</td>\n",
       "      <td>2392</td>\n",
       "      <td>1286</td>\n",
       "      <td>0</td>\n",
       "      <td>PF11799</td>\n",
       "      <td>CP002046_558</td>\n",
       "      <td>0.031875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2</td>\n",
       "      <td>2392</td>\n",
       "      <td>1286</td>\n",
       "      <td>0</td>\n",
       "      <td>PF10391</td>\n",
       "      <td>CP002046_558</td>\n",
       "      <td>0.031067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2</td>\n",
       "      <td>3288</td>\n",
       "      <td>2392</td>\n",
       "      <td>0</td>\n",
       "      <td>PF12388</td>\n",
       "      <td>CP002046_2647</td>\n",
       "      <td>0.021486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2</td>\n",
       "      <td>3288</td>\n",
       "      <td>2392</td>\n",
       "      <td>0</td>\n",
       "      <td>PF13582</td>\n",
       "      <td>CP002046_2647</td>\n",
       "      <td>0.015459</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2</td>\n",
       "      <td>3288</td>\n",
       "      <td>2392</td>\n",
       "      <td>0</td>\n",
       "      <td>PF10026</td>\n",
       "      <td>CP002046_2647</td>\n",
       "      <td>0.014615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2</td>\n",
       "      <td>3288</td>\n",
       "      <td>2392</td>\n",
       "      <td>0</td>\n",
       "      <td>PF13688</td>\n",
       "      <td>CP002046_2647</td>\n",
       "      <td>0.011562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2</td>\n",
       "      <td>3288</td>\n",
       "      <td>2392</td>\n",
       "      <td>0</td>\n",
       "      <td>PF13574</td>\n",
       "      <td>CP002046_2647</td>\n",
       "      <td>0.012259</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                contig_id  gene_end  gene_start  in_cluster  \\\n",
       "0  NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2      1286           0           0   \n",
       "1  NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2      2392        1286           0   \n",
       "2  NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2      2392        1286           0   \n",
       "3  NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2      2392        1286           0   \n",
       "4  NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2      2392        1286           0   \n",
       "5  NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2      3288        2392           0   \n",
       "6  NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2      3288        2392           0   \n",
       "7  NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2      3288        2392           0   \n",
       "8  NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2      3288        2392           0   \n",
       "9  NEG_FAKE_CLUSTER|CP002046.1|FJ477836.2      3288        2392           0   \n",
       "\n",
       "   pfam_id     protein_id  prediction  \n",
       "0  PF07690  CP002046_1414    0.102327  \n",
       "1  PF00817   CP002046_558    0.047054  \n",
       "2  PF11798   CP002046_558    0.036830  \n",
       "3  PF11799   CP002046_558    0.031875  \n",
       "4  PF10391   CP002046_558    0.031067  \n",
       "5  PF12388  CP002046_2647    0.021486  \n",
       "6  PF13582  CP002046_2647    0.015459  \n",
       "7  PF10026  CP002046_2647    0.014615  \n",
       "8  PF13688  CP002046_2647    0.011562  \n",
       "9  PF13574  CP002046_2647    0.012259  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ripp_lco = load_test_prediction('128lstm-100pfamdim-8pfamiter-posweighted', 'RiPP')\n",
    "ripp_lco.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Similarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "199"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(ripp_lco[ripp_lco['in_cluster'].astype(np.bool)]['contig_id'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BGC0000618.1\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZIAAACcCAYAAABV/y0+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAGXRJREFUeJzt3XmUXVWZ9/Hvj4SEKWEIwQZCqDAophWwiaCIAyLIHGaD4EIBgVZeRewFeV+VWRtopBsFgcgggkAAoVcCYWgkIKERkyBTDIESQhMiEKZMkA6Jz/vH3lcON3XvPamqW8VN/T5r3VVn2Hvfpwpyn7v3PmcfRQRmZmadtVpvB2BmZq3NicTMzLrEicTMzLrEicTMzLrEicTMzLrEicTMzLrEicSsxUg6XtJ/dFNbX5c0pTvaWsn3/X+SruhiGx+SNFPSwO6KyzrHicQ6RdJsSe9IWiTpTUl3SNqsqswoSbfn829J+rOkH0tav1BmY0lXSvqrpIWSnpZ0pqS18/k2SZMlvZ3PfanqPb4n6WVJ8yVdVfxQ6UrdfP67kp6XtDh/YH24EPMESXMlhaS2qnobSBov6bX8+o2kwTX+jgMk3ZL/niHpCw3+7gOAHwL/lvc3lPSQpNfz3/hhSZ+p18YHQUT8JCKO7WIbrwCTgeO6JyrrLCcS64r9ImIdYGPgFeDnlROSdgbuBx4CtomI9YA9gWXAdrnMBsDDwJrApyNiELA7sB6wZW7qBuBPwBDgB8Atkobm+l8GxgK7AW3AFsCZhfg6XVfSscAxwD7AOsC+wGv59N+Au4CDa/xdzgHWz21uCXwIOKNGWYApwJHAy3XKVIwGno6Il/L+IuBoYGh+z/OAiZL6l2hrVfAb4PjeDqLPiwi//FrpFzAb+FJhf2/gmcL+FODnDdo4B3gSWK3G+Q8D/wsMKhx7EDghb18P/KRwbjfg5W6ouxrwIrBbg/j7AwG0VR2/E/hWYf/bwN0l/qZzgC80KHMV8MMa51YD9ssxbVSjzBBgArAA+CNwNjClcH5nYCowP//cuXDu/vzf7L9JCWxibu83ub2pxb8FcFH+Oy4ApgOfLZw7A7gub7flmI8C/oeUsH9QKLsjMC238wpwYdV/g7eBzXv730RffrlHYl0maS3gK8Af8v7awKeB3zao+iXg1oj4W43z/wg8FxELC8cez8cr5x+vOvchSUO6WHdYfn1M0ot5eOtMSWX/vVwC7Ctp/TyMdzApuXSHjwOzqg9KegJYQkoSV0TEq3ViW0LqRR6dX5U2NgDuAH5GShAXAnfkv0nFGOBrwKak3tbDwNXABsBM4PRC2anA9vnc9cDNktao87vtAnyElNRPk/TRfPwi4KKIGJzf86ZKhYhYBrSTe7nWO5xIrCv+U9JbpG+Ku5PH7UlDLKtRGKqRdH4ew18s6Yf58BDgr3XaX4f0zbhoPjCoxvnK9qAu1h2Wt/cgfXDvChxOGuoq41FgAPB6fi0HflGybiPrAQurD0bEtsBg4Kuk3uAKJPUjJbXTImJxRDwFXFMosg/wbERcGxHLIuIG4GlSL6fi6oj4S0TMJyXHv0TEvfkD/WbgE4WYrouI13NbPwUGkhJFLWdGxDsR8TgpsVeSw7vAVpI2jIhFEfGHqnoL89/FeokTiXXFAZHmPgYCJwIPSPoH4E3SPMLGlYIRcUouextpOALSh+zG1LaI9OFYNJj3Pkirz1e2F3ax7jt5+/yIeCsiZgOXk4bvyrgZeIaUlAYDfwGuK1m3kTd5Lxm+T0QsyR/+YyV19A19KOlv/2Lh2AuF7U2q9ivnNy3sv1LYfqeD/XUqO5K+ny9SmJ+/cKwLbNjhb5UU54jeLrR1DGmo8mlJUyXtW1VvEPBWnXatyRomktzdNaspIpZHxK2kb967RMRi4BHgoAZV7wUOrDNkNAPYQlLxg3O7fLxyfruqc69ExOtdrDsLWEoat++M7YDL87f+RcBllE9CjTxB+lCtZ3XSRH+1eaSLHYpX1w0vbM8FNq+qMxx4iZUk6bPAqcBhwPr5S8R8QCvbVkQ8GxGHAxuRLia4pXBVX39gK94/TGk9rEyP5BFJN0vaW9JK/09gqz4lo0lDWjPz4VOAoyWNlbRRLjcMGFGoeiHpG/s1kjbPZTaVdKGkbSPiGeAx4HRJa0g6ENiW9+Zefg0cI2lknov4IfArgC7WfRsYD5wiaVCO+5vA7YXfeQ1STwxgYNXY/1TgWElrSlqTdHlqzQ86ScX6A3K8tf6tTQI+X6j7KUm75MuI15R0KukqsUeqK0bEcuBW4AxJa0kaSZrgLrb9YUlfldRf0leAkcXfeyUMIiWteUB/SaexYg+xFElHShqa59IqPY/l+eeOwOyIqO5JWQ8qk0g+DIwjTbC1S/pJ5Xp66/MmSlpEmiP5MXBURMwAiIgpwBeBzwHP5KGNu0hX/vw8l3mDdJXQu6QvLAuB35G+ubbn9xgDjCIN6ZwLHBIR83L9u4DzSfcSvJBfxcnertQ9kTT8NZc0oXw96YqpinfyeUjzCO8Uzh1NuhJpDunb/BbA1ysnJc2QdESh/Kxcf1Pg7rxd3TOomAhsI2mTvD+QNIH+en6vvYF9ImJujfonkoaMXiYlzqsrJ3JvbF/g+7m9U4B9I+K1FZtp6G7SHMozpL/tEt4/pLYy9gRm5P/XLgLGRMSSfO4IUo/PepEiyvfeJe1KGutdm/QNa2xEPNyk2MysA5KOA0ZGxEm9HUtvyj3dB4BPFBKL9YKGiSRf+nckqUfyCnAl6RLD7YGbI2JEnepmZraKK3P368PAtaQrdOYUjk+T5C6lmVkfV6ZHclhE3FR17NCIuLmpkZmZWUsok0gejYh/anTMzMz6pppDW5L2Il0BsqmknxVODSZd1mdmZlZ3jmQuaaG0/UkLrlUsBL7XzKC624YbbhhtbW29HYaZWUuZPn36axExtFG5MkNb/fM6Oi1r1KhRMW3atN4Ow8yspUiaHhGjGpWrN7R1U0QcBvxJ0grZJi8SZ2ZmfVy9oa3v5p/VC6SZmZn9Xc1EEhF/zT+9ho2ZmdVUb2hrIR2vfiog8kNmzKygp9Y1XZmljcyareaijRExKCIGd/Aa5CRi1rGVfUTp5qfe3qlHm5p9kNTrkQyOiAW1nkeSV241M7M+rt5k+/WkifbppCGuYp896PjBOWZm1sfUm2zfN//06r5mZlZTmdV/kXQQsAupJ/JgRPxnU6MyM7OWUeaZ7b8ATgCeBJ4CTpB0SbMDMzOz1lDmUbufB74cEVdHxNWkhRy/UKZxSXtKmiWpXdLYDs4PlDQ+n39EUls+PkTSZEmLJF1cVef+3OZj+bVRmVjMzKw5yiSSWcDwwv5mwBONKknqR3qW9F7ASOBwSSOrih0DvBkRWwH/DpyXjy8BfgT8S43mj4iI7fPr1RK/g5mZNUnNRCJpoqQJwBBgZu4JTAZmAg1XgwR2BNoj4rmIWArcCIyuKjMauCZv3wLsJkkRsTgippASipmZfYDVm2y/oIttbwq8WNifA+xUq0xELJM0n5S4XmvQ9tWSlgO/Bc4J36FlZtZr6l3++0AX2+5orYjqD/wyZaodEREvSRpESiRfA369wptLxwHHAQwfPrz6tJmZdZMyV219StLUPPG9VNJySQtKtD2HNJ9SMYz0sKwOy0jqD6wL1L1jPiJeyj8Xkm6a3LFGuXERMSoiRg0dWmYkzszMOqPMZPvFwOHAs8CawLH5WCNTga0ljZA0ABgDTKgqMwE4Km8fAtxXb5hKUn9JG+bt1Ul33j9VIhYzM2uSUjckRkS7pH4RsZw0P/HfJeosk3QicDfQD7gqImZIOguYFhETgCuBayW1k3oiYyr1Jc0mPR9+gKQDgD2AF4C7cxLpB9wL/LL8r2tmZt2tTCJ5O/coHpN0PvBXYO0yjUfEJGBS1bHTCttLgENr1G2r0ewOZd7bzMx6Rpmhra/lcicCi0lzGgc3MygzM2sdDXskEfFC7pG0AbcCs/J9IWZmZo0TiaR9gMuAv5Au1x0h6fiIuLPZwZmZ2QdfmTmSnwK7RkQ7gKQtgTsAJxIzMys1R/JqJYlkzwFe38rMzID6j9o9KG/OkDQJuIl01/mhpHtEzMzM6g5t7VfYfoW0nDzAPGD9pkVkZmYtpd5aW9/oyUDMzKw1lVlra5ik2yS9KukVSb+VNKwngjMzsw++MpPtV5PWxNqEtOz7xHzMzMysVCIZmh+zuyy/fkW5B1uZmVkfUCaRvCbpSEn98utI4PVmB2ZmZq2hTCI5GjgMeJm0YOMh+ZiZmVn9O9sl9QMOjoj9eygeMzNrMXV7JPn5I6N7KBYzM2tBZdbaekjSxcB40jLyAETEo02LyszMWkaZOZKdgX8EziIt4PhT4IIyjUvaU9IsSe2SxnZwfqCk8fn8I5La8vEhkibn58R3+FhfSRMk+TG7Zma9rMzzSHbtTMN5fuUSYHdgDjBV0oSI+HOh2DHAmxGxlaQxwHnAV4AlwI+Aj+VXddsHAYs6E5eZmXWvMne2D5H0M0mPSpou6SJJQ0q0vSPQHhHP5Qdh3ciK8y2jgWvy9i3AbpIUEYsjYgopoVTHsw5wMnBOiRjMzKzJygxt3UhaqPFg0qW/80jzJY1sCrxY2J+Tj3VYJiKWAfOBRknqbNLw2tslYjAzsyYrk0g2iIizI+L5/DoHWK9EPXVwLDpR5r3C0vbAVhFxW8M3l46TNE3StHnz5jUqbmZmnVTmqq3Jef7iprx/COkJiY3MATYr7A8D5tYoM0dSf2Bd4I06bX4a2EHSbFLsG0m6PyK+UF0wIsYB4wBGjRpVMzmZ1bLdmfcw/513m/4+bWPL/HPqvHXXXJ3HT9+jqe9hfVuZRHI8aU7i2rzfD1gs6WQgImJwjXpTga0ljQBeAsYAX60qMwE4CniYlKDui4iaH/oRcSlwKUC+wuv2jpKIWXeY/867zD53n94Oo8uanajMyly1NagzDUfEMkknAneTks9VETFD0lnAtIiYAFwJXCupndQTGVOpn3sdg4EBkg4A9qi64svMzD4AyvRIOi0iJgGTqo6dVtheQnp0b0d12xq0PZsOLg02M7OeVWay3czMrCYnEjMz65KaiUTSJyXt1cHx/SXt0NywzMysVdTrkfwbMLOD43/O58zMzOomkiF5Qvt9IqKdxnefm5lZH1EvkaxZ59za3R2ImZm1pnqJ5F5JP5b0vmVMJJ0J3NfcsMzMrFXUu4/k+8AVQLukx/Kx7YBpwLHNDszMzFpDzUQSEYuBwyVtQXqwFcCMiHiuRyIzM7OWUO/y336S1snPE5lIWj5+mKTPSerUsilmZrbqqTe0dR7wKnB+3r8emAGsATwKnNrc0MzMrBXUSyS7AZ8s7M+PiP3y5PuDzQ3LzMxaRb2rtlbLTy2sOBXSuvHAOk2NyszMWka9RDKgOBcSEfcASFqXNLxlZmZWN5H8EhgvaXjlgKTNgRvyOTMzs7qX/14o6W1giqTKneyLgHPzkwrNzMzqLyMfEZdFxHBgc6AtIjZfmSQiaU9JsyS1SxrbwfmBksbn84/kx+ciaYikyZIWSbq4qs5dkh6XNEPSZZL6lY3HzMy6X91EIunzkj4eEYuAvSRdLOl7kgY2ajh/wF8C7AWMJN3cOLKq2DHAmxGxFfDvpEuOAZYAPwL+pYOmD4uI7UhPRxxKjScsmplZz6g5tCXpEmBbYA1Js0hXat0F7AxcBRzRoO0dgfbKnfCSbgRGk5ahrxgNnJG3bwEulqR8V/0USVtVNxoRCwqxDwCiQRxmZtZE9e4j2TUiRkpaA3gJ2Cgilku6HHiiRNubAi8W9ucAO9UqExHLJM0nLVH/Wr2GJd1NSlR3khKQmZn1knpDW0sAImIJ8EJELM/7Abxbom11cKy691CmzIoFIr4MbAwMBL7Y4ZtLx0maJmnavHnzGjVpZmadVK9HspGkk0kf9pVt8v7QEm3PATYr7A8D5tYoM0dSf2Bd4I0ygUfEEkkTSMNj/9XB+XHAOIBRo0Z5+MvMrEka3UcyiDQ3Utmu7F9Rou2pwNaSRkgaAIwBJlSVmQAclbcPAe7LPZ4OSVpH0sZ5uz+wN/B0iVjMzKxJ6t1HcmZXGs5zHicCdwP9gKsiYoaks4BpETEBuBK4VlI7qScyplJf0mxgMOkO+wOAPYDXgQn5qrF+pAdsXdaVOM3MrGvqDW11WURMAiZVHTutsL2EGpfvRkRbjWY/WeO4mZn1grr3kZiZmTXiRGJmZl1S74bEk2udg7QWV/eHY2ZmrabeHEllCfmPkOYlKldc7Qf8vplBmZlZ62h41Zake4B/ioiFef8M4OYeic7MzD7wysyRDAeWFvaXAm1NicbMzFpOmct/rwX+KOk20vIlBwK/bmpUZmbWMhomkoj4saQ7gc/mQ9+IiD81Nyyz3jfoo2P5+DUrPEan5Qz6KMA+vR2GrcLK3pC4FrAgIq6WNFTSiIh4vpmBmfW2hTPPZfa5rf8B3Db2jt4OwVZxDedIJJ0OnAr833xodeC6ZgZlZmato8xk+4HA/sBigIiYy3uXBpuZWR9XJpEszSvyBoCktZsbkpmZtZIyieSm/FTE9SR9E7iXcsvIm5lZH1Dmqq0LJO0OLCDd5X5aRKzwICkzM+ubGiYSSedFxKkUnkJYOGZmZn1cmaGt3Ts4tld3B2JmZq2pZiKR9M+SngS2kfRE4fU88GSZxiXtKWmWpHZJK9zZJWmgpPH5/COS2vLxIZImS1ok6eJC+bUk3SHpaUkzJJ27sr+wmZl1r3pDW9cDdwL/ChSTwMKIeKNRw5L6AZeQejRzgKmSJkTEnwvFjgHejIitJI0BzgO+AiwBfgR8LL+KLoiIyfk58L+TtFdE3NkoHjMza46aPZKImB8Rs4GLgDci4oWIeAF4V9JOJdreEWiPiOciYilwIzC6qsxo4Jq8fQuwmyRFxOKImEJKKMWY3o6IyXl7KfAoMKxELGZm1iRl5kguBRYV9hfnY41sCrxY2J+Tj3VYJiKWAfOBISXaRtJ6pGej/K7G+eMkTZM0bd68eWWaNDOzTiiz1pbyDYkARMTfJJWq18Gx6ESZFRtO738D8LOIeK6jMhExDhgHMGrUqIZtmnVkVVinat01V+/tEGwVVyYhPCfpO7zXC/kW0OGHd5U5wGaF/WHA3Bpl5uTksC7QcP6FlCCejYj/KFHWrFN6YsHGtrF3rBILQ1rfVmZo6wRgZ+Al0gf/TsBxJepNBbaWNCJPjI/hvcf1VkwAjsrbhwD3FXs/HZF0DinhnFQiBjMza7Iyd7a/SkoCKyUilkk6Ebgb6AdcFREzJJ0FTIuICcCVwLWS2kk9kb+/j6TZwGBggKQDgD1Id9f/AHgaeFQSwMUR4SVbzMx6Sc1EIumUiDhf0s/pYN4iIr7TqPGImARMqjp2WmF7CXBojbpttUJr9L5mZtZz6vVIZuaf03oiEDMza001E0lETMw/r6lVxszMrN7Q1kTqXIobEfs3JSIzM2sp9Ya2Lsg/DwL+gfcer3s4MLuJMZmZWQupN7T1AICksyPic4VTEyX9vumRmZlZSyhzH8lQSVtUdiSNAIY2LyQzM2slZe5s/x5wv6TK3extwPFNi8jMzFpKmRsS75K0NbBNPvR0RPxvc8MyM7NWUeZRu2sBJwObR8Q3JW0t6SMRcXvzwzNrLXm1hZWrc97Kv0+DlYTMelSZOZKrgaXAp/P+HOCcpkVk1sIiokdeZh8kZRLJlhFxPvAuQES8g5cpMTOzrEwiWSppTfLNiZK2BDxHYmZmQLmrtk4H7gI2k/Qb4DPA15sZlJmZtQ7VG29VmjkcBrwNfIo0pPWHiHitZ8LrHpLmAS/0dhxmHdgQaKl/T9anbB4RDe8brJtIACRNj4gdui0sM/s7SdMiYlRvx2HWFWXmSP4g6ZNNj8TMzFpSmTmSXYET8hMLF5OGtyIitm1mYGZm1hrKJJK9mh6FWd81rrcDMOuqmnMkktYATgC2Ap4EroyIZT0Ym5mZtYB6iWQ86SbEB0m9khci4rs9GJuZmbWAepPtIyPiyIi4HDgE+GwPxWS2ypK0KP/cRNItDcqelNe6q+xPkrRes2M0W1n1Esm7lQ0PaZnVJqnfytaJiLkRcUiDYicBf08kEbF3RLy1su9l1mz1Esl2khbk10Jg28q2pAU9FaBZb5LUJulpSddIekLSLZLWkjRb0mmSpgCHStpS0l2Spkt6UNI2uf4ISQ9Lmirp7Kp2n8rb/SRdIOnJ/B7/R9J3gE2AyZIm53KzJW2Yt0+W9FR+nVRoc6akX0qaIemevLyRWVPVe9TuSn/LMltFfQQ4JiIeknQV8K18fElE7AIg6XfACRHxrKSdgF8AXwQuAi6NiF9L+naN9o8DRgCfiIhlkjaIiDcknQzsWr2ShKQdgG8AO5Eux39E0gPAm8DWwOH5kQ83AQcD13XbX8KsA2VuSDTr616MiIfy9nXALnl7PICkdYCdgZslPQZcDmycy3wGuCFvX1uj/S8Bl1WGkCPijQbx7ALcFhGLI2IRcCvvzWE+HxGP5e3ppCeamjVVmftIzPq66ksbK/uL88/VgLciYvuS9aupRJnq8rUUV+ZeDnhoy5rOPRKzxoZLqjzY7XBgSvFkRCwAnpd0KKTFTiVtl08/BIzJ20fUaP8e0uoR/XP9DfLxhcCgDsr/Hjggz9WsDRxIukzfrFc4kZg1NhM4StITwAbApR2UOQI4RtLjwAxgdD7+XeDbkqYC69Zo/wrgf4Ancv2v5uPjgDsrk+0VEfEo8Cvgj8AjwBUR8adO/m5mXdZw9V+zvkxSG3B7RHysl0Mx+8Byj8TMzLrEPRIzM+sS90jMzKxLnEjMzKxLnEjMzKxLnEjMzKxLnEjMzKxLnEjMzKxL/j+Wh4aCgZqkcQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114e76c50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BGC0000485.1\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114e76080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for contig_id in different['contig_id'][:2]:\n",
    "    print(contig_id)\n",
    "    outlier_domains = ripp_lco[ripp_lco['contig_id'] == contig_id]\n",
    "    ax = outlier_domains[['prediction']].plot.box(title='{} ({} domains)'.format(contig_id, len(outlier_domains)), figsize=(6,2))\n",
    "    ax.set_ylabel('Predicted BGC probability')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>gene_end</th>\n",
       "      <th>gene_start</th>\n",
       "      <th>in_cluster</th>\n",
       "      <th>pfam_id</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>233545</th>\n",
       "      <td>BGC0000618.1</td>\n",
       "      <td>147</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>PF04369</td>\n",
       "      <td>AFK59990.1</td>\n",
       "      <td>0.015449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>233546</th>\n",
       "      <td>BGC0000618.1</td>\n",
       "      <td>589</td>\n",
       "      <td>250</td>\n",
       "      <td>1</td>\n",
       "      <td>PF08951</td>\n",
       "      <td>AFK59991.1</td>\n",
       "      <td>0.012497</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>233547</th>\n",
       "      <td>BGC0000618.1</td>\n",
       "      <td>761</td>\n",
       "      <td>563</td>\n",
       "      <td>1</td>\n",
       "      <td>PF01721</td>\n",
       "      <td>AFK59992.1</td>\n",
       "      <td>0.011460</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           contig_id  gene_end  gene_start  in_cluster  pfam_id  protein_id  \\\n",
       "233545  BGC0000618.1       147           0           1  PF04369  AFK59990.1   \n",
       "233546  BGC0000618.1       589         250           1  PF08951  AFK59991.1   \n",
       "233547  BGC0000618.1       761         563           1  PF01721  AFK59992.1   \n",
       "\n",
       "        prediction  \n",
       "233545    0.015449  \n",
       "233546    0.012497  \n",
       "233547    0.011460  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ripp_lco[ripp_lco['contig_id'] == 'BGC0000618.1']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x114ed72b0>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114b3b898>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ripp_lco.loc[233535:233557].drop(['gene_end','gene_start'], axis=1).plot(figsize=(11, 2), drawstyle=\"steps\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# NRP LCO"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 246,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>gene_end</th>\n",
       "      <th>gene_start</th>\n",
       "      <th>in_cluster</th>\n",
       "      <th>pfam_id</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>HMM Reconstructed</th>\n",
       "      <th>HMM GeneBorder</th>\n",
       "      <th>LSTM 250 epochs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP006889.1|EF217405.1</td>\n",
       "      <td>1166</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>PF07687</td>\n",
       "      <td>CP006889_539</td>\n",
       "      <td>0.003142</td>\n",
       "      <td>1.342748e-04</td>\n",
       "      <td>0.031424</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP006889.1|EF217405.1</td>\n",
       "      <td>1166</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>PF01546</td>\n",
       "      <td>CP006889_539</td>\n",
       "      <td>0.000680</td>\n",
       "      <td>1.342748e-04</td>\n",
       "      <td>0.025055</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP006889.1|EF217405.1</td>\n",
       "      <td>1624</td>\n",
       "      <td>1166</td>\n",
       "      <td>0</td>\n",
       "      <td>PF01668</td>\n",
       "      <td>CP006889_1420</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>1.157587e-06</td>\n",
       "      <td>0.003387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP006889.1|EF217405.1</td>\n",
       "      <td>2076</td>\n",
       "      <td>1624</td>\n",
       "      <td>0</td>\n",
       "      <td>PF01618</td>\n",
       "      <td>CP006889_1421</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>1.465741e-07</td>\n",
       "      <td>0.001086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP006889.1|EF217405.1</td>\n",
       "      <td>4499</td>\n",
       "      <td>2076</td>\n",
       "      <td>0</td>\n",
       "      <td>PF01584</td>\n",
       "      <td>CP006889_1333</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>2.332568e-14</td>\n",
       "      <td>0.000189</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP006889.1|EF217405.1</td>\n",
       "      <td>4499</td>\n",
       "      <td>2076</td>\n",
       "      <td>0</td>\n",
       "      <td>PF00072</td>\n",
       "      <td>CP006889_1333</td>\n",
       "      <td>0.000603</td>\n",
       "      <td>2.332568e-14</td>\n",
       "      <td>0.000165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP006889.1|EF217405.1</td>\n",
       "      <td>4499</td>\n",
       "      <td>2076</td>\n",
       "      <td>0</td>\n",
       "      <td>PF01627</td>\n",
       "      <td>CP006889_1333</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>2.332568e-14</td>\n",
       "      <td>0.000199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP006889.1|EF217405.1</td>\n",
       "      <td>4499</td>\n",
       "      <td>2076</td>\n",
       "      <td>0</td>\n",
       "      <td>PF02895</td>\n",
       "      <td>CP006889_1333</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>2.332568e-14</td>\n",
       "      <td>0.000314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP006889.1|EF217405.1</td>\n",
       "      <td>4499</td>\n",
       "      <td>2076</td>\n",
       "      <td>0</td>\n",
       "      <td>PF02518</td>\n",
       "      <td>CP006889_1333</td>\n",
       "      <td>0.000335</td>\n",
       "      <td>2.332568e-14</td>\n",
       "      <td>0.000631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>NEG_FAKE_CLUSTER|CP006889.1|EF217405.1</td>\n",
       "      <td>4957</td>\n",
       "      <td>4499</td>\n",
       "      <td>0</td>\n",
       "      <td>PF01668</td>\n",
       "      <td>CP006889_1420</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>3.743382e-07</td>\n",
       "      <td>0.000384</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                contig_id  gene_end  gene_start  in_cluster  \\\n",
       "0  NEG_FAKE_CLUSTER|CP006889.1|EF217405.1      1166           0           0   \n",
       "1  NEG_FAKE_CLUSTER|CP006889.1|EF217405.1      1166           0           0   \n",
       "2  NEG_FAKE_CLUSTER|CP006889.1|EF217405.1      1624        1166           0   \n",
       "3  NEG_FAKE_CLUSTER|CP006889.1|EF217405.1      2076        1624           0   \n",
       "4  NEG_FAKE_CLUSTER|CP006889.1|EF217405.1      4499        2076           0   \n",
       "5  NEG_FAKE_CLUSTER|CP006889.1|EF217405.1      4499        2076           0   \n",
       "6  NEG_FAKE_CLUSTER|CP006889.1|EF217405.1      4499        2076           0   \n",
       "7  NEG_FAKE_CLUSTER|CP006889.1|EF217405.1      4499        2076           0   \n",
       "8  NEG_FAKE_CLUSTER|CP006889.1|EF217405.1      4499        2076           0   \n",
       "9  NEG_FAKE_CLUSTER|CP006889.1|EF217405.1      4957        4499           0   \n",
       "\n",
       "   pfam_id     protein_id  HMM Reconstructed  HMM GeneBorder  LSTM 250 epochs  \n",
       "0  PF07687   CP006889_539           0.003142    1.342748e-04         0.031424  \n",
       "1  PF01546   CP006889_539           0.000680    1.342748e-04         0.025055  \n",
       "2  PF01668  CP006889_1420           0.000009    1.157587e-06         0.003387  \n",
       "3  PF01618  CP006889_1421           0.000002    1.465741e-07         0.001086  \n",
       "4  PF01584  CP006889_1333           0.000002    2.332568e-14         0.000189  \n",
       "5  PF00072  CP006889_1333           0.000603    2.332568e-14         0.000165  \n",
       "6  PF01627  CP006889_1333           0.000001    2.332568e-14         0.000199  \n",
       "7  PF02895  CP006889_1333           0.000003    2.332568e-14         0.000314  \n",
       "8  PF02518  CP006889_1333           0.000335    2.332568e-14         0.000631  \n",
       "9  PF01668  CP006889_1420           0.000007    3.743382e-07         0.000384  "
      ]
     },
     "execution_count": 246,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nrp_lco = load_class('NRP')\n",
    "nrp_lco.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x13a5d8668>"
      ]
     },
     "execution_count": 247,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x13aa47b70>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "nrp_lco[['in_cluster','LSTM 250 epochs']].plot(figsize=(80, 2), drawstyle=\"steps\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Missed 2534/21417 = 11.8%\n"
     ]
    }
   ],
   "source": [
    "nrp_positives = nrp_lco[nrp_lco['in_cluster'] == 1]\n",
    "nrp_missed = nrp_positives[nrp_positives['LSTM 250 epochs'] < 0.01]\n",
    "print('Missed {}/{} = {:.1f}%'.format(len(nrp_missed), len(nrp_positives), len(nrp_missed) / len(nrp_positives) * 100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 249,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>gene_end</th>\n",
       "      <th>gene_start</th>\n",
       "      <th>in_cluster</th>\n",
       "      <th>pfam_id</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>HMM Reconstructed</th>\n",
       "      <th>HMM GeneBorder</th>\n",
       "      <th>LSTM 250 epochs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2201</th>\n",
       "      <td>BGC0000466.1</td>\n",
       "      <td>2208</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>PF01315</td>\n",
       "      <td>ADZ13533.1</td>\n",
       "      <td>0.000513</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2202</th>\n",
       "      <td>BGC0000466.1</td>\n",
       "      <td>2208</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>PF02738</td>\n",
       "      <td>ADZ13533.1</td>\n",
       "      <td>0.000611</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2203</th>\n",
       "      <td>BGC0000466.1</td>\n",
       "      <td>3194</td>\n",
       "      <td>2204</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00941</td>\n",
       "      <td>ADZ13534.1</td>\n",
       "      <td>0.001076</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>0.000321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2204</th>\n",
       "      <td>BGC0000466.1</td>\n",
       "      <td>3194</td>\n",
       "      <td>2204</td>\n",
       "      <td>1</td>\n",
       "      <td>PF03450</td>\n",
       "      <td>ADZ13534.1</td>\n",
       "      <td>0.001168</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>0.000366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2205</th>\n",
       "      <td>BGC0000466.1</td>\n",
       "      <td>3699</td>\n",
       "      <td>3180</td>\n",
       "      <td>1</td>\n",
       "      <td>PF01799</td>\n",
       "      <td>ADZ13535.1</td>\n",
       "      <td>0.000534</td>\n",
       "      <td>0.000030</td>\n",
       "      <td>0.000710</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         contig_id  gene_end  gene_start  in_cluster  pfam_id  protein_id  \\\n",
       "2201  BGC0000466.1      2208           0           1  PF01315  ADZ13533.1   \n",
       "2202  BGC0000466.1      2208           0           1  PF02738  ADZ13533.1   \n",
       "2203  BGC0000466.1      3194        2204           1  PF00941  ADZ13534.1   \n",
       "2204  BGC0000466.1      3194        2204           1  PF03450  ADZ13534.1   \n",
       "2205  BGC0000466.1      3699        3180           1  PF01799  ADZ13535.1   \n",
       "\n",
       "      HMM Reconstructed  HMM GeneBorder  LSTM 250 epochs  \n",
       "2201           0.000513        0.000008         0.000148  \n",
       "2202           0.000611        0.000008         0.000158  \n",
       "2203           0.001076        0.000027         0.000321  \n",
       "2204           0.001168        0.000027         0.000366  \n",
       "2205           0.000534        0.000030         0.000710  "
      ]
     },
     "execution_count": 249,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nrp_missed.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>locus_tag</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>gene_start</th>\n",
       "      <th>gene_end</th>\n",
       "      <th>gene_strand</th>\n",
       "      <th>pfam_id</th>\n",
       "      <th>domain_start</th>\n",
       "      <th>domain_end</th>\n",
       "      <th>evalue</th>\n",
       "      <th>bitscore</th>\n",
       "      <th>in_cluster</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>28593</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2142</td>\n",
       "      <td>AAY91416.1</td>\n",
       "      <td>0</td>\n",
       "      <td>1431</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF02321</td>\n",
       "      <td>6</td>\n",
       "      <td>188</td>\n",
       "      <td>1.600000e-28</td>\n",
       "      <td>99.7</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28594</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2143</td>\n",
       "      <td>AAY91417.1</td>\n",
       "      <td>1490</td>\n",
       "      <td>2291</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF00196</td>\n",
       "      <td>1</td>\n",
       "      <td>56</td>\n",
       "      <td>5.500000e-09</td>\n",
       "      <td>35.4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28595</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2145</td>\n",
       "      <td>AAY91419.3</td>\n",
       "      <td>2885</td>\n",
       "      <td>9251</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00501</td>\n",
       "      <td>0</td>\n",
       "      <td>423</td>\n",
       "      <td>3.100000e-96</td>\n",
       "      <td>322.4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28596</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2145</td>\n",
       "      <td>AAY91419.3</td>\n",
       "      <td>2885</td>\n",
       "      <td>9251</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13193</td>\n",
       "      <td>0</td>\n",
       "      <td>76</td>\n",
       "      <td>3.100000e-12</td>\n",
       "      <td>47.2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28597</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2145</td>\n",
       "      <td>AAY91419.3</td>\n",
       "      <td>2885</td>\n",
       "      <td>9251</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00550</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>1.000000e-11</td>\n",
       "      <td>44.9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28598</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2145</td>\n",
       "      <td>AAY91419.3</td>\n",
       "      <td>2885</td>\n",
       "      <td>9251</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00668</td>\n",
       "      <td>4</td>\n",
       "      <td>456</td>\n",
       "      <td>1.000000e-68</td>\n",
       "      <td>232.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28599</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2146</td>\n",
       "      <td>AAY91420.2</td>\n",
       "      <td>9568</td>\n",
       "      <td>22672</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00668</td>\n",
       "      <td>0</td>\n",
       "      <td>455</td>\n",
       "      <td>3.200000e-50</td>\n",
       "      <td>171.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28600</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2146</td>\n",
       "      <td>AAY91420.2</td>\n",
       "      <td>9568</td>\n",
       "      <td>22672</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13193</td>\n",
       "      <td>0</td>\n",
       "      <td>76</td>\n",
       "      <td>1.100000e-12</td>\n",
       "      <td>48.7</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28601</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2146</td>\n",
       "      <td>AAY91420.2</td>\n",
       "      <td>9568</td>\n",
       "      <td>22672</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00501</td>\n",
       "      <td>1</td>\n",
       "      <td>423</td>\n",
       "      <td>5.000000e-106</td>\n",
       "      <td>354.7</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28602</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2146</td>\n",
       "      <td>AAY91420.2</td>\n",
       "      <td>9568</td>\n",
       "      <td>22672</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00550</td>\n",
       "      <td>1</td>\n",
       "      <td>67</td>\n",
       "      <td>3.500000e-11</td>\n",
       "      <td>43.2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28603</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2147</td>\n",
       "      <td>AAY91421.3</td>\n",
       "      <td>22668</td>\n",
       "      <td>37374</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00501</td>\n",
       "      <td>0</td>\n",
       "      <td>423</td>\n",
       "      <td>5.100000e-105</td>\n",
       "      <td>351.4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28604</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2147</td>\n",
       "      <td>AAY91421.3</td>\n",
       "      <td>22668</td>\n",
       "      <td>37374</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13193</td>\n",
       "      <td>0</td>\n",
       "      <td>76</td>\n",
       "      <td>3.100000e-10</td>\n",
       "      <td>40.8</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28605</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2147</td>\n",
       "      <td>AAY91421.3</td>\n",
       "      <td>22668</td>\n",
       "      <td>37374</td>\n",
       "      <td>1</td>\n",
       "      <td>PF12697</td>\n",
       "      <td>0</td>\n",
       "      <td>203</td>\n",
       "      <td>1.200000e-07</td>\n",
       "      <td>32.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28606</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2147</td>\n",
       "      <td>AAY91421.3</td>\n",
       "      <td>22668</td>\n",
       "      <td>37374</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00550</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>2.300000e-12</td>\n",
       "      <td>47.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28607</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2147</td>\n",
       "      <td>AAY91421.3</td>\n",
       "      <td>22668</td>\n",
       "      <td>37374</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00561</td>\n",
       "      <td>1</td>\n",
       "      <td>106</td>\n",
       "      <td>1.400000e-03</td>\n",
       "      <td>18.2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28608</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2147</td>\n",
       "      <td>AAY91421.3</td>\n",
       "      <td>22668</td>\n",
       "      <td>37374</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00668</td>\n",
       "      <td>2</td>\n",
       "      <td>457</td>\n",
       "      <td>2.900000e-107</td>\n",
       "      <td>359.1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28609</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2147</td>\n",
       "      <td>AAY91421.3</td>\n",
       "      <td>22668</td>\n",
       "      <td>37374</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00975</td>\n",
       "      <td>2</td>\n",
       "      <td>217</td>\n",
       "      <td>9.500000e-23</td>\n",
       "      <td>81.4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28610</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2148</td>\n",
       "      <td>AAY91422.1</td>\n",
       "      <td>37524</td>\n",
       "      <td>38676</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13437</td>\n",
       "      <td>1</td>\n",
       "      <td>82</td>\n",
       "      <td>2.100000e-08</td>\n",
       "      <td>34.7</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28611</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2148</td>\n",
       "      <td>AAY91422.1</td>\n",
       "      <td>37524</td>\n",
       "      <td>38676</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13533</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "      <td>2.600000e-10</td>\n",
       "      <td>39.9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28612</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2148</td>\n",
       "      <td>AAY91422.1</td>\n",
       "      <td>37524</td>\n",
       "      <td>38676</td>\n",
       "      <td>1</td>\n",
       "      <td>PF16576</td>\n",
       "      <td>18</td>\n",
       "      <td>181</td>\n",
       "      <td>3.600000e-13</td>\n",
       "      <td>49.2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28613</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2148</td>\n",
       "      <td>AAY91422.1</td>\n",
       "      <td>37524</td>\n",
       "      <td>38676</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00364</td>\n",
       "      <td>41</td>\n",
       "      <td>73</td>\n",
       "      <td>5.400000e-04</td>\n",
       "      <td>19.7</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28614</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2149</td>\n",
       "      <td>AAY91423.1</td>\n",
       "      <td>38678</td>\n",
       "      <td>40637</td>\n",
       "      <td>1</td>\n",
       "      <td>PF12704</td>\n",
       "      <td>0</td>\n",
       "      <td>192</td>\n",
       "      <td>2.000000e-38</td>\n",
       "      <td>132.6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28615</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2149</td>\n",
       "      <td>AAY91423.1</td>\n",
       "      <td>38678</td>\n",
       "      <td>40637</td>\n",
       "      <td>1</td>\n",
       "      <td>PF00005</td>\n",
       "      <td>0</td>\n",
       "      <td>136</td>\n",
       "      <td>4.100000e-28</td>\n",
       "      <td>98.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28616</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2149</td>\n",
       "      <td>AAY91423.1</td>\n",
       "      <td>38678</td>\n",
       "      <td>40637</td>\n",
       "      <td>1</td>\n",
       "      <td>PF02687</td>\n",
       "      <td>1</td>\n",
       "      <td>96</td>\n",
       "      <td>1.300000e-11</td>\n",
       "      <td>44.7</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28617</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2149</td>\n",
       "      <td>AAY91423.1</td>\n",
       "      <td>38678</td>\n",
       "      <td>40637</td>\n",
       "      <td>1</td>\n",
       "      <td>PF14342</td>\n",
       "      <td>6</td>\n",
       "      <td>73</td>\n",
       "      <td>2.800000e-03</td>\n",
       "      <td>17.9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28618</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2149</td>\n",
       "      <td>AAY91423.1</td>\n",
       "      <td>38678</td>\n",
       "      <td>40637</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13555</td>\n",
       "      <td>16</td>\n",
       "      <td>39</td>\n",
       "      <td>2.000000e-04</td>\n",
       "      <td>20.9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28619</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2149</td>\n",
       "      <td>AAY91423.1</td>\n",
       "      <td>38678</td>\n",
       "      <td>40637</td>\n",
       "      <td>1</td>\n",
       "      <td>PF02463</td>\n",
       "      <td>25</td>\n",
       "      <td>213</td>\n",
       "      <td>7.400000e-03</td>\n",
       "      <td>15.7</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28620</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2150</td>\n",
       "      <td>AAY91424.1</td>\n",
       "      <td>40697</td>\n",
       "      <td>41486</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF00196</td>\n",
       "      <td>1</td>\n",
       "      <td>56</td>\n",
       "      <td>1.400000e-08</td>\n",
       "      <td>34.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28621</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2150</td>\n",
       "      <td>AAY91424.1</td>\n",
       "      <td>40697</td>\n",
       "      <td>41486</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF08281</td>\n",
       "      <td>25</td>\n",
       "      <td>54</td>\n",
       "      <td>1.700000e-03</td>\n",
       "      <td>17.9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28622</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2151</td>\n",
       "      <td>AAY91425.1</td>\n",
       "      <td>41713</td>\n",
       "      <td>42127</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF00903</td>\n",
       "      <td>0</td>\n",
       "      <td>126</td>\n",
       "      <td>2.200000e-03</td>\n",
       "      <td>18.1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28623</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2151</td>\n",
       "      <td>AAY91425.1</td>\n",
       "      <td>41713</td>\n",
       "      <td>42127</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF06983</td>\n",
       "      <td>1</td>\n",
       "      <td>115</td>\n",
       "      <td>1.300000e-11</td>\n",
       "      <td>45.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28624</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2152</td>\n",
       "      <td>AAY91426.1</td>\n",
       "      <td>42302</td>\n",
       "      <td>43811</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF13186</td>\n",
       "      <td>0</td>\n",
       "      <td>67</td>\n",
       "      <td>1.600000e-05</td>\n",
       "      <td>25.3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28625</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2152</td>\n",
       "      <td>AAY91426.1</td>\n",
       "      <td>42302</td>\n",
       "      <td>43811</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF04055</td>\n",
       "      <td>1</td>\n",
       "      <td>148</td>\n",
       "      <td>7.100000e-18</td>\n",
       "      <td>65.4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28626</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2152</td>\n",
       "      <td>AAY91426.1</td>\n",
       "      <td>42302</td>\n",
       "      <td>43811</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF13920</td>\n",
       "      <td>23</td>\n",
       "      <td>47</td>\n",
       "      <td>5.100000e-03</td>\n",
       "      <td>16.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28627</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2153</td>\n",
       "      <td>AAY91427.1</td>\n",
       "      <td>43901</td>\n",
       "      <td>44510</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF08898</td>\n",
       "      <td>0</td>\n",
       "      <td>52</td>\n",
       "      <td>8.300000e-22</td>\n",
       "      <td>77.3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28628</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2153</td>\n",
       "      <td>AAY91427.1</td>\n",
       "      <td>43901</td>\n",
       "      <td>44510</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF08896</td>\n",
       "      <td>1</td>\n",
       "      <td>114</td>\n",
       "      <td>1.200000e-44</td>\n",
       "      <td>151.1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28629</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2154</td>\n",
       "      <td>AAY91428.1</td>\n",
       "      <td>44561</td>\n",
       "      <td>45197</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF08898</td>\n",
       "      <td>0</td>\n",
       "      <td>52</td>\n",
       "      <td>3.100000e-20</td>\n",
       "      <td>72.3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28630</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2154</td>\n",
       "      <td>AAY91428.1</td>\n",
       "      <td>44561</td>\n",
       "      <td>45197</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF08896</td>\n",
       "      <td>1</td>\n",
       "      <td>114</td>\n",
       "      <td>1.500000e-46</td>\n",
       "      <td>157.3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28631</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2155</td>\n",
       "      <td>AAY91429.1</td>\n",
       "      <td>45233</td>\n",
       "      <td>45422</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF08898</td>\n",
       "      <td>0</td>\n",
       "      <td>52</td>\n",
       "      <td>5.600000e-22</td>\n",
       "      <td>77.9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28632</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2156</td>\n",
       "      <td>AAY91430.1</td>\n",
       "      <td>45739</td>\n",
       "      <td>46369</td>\n",
       "      <td>-1</td>\n",
       "      <td>PF13419</td>\n",
       "      <td>37</td>\n",
       "      <td>176</td>\n",
       "      <td>2.200000e-11</td>\n",
       "      <td>44.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28633</th>\n",
       "      <td>BGC0000399.1</td>\n",
       "      <td>PFL_2158</td>\n",
       "      <td>AAY91432.1</td>\n",
       "      <td>46834</td>\n",
       "      <td>47020</td>\n",
       "      <td>1</td>\n",
       "      <td>PF09954</td>\n",
       "      <td>0</td>\n",
       "      <td>46</td>\n",
       "      <td>1.500000e-04</td>\n",
       "      <td>21.8</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          contig_id locus_tag  protein_id  gene_start  gene_end  gene_strand  \\\n",
       "28593  BGC0000399.1  PFL_2142  AAY91416.1           0      1431           -1   \n",
       "28594  BGC0000399.1  PFL_2143  AAY91417.1        1490      2291           -1   \n",
       "28595  BGC0000399.1  PFL_2145  AAY91419.3        2885      9251            1   \n",
       "28596  BGC0000399.1  PFL_2145  AAY91419.3        2885      9251            1   \n",
       "28597  BGC0000399.1  PFL_2145  AAY91419.3        2885      9251            1   \n",
       "28598  BGC0000399.1  PFL_2145  AAY91419.3        2885      9251            1   \n",
       "28599  BGC0000399.1  PFL_2146  AAY91420.2        9568     22672            1   \n",
       "28600  BGC0000399.1  PFL_2146  AAY91420.2        9568     22672            1   \n",
       "28601  BGC0000399.1  PFL_2146  AAY91420.2        9568     22672            1   \n",
       "28602  BGC0000399.1  PFL_2146  AAY91420.2        9568     22672            1   \n",
       "28603  BGC0000399.1  PFL_2147  AAY91421.3       22668     37374            1   \n",
       "28604  BGC0000399.1  PFL_2147  AAY91421.3       22668     37374            1   \n",
       "28605  BGC0000399.1  PFL_2147  AAY91421.3       22668     37374            1   \n",
       "28606  BGC0000399.1  PFL_2147  AAY91421.3       22668     37374            1   \n",
       "28607  BGC0000399.1  PFL_2147  AAY91421.3       22668     37374            1   \n",
       "28608  BGC0000399.1  PFL_2147  AAY91421.3       22668     37374            1   \n",
       "28609  BGC0000399.1  PFL_2147  AAY91421.3       22668     37374            1   \n",
       "28610  BGC0000399.1  PFL_2148  AAY91422.1       37524     38676            1   \n",
       "28611  BGC0000399.1  PFL_2148  AAY91422.1       37524     38676            1   \n",
       "28612  BGC0000399.1  PFL_2148  AAY91422.1       37524     38676            1   \n",
       "28613  BGC0000399.1  PFL_2148  AAY91422.1       37524     38676            1   \n",
       "28614  BGC0000399.1  PFL_2149  AAY91423.1       38678     40637            1   \n",
       "28615  BGC0000399.1  PFL_2149  AAY91423.1       38678     40637            1   \n",
       "28616  BGC0000399.1  PFL_2149  AAY91423.1       38678     40637            1   \n",
       "28617  BGC0000399.1  PFL_2149  AAY91423.1       38678     40637            1   \n",
       "28618  BGC0000399.1  PFL_2149  AAY91423.1       38678     40637            1   \n",
       "28619  BGC0000399.1  PFL_2149  AAY91423.1       38678     40637            1   \n",
       "28620  BGC0000399.1  PFL_2150  AAY91424.1       40697     41486           -1   \n",
       "28621  BGC0000399.1  PFL_2150  AAY91424.1       40697     41486           -1   \n",
       "28622  BGC0000399.1  PFL_2151  AAY91425.1       41713     42127           -1   \n",
       "28623  BGC0000399.1  PFL_2151  AAY91425.1       41713     42127           -1   \n",
       "28624  BGC0000399.1  PFL_2152  AAY91426.1       42302     43811           -1   \n",
       "28625  BGC0000399.1  PFL_2152  AAY91426.1       42302     43811           -1   \n",
       "28626  BGC0000399.1  PFL_2152  AAY91426.1       42302     43811           -1   \n",
       "28627  BGC0000399.1  PFL_2153  AAY91427.1       43901     44510           -1   \n",
       "28628  BGC0000399.1  PFL_2153  AAY91427.1       43901     44510           -1   \n",
       "28629  BGC0000399.1  PFL_2154  AAY91428.1       44561     45197           -1   \n",
       "28630  BGC0000399.1  PFL_2154  AAY91428.1       44561     45197           -1   \n",
       "28631  BGC0000399.1  PFL_2155  AAY91429.1       45233     45422           -1   \n",
       "28632  BGC0000399.1  PFL_2156  AAY91430.1       45739     46369           -1   \n",
       "28633  BGC0000399.1  PFL_2158  AAY91432.1       46834     47020            1   \n",
       "\n",
       "       pfam_id  domain_start  domain_end         evalue  bitscore  in_cluster  \n",
       "28593  PF02321             6         188   1.600000e-28      99.7           1  \n",
       "28594  PF00196             1          56   5.500000e-09      35.4           1  \n",
       "28595  PF00501             0         423   3.100000e-96     322.4           1  \n",
       "28596  PF13193             0          76   3.100000e-12      47.2           1  \n",
       "28597  PF00550             1          66   1.000000e-11      44.9           1  \n",
       "28598  PF00668             4         456   1.000000e-68     232.0           1  \n",
       "28599  PF00668             0         455   3.200000e-50     171.0           1  \n",
       "28600  PF13193             0          76   1.100000e-12      48.7           1  \n",
       "28601  PF00501             1         423  5.000000e-106     354.7           1  \n",
       "28602  PF00550             1          67   3.500000e-11      43.2           1  \n",
       "28603  PF00501             0         423  5.100000e-105     351.4           1  \n",
       "28604  PF13193             0          76   3.100000e-10      40.8           1  \n",
       "28605  PF12697             0         203   1.200000e-07      32.5           1  \n",
       "28606  PF00550             1          66   2.300000e-12      47.0           1  \n",
       "28607  PF00561             1         106   1.400000e-03      18.2           1  \n",
       "28608  PF00668             2         457  2.900000e-107     359.1           1  \n",
       "28609  PF00975             2         217   9.500000e-23      81.4           1  \n",
       "28610  PF13437             1          82   2.100000e-08      34.7           1  \n",
       "28611  PF13533             1          46   2.600000e-10      39.9           1  \n",
       "28612  PF16576            18         181   3.600000e-13      49.2           1  \n",
       "28613  PF00364            41          73   5.400000e-04      19.7           1  \n",
       "28614  PF12704             0         192   2.000000e-38     132.6           1  \n",
       "28615  PF00005             0         136   4.100000e-28      98.5           1  \n",
       "28616  PF02687             1          96   1.300000e-11      44.7           1  \n",
       "28617  PF14342             6          73   2.800000e-03      17.9           1  \n",
       "28618  PF13555            16          39   2.000000e-04      20.9           1  \n",
       "28619  PF02463            25         213   7.400000e-03      15.7           1  \n",
       "28620  PF00196             1          56   1.400000e-08      34.0           1  \n",
       "28621  PF08281            25          54   1.700000e-03      17.9           1  \n",
       "28622  PF00903             0         126   2.200000e-03      18.1           1  \n",
       "28623  PF06983             1         115   1.300000e-11      45.0           1  \n",
       "28624  PF13186             0          67   1.600000e-05      25.3           1  \n",
       "28625  PF04055             1         148   7.100000e-18      65.4           1  \n",
       "28626  PF13920            23          47   5.100000e-03      16.5           1  \n",
       "28627  PF08898             0          52   8.300000e-22      77.3           1  \n",
       "28628  PF08896             1         114   1.200000e-44     151.1           1  \n",
       "28629  PF08898             0          52   3.100000e-20      72.3           1  \n",
       "28630  PF08896             1         114   1.500000e-46     157.3           1  \n",
       "28631  PF08898             0          52   5.600000e-22      77.9           1  \n",
       "28632  PF13419            37         176   2.200000e-11      44.0           1  \n",
       "28633  PF09954             0          46   1.500000e-04      21.8           1  "
      ]
     },
     "execution_count": 250,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "missed_example = domains[domains['contig_id'] == 'BGC0000399.1']\n",
    "missed_example"
   ]
  },
  {
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}