Manual.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "hideCode": false,
    "hidePrompt": false
   },
   "source": [
    "# ANUBIS\n",
    "\n",
    " \n",
    "- **Authors:** Miravet-Verde, Samuel\n",
    "- **Last update:** 04/29/2020 (MM/DD/YYYY)\n",
    "\n",
    "\n",
    "ANUBIS is a Python framework to analyze Tn-seq data.\n",
    "\n",
    "\n",
    "----\n",
    "\n",
    "## Index of Content:\n",
    "* [Requirements](#req)\n",
    "* [Tn-seq data processing](#proc)\n",
    "* [Installation](#req)\n",
    "* [Tn-seq data processing](#proc)\n",
    "\n",
    "----\n",
    "\n",
    "## Requirements <a class=\"anchor\" id=\"req\"></a>\n",
    "\n",
    "Specific libraries are required by ANUBIS. We provide a [requirements](./requirements.txt) file to install everything at once. To do so, you will need first to have [pip](https://pip.pypa.io/en/stable/installing/) installed and then run:\n",
    "\n",
    "```\n",
    "sudo apt-get install python-pip    # if you need to install pip\n",
    "pip install -r requirements.txt\n",
    "```\n",
    "\n",
    "----\n",
    "\n",
    "## Tn-seq data processing <a class=\"anchor\" id=\"proc\"></a>\n",
    "\n",
    "ANUBIS has been mainly developed to work with [FASTQINS](https://github.com/CRG-CNAG/fastqins) generated files. This is a standard pipeline to process raw fastq Tn-seq files to obtain insertion profiles that can be interpreted by ANUBIS (.wig or .qins extensions).\n",
    "\n",
    "----\n",
    "\n",
    "\n",
    "## Installation<a class=\"anchor\" id=\"install\"></a>\n",
    "\n",
    "Download ANUBIS repository and run:\n",
    "\n",
    "```\n",
    "python3 setup.py install\n",
    "```\n",
    "\n",
    "You may require to call it using sudo. Once installed, `anubis` will be integrated in your python distribution and can be called as:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "hideCode": false,
    "hidePrompt": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "import anubis \n",
    "\n",
    "# for jupyter:\n",
    "%matplotlib inline\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hideCode": false,
    "hidePrompt": false
   },
   "source": [
    "## Manual<a class=\"anchor\" id=\"man\"></a>\n",
    "\n",
    "### Load a dataset\n",
    "\n",
    "Load a dataset including the goldset as a two column file with the gene and the expected label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "hideCode": false,
    "hidePrompt": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['GC',\n",
       " 'NRL_for_annotation',\n",
       " 'QC',\n",
       " 'RAUC',\n",
       " '__class__',\n",
       " '__delattr__',\n",
       " '__dict__',\n",
       " '__dir__',\n",
       " '__doc__',\n",
       " '__eq__',\n",
       " '__format__',\n",
       " '__ge__',\n",
       " '__getattribute__',\n",
       " '__gt__',\n",
       " '__hash__',\n",
       " '__init__',\n",
       " '__init_subclass__',\n",
       " '__le__',\n",
       " '__lt__',\n",
       " '__module__',\n",
       " '__ne__',\n",
       " '__new__',\n",
       " '__reduce__',\n",
       " '__reduce_ex__',\n",
       " '__repr__',\n",
       " '__setattr__',\n",
       " '__sizeof__',\n",
       " '__str__',\n",
       " '__subclasshook__',\n",
       " '__weakref__',\n",
       " '_bin_metric_',\n",
       " '_bin_roll_',\n",
       " '_bin_window_size_',\n",
       " '_domains_evalue_',\n",
       " '_empty_profile_',\n",
       " '_metric_df',\n",
       " '_one_base_annotation_',\n",
       " '_roll_',\n",
       " '_set_profile_array_',\n",
       " '_single_metric_by',\n",
       " '_update_roll_',\n",
       " '_update_stats_',\n",
       " '_update_stats_from_array_',\n",
       " '_window_metric_',\n",
       " '_window_size_',\n",
       " 'addition',\n",
       " 'additional_annotation',\n",
       " 'annotation',\n",
       " 'annotation_per_base',\n",
       " 'apositions',\n",
       " 'aprofile',\n",
       " 'condition',\n",
       " 'copy',\n",
       " 'correct_TSD',\n",
       " 'correlate',\n",
       " 'correlate_metric',\n",
       " 'correlation_by_distance',\n",
       " 'coverage',\n",
       " 'cqn',\n",
       " 'dataframe',\n",
       " 'density',\n",
       " 'difference',\n",
       " 'dilution',\n",
       " 'domains',\n",
       " 'domains_location',\n",
       " 'doubling_time',\n",
       " 'filter',\n",
       " 'frame_metric',\n",
       " 'genome',\n",
       " 'genome_length',\n",
       " 'genome_sequence',\n",
       " 'goldE',\n",
       " 'goldNE',\n",
       " 'goldset',\n",
       " 'identifier',\n",
       " 'ignore',\n",
       " 'intergenic',\n",
       " 'intergenic_annotation',\n",
       " 'length',\n",
       " 'location',\n",
       " 'log',\n",
       " 'metric',\n",
       " 'metric_by',\n",
       " 'metric_ides',\n",
       " 'metric_lengths',\n",
       " 'metric_values',\n",
       " 'original_annotation',\n",
       " 'passage',\n",
       " 'plot_profile',\n",
       " 'plot_read_density',\n",
       " 'positions',\n",
       " 'prioritize_regions_by_metric',\n",
       " 'profile',\n",
       " 'profile_array',\n",
       " 'range_RAUC',\n",
       " 'reads',\n",
       " 'reduce',\n",
       " 'reduce_old',\n",
       " 'remove_positions',\n",
       " 'replicate',\n",
       " 'save_dataset',\n",
       " 'set_additional_annotation',\n",
       " 'set_domains',\n",
       " 'set_goldset',\n",
       " 'set_ignore',\n",
       " 'terminal_shortening',\n",
       " 'time',\n",
       " 'treatment',\n",
       " 'write_profile',\n",
       " 'zpositions',\n",
       " 'zprofile',\n",
       " 'zreads']"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d = anubis.dataset('./test/P02_R1_U0_PE1.qins', \n",
    "                   genome='./test/myco_pneumoniae_m129.gb', \n",
    "                   goldset='./test/goldsets.csv')\n",
    "d.annotation = {k:v for k, v in d.annotation.items() if len(k)==6}\n",
    "dir(d)    # dataset implements multiple functions to work with this kind of files"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hideCode": false,
    "hidePrompt": false
   },
   "source": [
    "### Calculate metrics\n",
    "\n",
    "Metric are calculated for annotations by default:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "hideCode": false,
    "hidePrompt": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/smiravet/crg/anubis/anubis/anubis/anubis_core.py:784: RuntimeWarning: invalid value encountered in true_divide\n",
      "  return multiplying_factor*(np.array(value)-min(value))/(max(value)-min(value))\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sum</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>median</th>\n",
       "      <th>count</th>\n",
       "      <th>L</th>\n",
       "      <th>min</th>\n",
       "      <th>max</th>\n",
       "      <th>I</th>\n",
       "      <th>dens</th>\n",
       "      <th>...</th>\n",
       "      <th>norm_count</th>\n",
       "      <th>norm_L</th>\n",
       "      <th>norm_min</th>\n",
       "      <th>norm_max</th>\n",
       "      <th>norm_I</th>\n",
       "      <th>norm_dens</th>\n",
       "      <th>norm_R</th>\n",
       "      <th>norm_RI</th>\n",
       "      <th>norm_CPM</th>\n",
       "      <th>norm_RPKM</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ides</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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",
       "      <td>MPN001</td>\n",
       "      <td>83.0</td>\n",
       "      <td>1.408811</td>\n",
       "      <td>0.080661</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1029</td>\n",
       "      <td>1029</td>\n",
       "      <td>0.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.008746</td>\n",
       "      <td>...</td>\n",
       "      <td>0.191549</td>\n",
       "      <td>0.191549</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.002234</td>\n",
       "      <td>0.004907</td>\n",
       "      <td>0.011222</td>\n",
       "      <td>0.000143</td>\n",
       "      <td>2.078326e-06</td>\n",
       "      <td>0.000143</td>\n",
       "      <td>0.000178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPN002</td>\n",
       "      <td>10311.0</td>\n",
       "      <td>49.043396</td>\n",
       "      <td>12.304296</td>\n",
       "      <td>0.0</td>\n",
       "      <td>838</td>\n",
       "      <td>838</td>\n",
       "      <td>0.0</td>\n",
       "      <td>673.0</td>\n",
       "      <td>266.0</td>\n",
       "      <td>0.317422</td>\n",
       "      <td>...</td>\n",
       "      <td>0.153478</td>\n",
       "      <td>0.153478</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.044212</td>\n",
       "      <td>0.145038</td>\n",
       "      <td>0.407281</td>\n",
       "      <td>0.017748</td>\n",
       "      <td>9.370154e-03</td>\n",
       "      <td>0.017748</td>\n",
       "      <td>0.027114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPN003</td>\n",
       "      <td>61.0</td>\n",
       "      <td>0.697141</td>\n",
       "      <td>0.034679</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1759</td>\n",
       "      <td>1759</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.005117</td>\n",
       "      <td>...</td>\n",
       "      <td>0.337054</td>\n",
       "      <td>0.337054</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.001511</td>\n",
       "      <td>0.004907</td>\n",
       "      <td>0.006565</td>\n",
       "      <td>0.000105</td>\n",
       "      <td>8.935418e-07</td>\n",
       "      <td>0.000105</td>\n",
       "      <td>0.000076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPN004</td>\n",
       "      <td>150.0</td>\n",
       "      <td>1.841861</td>\n",
       "      <td>0.066138</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2268</td>\n",
       "      <td>2268</td>\n",
       "      <td>0.0</td>\n",
       "      <td>81.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.005732</td>\n",
       "      <td>...</td>\n",
       "      <td>0.438509</td>\n",
       "      <td>0.438509</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.005321</td>\n",
       "      <td>0.007088</td>\n",
       "      <td>0.007355</td>\n",
       "      <td>0.000258</td>\n",
       "      <td>2.461500e-06</td>\n",
       "      <td>0.000258</td>\n",
       "      <td>0.000146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPN005</td>\n",
       "      <td>22.0</td>\n",
       "      <td>0.247368</td>\n",
       "      <td>0.019349</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1137</td>\n",
       "      <td>1137</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.008795</td>\n",
       "      <td>...</td>\n",
       "      <td>0.213076</td>\n",
       "      <td>0.213076</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000328</td>\n",
       "      <td>0.005453</td>\n",
       "      <td>0.011285</td>\n",
       "      <td>0.000038</td>\n",
       "      <td>5.539500e-07</td>\n",
       "      <td>0.000038</td>\n",
       "      <td>0.000043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPNt33</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>69</td>\n",
       "      <td>69</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000199</td>\n",
       "      <td>0.000199</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPNt34</td>\n",
       "      <td>15.0</td>\n",
       "      <td>1.779991</td>\n",
       "      <td>0.214286</td>\n",
       "      <td>0.0</td>\n",
       "      <td>70</td>\n",
       "      <td>70</td>\n",
       "      <td>0.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000985</td>\n",
       "      <td>0.000545</td>\n",
       "      <td>0.018330</td>\n",
       "      <td>0.000026</td>\n",
       "      <td>6.134816e-07</td>\n",
       "      <td>0.000026</td>\n",
       "      <td>0.000472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPNt35</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>78</td>\n",
       "      <td>78</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001993</td>\n",
       "      <td>0.001993</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPNt36</td>\n",
       "      <td>201.0</td>\n",
       "      <td>11.429597</td>\n",
       "      <td>2.716216</td>\n",
       "      <td>0.0</td>\n",
       "      <td>74</td>\n",
       "      <td>74</td>\n",
       "      <td>0.0</td>\n",
       "      <td>85.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.121622</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001196</td>\n",
       "      <td>0.001196</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.005584</td>\n",
       "      <td>0.004907</td>\n",
       "      <td>0.156051</td>\n",
       "      <td>0.000346</td>\n",
       "      <td>6.998665e-05</td>\n",
       "      <td>0.000346</td>\n",
       "      <td>0.005985</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPNt37</td>\n",
       "      <td>40.0</td>\n",
       "      <td>4.746642</td>\n",
       "      <td>0.571429</td>\n",
       "      <td>0.0</td>\n",
       "      <td>70</td>\n",
       "      <td>70</td>\n",
       "      <td>0.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.002628</td>\n",
       "      <td>0.000545</td>\n",
       "      <td>0.018330</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>1.635951e-06</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>0.001259</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>687 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            sum        std       mean  median  count     L  min    max      I  \\\n",
       "ides                                                                            \n",
       "MPN001     83.0   1.408811   0.080661     0.0   1029  1029  0.0   34.0    9.0   \n",
       "MPN002  10311.0  49.043396  12.304296     0.0    838   838  0.0  673.0  266.0   \n",
       "MPN003     61.0   0.697141   0.034679     0.0   1759  1759  0.0   23.0    9.0   \n",
       "MPN004    150.0   1.841861   0.066138     0.0   2268  2268  0.0   81.0   13.0   \n",
       "MPN005     22.0   0.247368   0.019349     0.0   1137  1137  0.0    5.0   10.0   \n",
       "...         ...        ...        ...     ...    ...   ...  ...    ...    ...   \n",
       "MPNt33      0.0   0.000000   0.000000     0.0     69    69  0.0    0.0    0.0   \n",
       "MPNt34     15.0   1.779991   0.214286     0.0     70    70  0.0   15.0    1.0   \n",
       "MPNt35      0.0   0.000000   0.000000     0.0     78    78  0.0    0.0    0.0   \n",
       "MPNt36    201.0  11.429597   2.716216     0.0     74    74  0.0   85.0    9.0   \n",
       "MPNt37     40.0   4.746642   0.571429     0.0     70    70  0.0   40.0    1.0   \n",
       "\n",
       "            dens  ...  norm_count    norm_L  norm_min  norm_max    norm_I  \\\n",
       "ides              ...                                                       \n",
       "MPN001  0.008746  ...    0.191549  0.191549       NaN  0.002234  0.004907   \n",
       "MPN002  0.317422  ...    0.153478  0.153478       NaN  0.044212  0.145038   \n",
       "MPN003  0.005117  ...    0.337054  0.337054       NaN  0.001511  0.004907   \n",
       "MPN004  0.005732  ...    0.438509  0.438509       NaN  0.005321  0.007088   \n",
       "MPN005  0.008795  ...    0.213076  0.213076       NaN  0.000328  0.005453   \n",
       "...          ...  ...         ...       ...       ...       ...       ...   \n",
       "MPNt33  0.000000  ...    0.000199  0.000199       NaN  0.000000  0.000000   \n",
       "MPNt34  0.014286  ...    0.000399  0.000399       NaN  0.000985  0.000545   \n",
       "MPNt35  0.000000  ...    0.001993  0.001993       NaN  0.000000  0.000000   \n",
       "MPNt36  0.121622  ...    0.001196  0.001196       NaN  0.005584  0.004907   \n",
       "MPNt37  0.014286  ...    0.000399  0.000399       NaN  0.002628  0.000545   \n",
       "\n",
       "        norm_dens    norm_R       norm_RI  norm_CPM  norm_RPKM  \n",
       "ides                                                            \n",
       "MPN001   0.011222  0.000143  2.078326e-06  0.000143   0.000178  \n",
       "MPN002   0.407281  0.017748  9.370154e-03  0.017748   0.027114  \n",
       "MPN003   0.006565  0.000105  8.935418e-07  0.000105   0.000076  \n",
       "MPN004   0.007355  0.000258  2.461500e-06  0.000258   0.000146  \n",
       "MPN005   0.011285  0.000038  5.539500e-07  0.000038   0.000043  \n",
       "...           ...       ...           ...       ...        ...  \n",
       "MPNt33   0.000000  0.000000  0.000000e+00  0.000000   0.000000  \n",
       "MPNt34   0.018330  0.000026  6.134816e-07  0.000026   0.000472  \n",
       "MPNt35   0.000000  0.000000  0.000000e+00  0.000000   0.000000  \n",
       "MPNt36   0.156051  0.000346  6.998665e-05  0.000346   0.005985  \n",
       "MPNt37   0.018330  0.000069  1.635951e-06  0.000069   0.001259  \n",
       "\n",
       "[687 rows x 28 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d.metric_by()   # compute all metrics by default, normalizing \n",
    "d.metric        # metrics are stored in a dataframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hideCode": false,
    "hidePrompt": false
   },
   "source": [
    "Metrics can be selected and computed for 'bin' or 'window':"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>I</th>\n",
       "      <th>norm_I</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ides</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>win000001</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>win000002</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>win000003</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>win000004</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>win000005</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>win816390</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>win816391</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>win816392</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>win816393</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>win816394</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>816394 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             I  norm_I\n",
       "ides                  \n",
       "win000001  0.0     0.0\n",
       "win000002  0.0     0.0\n",
       "win000003  0.0     0.0\n",
       "win000004  0.0     0.0\n",
       "win000005  0.0     0.0\n",
       "...        ...     ...\n",
       "win816390  0.0     0.0\n",
       "win816391  0.0     0.0\n",
       "win816392  0.0     0.0\n",
       "win816393  0.0     0.0\n",
       "win816394  0.0     0.0\n",
       "\n",
       "[816394 rows x 2 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d.metric_by(metric='I', by='window', window_size=20)   # extract insertions for a sliding window of size=20\n",
    "d.metric        # metrics are stored in a dataframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hideCode": false,
    "hidePrompt": false
   },
   "source": [
    "### Filtering reads\n",
    "\n",
    "Datasets can be filtered by read count values:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25.9449481500354 13.749243625014392\n"
     ]
    }
   ],
   "source": [
    "d1 = d.filter(filter_by='genes', inplace=False) # If inplace=True --> substitus orginal dataset\n",
    "print(d.coverage, d1.coverage) # see the coverages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25.9449481500354 19.969891988427158\n"
     ]
    }
   ],
   "source": [
    "d1 = d.filter(filter_by='tails', lpercentile=15, rpercentile=90) # Remove the <15th percentile and >10th percentile based on read count\n",
    "print(d.coverage, d1.coverage) # see the coverages"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hideCode": false,
    "hidePrompt": false
   },
   "source": [
    "### Explore sequence biases\n",
    "\n",
    "Sequencing biases can be detected by binning and visual exploration and corrected using Conditional Quantile Normalization (cqn)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "hideCode": false,
    "hidePrompt": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[27.586206896551722, 32.856926799758014, 38.127646702964306, 43.3983666061706, 48.66908650937689, 53.93980641258318, 60.21052631578947]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x576 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# It can be done for GC\n",
    "corrected = d.cqn(condition='GC', bins=6, same_size=False, normalization='minmax', annotation='all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "hideCode": false,
    "hidePrompt": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>dens</th>\n",
       "      <th>dens_by_GC</th>\n",
       "      <th>gc</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>MPN001</td>\n",
       "      <td>MPN001</td>\n",
       "      <td>0.008746</td>\n",
       "      <td>0.011222</td>\n",
       "      <td>33.916084</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPN002</td>\n",
       "      <td>MPN002</td>\n",
       "      <td>0.317422</td>\n",
       "      <td>0.454972</td>\n",
       "      <td>42.534909</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPN003</td>\n",
       "      <td>MPN003</td>\n",
       "      <td>0.005117</td>\n",
       "      <td>0.008730</td>\n",
       "      <td>45.649949</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPN004</td>\n",
       "      <td>MPN004</td>\n",
       "      <td>0.005732</td>\n",
       "      <td>0.009780</td>\n",
       "      <td>45.537485</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPN005</td>\n",
       "      <td>MPN005</td>\n",
       "      <td>0.008795</td>\n",
       "      <td>0.015007</td>\n",
       "      <td>45.411392</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPNt33</td>\n",
       "      <td>MPNt33</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>59.210526</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPNt34</td>\n",
       "      <td>MPNt34</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.053102</td>\n",
       "      <td>54.545455</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPNt35</td>\n",
       "      <td>MPNt35</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>56.321839</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPNt36</td>\n",
       "      <td>MPNt36</td>\n",
       "      <td>0.121622</td>\n",
       "      <td>0.452080</td>\n",
       "      <td>55.421687</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MPNt37</td>\n",
       "      <td>MPNt37</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.053102</td>\n",
       "      <td>58.441558</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>687 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          name      dens  dens_by_GC         gc  label\n",
       "MPN001  MPN001  0.008746    0.011222  33.916084      1\n",
       "MPN002  MPN002  0.317422    0.454972  42.534909      2\n",
       "MPN003  MPN003  0.005117    0.008730  45.649949      3\n",
       "MPN004  MPN004  0.005732    0.009780  45.537485      3\n",
       "MPN005  MPN005  0.008795    0.015007  45.411392      3\n",
       "...        ...       ...         ...        ...    ...\n",
       "MPNt33  MPNt33  0.000000    0.000000  59.210526      5\n",
       "MPNt34  MPNt34  0.014286    0.053102  54.545455      5\n",
       "MPNt35  MPNt35  0.000000    0.000000  56.321839      5\n",
       "MPNt36  MPNt36  0.121622    0.452080  55.421687      5\n",
       "MPNt37  MPNt37  0.014286    0.053102  58.441558      5\n",
       "\n",
       "[687 rows x 5 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corrected"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Correlate positions and TSD correction\n",
    "\n",
    "Exploring read count correlation between insertions and contiguous positions can unveil artifacts such as Transposon Duplicated Sites, that can increase read count signal artifactually (for example +7 in our case):\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "corr = d.correlation_by_distance(min_relative_distance=-20)\n",
    "\n",
    "plt.scatter(list(corr.keys()), list(corr.values()))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For correction we require reads separated by forward- and reverse-oriented (generated by default by FASTQINS):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25.9449481500354 24.486338704106107\n"
     ]
    }
   ],
   "source": [
    "d1 = d.correct_TSD(fw='./test/P02_R1_U0_PE1_fw.qins', rv='./test/P02_R1_U0_PE1_rv.qins')\n",
    "print(d.coverage, d1.coverage)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Differential Essentiality regions\n",
    "\n",
    "Most important factor are extended NE N' and C'-termini, repeated regions and protein domains."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[265243 265523 265608 265634]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(5.995203836930456, 26.858513189448445)"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# a metagene approach can be used to explore all genes together\n",
    "metaprofile = anubis.metagene(d) # by default computed by 100 bins per bin and dens as metric\n",
    "\n",
    "import ruptures as rpt\n",
    "\n",
    "\n",
    "st, en, _ = d.annotation['MPN214']\n",
    "d._update_roll_(10, metric='dens', overlapping=True)\n",
    "signal = d._roll_[st:en]\n",
    "\n",
    "# detection\n",
    "algo = rpt.Pelt(model=\"rbf\").fit(signal)\n",
    "result = algo.predict(pen=20)\n",
    "\n",
    "# display\n",
    "rpt.display(signal, result, result)\n",
    "plt.show()\n",
    "print(st+np.array(result))\n",
    "\n",
    "\n",
    "d.plot_profile('MPN214')\n",
    "anubis.index2perc(result[0], en-st+1), 100-anubis.index2perc(result[1], en-st+1)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also add domains from a HMMER prediction to plot below a profile:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[['PF09185.9|DUF1948', 34556, 34973, 2.9e-97]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "[<Figure size 1296x216 with 1 Axes>,\n",
       " <matplotlib.axes._subplots.AxesSubplot at 0x7f97bc4bf630>]"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d.set_domains('./test/perdom.out')\n",
    "print(d.domains['MPN030'])\n",
    "d.plot_profile('MPN030', show_domains=True, domain_levels=1, domain_colors=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This can be done with a list of positions considered repeated in the genome."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Reference essentiality estimates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/smiravet/crg/anubis/anubis/anubis/anubis_core.py:784: RuntimeWarning: invalid value encountered in true_divide\n",
      "  return multiplying_factor*(np.array(value)-min(value))/(max(value)-min(value))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "111\n",
      "111\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>identifier</th>\n",
       "      <th>left ann</th>\n",
       "      <th>right ann</th>\n",
       "      <th>strand</th>\n",
       "      <th>len</th>\n",
       "      <th>len_considered</th>\n",
       "      <th>goldset</th>\n",
       "      <th>dens</th>\n",
       "      <th>P(no insertions)</th>\n",
       "      <th>P(saturation)</th>\n",
       "      <th>P(E)</th>\n",
       "      <th>P(N)</th>\n",
       "      <th>LLR [log2(E/N)]</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>MPN001</td>\n",
       "      <td>692</td>\n",
       "      <td>1834</td>\n",
       "      <td>+</td>\n",
       "      <td>1143</td>\n",
       "      <td>1029</td>\n",
       "      <td>E</td>\n",
       "      <td>0.008746</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.017</td>\n",
       "      <td>2.377916e-01</td>\n",
       "      <td>2.642257e-228</td>\n",
       "      <td>753.925604</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>MPN002</td>\n",
       "      <td>1838</td>\n",
       "      <td>2767</td>\n",
       "      <td>+</td>\n",
       "      <td>930</td>\n",
       "      <td>838</td>\n",
       "      <td>u</td>\n",
       "      <td>0.317422</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.019</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.814571e-23</td>\n",
       "      <td>-955.575508</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>MPN003</td>\n",
       "      <td>2869</td>\n",
       "      <td>4821</td>\n",
       "      <td>+</td>\n",
       "      <td>1953</td>\n",
       "      <td>1759</td>\n",
       "      <td>u</td>\n",
       "      <td>0.005117</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.013</td>\n",
       "      <td>8.361796e-01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>inf</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>MPN004</td>\n",
       "      <td>4821</td>\n",
       "      <td>7340</td>\n",
       "      <td>+</td>\n",
       "      <td>2520</td>\n",
       "      <td>2268</td>\n",
       "      <td>u</td>\n",
       "      <td>0.005732</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.011</td>\n",
       "      <td>7.857050e-01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>inf</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>MPN005</td>\n",
       "      <td>7312</td>\n",
       "      <td>8574</td>\n",
       "      <td>+</td>\n",
       "      <td>1263</td>\n",
       "      <td>1137</td>\n",
       "      <td>E</td>\n",
       "      <td>0.008795</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.016</td>\n",
       "      <td>2.281586e-01</td>\n",
       "      <td>5.071284e-252</td>\n",
       "      <td>832.651638</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>682</td>\n",
       "      <td>MPNt33</td>\n",
       "      <td>562556</td>\n",
       "      <td>562630</td>\n",
       "      <td>-</td>\n",
       "      <td>75</td>\n",
       "      <td>69</td>\n",
       "      <td>u</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.065</td>\n",
       "      <td>4.001868e-01</td>\n",
       "      <td>2.815938e-17</td>\n",
       "      <td>53.657908</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>683</td>\n",
       "      <td>MPNt34</td>\n",
       "      <td>643913</td>\n",
       "      <td>643988</td>\n",
       "      <td>+</td>\n",
       "      <td>76</td>\n",
       "      <td>70</td>\n",
       "      <td>u</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.064</td>\n",
       "      <td>9.587920e-02</td>\n",
       "      <td>6.428737e-16</td>\n",
       "      <td>47.083676</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>684</td>\n",
       "      <td>MPNt35</td>\n",
       "      <td>644970</td>\n",
       "      <td>645055</td>\n",
       "      <td>-</td>\n",
       "      <td>86</td>\n",
       "      <td>78</td>\n",
       "      <td>u</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.061</td>\n",
       "      <td>4.388725e-01</td>\n",
       "      <td>1.953792e-19</td>\n",
       "      <td>60.962231</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>685</td>\n",
       "      <td>MPNt36</td>\n",
       "      <td>648912</td>\n",
       "      <td>648993</td>\n",
       "      <td>+</td>\n",
       "      <td>82</td>\n",
       "      <td>74</td>\n",
       "      <td>u</td>\n",
       "      <td>0.121622</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.062</td>\n",
       "      <td>4.106478e-10</td>\n",
       "      <td>1.984611e-09</td>\n",
       "      <td>-2.272883</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>686</td>\n",
       "      <td>MPNt37</td>\n",
       "      <td>680400</td>\n",
       "      <td>680475</td>\n",
       "      <td>-</td>\n",
       "      <td>76</td>\n",
       "      <td>70</td>\n",
       "      <td>u</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.064</td>\n",
       "      <td>9.587920e-02</td>\n",
       "      <td>6.428737e-16</td>\n",
       "      <td>47.083676</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>687 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    identifier  left ann  right ann strand   len  len_considered goldset  \\\n",
       "0       MPN001       692       1834      +  1143            1029       E   \n",
       "1       MPN002      1838       2767      +   930             838       u   \n",
       "2       MPN003      2869       4821      +  1953            1759       u   \n",
       "3       MPN004      4821       7340      +  2520            2268       u   \n",
       "4       MPN005      7312       8574      +  1263            1137       E   \n",
       "..         ...       ...        ...    ...   ...             ...     ...   \n",
       "682     MPNt33    562556     562630      -    75              69       u   \n",
       "683     MPNt34    643913     643988      +    76              70       u   \n",
       "684     MPNt35    644970     645055      -    86              78       u   \n",
       "685     MPNt36    648912     648993      +    82              74       u   \n",
       "686     MPNt37    680400     680475      -    76              70       u   \n",
       "\n",
       "         dens  P(no insertions)  P(saturation)          P(E)           P(N)  \\\n",
       "0    0.008746               0.0          0.017  2.377916e-01  2.642257e-228   \n",
       "1    0.317422               0.0          0.019  0.000000e+00   1.814571e-23   \n",
       "2    0.005117               0.0          0.013  8.361796e-01   0.000000e+00   \n",
       "3    0.005732               0.0          0.011  7.857050e-01   0.000000e+00   \n",
       "4    0.008795               0.0          0.016  2.281586e-01  5.071284e-252   \n",
       "..        ...               ...            ...           ...            ...   \n",
       "682  0.000000               0.0          0.065  4.001868e-01   2.815938e-17   \n",
       "683  0.014286               0.0          0.064  9.587920e-02   6.428737e-16   \n",
       "684  0.000000               0.0          0.061  4.388725e-01   1.953792e-19   \n",
       "685  0.121622               0.0          0.062  4.106478e-10   1.984611e-09   \n",
       "686  0.014286               0.0          0.064  9.587920e-02   6.428737e-16   \n",
       "\n",
       "     LLR [log2(E/N)] class  \n",
       "0         753.925604     F  \n",
       "1        -955.575508     N  \n",
       "2                inf     E  \n",
       "3                inf     E  \n",
       "4         832.651638     F  \n",
       "..               ...   ...  \n",
       "682        53.657908     E  \n",
       "683        47.083676     F  \n",
       "684        60.962231     E  \n",
       "685        -2.272883     F  \n",
       "686        47.083676     F  \n",
       "\n",
       "[687 rows x 14 columns]"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Poisson, absolute criteria\n",
    "\n",
    "d.metric_by()\n",
    "anubis.Poisson(d).predict(criteria='absolute')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "111\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>identifier</th>\n",
       "      <th>left ann</th>\n",
       "      <th>right ann</th>\n",
       "      <th>strand</th>\n",
       "      <th>len</th>\n",
       "      <th>len_considered</th>\n",
       "      <th>goldset</th>\n",
       "      <th>dens</th>\n",
       "      <th>P(no insertions)</th>\n",
       "      <th>P(saturation)</th>\n",
       "      <th>P(E)</th>\n",
       "      <th>P(N)</th>\n",
       "      <th>LLR [log2(E/N)]</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>MPN001</td>\n",
       "      <td>692</td>\n",
       "      <td>1834</td>\n",
       "      <td>+</td>\n",
       "      <td>1143</td>\n",
       "      <td>1029</td>\n",
       "      <td>E</td>\n",
       "      <td>0.008746</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.018</td>\n",
       "      <td>0.222938</td>\n",
       "      <td>5.216849e-207</td>\n",
       "      <td>683.090654</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>MPN002</td>\n",
       "      <td>1838</td>\n",
       "      <td>2767</td>\n",
       "      <td>+</td>\n",
       "      <td>930</td>\n",
       "      <td>838</td>\n",
       "      <td>u</td>\n",
       "      <td>0.317422</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.020</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.219115e-56</td>\n",
       "      <td>-881.992728</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>MPN003</td>\n",
       "      <td>2869</td>\n",
       "      <td>4821</td>\n",
       "      <td>+</td>\n",
       "      <td>1953</td>\n",
       "      <td>1759</td>\n",
       "      <td>u</td>\n",
       "      <td>0.005117</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.014</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>inf</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>MPN004</td>\n",
       "      <td>4821</td>\n",
       "      <td>7340</td>\n",
       "      <td>+</td>\n",
       "      <td>2520</td>\n",
       "      <td>2268</td>\n",
       "      <td>u</td>\n",
       "      <td>0.005732</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.012</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>inf</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>MPN005</td>\n",
       "      <td>7312</td>\n",
       "      <td>8574</td>\n",
       "      <td>+</td>\n",
       "      <td>1263</td>\n",
       "      <td>1137</td>\n",
       "      <td>E</td>\n",
       "      <td>0.008795</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.017</td>\n",
       "      <td>0.186591</td>\n",
       "      <td>1.232578e-228</td>\n",
       "      <td>754.675880</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>682</td>\n",
       "      <td>MPNt33</td>\n",
       "      <td>562556</td>\n",
       "      <td>562630</td>\n",
       "      <td>-</td>\n",
       "      <td>75</td>\n",
       "      <td>69</td>\n",
       "      <td>u</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.070</td>\n",
       "      <td>0.811222</td>\n",
       "      <td>8.045324e-15</td>\n",
       "      <td>46.518940</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>683</td>\n",
       "      <td>MPNt34</td>\n",
       "      <td>643913</td>\n",
       "      <td>643988</td>\n",
       "      <td>+</td>\n",
       "      <td>76</td>\n",
       "      <td>70</td>\n",
       "      <td>u</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.069</td>\n",
       "      <td>0.460916</td>\n",
       "      <td>1.366377e-14</td>\n",
       "      <td>44.939213</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>684</td>\n",
       "      <td>MPNt35</td>\n",
       "      <td>644970</td>\n",
       "      <td>645055</td>\n",
       "      <td>-</td>\n",
       "      <td>86</td>\n",
       "      <td>78</td>\n",
       "      <td>u</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.066</td>\n",
       "      <td>0.831720</td>\n",
       "      <td>1.167176e-16</td>\n",
       "      <td>52.661998</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>685</td>\n",
       "      <td>MPNt36</td>\n",
       "      <td>648912</td>\n",
       "      <td>648993</td>\n",
       "      <td>+</td>\n",
       "      <td>82</td>\n",
       "      <td>74</td>\n",
       "      <td>u</td>\n",
       "      <td>0.121622</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.068</td>\n",
       "      <td>0.000213</td>\n",
       "      <td>6.206638e-12</td>\n",
       "      <td>25.035766</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>686</td>\n",
       "      <td>MPNt37</td>\n",
       "      <td>680400</td>\n",
       "      <td>680475</td>\n",
       "      <td>-</td>\n",
       "      <td>76</td>\n",
       "      <td>70</td>\n",
       "      <td>u</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.069</td>\n",
       "      <td>0.460916</td>\n",
       "      <td>1.366377e-14</td>\n",
       "      <td>44.939213</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>687 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    identifier  left ann  right ann strand   len  len_considered goldset  \\\n",
       "0       MPN001       692       1834      +  1143            1029       E   \n",
       "1       MPN002      1838       2767      +   930             838       u   \n",
       "2       MPN003      2869       4821      +  1953            1759       u   \n",
       "3       MPN004      4821       7340      +  2520            2268       u   \n",
       "4       MPN005      7312       8574      +  1263            1137       E   \n",
       "..         ...       ...        ...    ...   ...             ...     ...   \n",
       "682     MPNt33    562556     562630      -    75              69       u   \n",
       "683     MPNt34    643913     643988      +    76              70       u   \n",
       "684     MPNt35    644970     645055      -    86              78       u   \n",
       "685     MPNt36    648912     648993      +    82              74       u   \n",
       "686     MPNt37    680400     680475      -    76              70       u   \n",
       "\n",
       "         dens  P(no insertions)  P(saturation)      P(E)           P(N)  \\\n",
       "0    0.008746               0.0          0.018  0.222938  5.216849e-207   \n",
       "1    0.317422               0.0          0.020  0.000000   2.219115e-56   \n",
       "2    0.005117               0.0          0.014  1.000000   0.000000e+00   \n",
       "3    0.005732               0.0          0.012  1.000000   0.000000e+00   \n",
       "4    0.008795               0.0          0.017  0.186591  1.232578e-228   \n",
       "..        ...               ...            ...       ...            ...   \n",
       "682  0.000000               0.0          0.070  0.811222   8.045324e-15   \n",
       "683  0.014286               0.0          0.069  0.460916   1.366377e-14   \n",
       "684  0.000000               0.0          0.066  0.831720   1.167176e-16   \n",
       "685  0.121622               0.0          0.068  0.000213   6.206638e-12   \n",
       "686  0.014286               0.0          0.069  0.460916   1.366377e-14   \n",
       "\n",
       "     LLR [log2(E/N)] class  \n",
       "0         683.090654     E  \n",
       "1        -881.992728     F  \n",
       "2                inf     E  \n",
       "3                inf     E  \n",
       "4         754.675880     E  \n",
       "..               ...   ...  \n",
       "682        46.518940     E  \n",
       "683        44.939213     E  \n",
       "684        52.661998     E  \n",
       "685        25.035766     F  \n",
       "686        44.939213     E  \n",
       "\n",
       "[687 rows x 14 columns]"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Gumbel, threshold criteria, NE gold set --> non-coding regions\n",
    "\n",
    "d.metric_by()\n",
    "anubis.Gumbel(d).predict(criteria='threshold', thr=0.01, prior={'E':'goldset', 'N':'intergenic'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>identifier</th>\n",
       "      <th>left ann</th>\n",
       "      <th>right ann</th>\n",
       "      <th>strand</th>\n",
       "      <th>len</th>\n",
       "      <th>len_considered</th>\n",
       "      <th>goldset</th>\n",
       "      <th>dens</th>\n",
       "      <th>P(no insertions)</th>\n",
       "      <th>P(saturation)</th>\n",
       "      <th>P(E)</th>\n",
       "      <th>P(N)</th>\n",
       "      <th>LLR [log2(E/N)]</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>MPN001</td>\n",
       "      <td>692</td>\n",
       "      <td>1834</td>\n",
       "      <td>+</td>\n",
       "      <td>1143</td>\n",
       "      <td>1029</td>\n",
       "      <td>u</td>\n",
       "      <td>0.008746</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.018</td>\n",
       "      <td>0.000123</td>\n",
       "      <td>5.216849e-207</td>\n",
       "      <td>672.271593</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>MPN002</td>\n",
       "      <td>1838</td>\n",
       "      <td>2767</td>\n",
       "      <td>+</td>\n",
       "      <td>930</td>\n",
       "      <td>838</td>\n",
       "      <td>u</td>\n",
       "      <td>0.317422</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.020</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.219115e-56</td>\n",
       "      <td>-881.992728</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>MPN003</td>\n",
       "      <td>2869</td>\n",
       "      <td>4821</td>\n",
       "      <td>+</td>\n",
       "      <td>1953</td>\n",
       "      <td>1759</td>\n",
       "      <td>u</td>\n",
       "      <td>0.005117</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.014</td>\n",
       "      <td>0.000123</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>inf</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>MPN004</td>\n",
       "      <td>4821</td>\n",
       "      <td>7340</td>\n",
       "      <td>+</td>\n",
       "      <td>2520</td>\n",
       "      <td>2268</td>\n",
       "      <td>u</td>\n",
       "      <td>0.005732</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.012</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>inf</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>MPN005</td>\n",
       "      <td>7312</td>\n",
       "      <td>8574</td>\n",
       "      <td>+</td>\n",
       "      <td>1263</td>\n",
       "      <td>1137</td>\n",
       "      <td>u</td>\n",
       "      <td>0.008795</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.017</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>1.232578e-228</td>\n",
       "      <td>742.670944</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>682</td>\n",
       "      <td>MPNt33</td>\n",
       "      <td>562556</td>\n",
       "      <td>562630</td>\n",
       "      <td>-</td>\n",
       "      <td>75</td>\n",
       "      <td>69</td>\n",
       "      <td>u</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.070</td>\n",
       "      <td>0.632121</td>\n",
       "      <td>8.045324e-15</td>\n",
       "      <td>46.159043</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>683</td>\n",
       "      <td>MPNt34</td>\n",
       "      <td>643913</td>\n",
       "      <td>643988</td>\n",
       "      <td>+</td>\n",
       "      <td>76</td>\n",
       "      <td>70</td>\n",
       "      <td>u</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.069</td>\n",
       "      <td>0.307799</td>\n",
       "      <td>1.366377e-14</td>\n",
       "      <td>44.356700</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>684</td>\n",
       "      <td>MPNt35</td>\n",
       "      <td>644970</td>\n",
       "      <td>645055</td>\n",
       "      <td>-</td>\n",
       "      <td>86</td>\n",
       "      <td>78</td>\n",
       "      <td>u</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.066</td>\n",
       "      <td>0.632121</td>\n",
       "      <td>1.167176e-16</td>\n",
       "      <td>52.266100</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>685</td>\n",
       "      <td>MPNt36</td>\n",
       "      <td>648912</td>\n",
       "      <td>648993</td>\n",
       "      <td>+</td>\n",
       "      <td>82</td>\n",
       "      <td>74</td>\n",
       "      <td>u</td>\n",
       "      <td>0.121622</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.068</td>\n",
       "      <td>0.000123</td>\n",
       "      <td>6.206638e-12</td>\n",
       "      <td>24.244981</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>686</td>\n",
       "      <td>MPNt37</td>\n",
       "      <td>680400</td>\n",
       "      <td>680475</td>\n",
       "      <td>-</td>\n",
       "      <td>76</td>\n",
       "      <td>70</td>\n",
       "      <td>u</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.069</td>\n",
       "      <td>0.307799</td>\n",
       "      <td>1.366377e-14</td>\n",
       "      <td>44.356700</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>687 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    identifier  left ann  right ann strand   len  len_considered goldset  \\\n",
       "0       MPN001       692       1834      +  1143            1029       u   \n",
       "1       MPN002      1838       2767      +   930             838       u   \n",
       "2       MPN003      2869       4821      +  1953            1759       u   \n",
       "3       MPN004      4821       7340      +  2520            2268       u   \n",
       "4       MPN005      7312       8574      +  1263            1137       u   \n",
       "..         ...       ...        ...    ...   ...             ...     ...   \n",
       "682     MPNt33    562556     562630      -    75              69       u   \n",
       "683     MPNt34    643913     643988      +    76              70       u   \n",
       "684     MPNt35    644970     645055      -    86              78       u   \n",
       "685     MPNt36    648912     648993      +    82              74       u   \n",
       "686     MPNt37    680400     680475      -    76              70       u   \n",
       "\n",
       "         dens  P(no insertions)  P(saturation)      P(E)           P(N)  \\\n",
       "0    0.008746               0.0          0.018  0.000123  5.216849e-207   \n",
       "1    0.317422               0.0          0.020  0.000000   2.219115e-56   \n",
       "2    0.005117               0.0          0.014  0.000123   0.000000e+00   \n",
       "3    0.005732               0.0          0.012  0.000002   0.000000e+00   \n",
       "4    0.008795               0.0          0.017  0.000045  1.232578e-228   \n",
       "..        ...               ...            ...       ...            ...   \n",
       "682  0.000000               0.0          0.070  0.632121   8.045324e-15   \n",
       "683  0.014286               0.0          0.069  0.307799   1.366377e-14   \n",
       "684  0.000000               0.0          0.066  0.632121   1.167176e-16   \n",
       "685  0.121622               0.0          0.068  0.000123   6.206638e-12   \n",
       "686  0.014286               0.0          0.069  0.307799   1.366377e-14   \n",
       "\n",
       "     LLR [log2(E/N)] class  \n",
       "0         672.271593     E  \n",
       "1        -881.992728     N  \n",
       "2                inf     E  \n",
       "3                inf     E  \n",
       "4         742.670944     E  \n",
       "..               ...   ...  \n",
       "682        46.159043     E  \n",
       "683        44.356700     E  \n",
       "684        52.266100     E  \n",
       "685        24.244981     E  \n",
       "686        44.356700     E  \n",
       "\n",
       "[687 rows x 14 columns]"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Gamma, foldchange criteria with threshold>3. NE gold set --> non-coding regions and assume E genes linear density=0\n",
    "\n",
    "d.metric_by()\n",
    "anubis.Gumbel(d).predict(criteria='foldchange', foldchange=3, prior={'E':'0', 'N':'intergenic'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mixture models\n",
    "\n",
    "These model have as an advantage the minimal parameterization required:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/smiravet/crg/anubis/anubis/anubis/anubis_core.py:784: RuntimeWarning: invalid value encountered in true_divide\n",
      "  return multiplying_factor*(np.array(value)-min(value))/(max(value)-min(value))\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>identifier</th>\n",
       "      <th>left ann</th>\n",
       "      <th>right ann</th>\n",
       "      <th>strand</th>\n",
       "      <th>len</th>\n",
       "      <th>len_considered</th>\n",
       "      <th>dens</th>\n",
       "      <th>P(no insertions)</th>\n",
       "      <th>P(saturation)</th>\n",
       "      <th>P(E)</th>\n",
       "      <th>P(F)</th>\n",
       "      <th>P(N)</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>MPN001</td>\n",
       "      <td>692</td>\n",
       "      <td>1834</td>\n",
       "      <td>+</td>\n",
       "      <td>1143</td>\n",
       "      <td>1029</td>\n",
       "      <td>0.008746</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.018</td>\n",
       "      <td>31.837</td>\n",
       "      <td>1.016</td>\n",
       "      <td>0.012</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>MPN002</td>\n",
       "      <td>1838</td>\n",
       "      <td>2767</td>\n",
       "      <td>+</td>\n",
       "      <td>930</td>\n",
       "      <td>838</td>\n",
       "      <td>0.317422</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.020</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.677</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>MPN003</td>\n",
       "      <td>2869</td>\n",
       "      <td>4821</td>\n",
       "      <td>+</td>\n",
       "      <td>1953</td>\n",
       "      <td>1759</td>\n",
       "      <td>0.005117</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.014</td>\n",
       "      <td>28.118</td>\n",
       "      <td>0.903</td>\n",
       "      <td>0.011</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>MPN004</td>\n",
       "      <td>4821</td>\n",
       "      <td>7340</td>\n",
       "      <td>+</td>\n",
       "      <td>2520</td>\n",
       "      <td>2268</td>\n",
       "      <td>0.005732</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.012</td>\n",
       "      <td>30.215</td>\n",
       "      <td>0.922</td>\n",
       "      <td>0.011</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>MPN005</td>\n",
       "      <td>7312</td>\n",
       "      <td>8574</td>\n",
       "      <td>+</td>\n",
       "      <td>1263</td>\n",
       "      <td>1137</td>\n",
       "      <td>0.008795</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.017</td>\n",
       "      <td>31.734</td>\n",
       "      <td>1.018</td>\n",
       "      <td>0.012</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>682</td>\n",
       "      <td>MPNt33</td>\n",
       "      <td>562556</td>\n",
       "      <td>562630</td>\n",
       "      <td>-</td>\n",
       "      <td>75</td>\n",
       "      <td>69</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.070</td>\n",
       "      <td>6.920</td>\n",
       "      <td>0.755</td>\n",
       "      <td>0.010</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>683</td>\n",
       "      <td>MPNt34</td>\n",
       "      <td>643913</td>\n",
       "      <td>643988</td>\n",
       "      <td>+</td>\n",
       "      <td>76</td>\n",
       "      <td>70</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.070</td>\n",
       "      <td>9.560</td>\n",
       "      <td>1.196</td>\n",
       "      <td>0.013</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>684</td>\n",
       "      <td>MPNt35</td>\n",
       "      <td>644970</td>\n",
       "      <td>645055</td>\n",
       "      <td>-</td>\n",
       "      <td>86</td>\n",
       "      <td>78</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.066</td>\n",
       "      <td>6.920</td>\n",
       "      <td>0.755</td>\n",
       "      <td>0.010</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>685</td>\n",
       "      <td>MPNt36</td>\n",
       "      <td>648912</td>\n",
       "      <td>648993</td>\n",
       "      <td>+</td>\n",
       "      <td>82</td>\n",
       "      <td>74</td>\n",
       "      <td>0.121622</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.068</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.625</td>\n",
       "      <td>0.083</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>686</td>\n",
       "      <td>MPNt37</td>\n",
       "      <td>680400</td>\n",
       "      <td>680475</td>\n",
       "      <td>-</td>\n",
       "      <td>76</td>\n",
       "      <td>70</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.070</td>\n",
       "      <td>9.560</td>\n",
       "      <td>1.196</td>\n",
       "      <td>0.013</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>687 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    identifier  left ann  right ann strand   len  len_considered      dens  \\\n",
       "0       MPN001       692       1834      +  1143            1029  0.008746   \n",
       "1       MPN002      1838       2767      +   930             838  0.317422   \n",
       "2       MPN003      2869       4821      +  1953            1759  0.005117   \n",
       "3       MPN004      4821       7340      +  2520            2268  0.005732   \n",
       "4       MPN005      7312       8574      +  1263            1137  0.008795   \n",
       "..         ...       ...        ...    ...   ...             ...       ...   \n",
       "682     MPNt33    562556     562630      -    75              69  0.000000   \n",
       "683     MPNt34    643913     643988      +    76              70  0.014286   \n",
       "684     MPNt35    644970     645055      -    86              78  0.000000   \n",
       "685     MPNt36    648912     648993      +    82              74  0.121622   \n",
       "686     MPNt37    680400     680475      -    76              70  0.014286   \n",
       "\n",
       "     P(no insertions)  P(saturation)    P(E)   P(F)   P(N) class  \n",
       "0                 0.0          0.018  31.837  1.016  0.012     E  \n",
       "1                 0.0          0.020   0.000  0.000  0.677     N  \n",
       "2                 0.0          0.014  28.118  0.903  0.011     E  \n",
       "3                 0.0          0.012  30.215  0.922  0.011     E  \n",
       "4                 0.0          0.017  31.734  1.018  0.012     E  \n",
       "..                ...            ...     ...    ...    ...   ...  \n",
       "682               0.0          0.070   6.920  0.755  0.010     E  \n",
       "683               0.0          0.070   9.560  1.196  0.013     E  \n",
       "684               0.0          0.066   6.920  0.755  0.010     E  \n",
       "685               0.0          0.068   0.000  0.625  0.083     F  \n",
       "686               0.0          0.070   9.560  1.196  0.013     E  \n",
       "\n",
       "[687 rows x 13 columns]"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d.metric_by()\n",
    "anubis.GaussianMixture(d).predict()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "## Contact\n",
    "\n",
    "This project has been fully developed at [Centre for Genomic Regulation](http://www.crg.eu/) at the group of [Design of Biological Systems](http://www.crg.eu/en/luis_serrano).\n",
    "\n",
    "If you experience any problem at any step involving the program, you can use the 'Issues' page of this repository or contact:\n",
    "\n",
    "[Miravet-Verde, Samuel](mailto:samuel.miravet@crg.eu)       \n",
    "[Lluch-Senar, Maria](mailto:maria.lluch@crg.eu)       \n",
    "[Serrano, Luis](mailto:luis.serrano@crg.eu)"
   ]
  }
 ],
 "metadata": {
  "hide_code_all_hidden": false,
  "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}