notebook/Codon Bias vs Pfam.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import collections\n",
    "import os\n",
    "import json\n",
    "import logging\n",
    "import string\n",
    "import re\n",
    "\n",
    "from scipy.stats import entropy\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sqlalchemy import create_engine\n",
    "import statsmodels.api as sm\n",
    "from statsmodels.sandbox.regression.predstd import wls_prediction_std\n",
    "import networkx as nx\n",
    "\n",
    "if os.getcwd().endswith('notebook'):\n",
    "    os.chdir('..')\n",
    "\n",
    "from rna_learn.codon_bias.process_domains import compute_protein_domain_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.set(palette='colorblind', font_scale=1.3)\n",
    "palette = sns.color_palette()\n",
    "logging.basicConfig(level=logging.INFO, format=\"%(asctime)s (%(levelname)s) %(message)s\")\n",
    "logger = logging.getLogger(__name__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "db_path = os.path.join(os.getcwd(), 'data/db/seq.db')\n",
    "engine = create_engine(f'sqlite+pysqlite:///{db_path}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Match"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_match_score(engine):\n",
    "    q = \"\"\"\n",
    "    select assembly_accession from assembly_source\n",
    "    \"\"\"\n",
    "    assembly_accessions = pd.read_sql(q, engine)['assembly_accession'].values\n",
    "    \n",
    "    count = 0\n",
    "    for assembly in assembly_accessions:\n",
    "        protein_domains_path = os.path.join(\n",
    "            os.getcwd(), \n",
    "            f'data/domains/codon_bias/pfam/{assembly}_protein_domains.csv',\n",
    "        )\n",
    "        if os.path.isfile(protein_domains_path):\n",
    "            count += 1\n",
    "            \n",
    "    return count, 100 * count / len(assembly_accessions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1217, 37.08104814137721)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_match_score(engine)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Thermotoga maritima"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "species_taxid = 2336"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'GCA_000008545.1'"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "q = \"\"\"\n",
    "select assembly_accession, species_taxid from assembly_source\n",
    "where species_taxid = ?\n",
    "\"\"\"\n",
    "assembly_accession = pd.read_sql(q, engine, params=(species_taxid,))['assembly_accession'].iloc[0]\n",
    "assembly_accession"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "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>assembly_accession</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>record_type</th>\n",
       "      <th>pfam_query</th>\n",
       "      <th>pfam_accession</th>\n",
       "      <th>protein_label</th>\n",
       "      <th>below_threshold</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD36281.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MrpF_PhaF</td>\n",
       "      <td>PF04066.13</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD36013.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>Mrr_cat</td>\n",
       "      <td>PF04471.12</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35343.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>Mrr_cat</td>\n",
       "      <td>PF04471.12</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35100.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MR_MLE_C</td>\n",
       "      <td>PF13378.6</td>\n",
       "      <td>muconate-cycloisomerase</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35100.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MR_MLE_N</td>\n",
       "      <td>PF02746.16</td>\n",
       "      <td>muconate-cycloisomerase</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD36629.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MS_channel</td>\n",
       "      <td>PF00924.18</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35703.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTES_1575</td>\n",
       "      <td>PF18741.1</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35359.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTHFR</td>\n",
       "      <td>PF02219.17</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD36731.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD36156.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>ribosomal-protein-L11-methyltransferase,-putative</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35573.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>hemK-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35773.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD36458.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD36019.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35835.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>ubiquinone/menaquinone-biosynthesis-methyltran...</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD36764.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35789.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>glucose-inhibited-division-protein-B</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35954.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD35829.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>MTS</td>\n",
       "      <td>PF05175.14</td>\n",
       "      <td>conserved-hypothetical-protein</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>GCA_000008545.1</td>\n",
       "      <td>AAD36681.1</td>\n",
       "      <td>pfam</td>\n",
       "      <td>Mt_ATP-synt_B</td>\n",
       "      <td>PF05405.14</td>\n",
       "      <td>ATP-synthase-F0,-subunit-b</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   assembly_accession  protein_id record_type     pfam_query pfam_accession  \\\n",
       "0     GCA_000008545.1  AAD36281.1        pfam      MrpF_PhaF     PF04066.13   \n",
       "1     GCA_000008545.1  AAD36013.1        pfam        Mrr_cat     PF04471.12   \n",
       "2     GCA_000008545.1  AAD35343.1        pfam        Mrr_cat     PF04471.12   \n",
       "3     GCA_000008545.1  AAD35100.1        pfam       MR_MLE_C      PF13378.6   \n",
       "4     GCA_000008545.1  AAD35100.1        pfam       MR_MLE_N     PF02746.16   \n",
       "5     GCA_000008545.1  AAD36629.1        pfam     MS_channel     PF00924.18   \n",
       "6     GCA_000008545.1  AAD35703.1        pfam      MTES_1575      PF18741.1   \n",
       "7     GCA_000008545.1  AAD35359.1        pfam          MTHFR     PF02219.17   \n",
       "8     GCA_000008545.1  AAD36731.1        pfam            MTS     PF05175.14   \n",
       "9     GCA_000008545.1  AAD36156.1        pfam            MTS     PF05175.14   \n",
       "10    GCA_000008545.1  AAD35573.1        pfam            MTS     PF05175.14   \n",
       "11    GCA_000008545.1  AAD35773.1        pfam            MTS     PF05175.14   \n",
       "12    GCA_000008545.1  AAD36458.1        pfam            MTS     PF05175.14   \n",
       "13    GCA_000008545.1  AAD36019.1        pfam            MTS     PF05175.14   \n",
       "14    GCA_000008545.1  AAD35835.1        pfam            MTS     PF05175.14   \n",
       "15    GCA_000008545.1  AAD36764.1        pfam            MTS     PF05175.14   \n",
       "16    GCA_000008545.1  AAD35789.1        pfam            MTS     PF05175.14   \n",
       "17    GCA_000008545.1  AAD35954.1        pfam            MTS     PF05175.14   \n",
       "18    GCA_000008545.1  AAD35829.1        pfam            MTS     PF05175.14   \n",
       "19    GCA_000008545.1  AAD36681.1        pfam  Mt_ATP-synt_B     PF05405.14   \n",
       "\n",
       "                                        protein_label  below_threshold  \n",
       "0                      conserved-hypothetical-protein            False  \n",
       "1                      conserved-hypothetical-protein            False  \n",
       "2                      conserved-hypothetical-protein            False  \n",
       "3                             muconate-cycloisomerase            False  \n",
       "4                             muconate-cycloisomerase            False  \n",
       "5                      conserved-hypothetical-protein            False  \n",
       "6                      conserved-hypothetical-protein            False  \n",
       "7                      conserved-hypothetical-protein            False  \n",
       "8                      conserved-hypothetical-protein            False  \n",
       "9   ribosomal-protein-L11-methyltransferase,-putative            False  \n",
       "10                                       hemK-protein            False  \n",
       "11                     conserved-hypothetical-protein            False  \n",
       "12                     conserved-hypothetical-protein            False  \n",
       "13                     conserved-hypothetical-protein            False  \n",
       "14  ubiquinone/menaquinone-biosynthesis-methyltran...            False  \n",
       "15                     conserved-hypothetical-protein            False  \n",
       "16               glucose-inhibited-division-protein-B            False  \n",
       "17                     conserved-hypothetical-protein            False  \n",
       "18                     conserved-hypothetical-protein            False  \n",
       "19                         ATP-synthase-F0,-subunit-b            False  "
      ]
     },
     "execution_count": 161,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "protein_domains_path = os.path.join(\n",
    "    os.getcwd(), \n",
    "    f'data/domains/codon_bias/pfam/{assembly_accession}_protein_domains.csv',\n",
    ")\n",
    "thermotoga_domains = pd.read_csv(protein_domains_path)\n",
    "thermotoga_domains.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6.539035305818001"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "v = thermotoga_domains[thermotoga_domains['below_threshold']]\n",
    "100 * len(v) / len(thermotoga_domains)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "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>count_all</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pfam_query</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ABC_tran</th>\n",
       "      <td>65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BPD_transp_1</th>\n",
       "      <td>48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_21</th>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HD</th>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Radical_SAM</th>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              count_all\n",
       "pfam_query             \n",
       "ABC_tran             65\n",
       "BPD_transp_1         48\n",
       "AAA_21               41\n",
       "HD                   36\n",
       "Radical_SAM          33"
      ]
     },
     "execution_count": 163,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_counts = thermotoga_domains[['pfam_query', 'pfam_accession']].groupby('pfam_query').count()\n",
    "all_counts.columns = ['count_all']\n",
    "all_counts = all_counts.sort_values('count_all', ascending=False)\n",
    "all_counts.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "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>count_below</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pfam_query</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Helicase_C</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DHH</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DHHA1</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HD</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            count_below\n",
       "pfam_query             \n",
       "Helicase_C            6\n",
       "DHH                   4\n",
       "DHHA1                 4\n",
       "AAA                   4\n",
       "HD                    4"
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "below_threshold_counts = thermotoga_domains[\n",
    "    thermotoga_domains['below_threshold']\n",
    "][['pfam_query', 'pfam_accession']].groupby('pfam_query').count()\n",
    "below_threshold_counts.columns = ['count_below']\n",
    "below_threshold_counts = below_threshold_counts.sort_values('count_below', ascending=False)\n",
    "below_threshold_counts.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "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>count_all</th>\n",
       "      <th>count_below</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pfam_query</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ABC_tran</th>\n",
       "      <td>65</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BPD_transp_1</th>\n",
       "      <td>48</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_21</th>\n",
       "      <td>41</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HD</th>\n",
       "      <td>36</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Radical_SAM</th>\n",
       "      <td>33</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              count_all  count_below\n",
       "pfam_query                          \n",
       "ABC_tran             65            3\n",
       "BPD_transp_1         48            0\n",
       "AAA_21               41            2\n",
       "HD                   36            4\n",
       "Radical_SAM          33            1"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts = pd.merge(\n",
    "    all_counts,\n",
    "    below_threshold_counts,\n",
    "    how='left',\n",
    "    on='pfam_query',\n",
    ")\n",
    "counts['count_below'] = counts['count_below'].fillna(0).astype(int)\n",
    "counts.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [],
   "source": [
    "counts['frequency_weight'] = counts['count_below'] / counts['count_all']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [],
   "source": [
    "counts['species_score'] = np.sqrt(counts['count_below']) * counts['frequency_weight']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "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>count_all</th>\n",
       "      <th>count_below</th>\n",
       "      <th>frequency_weight</th>\n",
       "      <th>species_score</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pfam_query</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DHH</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DHHA1</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Glyco_hydro_38N</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.414214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Glycos_transf_3</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.414214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Glycos_trans_3N</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.414214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HD_4</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.414214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAS_8</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.414214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GMC_oxred_N</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.414214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RNA_pol_Rpb2_45</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MdcE</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LON_substr_bdg</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YscJ_FliF</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DUF1156</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gate</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DNA_topoisoIV</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PYNP_C</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RelA_SpoT</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Glyco_hydro_35</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FeoB_Cyto</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Glyco_hydro_36</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 count_all  count_below  frequency_weight  species_score\n",
       "pfam_query                                                              \n",
       "DHH                      4            4               1.0       2.000000\n",
       "DHHA1                    4            4               1.0       2.000000\n",
       "Glyco_hydro_38N          2            2               1.0       1.414214\n",
       "Glycos_transf_3          2            2               1.0       1.414214\n",
       "Glycos_trans_3N          2            2               1.0       1.414214\n",
       "HD_4                     2            2               1.0       1.414214\n",
       "PAS_8                    2            2               1.0       1.414214\n",
       "GMC_oxred_N              2            2               1.0       1.414214\n",
       "RNA_pol_Rpb2_45          1            1               1.0       1.000000\n",
       "MdcE                     1            1               1.0       1.000000\n",
       "LON_substr_bdg           1            1               1.0       1.000000\n",
       "YscJ_FliF                1            1               1.0       1.000000\n",
       "DUF1156                  1            1               1.0       1.000000\n",
       "Gate                     1            1               1.0       1.000000\n",
       "DNA_topoisoIV            1            1               1.0       1.000000\n",
       "PYNP_C                   1            1               1.0       1.000000\n",
       "RelA_SpoT                1            1               1.0       1.000000\n",
       "Glyco_hydro_35           1            1               1.0       1.000000\n",
       "FeoB_Cyto                1            1               1.0       1.000000\n",
       "Glyco_hydro_36           1            1               1.0       1.000000"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts.sort_values('species_score', ascending=False).head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## All species"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_protein_domain_score(engine, query_type):\n",
    "    if query_type not in ('pfam', 'tigr'):\n",
    "        raise ValueError('Query type must be one of (pfam, tigr)')\n",
    "        \n",
    "    q = \"\"\"\n",
    "    select a.assembly_accession, s.phylum from assembly_source as a\n",
    "    left join species_traits as s on s.species_taxid = a.species_taxid\n",
    "    \"\"\"\n",
    "    df = pd.read_sql(q, engine)\n",
    "    assembly_accessions = df['assembly_accession'].values\n",
    "    phyla = df['phylum'].values\n",
    "    \n",
    "    logger.info(f'Counting protein domains for {len(assembly_accessions):,} assemblies')\n",
    "    \n",
    "    n_assemblies = len(assembly_accessions)\n",
    "    n_assemblies_present = 0\n",
    "    \n",
    "    domain_to_phyla = collections.defaultdict(set)\n",
    "    domain_to_score = collections.defaultdict(int)\n",
    "    domain_count = collections.defaultdict(int)\n",
    "    domain_count_top = collections.defaultdict(int)\n",
    "    for i, assembly in enumerate(assembly_accessions):\n",
    "        if (i+1) % 200 == 0:\n",
    "            logger.info(f'{i+1:,} / {n_assemblies:,}')\n",
    "            \n",
    "        protein_domains_path = os.path.join(\n",
    "            os.getcwd(), \n",
    "            f'data/domains/codon_bias/{query_type}/{assembly}_protein_domains.csv',\n",
    "        )\n",
    "        if not os.path.isfile(protein_domains_path):\n",
    "            continue\n",
    "        else:\n",
    "            n_assemblies_present += 1\n",
    "        \n",
    "        protein_domains = pd.read_csv(protein_domains_path)\n",
    "        \n",
    "        phylum = phyla[i]\n",
    "        \n",
    "        all_counts = protein_domains[['pfam_query', 'pfam_accession']].groupby('pfam_query').count()\n",
    "        all_counts.columns = ['count_all']\n",
    "        \n",
    "        below_threshold_counts = protein_domains[\n",
    "            protein_domains['below_threshold']\n",
    "        ][['pfam_query', 'pfam_accession']].groupby('pfam_query').count()\n",
    "        below_threshold_counts.columns = ['count_below']\n",
    "        \n",
    "        counts = pd.merge(\n",
    "            all_counts,\n",
    "            below_threshold_counts,\n",
    "            how='left',\n",
    "            on='pfam_query',\n",
    "        )\n",
    "        counts['count_below'] = counts['count_below'].fillna(0).astype(int)\n",
    "        \n",
    "        counts['frequency_weight'] = counts['count_below'] / counts['count_all']\n",
    "        counts['assembly_score'] = np.sqrt(counts['count_below']) * counts['frequency_weight']\n",
    "        \n",
    "        for pfam_query in counts.index:\n",
    "            domain_to_phyla[pfam_query].add(phylum)\n",
    "            domain_to_score[pfam_query] += counts.loc[pfam_query, 'assembly_score']\n",
    "            domain_count[pfam_query] += 1\n",
    "            if counts.loc[pfam_query, 'count_below'] > 0:\n",
    "                domain_count_top[pfam_query] += 1\n",
    "        \n",
    "    query_key = f'{query_type}_query'\n",
    "    sorted_queries = sorted(domain_to_score.keys())\n",
    "    data = {\n",
    "        query_key: sorted_queries,\n",
    "        'n_phylum': [len(domain_to_phyla[k]) for k in sorted_queries],\n",
    "        'assembly_score_sum': [domain_to_score[k] for k in sorted_queries],\n",
    "        'assembly_count': [domain_count[k] for k in sorted_queries],\n",
    "        'assembly_count_top': [domain_count_top[k] for k in sorted_queries],\n",
    "        'score': [domain_to_score[k] / n_assemblies_present for k in sorted_queries],\n",
    "    }\n",
    "    output_df = pd.DataFrame.from_dict(data).set_index(query_key)\n",
    "    \n",
    "    return output_df.sort_values(['score', 'assembly_count'], ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_query_to_most_common_label(engine, query_type):\n",
    "    if query_type not in ('pfam', 'tigr'):\n",
    "        raise ValueError('Query type must be one of (pfam, tigr)')\n",
    "        \n",
    "    q = \"\"\"\n",
    "    select assembly_accession from assembly_source\n",
    "    \"\"\"\n",
    "    assembly_accessions = pd.read_sql(q, engine)['assembly_accession'].values\n",
    "    \n",
    "    logger.info(f'Finding most common protein labels per query for {len(assembly_accessions):,} assemblies')\n",
    "    \n",
    "    query_to_protein_labels = {}\n",
    "    for i, assembly in enumerate(assembly_accessions):\n",
    "        if (i+1) % 200 == 0:\n",
    "            logger.info(f'{i+1:,} / {len(assembly_accessions):,}')\n",
    "            \n",
    "        protein_domains_path = os.path.join(\n",
    "            os.getcwd(), \n",
    "            f'data/domains/codon_bias/{query_type}/{assembly}_protein_domains.csv',\n",
    "        )\n",
    "        if not os.path.isfile(protein_domains_path):\n",
    "            continue\n",
    "        \n",
    "        protein_domains = pd.read_csv(protein_domains_path)\n",
    "        \n",
    "        for tpl in protein_domains.itertuples():\n",
    "            query, label = tpl.pfam_query, tpl.protein_label\n",
    "            \n",
    "            if pd.isnull(label):\n",
    "                label = 'Unknown'\n",
    "                \n",
    "            label = label.strip()\n",
    "                \n",
    "            if query not in query_to_protein_labels:\n",
    "                query_to_protein_labels[query] = {\n",
    "                    label: 1,\n",
    "                }\n",
    "            elif label not in query_to_protein_labels[query]:\n",
    "                query_to_protein_labels[query][label] = 1\n",
    "            else:\n",
    "                query_to_protein_labels[query][label] += 1\n",
    "    \n",
    "    query_to_most_common_label = {}\n",
    "    for query in sorted(query_to_protein_labels.keys()):\n",
    "        label_counts = [(k, v) for k, v in query_to_protein_labels[query].items()]\n",
    "        \n",
    "        sorted_labels = sorted(label_counts, key=lambda t: t[1], reverse=True)\n",
    "        \n",
    "        query_to_most_common_label[query] = sorted_labels[0][0]\n",
    "        \n",
    "    return query_to_most_common_label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-12-07 14:01:44,412 (INFO) Counting protein domains for 3,282 assemblies\n",
      "2020-12-07 14:01:52,142 (INFO) 200 / 3,282\n",
      "2020-12-07 14:02:00,214 (INFO) 400 / 3,282\n",
      "2020-12-07 14:02:07,345 (INFO) 600 / 3,282\n",
      "2020-12-07 14:02:15,108 (INFO) 800 / 3,282\n",
      "2020-12-07 14:02:20,804 (INFO) 1,000 / 3,282\n",
      "2020-12-07 14:02:26,130 (INFO) 1,200 / 3,282\n",
      "2020-12-07 14:02:31,222 (INFO) 1,400 / 3,282\n",
      "2020-12-07 14:02:35,316 (INFO) 1,600 / 3,282\n",
      "2020-12-07 14:02:39,263 (INFO) 1,800 / 3,282\n",
      "2020-12-07 14:02:42,761 (INFO) 2,000 / 3,282\n",
      "2020-12-07 14:02:45,631 (INFO) 2,200 / 3,282\n",
      "2020-12-07 14:02:48,356 (INFO) 2,400 / 3,282\n",
      "2020-12-07 14:02:49,304 (INFO) 2,600 / 3,282\n",
      "2020-12-07 14:02:49,310 (INFO) 2,800 / 3,282\n",
      "2020-12-07 14:02:51,095 (INFO) 3,000 / 3,282\n",
      "2020-12-07 14:02:54,820 (INFO) 3,200 / 3,282\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1min 5s, sys: 2.72 s, total: 1min 8s\n",
      "Wall time: 1min 11s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "pfam_counts = compute_protein_domain_score(engine, query_type='pfam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-12-07 14:02:55,655 (INFO) Finding most common protein labels per query for 3,282 assemblies\n",
      "2020-12-07 14:02:58,728 (INFO) 200 / 3,282\n",
      "2020-12-07 14:03:02,201 (INFO) 400 / 3,282\n",
      "2020-12-07 14:03:05,323 (INFO) 600 / 3,282\n",
      "2020-12-07 14:03:08,473 (INFO) 800 / 3,282\n",
      "2020-12-07 14:03:11,025 (INFO) 1,000 / 3,282\n",
      "2020-12-07 14:03:13,288 (INFO) 1,200 / 3,282\n",
      "2020-12-07 14:03:15,415 (INFO) 1,400 / 3,282\n",
      "2020-12-07 14:03:17,200 (INFO) 1,600 / 3,282\n",
      "2020-12-07 14:03:18,968 (INFO) 1,800 / 3,282\n",
      "2020-12-07 14:03:20,582 (INFO) 2,000 / 3,282\n",
      "2020-12-07 14:03:21,953 (INFO) 2,200 / 3,282\n",
      "2020-12-07 14:03:23,217 (INFO) 2,400 / 3,282\n",
      "2020-12-07 14:03:23,662 (INFO) 2,600 / 3,282\n",
      "2020-12-07 14:03:23,669 (INFO) 2,800 / 3,282\n",
      "2020-12-07 14:03:24,465 (INFO) 3,000 / 3,282\n",
      "2020-12-07 14:03:26,171 (INFO) 3,200 / 3,282\n"
     ]
    }
   ],
   "source": [
    "pfam_query_to_most_common_label = compute_query_to_most_common_label(engine, query_type='pfam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "pfam_labels = [pfam_query_to_most_common_label[k] for k in pfam_counts.index]\n",
    "pfam_counts['label'] = pfam_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": 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>n_phylum</th>\n",
       "      <th>assembly_score_sum</th>\n",
       "      <th>assembly_count</th>\n",
       "      <th>assembly_count_top</th>\n",
       "      <th>score</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pfam_query</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",
       "      <th>HHH_6</th>\n",
       "      <td>31</td>\n",
       "      <td>1285.177049</td>\n",
       "      <td>1212</td>\n",
       "      <td>921</td>\n",
       "      <td>1.056021</td>\n",
       "      <td>DNA-polymerase-III,-alpha-subunit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Helicase_C</th>\n",
       "      <td>31</td>\n",
       "      <td>1253.404987</td>\n",
       "      <td>1217</td>\n",
       "      <td>1160</td>\n",
       "      <td>1.029914</td>\n",
       "      <td>DEAD/DEAH-box-helicase-domain-protein</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DNA_pol3_alpha</th>\n",
       "      <td>31</td>\n",
       "      <td>954.846423</td>\n",
       "      <td>1211</td>\n",
       "      <td>927</td>\n",
       "      <td>0.784590</td>\n",
       "      <td>DNA-polymerase-III,-alpha-subunit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DNA_pol3_finger</th>\n",
       "      <td>31</td>\n",
       "      <td>954.682354</td>\n",
       "      <td>1211</td>\n",
       "      <td>926</td>\n",
       "      <td>0.784456</td>\n",
       "      <td>DNA-polymerase-III,-alpha-subunit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HATPase_c</th>\n",
       "      <td>31</td>\n",
       "      <td>866.584429</td>\n",
       "      <td>1217</td>\n",
       "      <td>1078</td>\n",
       "      <td>0.712066</td>\n",
       "      <td>histidine-kinase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DEAD</th>\n",
       "      <td>31</td>\n",
       "      <td>847.189268</td>\n",
       "      <td>1217</td>\n",
       "      <td>1145</td>\n",
       "      <td>0.696129</td>\n",
       "      <td>DEAD/DEAH-box-helicase-domain-protein</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ResIII</th>\n",
       "      <td>31</td>\n",
       "      <td>805.120859</td>\n",
       "      <td>1217</td>\n",
       "      <td>1149</td>\n",
       "      <td>0.661562</td>\n",
       "      <td>ATP-dependent-DNA-helicase-RecG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HisKA</th>\n",
       "      <td>31</td>\n",
       "      <td>772.113047</td>\n",
       "      <td>1184</td>\n",
       "      <td>956</td>\n",
       "      <td>0.634440</td>\n",
       "      <td>histidine-kinase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HD_4</th>\n",
       "      <td>29</td>\n",
       "      <td>739.445267</td>\n",
       "      <td>1181</td>\n",
       "      <td>623</td>\n",
       "      <td>0.607597</td>\n",
       "      <td>GTP-pyrophosphokinase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UvrD_C</th>\n",
       "      <td>31</td>\n",
       "      <td>735.901076</td>\n",
       "      <td>1215</td>\n",
       "      <td>975</td>\n",
       "      <td>0.604685</td>\n",
       "      <td>UvrD/REP-helicase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TRCF</th>\n",
       "      <td>31</td>\n",
       "      <td>735.242641</td>\n",
       "      <td>1153</td>\n",
       "      <td>735</td>\n",
       "      <td>0.604144</td>\n",
       "      <td>transcription-repair-coupling-factor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PDDEXK_1</th>\n",
       "      <td>31</td>\n",
       "      <td>706.002791</td>\n",
       "      <td>1145</td>\n",
       "      <td>803</td>\n",
       "      <td>0.580117</td>\n",
       "      <td>CRISPR-associated-protein-Cas4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HHH_2</th>\n",
       "      <td>30</td>\n",
       "      <td>699.163542</td>\n",
       "      <td>1189</td>\n",
       "      <td>642</td>\n",
       "      <td>0.574498</td>\n",
       "      <td>DNA-ligase,-NAD-dependent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UvrD-helicase</th>\n",
       "      <td>31</td>\n",
       "      <td>696.531793</td>\n",
       "      <td>1214</td>\n",
       "      <td>976</td>\n",
       "      <td>0.572335</td>\n",
       "      <td>UvrD/REP-helicase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UvrA_DNA-bind</th>\n",
       "      <td>31</td>\n",
       "      <td>693.821033</td>\n",
       "      <td>1210</td>\n",
       "      <td>729</td>\n",
       "      <td>0.570108</td>\n",
       "      <td>excinuclease-ABC-subunit-A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UvrD_C_2</th>\n",
       "      <td>31</td>\n",
       "      <td>680.867160</td>\n",
       "      <td>1215</td>\n",
       "      <td>977</td>\n",
       "      <td>0.559464</td>\n",
       "      <td>UvrD/REP-helicase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UvrA_inter</th>\n",
       "      <td>31</td>\n",
       "      <td>680.347130</td>\n",
       "      <td>1209</td>\n",
       "      <td>697</td>\n",
       "      <td>0.559036</td>\n",
       "      <td>excinuclease-ABC-subunit-A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAS_4</th>\n",
       "      <td>31</td>\n",
       "      <td>676.698027</td>\n",
       "      <td>1013</td>\n",
       "      <td>809</td>\n",
       "      <td>0.556038</td>\n",
       "      <td>PAS/PAC-sensor-signal-transduction-histidine-k...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAS</th>\n",
       "      <td>31</td>\n",
       "      <td>658.108907</td>\n",
       "      <td>1045</td>\n",
       "      <td>816</td>\n",
       "      <td>0.540763</td>\n",
       "      <td>PAS/PAC-sensor-signal-transduction-histidine-k...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_19</th>\n",
       "      <td>31</td>\n",
       "      <td>648.212791</td>\n",
       "      <td>1216</td>\n",
       "      <td>1035</td>\n",
       "      <td>0.532632</td>\n",
       "      <td>UvrD/REP-helicase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>E1-E2_ATPase</th>\n",
       "      <td>31</td>\n",
       "      <td>645.665463</td>\n",
       "      <td>1182</td>\n",
       "      <td>918</td>\n",
       "      <td>0.530539</td>\n",
       "      <td>heavy-metal-translocating-P-type-ATPase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAS_9</th>\n",
       "      <td>31</td>\n",
       "      <td>642.209875</td>\n",
       "      <td>969</td>\n",
       "      <td>767</td>\n",
       "      <td>0.527699</td>\n",
       "      <td>PAS-domain-S-box-containing-protein</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Transpeptidase</th>\n",
       "      <td>30</td>\n",
       "      <td>627.626578</td>\n",
       "      <td>1175</td>\n",
       "      <td>943</td>\n",
       "      <td>0.515716</td>\n",
       "      <td>penicillin-binding-protein</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FtsK_alpha</th>\n",
       "      <td>27</td>\n",
       "      <td>623.682083</td>\n",
       "      <td>1133</td>\n",
       "      <td>632</td>\n",
       "      <td>0.512475</td>\n",
       "      <td>cell-division-protein-FtsK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PHP</th>\n",
       "      <td>31</td>\n",
       "      <td>619.833305</td>\n",
       "      <td>1216</td>\n",
       "      <td>927</td>\n",
       "      <td>0.509312</td>\n",
       "      <td>DNA-polymerase-III,-alpha-subunit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAS_8</th>\n",
       "      <td>31</td>\n",
       "      <td>618.235339</td>\n",
       "      <td>972</td>\n",
       "      <td>737</td>\n",
       "      <td>0.507999</td>\n",
       "      <td>PAS/PAC-sensor-signal-transduction-histidine-k...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FtsK_gamma</th>\n",
       "      <td>26</td>\n",
       "      <td>614.244180</td>\n",
       "      <td>1130</td>\n",
       "      <td>629</td>\n",
       "      <td>0.504720</td>\n",
       "      <td>cell-division-protein-FtsK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAS_3</th>\n",
       "      <td>30</td>\n",
       "      <td>611.126491</td>\n",
       "      <td>870</td>\n",
       "      <td>667</td>\n",
       "      <td>0.502158</td>\n",
       "      <td>diguanylate-cyclase/phosphodiesterase-with-PAS...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FtsK_SpoIIIE</th>\n",
       "      <td>27</td>\n",
       "      <td>605.743321</td>\n",
       "      <td>1143</td>\n",
       "      <td>694</td>\n",
       "      <td>0.497735</td>\n",
       "      <td>cell-division-protein-FtsK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UvrB_inter</th>\n",
       "      <td>31</td>\n",
       "      <td>589.068602</td>\n",
       "      <td>1214</td>\n",
       "      <td>807</td>\n",
       "      <td>0.484033</td>\n",
       "      <td>transcription-repair-coupling-factor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DNA_pol_A</th>\n",
       "      <td>31</td>\n",
       "      <td>587.270647</td>\n",
       "      <td>1194</td>\n",
       "      <td>621</td>\n",
       "      <td>0.482556</td>\n",
       "      <td>DNA-polymerase-I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CarD_CdnL_TRCF</th>\n",
       "      <td>31</td>\n",
       "      <td>566.578427</td>\n",
       "      <td>1162</td>\n",
       "      <td>737</td>\n",
       "      <td>0.465553</td>\n",
       "      <td>transcription-repair-coupling-factor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MutS_V</th>\n",
       "      <td>31</td>\n",
       "      <td>565.436599</td>\n",
       "      <td>1023</td>\n",
       "      <td>651</td>\n",
       "      <td>0.464615</td>\n",
       "      <td>DNA-mismatch-repair-protein-MutS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GXGXG</th>\n",
       "      <td>28</td>\n",
       "      <td>561.463566</td>\n",
       "      <td>849</td>\n",
       "      <td>422</td>\n",
       "      <td>0.461351</td>\n",
       "      <td>glutamate-synthase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MutS_I</th>\n",
       "      <td>30</td>\n",
       "      <td>550.990188</td>\n",
       "      <td>984</td>\n",
       "      <td>552</td>\n",
       "      <td>0.452745</td>\n",
       "      <td>DNA-mismatch-repair-protein-MutS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5_3_exonuc</th>\n",
       "      <td>31</td>\n",
       "      <td>548.909307</td>\n",
       "      <td>1208</td>\n",
       "      <td>612</td>\n",
       "      <td>0.451035</td>\n",
       "      <td>DNA-polymerase-I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5_3_exonuc_N</th>\n",
       "      <td>31</td>\n",
       "      <td>548.409307</td>\n",
       "      <td>1209</td>\n",
       "      <td>612</td>\n",
       "      <td>0.450624</td>\n",
       "      <td>DNA-polymerase-I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Anticodon_1</th>\n",
       "      <td>31</td>\n",
       "      <td>547.675312</td>\n",
       "      <td>1216</td>\n",
       "      <td>730</td>\n",
       "      <td>0.450021</td>\n",
       "      <td>leucyl-tRNA-synthetase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hpt</th>\n",
       "      <td>27</td>\n",
       "      <td>546.242679</td>\n",
       "      <td>830</td>\n",
       "      <td>510</td>\n",
       "      <td>0.448844</td>\n",
       "      <td>CheA-signal-transduction-histidine-kinase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Topoisom_bac</th>\n",
       "      <td>31</td>\n",
       "      <td>544.492580</td>\n",
       "      <td>1199</td>\n",
       "      <td>639</td>\n",
       "      <td>0.447406</td>\n",
       "      <td>DNA-topoisomerase-I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MutS_III</th>\n",
       "      <td>30</td>\n",
       "      <td>542.707520</td>\n",
       "      <td>985</td>\n",
       "      <td>559</td>\n",
       "      <td>0.445939</td>\n",
       "      <td>DNA-mismatch-repair-protein-MutS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MutS_IV</th>\n",
       "      <td>30</td>\n",
       "      <td>542.242641</td>\n",
       "      <td>968</td>\n",
       "      <td>543</td>\n",
       "      <td>0.445557</td>\n",
       "      <td>DNA-mismatch-repair-protein-MutS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_16</th>\n",
       "      <td>31</td>\n",
       "      <td>538.848836</td>\n",
       "      <td>1215</td>\n",
       "      <td>909</td>\n",
       "      <td>0.442768</td>\n",
       "      <td>ATP-dependent-chaperone-ClpB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Viral_helicase1</th>\n",
       "      <td>31</td>\n",
       "      <td>536.158880</td>\n",
       "      <td>1196</td>\n",
       "      <td>810</td>\n",
       "      <td>0.440558</td>\n",
       "      <td>UvrD/REP-helicase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ACR_tran</th>\n",
       "      <td>29</td>\n",
       "      <td>534.647959</td>\n",
       "      <td>940</td>\n",
       "      <td>720</td>\n",
       "      <td>0.439316</td>\n",
       "      <td>acriflavin-resistance-protein</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PriA_3primeBD</th>\n",
       "      <td>30</td>\n",
       "      <td>531.914214</td>\n",
       "      <td>1138</td>\n",
       "      <td>532</td>\n",
       "      <td>0.437070</td>\n",
       "      <td>primosomal-protein-N'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MutS_II</th>\n",
       "      <td>29</td>\n",
       "      <td>530.685450</td>\n",
       "      <td>948</td>\n",
       "      <td>530</td>\n",
       "      <td>0.436060</td>\n",
       "      <td>DNA-mismatch-repair-protein-MutS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Molybdopterin</th>\n",
       "      <td>27</td>\n",
       "      <td>525.168247</td>\n",
       "      <td>901</td>\n",
       "      <td>674</td>\n",
       "      <td>0.431527</td>\n",
       "      <td>molybdopterin-oxidoreductase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>B5</th>\n",
       "      <td>31</td>\n",
       "      <td>514.914214</td>\n",
       "      <td>1201</td>\n",
       "      <td>515</td>\n",
       "      <td>0.423101</td>\n",
       "      <td>phenylalanyl-tRNA-synthetase,-beta-subunit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FDX-ACB</th>\n",
       "      <td>31</td>\n",
       "      <td>513.414214</td>\n",
       "      <td>1176</td>\n",
       "      <td>514</td>\n",
       "      <td>0.421869</td>\n",
       "      <td>phenylalanyl-tRNA-synthetase,-beta-subunit</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 n_phylum  assembly_score_sum  assembly_count  \\\n",
       "pfam_query                                                      \n",
       "HHH_6                  31         1285.177049            1212   \n",
       "Helicase_C             31         1253.404987            1217   \n",
       "DNA_pol3_alpha         31          954.846423            1211   \n",
       "DNA_pol3_finger        31          954.682354            1211   \n",
       "HATPase_c              31          866.584429            1217   \n",
       "DEAD                   31          847.189268            1217   \n",
       "ResIII                 31          805.120859            1217   \n",
       "HisKA                  31          772.113047            1184   \n",
       "HD_4                   29          739.445267            1181   \n",
       "UvrD_C                 31          735.901076            1215   \n",
       "TRCF                   31          735.242641            1153   \n",
       "PDDEXK_1               31          706.002791            1145   \n",
       "HHH_2                  30          699.163542            1189   \n",
       "UvrD-helicase          31          696.531793            1214   \n",
       "UvrA_DNA-bind          31          693.821033            1210   \n",
       "UvrD_C_2               31          680.867160            1215   \n",
       "UvrA_inter             31          680.347130            1209   \n",
       "PAS_4                  31          676.698027            1013   \n",
       "PAS                    31          658.108907            1045   \n",
       "AAA_19                 31          648.212791            1216   \n",
       "E1-E2_ATPase           31          645.665463            1182   \n",
       "PAS_9                  31          642.209875             969   \n",
       "Transpeptidase         30          627.626578            1175   \n",
       "FtsK_alpha             27          623.682083            1133   \n",
       "PHP                    31          619.833305            1216   \n",
       "PAS_8                  31          618.235339             972   \n",
       "FtsK_gamma             26          614.244180            1130   \n",
       "PAS_3                  30          611.126491             870   \n",
       "FtsK_SpoIIIE           27          605.743321            1143   \n",
       "UvrB_inter             31          589.068602            1214   \n",
       "DNA_pol_A              31          587.270647            1194   \n",
       "CarD_CdnL_TRCF         31          566.578427            1162   \n",
       "MutS_V                 31          565.436599            1023   \n",
       "GXGXG                  28          561.463566             849   \n",
       "MutS_I                 30          550.990188             984   \n",
       "5_3_exonuc             31          548.909307            1208   \n",
       "5_3_exonuc_N           31          548.409307            1209   \n",
       "Anticodon_1            31          547.675312            1216   \n",
       "Hpt                    27          546.242679             830   \n",
       "Topoisom_bac           31          544.492580            1199   \n",
       "MutS_III               30          542.707520             985   \n",
       "MutS_IV                30          542.242641             968   \n",
       "AAA_16                 31          538.848836            1215   \n",
       "Viral_helicase1        31          536.158880            1196   \n",
       "ACR_tran               29          534.647959             940   \n",
       "PriA_3primeBD          30          531.914214            1138   \n",
       "MutS_II                29          530.685450             948   \n",
       "Molybdopterin          27          525.168247             901   \n",
       "B5                     31          514.914214            1201   \n",
       "FDX-ACB                31          513.414214            1176   \n",
       "\n",
       "                 assembly_count_top     score  \\\n",
       "pfam_query                                      \n",
       "HHH_6                           921  1.056021   \n",
       "Helicase_C                     1160  1.029914   \n",
       "DNA_pol3_alpha                  927  0.784590   \n",
       "DNA_pol3_finger                 926  0.784456   \n",
       "HATPase_c                      1078  0.712066   \n",
       "DEAD                           1145  0.696129   \n",
       "ResIII                         1149  0.661562   \n",
       "HisKA                           956  0.634440   \n",
       "HD_4                            623  0.607597   \n",
       "UvrD_C                          975  0.604685   \n",
       "TRCF                            735  0.604144   \n",
       "PDDEXK_1                        803  0.580117   \n",
       "HHH_2                           642  0.574498   \n",
       "UvrD-helicase                   976  0.572335   \n",
       "UvrA_DNA-bind                   729  0.570108   \n",
       "UvrD_C_2                        977  0.559464   \n",
       "UvrA_inter                      697  0.559036   \n",
       "PAS_4                           809  0.556038   \n",
       "PAS                             816  0.540763   \n",
       "AAA_19                         1035  0.532632   \n",
       "E1-E2_ATPase                    918  0.530539   \n",
       "PAS_9                           767  0.527699   \n",
       "Transpeptidase                  943  0.515716   \n",
       "FtsK_alpha                      632  0.512475   \n",
       "PHP                             927  0.509312   \n",
       "PAS_8                           737  0.507999   \n",
       "FtsK_gamma                      629  0.504720   \n",
       "PAS_3                           667  0.502158   \n",
       "FtsK_SpoIIIE                    694  0.497735   \n",
       "UvrB_inter                      807  0.484033   \n",
       "DNA_pol_A                       621  0.482556   \n",
       "CarD_CdnL_TRCF                  737  0.465553   \n",
       "MutS_V                          651  0.464615   \n",
       "GXGXG                           422  0.461351   \n",
       "MutS_I                          552  0.452745   \n",
       "5_3_exonuc                      612  0.451035   \n",
       "5_3_exonuc_N                    612  0.450624   \n",
       "Anticodon_1                     730  0.450021   \n",
       "Hpt                             510  0.448844   \n",
       "Topoisom_bac                    639  0.447406   \n",
       "MutS_III                        559  0.445939   \n",
       "MutS_IV                         543  0.445557   \n",
       "AAA_16                          909  0.442768   \n",
       "Viral_helicase1                 810  0.440558   \n",
       "ACR_tran                        720  0.439316   \n",
       "PriA_3primeBD                   532  0.437070   \n",
       "MutS_II                         530  0.436060   \n",
       "Molybdopterin                   674  0.431527   \n",
       "B5                              515  0.423101   \n",
       "FDX-ACB                         514  0.421869   \n",
       "\n",
       "                                                             label  \n",
       "pfam_query                                                          \n",
       "HHH_6                            DNA-polymerase-III,-alpha-subunit  \n",
       "Helicase_C                   DEAD/DEAH-box-helicase-domain-protein  \n",
       "DNA_pol3_alpha                   DNA-polymerase-III,-alpha-subunit  \n",
       "DNA_pol3_finger                  DNA-polymerase-III,-alpha-subunit  \n",
       "HATPase_c                                         histidine-kinase  \n",
       "DEAD                         DEAD/DEAH-box-helicase-domain-protein  \n",
       "ResIII                             ATP-dependent-DNA-helicase-RecG  \n",
       "HisKA                                             histidine-kinase  \n",
       "HD_4                                         GTP-pyrophosphokinase  \n",
       "UvrD_C                                           UvrD/REP-helicase  \n",
       "TRCF                          transcription-repair-coupling-factor  \n",
       "PDDEXK_1                            CRISPR-associated-protein-Cas4  \n",
       "HHH_2                                    DNA-ligase,-NAD-dependent  \n",
       "UvrD-helicase                                    UvrD/REP-helicase  \n",
       "UvrA_DNA-bind                           excinuclease-ABC-subunit-A  \n",
       "UvrD_C_2                                         UvrD/REP-helicase  \n",
       "UvrA_inter                              excinuclease-ABC-subunit-A  \n",
       "PAS_4            PAS/PAC-sensor-signal-transduction-histidine-k...  \n",
       "PAS              PAS/PAC-sensor-signal-transduction-histidine-k...  \n",
       "AAA_19                                           UvrD/REP-helicase  \n",
       "E1-E2_ATPase               heavy-metal-translocating-P-type-ATPase  \n",
       "PAS_9                          PAS-domain-S-box-containing-protein  \n",
       "Transpeptidase                          penicillin-binding-protein  \n",
       "FtsK_alpha                              cell-division-protein-FtsK  \n",
       "PHP                              DNA-polymerase-III,-alpha-subunit  \n",
       "PAS_8            PAS/PAC-sensor-signal-transduction-histidine-k...  \n",
       "FtsK_gamma                              cell-division-protein-FtsK  \n",
       "PAS_3            diguanylate-cyclase/phosphodiesterase-with-PAS...  \n",
       "FtsK_SpoIIIE                            cell-division-protein-FtsK  \n",
       "UvrB_inter                    transcription-repair-coupling-factor  \n",
       "DNA_pol_A                                         DNA-polymerase-I  \n",
       "CarD_CdnL_TRCF                transcription-repair-coupling-factor  \n",
       "MutS_V                            DNA-mismatch-repair-protein-MutS  \n",
       "GXGXG                                           glutamate-synthase  \n",
       "MutS_I                            DNA-mismatch-repair-protein-MutS  \n",
       "5_3_exonuc                                        DNA-polymerase-I  \n",
       "5_3_exonuc_N                                      DNA-polymerase-I  \n",
       "Anticodon_1                                 leucyl-tRNA-synthetase  \n",
       "Hpt                      CheA-signal-transduction-histidine-kinase  \n",
       "Topoisom_bac                                   DNA-topoisomerase-I  \n",
       "MutS_III                          DNA-mismatch-repair-protein-MutS  \n",
       "MutS_IV                           DNA-mismatch-repair-protein-MutS  \n",
       "AAA_16                                ATP-dependent-chaperone-ClpB  \n",
       "Viral_helicase1                                  UvrD/REP-helicase  \n",
       "ACR_tran                             acriflavin-resistance-protein  \n",
       "PriA_3primeBD                                primosomal-protein-N'  \n",
       "MutS_II                           DNA-mismatch-repair-protein-MutS  \n",
       "Molybdopterin                         molybdopterin-oxidoreductase  \n",
       "B5                      phenylalanyl-tRNA-synthetase,-beta-subunit  \n",
       "FDX-ACB                 phenylalanyl-tRNA-synthetase,-beta-subunit  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pfam_counts.head(50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4.816451700189411"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "100 * len(pfam_counts[pfam_counts['score'] >= 0.1]) / len(pfam_counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_, ax = plt.subplots(1, 1, figsize=(12, 6))\n",
    "ax.hist(pfam_counts['score'].values, bins=50, log=True);\n",
    "pfam_threshold = np.percentile(pfam_counts['score'].values, 95)\n",
    "ax.axvline(pfam_threshold, color='red')\n",
    "ax.set_xlabel('Score (%)')\n",
    "ax.set_ylabel('Pfam entry count')\n",
    "ax.set_title('Distribution of scores for all Pfam entries');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.09701828073063584"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-12-07 14:03:27,845 (INFO) Counting protein domains for 3,282 assemblies\n",
      "2020-12-07 14:03:31,651 (INFO) 200 / 3,282\n",
      "2020-12-07 14:03:35,309 (INFO) 400 / 3,282\n",
      "2020-12-07 14:03:38,640 (INFO) 600 / 3,282\n",
      "2020-12-07 14:03:41,973 (INFO) 800 / 3,282\n",
      "2020-12-07 14:03:44,667 (INFO) 1,000 / 3,282\n",
      "2020-12-07 14:03:47,189 (INFO) 1,200 / 3,282\n",
      "2020-12-07 14:03:49,559 (INFO) 1,400 / 3,282\n",
      "2020-12-07 14:03:51,417 (INFO) 1,600 / 3,282\n",
      "2020-12-07 14:03:53,149 (INFO) 1,800 / 3,282\n",
      "2020-12-07 14:03:54,717 (INFO) 2,000 / 3,282\n",
      "2020-12-07 14:03:55,988 (INFO) 2,200 / 3,282\n",
      "2020-12-07 14:03:57,172 (INFO) 2,400 / 3,282\n",
      "2020-12-07 14:03:57,568 (INFO) 2,600 / 3,282\n",
      "2020-12-07 14:03:57,575 (INFO) 2,800 / 3,282\n",
      "2020-12-07 14:03:58,356 (INFO) 3,000 / 3,282\n",
      "2020-12-07 14:03:59,967 (INFO) 3,200 / 3,282\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 30.7 s, sys: 752 ms, total: 31.4 s\n",
      "Wall time: 32.4 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "tigr_counts = compute_protein_domain_score(engine, query_type='tigr')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.6013187927973624"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "100 * len(tigr_counts[tigr_counts['score'] > 0.1]) / len(tigr_counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-12-07 14:04:00,269 (INFO) Finding most common protein labels per query for 3,282 assemblies\n",
      "2020-12-07 14:04:01,223 (INFO) 200 / 3,282\n",
      "2020-12-07 14:04:02,186 (INFO) 400 / 3,282\n",
      "2020-12-07 14:04:03,026 (INFO) 600 / 3,282\n",
      "2020-12-07 14:04:03,900 (INFO) 800 / 3,282\n",
      "2020-12-07 14:04:04,585 (INFO) 1,000 / 3,282\n",
      "2020-12-07 14:04:05,246 (INFO) 1,200 / 3,282\n",
      "2020-12-07 14:04:05,907 (INFO) 1,400 / 3,282\n",
      "2020-12-07 14:04:06,389 (INFO) 1,600 / 3,282\n",
      "2020-12-07 14:04:06,881 (INFO) 1,800 / 3,282\n",
      "2020-12-07 14:04:07,348 (INFO) 2,000 / 3,282\n",
      "2020-12-07 14:04:07,736 (INFO) 2,200 / 3,282\n",
      "2020-12-07 14:04:08,086 (INFO) 2,400 / 3,282\n",
      "2020-12-07 14:04:08,212 (INFO) 2,600 / 3,282\n",
      "2020-12-07 14:04:08,219 (INFO) 2,800 / 3,282\n",
      "2020-12-07 14:04:08,455 (INFO) 3,000 / 3,282\n",
      "2020-12-07 14:04:08,934 (INFO) 3,200 / 3,282\n"
     ]
    }
   ],
   "source": [
    "tigr_query_to_most_common_label = compute_query_to_most_common_label(engine, query_type='tigr')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "tigr_labels = [tigr_query_to_most_common_label[k] for k in tigr_counts.index]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>n_phylum</th>\n",
       "      <th>assembly_score_sum</th>\n",
       "      <th>assembly_count</th>\n",
       "      <th>assembly_count_top</th>\n",
       "      <th>score</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tigr_query</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",
       "      <th>TIGR00594</th>\n",
       "      <td>31</td>\n",
       "      <td>845.086076</td>\n",
       "      <td>1190</td>\n",
       "      <td>870</td>\n",
       "      <td>0.694401</td>\n",
       "      <td>DNA-polymerase-III,-alpha-subunit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00580</th>\n",
       "      <td>31</td>\n",
       "      <td>728.000000</td>\n",
       "      <td>1138</td>\n",
       "      <td>728</td>\n",
       "      <td>0.598192</td>\n",
       "      <td>transcription-repair-coupling-factor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00229</th>\n",
       "      <td>31</td>\n",
       "      <td>704.536214</td>\n",
       "      <td>1033</td>\n",
       "      <td>831</td>\n",
       "      <td>0.578912</td>\n",
       "      <td>PAS-domain-S-box-containing-protein</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00630</th>\n",
       "      <td>31</td>\n",
       "      <td>694.991743</td>\n",
       "      <td>1209</td>\n",
       "      <td>722</td>\n",
       "      <td>0.571070</td>\n",
       "      <td>excinuclease-ABC-subunit-A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR01494</th>\n",
       "      <td>31</td>\n",
       "      <td>644.297111</td>\n",
       "      <td>1182</td>\n",
       "      <td>918</td>\n",
       "      <td>0.529414</td>\n",
       "      <td>heavy-metal-translocating-P-type-ATPase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00593</th>\n",
       "      <td>31</td>\n",
       "      <td>611.000000</td>\n",
       "      <td>1181</td>\n",
       "      <td>611</td>\n",
       "      <td>0.502054</td>\n",
       "      <td>DNA-polymerase-I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00691</th>\n",
       "      <td>27</td>\n",
       "      <td>559.426407</td>\n",
       "      <td>1171</td>\n",
       "      <td>608</td>\n",
       "      <td>0.459677</td>\n",
       "      <td>GTP-pyrophosphokinase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR01070</th>\n",
       "      <td>30</td>\n",
       "      <td>542.828427</td>\n",
       "      <td>964</td>\n",
       "      <td>543</td>\n",
       "      <td>0.446038</td>\n",
       "      <td>DNA-mismatch-repair-protein-MutS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR02168</th>\n",
       "      <td>31</td>\n",
       "      <td>527.000000</td>\n",
       "      <td>906</td>\n",
       "      <td>527</td>\n",
       "      <td>0.433032</td>\n",
       "      <td>chromosome-segregation-protein-SMC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00472</th>\n",
       "      <td>31</td>\n",
       "      <td>514.914214</td>\n",
       "      <td>1179</td>\n",
       "      <td>515</td>\n",
       "      <td>0.423101</td>\n",
       "      <td>phenylalanyl-tRNA-synthetase,-beta-subunit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00575</th>\n",
       "      <td>29</td>\n",
       "      <td>507.914214</td>\n",
       "      <td>1176</td>\n",
       "      <td>510</td>\n",
       "      <td>0.417349</td>\n",
       "      <td>DNA-ligase,-NAD-dependent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR01525</th>\n",
       "      <td>31</td>\n",
       "      <td>503.028299</td>\n",
       "      <td>1137</td>\n",
       "      <td>719</td>\n",
       "      <td>0.413335</td>\n",
       "      <td>heavy-metal-translocating-P-type-ATPase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR01512</th>\n",
       "      <td>31</td>\n",
       "      <td>500.463291</td>\n",
       "      <td>1130</td>\n",
       "      <td>707</td>\n",
       "      <td>0.411227</td>\n",
       "      <td>heavy-metal-translocating-P-type-ATPase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR01051</th>\n",
       "      <td>29</td>\n",
       "      <td>494.833333</td>\n",
       "      <td>1113</td>\n",
       "      <td>498</td>\n",
       "      <td>0.406601</td>\n",
       "      <td>DNA-topoisomerase-I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR01511</th>\n",
       "      <td>31</td>\n",
       "      <td>494.121129</td>\n",
       "      <td>1115</td>\n",
       "      <td>642</td>\n",
       "      <td>0.406016</td>\n",
       "      <td>heavy-metal-translocating-P-type-ATPase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00595</th>\n",
       "      <td>28</td>\n",
       "      <td>491.414214</td>\n",
       "      <td>991</td>\n",
       "      <td>491</td>\n",
       "      <td>0.403791</td>\n",
       "      <td>primosomal-protein-N'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00344</th>\n",
       "      <td>31</td>\n",
       "      <td>485.000000</td>\n",
       "      <td>1209</td>\n",
       "      <td>485</td>\n",
       "      <td>0.398521</td>\n",
       "      <td>alanyl-tRNA-synthetase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00392</th>\n",
       "      <td>31</td>\n",
       "      <td>462.985281</td>\n",
       "      <td>1206</td>\n",
       "      <td>461</td>\n",
       "      <td>0.380432</td>\n",
       "      <td>isoleucyl-tRNA-synthetase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR00254</th>\n",
       "      <td>31</td>\n",
       "      <td>444.075262</td>\n",
       "      <td>894</td>\n",
       "      <td>696</td>\n",
       "      <td>0.364893</td>\n",
       "      <td>diguanylate-cyclase</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIGR01369</th>\n",
       "      <td>28</td>\n",
       "      <td>430.369216</td>\n",
       "      <td>1076</td>\n",
       "      <td>447</td>\n",
       "      <td>0.353631</td>\n",
       "      <td>carbamoyl-phosphate-synthase-large-subunit</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            n_phylum  assembly_score_sum  assembly_count  assembly_count_top  \\\n",
       "tigr_query                                                                     \n",
       "TIGR00594         31          845.086076            1190                 870   \n",
       "TIGR00580         31          728.000000            1138                 728   \n",
       "TIGR00229         31          704.536214            1033                 831   \n",
       "TIGR00630         31          694.991743            1209                 722   \n",
       "TIGR01494         31          644.297111            1182                 918   \n",
       "TIGR00593         31          611.000000            1181                 611   \n",
       "TIGR00691         27          559.426407            1171                 608   \n",
       "TIGR01070         30          542.828427             964                 543   \n",
       "TIGR02168         31          527.000000             906                 527   \n",
       "TIGR00472         31          514.914214            1179                 515   \n",
       "TIGR00575         29          507.914214            1176                 510   \n",
       "TIGR01525         31          503.028299            1137                 719   \n",
       "TIGR01512         31          500.463291            1130                 707   \n",
       "TIGR01051         29          494.833333            1113                 498   \n",
       "TIGR01511         31          494.121129            1115                 642   \n",
       "TIGR00595         28          491.414214             991                 491   \n",
       "TIGR00344         31          485.000000            1209                 485   \n",
       "TIGR00392         31          462.985281            1206                 461   \n",
       "TIGR00254         31          444.075262             894                 696   \n",
       "TIGR01369         28          430.369216            1076                 447   \n",
       "\n",
       "               score                                       label  \n",
       "tigr_query                                                        \n",
       "TIGR00594   0.694401           DNA-polymerase-III,-alpha-subunit  \n",
       "TIGR00580   0.598192        transcription-repair-coupling-factor  \n",
       "TIGR00229   0.578912         PAS-domain-S-box-containing-protein  \n",
       "TIGR00630   0.571070                  excinuclease-ABC-subunit-A  \n",
       "TIGR01494   0.529414     heavy-metal-translocating-P-type-ATPase  \n",
       "TIGR00593   0.502054                            DNA-polymerase-I  \n",
       "TIGR00691   0.459677                       GTP-pyrophosphokinase  \n",
       "TIGR01070   0.446038            DNA-mismatch-repair-protein-MutS  \n",
       "TIGR02168   0.433032          chromosome-segregation-protein-SMC  \n",
       "TIGR00472   0.423101  phenylalanyl-tRNA-synthetase,-beta-subunit  \n",
       "TIGR00575   0.417349                   DNA-ligase,-NAD-dependent  \n",
       "TIGR01525   0.413335     heavy-metal-translocating-P-type-ATPase  \n",
       "TIGR01512   0.411227     heavy-metal-translocating-P-type-ATPase  \n",
       "TIGR01051   0.406601                         DNA-topoisomerase-I  \n",
       "TIGR01511   0.406016     heavy-metal-translocating-P-type-ATPase  \n",
       "TIGR00595   0.403791                       primosomal-protein-N'  \n",
       "TIGR00344   0.398521                      alanyl-tRNA-synthetase  \n",
       "TIGR00392   0.380432                   isoleucyl-tRNA-synthetase  \n",
       "TIGR00254   0.364893                         diguanylate-cyclase  \n",
       "TIGR01369   0.353631  carbamoyl-phosphate-synthase-large-subunit  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tigr_counts['label'] = tigr_labels\n",
    "tigr_counts.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_, ax = plt.subplots(1, 1, figsize=(12, 6))\n",
    "ax.hist(tigr_counts['score'].values, bins=20, log=True);\n",
    "tigr_thresold = np.percentile(tigr_counts['score'].values, 95)\n",
    "ax.axvline(tigr_thresold, color='red')\n",
    "ax.set_xlabel('Score (%)')\n",
    "ax.set_ylabel('TIGR entry count')\n",
    "ax.set_title('Distribution of scores for all TIGR entries');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "score_threshold = 0.05\n",
    "base_path = os.path.join(os.getcwd(), 'data/domains/codon_bias/')\n",
    "pfam_counts[pfam_counts['score'] >= pfam_threshold].to_excel(os.path.join(base_path, 'pfam_top.xlsx'))\n",
    "tigr_counts[tigr_counts['score'] >= tigr_thresold].to_excel(os.path.join(base_path, 'tigr_top.xlsx'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Validation: Protein id match\n",
    "\n",
    "Let's make sure IDs properly match and we are not simply seeing an artefact of proteins that joined"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "def check_protein_matching(engine, query_type):\n",
    "    if query_type not in ('pfam', 'tigr'):\n",
    "        raise ValueError('Query type must be one of (pfam, tigr)')\n",
    "        \n",
    "    q = \"\"\"\n",
    "    select assembly_accession from assembly_source\n",
    "    \"\"\"\n",
    "    assembly_accessions = pd.read_sql(q, engine)['assembly_accession'].values\n",
    "    \n",
    "    logger.info(f'Checking {query_type} protein ID match for {len(assembly_accessions):,} assemblies')\n",
    "    \n",
    "    matching_scores = {}\n",
    "    for i, assembly in enumerate(assembly_accessions):\n",
    "        if (i+1) % 200 == 0:\n",
    "            logger.info(f'{i+1:,} / {len(assembly_accessions):,}')\n",
    "            \n",
    "        protein_domains_path = os.path.join(\n",
    "            os.getcwd(), \n",
    "            f'data/domains/codon_bias/{query_type}/{assembly}_protein_domains.csv',\n",
    "        )\n",
    "        if not os.path.isfile(protein_domains_path):\n",
    "            continue\n",
    "        \n",
    "        protein_domains = pd.read_csv(protein_domains_path)\n",
    "        \n",
    "        protein_query = \"\"\"\n",
    "        select metadata_json from sequences where sequence_type = 'CDS' and assembly_accession = ?\n",
    "        \"\"\"\n",
    "        cds_metadata_df = pd.read_sql(protein_query, engine, params=(assembly,))\n",
    "        metadata = [json.loads(v) for v in cds_metadata_df['metadata_json'].values if not pd.isnull(v)]\n",
    "        \n",
    "        cds_protein_ids = {\n",
    "            m['protein_id'].strip() for m in metadata\n",
    "            if m.get('protein_id') is not None\n",
    "        }\n",
    "        query_protein_ids = set([p.strip() for p in protein_domains['protein_id'].values if not pd.isnull(p)])\n",
    "        \n",
    "        matching_score = 100 * len(cds_protein_ids & query_protein_ids) / len(query_protein_ids)\n",
    "        matching_scores[assembly] = matching_score\n",
    "        \n",
    "    return matching_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-12-07 14:04:10,201 (INFO) Checking pfam protein ID match for 3,282 assemblies\n",
      "2020-12-07 14:04:18,377 (INFO) 200 / 3,282\n",
      "2020-12-07 14:04:27,256 (INFO) 400 / 3,282\n",
      "2020-12-07 14:04:35,071 (INFO) 600 / 3,282\n",
      "2020-12-07 14:04:43,468 (INFO) 800 / 3,282\n",
      "2020-12-07 14:04:50,303 (INFO) 1,000 / 3,282\n",
      "2020-12-07 14:04:56,335 (INFO) 1,200 / 3,282\n",
      "2020-12-07 14:05:02,157 (INFO) 1,400 / 3,282\n",
      "2020-12-07 14:05:06,842 (INFO) 1,600 / 3,282\n",
      "2020-12-07 14:05:11,228 (INFO) 1,800 / 3,282\n",
      "2020-12-07 14:05:15,209 (INFO) 2,000 / 3,282\n",
      "2020-12-07 14:05:18,400 (INFO) 2,200 / 3,282\n",
      "2020-12-07 14:05:21,498 (INFO) 2,400 / 3,282\n",
      "2020-12-07 14:05:22,629 (INFO) 2,600 / 3,282\n",
      "2020-12-07 14:05:22,635 (INFO) 2,800 / 3,282\n",
      "2020-12-07 14:05:24,594 (INFO) 3,000 / 3,282\n",
      "2020-12-07 14:05:28,438 (INFO) 3,200 / 3,282\n"
     ]
    }
   ],
   "source": [
    "pfam_matching_scores = check_protein_matching(engine, query_type='pfam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-12-07 14:05:29,092 (INFO) Checking tigr protein ID match for 3,282 assemblies\n",
      "2020-12-07 14:05:36,206 (INFO) 200 / 3,282\n",
      "2020-12-07 14:05:44,101 (INFO) 400 / 3,282\n",
      "2020-12-07 14:05:50,840 (INFO) 600 / 3,282\n",
      "2020-12-07 14:05:58,058 (INFO) 800 / 3,282\n",
      "2020-12-07 14:06:03,764 (INFO) 1,000 / 3,282\n",
      "2020-12-07 14:06:09,050 (INFO) 1,200 / 3,282\n",
      "2020-12-07 14:06:14,001 (INFO) 1,400 / 3,282\n",
      "2020-12-07 14:06:18,021 (INFO) 1,600 / 3,282\n",
      "2020-12-07 14:06:21,759 (INFO) 1,800 / 3,282\n",
      "2020-12-07 14:06:25,215 (INFO) 2,000 / 3,282\n",
      "2020-12-07 14:06:27,934 (INFO) 2,200 / 3,282\n",
      "2020-12-07 14:06:30,548 (INFO) 2,400 / 3,282\n",
      "2020-12-07 14:06:31,517 (INFO) 2,600 / 3,282\n",
      "2020-12-07 14:06:31,524 (INFO) 2,800 / 3,282\n",
      "2020-12-07 14:06:33,182 (INFO) 3,000 / 3,282\n",
      "2020-12-07 14:06:36,440 (INFO) 3,200 / 3,282\n"
     ]
    }
   ],
   "source": [
    "tigr_matching_scores = check_protein_matching(engine, query_type='tigr')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "33"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outlier_threshold = 90\n",
    "outlier_assemblies = {a for a in pfam_matching_scores.keys() if pfam_matching_scores[a] < outlier_threshold}\n",
    "outlier_assemblies |= {a for a in tigr_matching_scores.keys() if tigr_matching_scores[a] < outlier_threshold}\n",
    "len(outlier_assemblies)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['GCA_000011225.1',\n",
       " 'GCA_000011445.1',\n",
       " 'GCA_000012765.1',\n",
       " 'GCA_000063605.1',\n",
       " 'GCA_000092585.1',\n",
       " 'GCA_000183365.1',\n",
       " 'GCA_000183385.1',\n",
       " 'GCA_000195875.1',\n",
       " 'GCA_000328725.1',\n",
       " 'GCA_000400935.1',\n",
       " 'GCA_000400955.1',\n",
       " 'GCA_000439435.1',\n",
       " 'GCA_000439455.1',\n",
       " 'GCA_000500935.1',\n",
       " 'GCA_000565175.1',\n",
       " 'GCA_000565215.1',\n",
       " 'GCA_001029265.1',\n",
       " 'GCA_001262715.1',\n",
       " 'GCA_001267155.1',\n",
       " 'GCA_001274875.1',\n",
       " 'GCA_001281045.1',\n",
       " 'GCA_001715535.1',\n",
       " 'GCA_001886855.1',\n",
       " 'GCA_001900245.1',\n",
       " 'GCA_001998865.1',\n",
       " 'GCA_002163585.1',\n",
       " 'GCA_002237575.1',\n",
       " 'GCA_002803985.1',\n",
       " 'GCA_002804105.1',\n",
       " 'GCA_002813555.1',\n",
       " 'GCA_002865545.1',\n",
       " 'GCA_003339775.1',\n",
       " 'GCA_003363775.1']"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted(outlier_assemblies)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "36"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "q = \"\"\"\n",
    "select a.assembly_accession, s.species_taxid, s.species, s.phylum from assembly_source as a\n",
    "left join species_traits as s on s.species_taxid = a.species_taxid\n",
    "\"\"\"\n",
    "df = pd.read_sql(q, engine, index_col='assembly_accession')\n",
    "ix = set(df.index.tolist()) - set(outlier_assemblies)\n",
    "phyla = df.loc[ix]['phylum'].unique()\n",
    "len(phyla)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "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>species_taxid</th>\n",
       "      <th>species</th>\n",
       "      <th>phylum</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assembly_accession</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>GCA_000025005.1</th>\n",
       "      <td>166501</td>\n",
       "      <td>Thermobaculum terrenum</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    species_taxid                 species phylum\n",
       "assembly_accession                                              \n",
       "GCA_000025005.1            166501  Thermobaculum terrenum   None"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['phylum'].isnull()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Top Gene Ontology (GO) categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "pfam2go_path = os.path.join(os.getcwd(), 'data/domains/Pfam2go.txt')\n",
    "pfam_results = pd.read_excel(\n",
    "    os.path.join(os.getcwd(), f'data/domains/codon_bias/pfam_top.xlsx'),\n",
    "    index_col='pfam_query',\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "def parse_pfam_to_go_file(path):\n",
    "    line_re = r'^Pfam:([^\\s]+) ([^>]+) > GO:([^;]+) ; GO:([0-9]+)$'\n",
    "    domain_to_go = collections.defaultdict(list)\n",
    "    with open(path, 'r') as f:\n",
    "        for line in f:\n",
    "            if not line.strip() or line.startswith('!'):\n",
    "                continue\n",
    "                \n",
    "            m = re.match(line_re, line)\n",
    "            if m:\n",
    "                pfam_id = m[1].strip()\n",
    "                query =  m[2].strip()\n",
    "                go_label = m[3].strip()\n",
    "                go_id = m[4].strip()\n",
    "                \n",
    "                domain_to_go[query].append((go_id, go_label))\n",
    "                \n",
    "    return dict(domain_to_go)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "domain_to_go = parse_pfam_to_go_file(pfam2go_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'domain_to_go' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-9-316f32008b92>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mdomain_to_go\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'Helicase_C_2'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'domain_to_go' is not defined"
     ]
    }
   ],
   "source": [
    "domain_to_go['Helicase_C_2']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_top_go_categories(pfam_results, domain_to_go):\n",
    "    data = {\n",
    "        'go_id': [],\n",
    "        'go_label': [],\n",
    "        'count': [],\n",
    "    }\n",
    "    matching = 0\n",
    "    go_id_count = collections.defaultdict(int)\n",
    "    go_id_to_label = {}\n",
    "    for domain in pfam_results.index:\n",
    "        if domain not in domain_to_go:\n",
    "            continue\n",
    "        else:\n",
    "            matching += 1\n",
    "            \n",
    "        go_data = domain_to_go[domain]\n",
    "        \n",
    "        for go_id, go_label in go_data:\n",
    "            go_id_count[go_id] += 1\n",
    "            go_id_to_label[go_id] = go_label\n",
    "            \n",
    "    for go_id in sorted(go_id_count.keys()):\n",
    "        data['go_id'].append(go_id)\n",
    "        data['go_label'].append(go_id_to_label[go_id])\n",
    "        data['count'].append(go_id_count[go_id])\n",
    "            \n",
    "    print(f'{matching} / {len(pfam_results)} ({100 * matching / len(pfam_results):.0f}%) matching domains with go')\n",
    "    \n",
    "    return pd.DataFrame.from_dict(data).set_index('go_id').sort_values('count', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "245 / 553 (44%) matching domains with go\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>go_label</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>go_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0005524</th>\n",
       "      <td>ATP binding</td>\n",
       "      <td>53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0003677</th>\n",
       "      <td>DNA binding</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0055114</th>\n",
       "      <td>oxidation-reduction process</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0005975</th>\n",
       "      <td>carbohydrate metabolic process</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0016491</th>\n",
       "      <td>oxidoreductase activity</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0003824</th>\n",
       "      <td>catalytic activity</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0005515</th>\n",
       "      <td>protein binding</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0003899</th>\n",
       "      <td>DNA-directed 5'-3' RNA polymerase activity</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0006351</th>\n",
       "      <td>transcription, DNA-templated</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0000166</th>\n",
       "      <td>nucleotide binding</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0016021</th>\n",
       "      <td>integral component of membrane</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0006260</th>\n",
       "      <td>DNA replication</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0003723</th>\n",
       "      <td>RNA binding</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0004553</th>\n",
       "      <td>hydrolase activity, hydrolyzing O-glycosyl com...</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0003676</th>\n",
       "      <td>nucleic acid binding</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0006298</th>\n",
       "      <td>mismatch repair</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0006508</th>\n",
       "      <td>proteolysis</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0005737</th>\n",
       "      <td>cytoplasm</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0030983</th>\n",
       "      <td>mismatched DNA binding</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0006265</th>\n",
       "      <td>DNA topological change</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                  go_label  count\n",
       "go_id                                                            \n",
       "0005524                                        ATP binding     53\n",
       "0003677                                        DNA binding     26\n",
       "0055114                        oxidation-reduction process     26\n",
       "0005975                     carbohydrate metabolic process     21\n",
       "0016491                            oxidoreductase activity     15\n",
       "0003824                                 catalytic activity     14\n",
       "0005515                                    protein binding     13\n",
       "0003899         DNA-directed 5'-3' RNA polymerase activity     11\n",
       "0006351                       transcription, DNA-templated     11\n",
       "0000166                                 nucleotide binding     10\n",
       "0016021                     integral component of membrane      8\n",
       "0006260                                    DNA replication      8\n",
       "0003723                                        RNA binding      8\n",
       "0004553  hydrolase activity, hydrolyzing O-glycosyl com...      8\n",
       "0003676                               nucleic acid binding      7\n",
       "0006298                                    mismatch repair      7\n",
       "0006508                                        proteolysis      7\n",
       "0005737                                          cytoplasm      6\n",
       "0030983                             mismatched DNA binding      6\n",
       "0006265                             DNA topological change      6"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "go_df = compute_top_go_categories(pfam_results, domain_to_go)\n",
    "go_df.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [],
   "source": [
    "go_df.to_excel(os.path.join(os.getcwd(), 'data/domains/codon_bias/go_labels.xlsx'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## tRNA adaptation index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "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>assembly_accession</th>\n",
       "      <th>species_taxid</th>\n",
       "      <th>row_id</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>protein</th>\n",
       "      <th>gene</th>\n",
       "      <th>adaptation_index</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>GCA_000005825.2</td>\n",
       "      <td>79885</td>\n",
       "      <td>253459</td>\n",
       "      <td>ADC51616.1</td>\n",
       "      <td>hypothetical protein</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>GCA_000005825.2</td>\n",
       "      <td>79885</td>\n",
       "      <td>252123</td>\n",
       "      <td>ADC50280.1</td>\n",
       "      <td>hypothetical protein</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.901509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>GCA_000005825.2</td>\n",
       "      <td>79885</td>\n",
       "      <td>252052</td>\n",
       "      <td>ADC50209.1</td>\n",
       "      <td>hypothetical protein</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.898870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>GCA_000005825.2</td>\n",
       "      <td>79885</td>\n",
       "      <td>253277</td>\n",
       "      <td>ADC51434.1</td>\n",
       "      <td>cold-shock DNA-binding protein family protein</td>\n",
       "      <td>cspC3</td>\n",
       "      <td>0.871788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GCA_000005825.2</td>\n",
       "      <td>79885</td>\n",
       "      <td>253078</td>\n",
       "      <td>ADC51235.1</td>\n",
       "      <td>hypothetical protein</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.842573</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  assembly_accession  species_taxid  row_id  protein_id  \\\n",
       "0    GCA_000005825.2          79885  253459  ADC51616.1   \n",
       "1    GCA_000005825.2          79885  252123  ADC50280.1   \n",
       "2    GCA_000005825.2          79885  252052  ADC50209.1   \n",
       "3    GCA_000005825.2          79885  253277  ADC51434.1   \n",
       "4    GCA_000005825.2          79885  253078  ADC51235.1   \n",
       "\n",
       "                                         protein   gene  adaptation_index  \n",
       "0                           hypothetical protein    NaN          1.000000  \n",
       "1                           hypothetical protein    NaN          0.901509  \n",
       "2                           hypothetical protein    NaN          0.898870  \n",
       "3  cold-shock DNA-binding protein family protein  cspC3          0.871788  \n",
       "4                           hypothetical protein    NaN          0.842573  "
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trnai = pd.read_csv(os.path.join(os.getcwd(), 'data/trn_adaptation_index/GCA_000005825.2_tai.csv'))\n",
    "trnai.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [],
   "source": [
    "def score_fn(trnai):\n",
    "    mean = trnai['adaptation_index'].mean()\n",
    "    std = trnai['adaptation_index'].std()\n",
    "    \n",
    "    def fn(adaptation_index):\n",
    "        if adaptation_index > mean + std:\n",
    "            return 'over expressed'\n",
    "        elif adaptation_index < mean - std:\n",
    "            return 'under expressed'\n",
    "        else:\n",
    "            return 'normally expressed'\n",
    "    \n",
    "    return fn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "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>assembly_accession</th>\n",
       "      <th>species_taxid</th>\n",
       "      <th>row_id</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>protein</th>\n",
       "      <th>gene</th>\n",
       "      <th>adaptation_index</th>\n",
       "      <th>percentile</th>\n",
       "      <th>expression</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>GCA_000005825.2</td>\n",
       "      <td>79885</td>\n",
       "      <td>253459</td>\n",
       "      <td>ADC51616.1</td>\n",
       "      <td>hypothetical protein</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>100</td>\n",
       "      <td>over expressed</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>GCA_000005825.2</td>\n",
       "      <td>79885</td>\n",
       "      <td>252123</td>\n",
       "      <td>ADC50280.1</td>\n",
       "      <td>hypothetical protein</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.901509</td>\n",
       "      <td>100</td>\n",
       "      <td>over expressed</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>GCA_000005825.2</td>\n",
       "      <td>79885</td>\n",
       "      <td>252052</td>\n",
       "      <td>ADC50209.1</td>\n",
       "      <td>hypothetical protein</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.898870</td>\n",
       "      <td>100</td>\n",
       "      <td>over expressed</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>GCA_000005825.2</td>\n",
       "      <td>79885</td>\n",
       "      <td>253277</td>\n",
       "      <td>ADC51434.1</td>\n",
       "      <td>cold-shock DNA-binding protein family protein</td>\n",
       "      <td>cspC3</td>\n",
       "      <td>0.871788</td>\n",
       "      <td>100</td>\n",
       "      <td>over expressed</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GCA_000005825.2</td>\n",
       "      <td>79885</td>\n",
       "      <td>253078</td>\n",
       "      <td>ADC51235.1</td>\n",
       "      <td>hypothetical protein</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.842573</td>\n",
       "      <td>100</td>\n",
       "      <td>over expressed</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  assembly_accession  species_taxid  row_id  protein_id  \\\n",
       "0    GCA_000005825.2          79885  253459  ADC51616.1   \n",
       "1    GCA_000005825.2          79885  252123  ADC50280.1   \n",
       "2    GCA_000005825.2          79885  252052  ADC50209.1   \n",
       "3    GCA_000005825.2          79885  253277  ADC51434.1   \n",
       "4    GCA_000005825.2          79885  253078  ADC51235.1   \n",
       "\n",
       "                                         protein   gene  adaptation_index  \\\n",
       "0                           hypothetical protein    NaN          1.000000   \n",
       "1                           hypothetical protein    NaN          0.901509   \n",
       "2                           hypothetical protein    NaN          0.898870   \n",
       "3  cold-shock DNA-binding protein family protein  cspC3          0.871788   \n",
       "4                           hypothetical protein    NaN          0.842573   \n",
       "\n",
       "   percentile      expression  \n",
       "0         100  over expressed  \n",
       "1         100  over expressed  \n",
       "2         100  over expressed  \n",
       "3         100  over expressed  \n",
       "4         100  over expressed  "
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trnai['expression'] = trnai['adaptation_index'].apply(score_fn(trnai))\n",
    "trnai.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "trnai['expression'].hist();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## AAA domains scale in numbers with genome size\n",
    "\n",
    "How does it affect our scoring? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>species</th>\n",
       "      <th>genome_size</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>species_taxid</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Azorhizobium caulinodans</td>\n",
       "      <td>5369771.500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Buchnera aphidicola</td>\n",
       "      <td>601699.243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Cellulomonas gilvus</td>\n",
       "      <td>3526440.800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Dictyoglomus thermophilum</td>\n",
       "      <td>1959987.600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Pelobacter carbinolicus</td>\n",
       "      <td>3722544.667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 species  genome_size\n",
       "species_taxid                                        \n",
       "7               Azorhizobium caulinodans  5369771.500\n",
       "9                    Buchnera aphidicola   601699.243\n",
       "11                   Cellulomonas gilvus  3526440.800\n",
       "14             Dictyoglomus thermophilum  1959987.600\n",
       "19               Pelobacter carbinolicus  3722544.667"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "species_traits = pd.read_sql(\n",
    "    'select species_taxid, species, genome_size from species_traits', \n",
    "    engine,\n",
    "    index_col='species_taxid',\n",
    ")\n",
    "species_traits.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>species</th>\n",
       "      <th>genome_size</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>species_taxid</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2336</th>\n",
       "      <td>Thermotoga maritima</td>\n",
       "      <td>1865969.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           species  genome_size\n",
       "species_taxid                                  \n",
       "2336           Thermotoga maritima    1865969.5"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "species_traits.loc[[2336]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>assembly_score_sum</th>\n",
       "      <th>assembly_count</th>\n",
       "      <th>assembly_count_top</th>\n",
       "      <th>score</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pfam_query</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AAA_assoc_2</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_PrkA</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_32</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_27</th>\n",
       "      <td>0.250000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_3</th>\n",
       "      <td>0.111111</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.111111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_5</th>\n",
       "      <td>0.092239</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>0.092239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_15</th>\n",
       "      <td>0.062500</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0.062500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA</th>\n",
       "      <td>0.060491</td>\n",
       "      <td>23</td>\n",
       "      <td>4</td>\n",
       "      <td>0.060491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_23</th>\n",
       "      <td>0.040000</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.040000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_21</th>\n",
       "      <td>0.003365</td>\n",
       "      <td>41</td>\n",
       "      <td>2</td>\n",
       "      <td>0.003365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_19</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_33</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_lid_4</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_lid_3</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_lid_2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_12</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_14</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_7</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_16</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_31</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_17</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_30</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_18</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_29</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_11</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_25</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_22</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAA_lid_9</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             assembly_score_sum  assembly_count  assembly_count_top     score\n",
       "pfam_query                                                                   \n",
       "AAA_assoc_2            1.000000               1                   1  1.000000\n",
       "AAA_PrkA               1.000000               1                   1  1.000000\n",
       "AAA_32                 1.000000               1                   1  1.000000\n",
       "AAA_27                 0.250000               2                   1  0.250000\n",
       "AAA_3                  0.111111               3                   1  0.111111\n",
       "AAA_5                  0.092239              13                   3  0.092239\n",
       "AAA_15                 0.062500               4                   1  0.062500\n",
       "AAA                    0.060491              23                   4  0.060491\n",
       "AAA_23                 0.040000               5                   1  0.040000\n",
       "AAA_21                 0.003365              41                   2  0.003365\n",
       "AAA_19                 0.000000               6                   0  0.000000\n",
       "AAA_33                 0.000000               2                   0  0.000000\n",
       "AAA_lid_4              0.000000               1                   0  0.000000\n",
       "AAA_lid_3              0.000000               1                   0  0.000000\n",
       "AAA_lid_2              0.000000               1                   0  0.000000\n",
       "AAA_12                 0.000000               4                   0  0.000000\n",
       "AAA_14                 0.000000               6                   0  0.000000\n",
       "AAA_7                  0.000000               1                   0  0.000000\n",
       "AAA_16                 0.000000               3                   0  0.000000\n",
       "AAA_31                 0.000000               5                   0  0.000000\n",
       "AAA_17                 0.000000               3                   0  0.000000\n",
       "AAA_2                  0.000000               4                   0  0.000000\n",
       "AAA_30                 0.000000               6                   0  0.000000\n",
       "AAA_18                 0.000000               4                   0  0.000000\n",
       "AAA_29                 0.000000               1                   0  0.000000\n",
       "AAA_11                 0.000000               2                   0  0.000000\n",
       "AAA_25                 0.000000               3                   0  0.000000\n",
       "AAA_22                 0.000000              12                   0  0.000000\n",
       "AAA_lid_9              0.000000               2                   0  0.000000"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "thermotoga_maritima_domains = compute_protein_domain_scores(engine, ['GCA_000008545.1'], 'pfam')\n",
    "aaa_domains = sorted([d for d in thermotoga_maritima_domains.index if 'AAA' in d])\n",
    "thermotoga_maritima_domains.loc[aaa_domains].sort_values('score', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Count unique Pfam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "11087\n",
      "CPU times: user 12.3 s, sys: 2.11 s, total: 14.4 s\n",
      "Wall time: 15.8 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "import pathlib\n",
    "pfam_folder = '/Users/srom/workspace/rna_learn/data/domains/tri_nucleotide_bias/pfam'\n",
    "\n",
    "protein_domains = set()\n",
    "paths = pathlib.Path(pfam_folder).glob('*.csv')\n",
    "for p in paths:\n",
    "    with p.open() as f:\n",
    "        df = pd.read_csv(f)\n",
    "        protein_domains |= set(df['pfam_query'].unique())\n",
    "        \n",
    "print(len(protein_domains))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_pfam_domains = len(protein_domains)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.1646973933435554"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "100 * 240 / n_pfam_domains"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}