Verrou All.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c9eb32da",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import csv\n",
    "import matplotlib.pyplot as plt\n",
    "import pickle\n",
    "from statistics import mean, stdev"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "674b2183",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Directory variables\n",
    "\n",
    "PROTEIN_SCORE_DIR = 'verrou_all_results'\n",
    "IEEE_DIR = 'IEEE'\n",
    "IEEE_PROTEIN_SCORE_REF = 'ieee_ref.pkl'\n",
    "INSTR_LIB = 'all'\n",
    "THRESHOLD = 0.3"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ef54ef1",
   "metadata": {},
   "source": [
    "## Calculate protein stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ece5b4c3",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "docker_pred = {}\n",
    "for i in range(1,11):\n",
    "    for j in range(0,646):\n",
    "        f = open(PROTEIN_SCORE_DIR + '/%s_%s.pkl'%(i,j), 'rb')\n",
    "        temp = pickle.load(f)\n",
    "                \n",
    "        pkl_dict = {}\n",
    "        for prot in temp:\n",
    "            pkl_dict[prot] = {}\n",
    "            for go_id in temp[prot]:\n",
    "                if temp[prot][go_id] >= 0.3:\n",
    "                    pkl_dict[prot][go_id] = np.float64(temp[prot][go_id])\n",
    "  \n",
    "        for protein in pkl_dict.keys():\n",
    "            if i == 1:\n",
    "                docker_pred[protein] = {}\n",
    "                for go_id in pkl_dict[protein]:\n",
    "                    docker_pred[protein][go_id] = np.array([pkl_dict[protein][go_id]])\n",
    "            else:\n",
    "\n",
    "                for go_id in pkl_dict[protein]:\n",
    "                    docker_pred[protein][go_id] = np.append(docker_pred[protein][go_id], pkl_dict[protein][go_id])\n",
    "\n",
    "original_file = open(IEEE_DIR + IEEE_PROTEIN_SCORE_REF, 'rb')\n",
    "pkl_dict = pickle.load(original_file)\n",
    "original_results = {}\n",
    "\n",
    "for protein in pkl_dict:\n",
    "    original_results[protein] = pkl_dict[protein]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d2cb1744",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# protein stats\n",
    "stats = {}\n",
    "dfs = {}\n",
    "proteins = []\n",
    "for name in list(docker_pred.keys()):\n",
    "    proteins.append(name)\n",
    "#     print(name, len(docker_pred[name]))\n",
    "    stats[name] = {'GO term':[], 'Standard deviation':[], 'Mean': [], 'Standard error': [], 'Original Result': []}\n",
    "    for i in docker_pred[name].keys():\n",
    "        stats[name]['GO term'] = np.append(stats[name]['GO term'], i)\n",
    "        stats[name]['Standard deviation'] = np.append(stats[name]['Standard deviation'], np.std(docker_pred[name][i].astype(np.float64), dtype=np.float64))\n",
    "        stats[name]['Mean'] = np.append(stats[name]['Mean'], np.mean(docker_pred[name][i].astype(np.float64), dtype=np.float64))\n",
    "        stats[name]['Standard error'] = np.append(stats[name]['Standard error'], (np.std(docker_pred[name][i].astype(np.float64), dtype=np.float64) / np.sqrt(len(docker_pred[name][i])).astype(np.float64)))\n",
    "        if i in original_results[name].keys():\n",
    "            stats[name]['Original Result'] = np.append(stats[name]['Original Result'], original_results[name][i])\n",
    "        \n",
    "    dfs[name] = pd.DataFrame(stats[name])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "795a8b06",
   "metadata": {},
   "source": [
    "### Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "23ae725e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#metric stats\n",
    "metrics ={'bp': {'threshold':[], 'Smin':[], 'Fmax':[], 'AUPR':[]}, 'cc':{'threshold':[], 'Smin':[], 'Fmax':[], 'AUPR':[]}, 'mf':{'threshold':[], 'Smin':[], 'Fmax':[], 'AUPR':[]}}\n",
    "metric_stats ={'bp': {'threshold':{}, 'Smin':{}, 'Fmax':{}, 'AUPR':{}}, 'cc':{'threshold':{}, 'Smin':{}, 'Fmax':{}, 'AUPR':{}}, 'mf':{'threshold':{}, 'Smin':{}, 'Fmax':{}, 'AUPR':{}}}\n",
    "df_metric = {'bp':{}, 'cc':{}, 'mf':{}}\n",
    "\n",
    "for category in ['bp', 'cc', 'mf']:\n",
    "    for i in range(1,11): #\n",
    "\n",
    "        metrics_file = open(PROTEIN_SCORE_DIR + \"/%s_%s.txt\"%(category, i))\n",
    "        read_file = csv.reader(metrics_file)\n",
    "        next(read_file)\n",
    "\n",
    "        for row in read_file: \n",
    "#             if 'Smin threshold' in row[0] or 'Precision' in row[0] or 'Recall' in row[0] or 'Positive' in row[0] or 'Negative' in row[0]: continue\n",
    "            if 'Length' in row[0]: continue\n",
    "            metric, score = row[0].split(': ')\n",
    "            metrics[category][metric].append(np.float64(score))\n",
    "    \n",
    "    metrics_file = csv.reader(open(IEEE_DIR + \"/%s.txt\"%category))\n",
    "    next(metrics_file)\n",
    "    for row in metrics_file:\n",
    "#         if 'Smin threshold' in row[0] or 'Precision' in row[0] or 'Recall' in row[0] or 'Positive' in row[0] or 'Negative' in row[0]: continue\n",
    "        if 'Length' in row[0]: continue\n",
    "        term, score = row[0].split(': ')\n",
    "        metric_stats[category][term]['Original Result'] = [np.float64(score)]\n",
    "\n",
    "    for i in metrics[category]: \n",
    "        metric_stats[category][i]['Standard Deviation'] = [stdev(metrics[category][i])]\n",
    "        metric_stats[category][i]['Mean'] = [mean(metrics[category][i])]\n",
    "        metric_stats[category][i]['Standard Error'] = [stdev(metrics[category][i]) / np.sqrt(len(metrics[category][i]), dtype=np.float64)]\n",
    "\n",
    "        df_metric[category]['%s'%i] = pd.DataFrame(metric_stats[category][i])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73f9d8da",
   "metadata": {},
   "source": [
    "## Significant Digits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "2bb0a200",
   "metadata": {},
   "outputs": [],
   "source": [
    "count = 0\n",
    "sig_file = open('References/sig_proteins_%s.txt'%INSTR_LIB)\n",
    "protein_sig = {}\n",
    "for row in sig_file:\n",
    "    if 'NEW PROTEIN' in row:\n",
    "        p_name = proteins[count]\n",
    "        protein_sig[p_name] = {}\n",
    "        count += 1\n",
    "        continue\n",
    "        \n",
    "    if len(row.split(',')) == 1:\n",
    "        name = row[:-2]\n",
    "\n",
    "    else:\n",
    "        if 'Relative' not in name and 'CNH' not in name:\n",
    "            scores = row.split(',')[:-1]\n",
    "            protein_sig[p_name][name + ' Mean'] = np.mean(np.array(scores).astype(float))\n",
    "            protein_sig[p_name][name + ' Standard Deviation'] = np.std(np.array(scores).astype(float))\n",
    "            \n",
    "protein_sig = pd.DataFrame(protein_sig).T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ec269ed8",
   "metadata": {},
   "outputs": [],
   "source": [
    "sig_metrics = {'bp': {'threshold':{}, 'Smin':{}, 'Fmax':{}, 'AUPR':{}}, 'cc':{'threshold':{}, 'Smin':{}, 'Fmax':{}, 'AUPR':{}}, 'mf':{'threshold':{}, 'Smin':{}, 'Fmax':{}, 'AUPR':{}}}\n",
    "\n",
    "for category in ['bp', 'cc', 'mf']:\n",
    "    sig_file = open('References/sig_%s_%s.txt'%(category, INSTR_LIB))\n",
    "\n",
    "    for row in sig_file:\n",
    "        if len(row.split(',')) == 1:\n",
    "            name = row[:-1]\n",
    "        else:\n",
    "            if 'Relative' not in name and 'CNH' not in name:\n",
    "                scores = row.split(',')\n",
    "                sig_metrics[category]['threshold'][name] = scores[0]\n",
    "                sig_metrics[category]['Smin'][name] = scores[1]\n",
    "                sig_metrics[category]['Fmax'][name] = scores[2]\n",
    "                sig_metrics[category]['AUPR'][name] = scores[3]\n",
    "\n",
    "for category in ['bp', 'cc', 'mf']:\n",
    "    for metric in sig_metrics[category]:\n",
    "        for column in sig_metrics[category][metric]:\n",
    "#             print(column)\n",
    "            df_metric[category][metric][column] = sig_metrics[category][metric][column]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e9a25eaf",
   "metadata": {},
   "source": [
    "## Results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14be7f68",
   "metadata": {},
   "source": [
    "#### Significant Digits of Test Proteins"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c6056afb",
   "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>[General] Absolute  significant Mean</th>\n",
       "      <th>[General] Absolute  significant Standard Deviation</th>\n",
       "      <th>[General] Absolute contributing Mean</th>\n",
       "      <th>[General] Absolute contributing Standard Deviation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3874.000000</td>\n",
       "      <td>3874.000000</td>\n",
       "      <td>3874.000000</td>\n",
       "      <td>3874.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>6.790122</td>\n",
       "      <td>0.156376</td>\n",
       "      <td>6.797097</td>\n",
       "      <td>0.171162</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.145018</td>\n",
       "      <td>0.128039</td>\n",
       "      <td>0.145653</td>\n",
       "      <td>0.135745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2.665878</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.446208</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.761597</td>\n",
       "      <td>0.132558</td>\n",
       "      <td>6.773175</td>\n",
       "      <td>0.138693</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>6.809155</td>\n",
       "      <td>0.147474</td>\n",
       "      <td>6.812784</td>\n",
       "      <td>0.159314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.845676</td>\n",
       "      <td>0.169167</td>\n",
       "      <td>6.844471</td>\n",
       "      <td>0.188121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6.923690</td>\n",
       "      <td>2.826399</td>\n",
       "      <td>6.923690</td>\n",
       "      <td>2.976401</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       [General] Absolute  significant Mean  \\\n",
       "count                           3874.000000   \n",
       "mean                               6.790122   \n",
       "std                                0.145018   \n",
       "min                                2.665878   \n",
       "25%                                6.761597   \n",
       "50%                                6.809155   \n",
       "75%                                6.845676   \n",
       "max                                6.923690   \n",
       "\n",
       "       [General] Absolute  significant Standard Deviation  \\\n",
       "count                                        3874.000000    \n",
       "mean                                            0.156376    \n",
       "std                                             0.128039    \n",
       "min                                             0.000000    \n",
       "25%                                             0.132558    \n",
       "50%                                             0.147474    \n",
       "75%                                             0.169167    \n",
       "max                                             2.826399    \n",
       "\n",
       "       [General] Absolute contributing Mean  \\\n",
       "count                           3874.000000   \n",
       "mean                               6.797097   \n",
       "std                                0.145653   \n",
       "min                                2.446208   \n",
       "25%                                6.773175   \n",
       "50%                                6.812784   \n",
       "75%                                6.844471   \n",
       "max                                6.923690   \n",
       "\n",
       "       [General] Absolute contributing Standard Deviation  \n",
       "count                                        3874.000000   \n",
       "mean                                            0.171162   \n",
       "std                                             0.135745   \n",
       "min                                             0.000000   \n",
       "25%                                             0.138693   \n",
       "50%                                             0.159314   \n",
       "75%                                             0.188121   \n",
       "max                                             2.976401   "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "protein_sig.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7351dc9",
   "metadata": {},
   "source": [
    "#### Stats on Protein Metrics for 3 Principal Protein Ontology Classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "cd48269d",
   "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>Original Result</th>\n",
       "      <th>Standard Deviation</th>\n",
       "      <th>Mean</th>\n",
       "      <th>Standard Error</th>\n",
       "      <th>[General] Absolute  significant:</th>\n",
       "      <th>[General] Absolute contributing:</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Smin</th>\n",
       "      <td>9.597233</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.597233</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fmax</th>\n",
       "      <td>0.663958</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.663958</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AUPR</th>\n",
       "      <td>0.670708</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.670708</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fmax threshold</th>\n",
       "      <td>0.180000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.180000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Original Result  Standard Deviation      Mean  Standard Error  \\\n",
       "Smin                   9.597233                 0.0  9.597233             0.0   \n",
       "Fmax                   0.663958                 0.0  0.663958             0.0   \n",
       "AUPR                   0.670708                 0.0  0.670708             0.0   \n",
       "Fmax threshold         0.180000                 0.0  0.180000             0.0   \n",
       "\n",
       "               [General] Absolute  significant:  \\\n",
       "Smin                          6.923689900271567   \n",
       "Fmax                          6.923689900271567   \n",
       "AUPR                          6.923689900271567   \n",
       "Fmax threshold                6.923689900271567   \n",
       "\n",
       "               [General] Absolute contributing:  \n",
       "Smin                          6.923689900271567  \n",
       "Fmax                          6.923689900271567  \n",
       "AUPR                          6.923689900271567  \n",
       "Fmax threshold                6.923689900271567  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_mf = df_metric['mf']['Smin'].append(df_metric['mf']['Fmax']).append(df_metric['mf']['AUPR']).append(df_metric['mf']['threshold']).set_index([pd.Index(['Smin', 'Fmax', 'AUPR', 'Fmax threshold'])])\n",
    "# with pd.option_context('display.precision', 20): display(df_bp)\n",
    "df_mf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "2d99e88a",
   "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>Original Result</th>\n",
       "      <th>Standard Deviation</th>\n",
       "      <th>Mean</th>\n",
       "      <th>Standard Error</th>\n",
       "      <th>[General] Absolute  significant:</th>\n",
       "      <th>[General] Absolute contributing:</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Smin</th>\n",
       "      <td>10.448033</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10.448033</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fmax</th>\n",
       "      <td>0.694438</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.694438</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AUPR</th>\n",
       "      <td>0.737946</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.737946</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fmax threshold</th>\n",
       "      <td>0.240000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.240000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Original Result  Standard Deviation       Mean  \\\n",
       "Smin                  10.448033                 0.0  10.448033   \n",
       "Fmax                   0.694438                 0.0   0.694438   \n",
       "AUPR                   0.737946                 0.0   0.737946   \n",
       "Fmax threshold         0.240000                 0.0   0.240000   \n",
       "\n",
       "                Standard Error [General] Absolute  significant:  \\\n",
       "Smin                       0.0                6.923689900271567   \n",
       "Fmax                       0.0                6.923689900271567   \n",
       "AUPR                       0.0                6.923689900271567   \n",
       "Fmax threshold             0.0                6.923689900271567   \n",
       "\n",
       "               [General] Absolute contributing:  \n",
       "Smin                          6.923689900271567  \n",
       "Fmax                          6.923689900271567  \n",
       "AUPR                          6.923689900271567  \n",
       "Fmax threshold                6.923689900271567  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_cc = df_metric['cc']['Smin'].append(df_metric['cc']['Fmax']).append(df_metric['cc']['AUPR']).append(df_metric['cc']['threshold']).set_index([pd.Index(['Smin', 'Fmax', 'AUPR', 'Fmax threshold'])])\n",
    "# with pd.option_context('display.precision', 20): display(df_bp)\n",
    "df_cc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "3d181566",
   "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>Original Result</th>\n",
       "      <th>Standard Deviation</th>\n",
       "      <th>Mean</th>\n",
       "      <th>Standard Error</th>\n",
       "      <th>[General] Absolute  significant:</th>\n",
       "      <th>[General] Absolute contributing:</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Smin</th>\n",
       "      <td>44.377911</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>44.377911</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fmax</th>\n",
       "      <td>0.530160</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.530160</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AUPR</th>\n",
       "      <td>0.495861</td>\n",
       "      <td>3.980676e-08</td>\n",
       "      <td>0.495861</td>\n",
       "      <td>1.258800e-08</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fmax threshold</th>\n",
       "      <td>0.210000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.210000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>6.923689900271567</td>\n",
       "      <td>6.923689900271567</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Original Result  Standard Deviation       Mean  \\\n",
       "Smin                  44.377911        0.000000e+00  44.377911   \n",
       "Fmax                   0.530160        0.000000e+00   0.530160   \n",
       "AUPR                   0.495861        3.980676e-08   0.495861   \n",
       "Fmax threshold         0.210000        0.000000e+00   0.210000   \n",
       "\n",
       "                Standard Error [General] Absolute  significant:  \\\n",
       "Smin              0.000000e+00                6.923689900271567   \n",
       "Fmax              0.000000e+00                6.923689900271567   \n",
       "AUPR              1.258800e-08                6.923689900271567   \n",
       "Fmax threshold    0.000000e+00                6.923689900271567   \n",
       "\n",
       "               [General] Absolute contributing:  \n",
       "Smin                          6.923689900271567  \n",
       "Fmax                          6.923689900271567  \n",
       "AUPR                          6.923689900271567  \n",
       "Fmax threshold                6.923689900271567  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_bp = df_metric['bp']['Smin'].append(df_metric['bp']['Fmax']).append(df_metric['bp']['AUPR']).append(df_metric['bp']['threshold']).set_index([pd.Index(['Smin', 'Fmax', 'AUPR', 'Fmax threshold'])])\n",
    "# with pd.option_context('display.precision', 20): display(df_bp)\n",
    "df_bp"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e6ab136",
   "metadata": {},
   "source": [
    "#### Stats on Individual Protein Classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "0889b29e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7.205816996988545e-07"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "max_val = []\n",
    "for i in np.array(list(dfs.keys())):\n",
    "\n",
    "    if len(dfs[i]['Standard deviation']) != 0:\n",
    "        a = max(dfs[i]['Standard deviation'])\n",
    "        max_val.append(a)\n",
    "\n",
    "max(max_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "a216abd5",
   "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>GO term</th>\n",
       "      <th>Standard deviation</th>\n",
       "      <th>Mean</th>\n",
       "      <th>Standard error</th>\n",
       "      <th>Original Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>GO:0005575</td>\n",
       "      <td>1.047754e-07</td>\n",
       "      <td>0.776371</td>\n",
       "      <td>3.313289e-08</td>\n",
       "      <td>0.776371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>GO:0110165</td>\n",
       "      <td>1.047754e-07</td>\n",
       "      <td>0.776371</td>\n",
       "      <td>3.313289e-08</td>\n",
       "      <td>0.776371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>GO:0003674</td>\n",
       "      <td>1.137184e-07</td>\n",
       "      <td>0.620349</td>\n",
       "      <td>3.596092e-08</td>\n",
       "      <td>0.620349</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>GO:0008150</td>\n",
       "      <td>9.253129e-08</td>\n",
       "      <td>0.933437</td>\n",
       "      <td>2.926096e-08</td>\n",
       "      <td>0.933437</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GO:0043229</td>\n",
       "      <td>9.253129e-08</td>\n",
       "      <td>0.530498</td>\n",
       "      <td>2.926096e-08</td>\n",
       "      <td>0.530498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>GO:0043231</td>\n",
       "      <td>6.529362e-08</td>\n",
       "      <td>0.410556</td>\n",
       "      <td>2.064765e-08</td>\n",
       "      <td>0.410556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>GO:0016020</td>\n",
       "      <td>6.114925e-08</td>\n",
       "      <td>0.468064</td>\n",
       "      <td>1.933709e-08</td>\n",
       "      <td>0.468064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>GO:0043227</td>\n",
       "      <td>8.429370e-08</td>\n",
       "      <td>0.421303</td>\n",
       "      <td>2.665601e-08</td>\n",
       "      <td>0.421303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>GO:0005622</td>\n",
       "      <td>1.067905e-07</td>\n",
       "      <td>0.764394</td>\n",
       "      <td>3.377012e-08</td>\n",
       "      <td>0.764394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>GO:0043226</td>\n",
       "      <td>9.253129e-08</td>\n",
       "      <td>0.530498</td>\n",
       "      <td>2.926096e-08</td>\n",
       "      <td>0.530498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>GO:0005634</td>\n",
       "      <td>8.476652e-08</td>\n",
       "      <td>0.362152</td>\n",
       "      <td>2.680553e-08</td>\n",
       "      <td>0.362152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>GO:0005737</td>\n",
       "      <td>9.684608e-08</td>\n",
       "      <td>0.540246</td>\n",
       "      <td>3.062542e-08</td>\n",
       "      <td>0.540246</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>GO:0009987</td>\n",
       "      <td>8.534086e-08</td>\n",
       "      <td>0.872470</td>\n",
       "      <td>2.698715e-08</td>\n",
       "      <td>0.872470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>GO:0005829</td>\n",
       "      <td>6.114925e-08</td>\n",
       "      <td>0.332176</td>\n",
       "      <td>1.933709e-08</td>\n",
       "      <td>0.332176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>GO:0008152</td>\n",
       "      <td>1.130919e-07</td>\n",
       "      <td>0.678671</td>\n",
       "      <td>3.576279e-08</td>\n",
       "      <td>0.678671</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>GO:0071840</td>\n",
       "      <td>1.004475e-07</td>\n",
       "      <td>0.448246</td>\n",
       "      <td>3.176430e-08</td>\n",
       "      <td>0.448246</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>GO:0016043</td>\n",
       "      <td>1.004475e-07</td>\n",
       "      <td>0.448246</td>\n",
       "      <td>3.176430e-08</td>\n",
       "      <td>0.448246</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>GO:0071704</td>\n",
       "      <td>7.152557e-08</td>\n",
       "      <td>0.666036</td>\n",
       "      <td>2.261837e-08</td>\n",
       "      <td>0.666036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>GO:0043170</td>\n",
       "      <td>1.355596e-07</td>\n",
       "      <td>0.491758</td>\n",
       "      <td>4.286772e-08</td>\n",
       "      <td>0.491758</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>GO:1901564</td>\n",
       "      <td>7.183534e-08</td>\n",
       "      <td>0.309629</td>\n",
       "      <td>2.271633e-08</td>\n",
       "      <td>0.309629</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>GO:0006807</td>\n",
       "      <td>1.251698e-07</td>\n",
       "      <td>0.503289</td>\n",
       "      <td>3.958215e-08</td>\n",
       "      <td>0.503289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>GO:0044238</td>\n",
       "      <td>1.081130e-07</td>\n",
       "      <td>0.557016</td>\n",
       "      <td>3.418834e-08</td>\n",
       "      <td>0.557016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>GO:0044237</td>\n",
       "      <td>1.390207e-07</td>\n",
       "      <td>0.563157</td>\n",
       "      <td>4.396221e-08</td>\n",
       "      <td>0.563157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>GO:0003824</td>\n",
       "      <td>9.798576e-08</td>\n",
       "      <td>0.372867</td>\n",
       "      <td>3.098582e-08</td>\n",
       "      <td>0.372867</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       GO term  Standard deviation      Mean  Standard error  Original Result\n",
       "0   GO:0005575        1.047754e-07  0.776371    3.313289e-08         0.776371\n",
       "1   GO:0110165        1.047754e-07  0.776371    3.313289e-08         0.776371\n",
       "2   GO:0003674        1.137184e-07  0.620349    3.596092e-08         0.620349\n",
       "3   GO:0008150        9.253129e-08  0.933437    2.926096e-08         0.933437\n",
       "4   GO:0043229        9.253129e-08  0.530498    2.926096e-08         0.530498\n",
       "5   GO:0043231        6.529362e-08  0.410556    2.064765e-08         0.410556\n",
       "6   GO:0016020        6.114925e-08  0.468064    1.933709e-08         0.468064\n",
       "7   GO:0043227        8.429370e-08  0.421303    2.665601e-08         0.421303\n",
       "8   GO:0005622        1.067905e-07  0.764394    3.377012e-08         0.764394\n",
       "9   GO:0043226        9.253129e-08  0.530498    2.926096e-08         0.530498\n",
       "10  GO:0005634        8.476652e-08  0.362152    2.680553e-08         0.362152\n",
       "11  GO:0005737        9.684608e-08  0.540246    3.062542e-08         0.540246\n",
       "12  GO:0009987        8.534086e-08  0.872470    2.698715e-08         0.872470\n",
       "13  GO:0005829        6.114925e-08  0.332176    1.933709e-08         0.332176\n",
       "14  GO:0008152        1.130919e-07  0.678671    3.576279e-08         0.678671\n",
       "15  GO:0071840        1.004475e-07  0.448246    3.176430e-08         0.448246\n",
       "16  GO:0016043        1.004475e-07  0.448246    3.176430e-08         0.448246\n",
       "17  GO:0071704        7.152557e-08  0.666036    2.261837e-08         0.666036\n",
       "18  GO:0043170        1.355596e-07  0.491758    4.286772e-08         0.491758\n",
       "19  GO:1901564        7.183534e-08  0.309629    2.271633e-08         0.309629\n",
       "20  GO:0006807        1.251698e-07  0.503289    3.958215e-08         0.503289\n",
       "21  GO:0044238        1.081130e-07  0.557016    3.418834e-08         0.557016\n",
       "22  GO:0044237        1.390207e-07  0.563157    4.396221e-08         0.563157\n",
       "23  GO:0003824        9.798576e-08  0.372867    3.098582e-08         0.372867"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfs['ACK1_YEAST'] "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e4d402df",
   "metadata": {},
   "source": [
    "## Graphics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "80e298a5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "protein_class_std = []\n",
    "for protein in dfs.values():\n",
    "    for value in protein['Standard deviation']:\n",
    "        protein_class_std.append(value)\n",
    "\n",
    "plt.figure(figsize=(20,15))\n",
    "_, _, bars = plt.hist(protein_class_std, bins=10)\n",
    "plt.bar_label(bars, fontsize=15), plt.title('Standard Deviation Across All Protein Classes', fontsize=20)\n",
    "plt.xlabel('Standard Deviation', fontsize=15), plt.ylabel('Protein Class Count', fontsize=15)\n",
    "plt.yscale('log'); "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "55821c0f",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}