[GO set]Count_genes_in_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## This Notebook calculates the number of genes mentioned in the GPT 4 analysis paragraph "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_LLM_gene_count(genes_fixed, LLM_analysis):\n",
    "    system_genes_set = set(genes_fixed.split(\" \"))\n",
    "    # Count the number of genes in the set that are mentioned in the LLM analysis\n",
    "    overlaps = sum(1 for system_gene in system_genes_set if system_gene in LLM_analysis) \n",
    "    return overlaps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "df = pd.read_csv('./data/GO_term_analysis/simrank_pval_LLM_processed_selected_1000_go_terms.tsv', sep='\\t')\n",
    "\n",
    "perc_gene_mentioned = []\n",
    "counts = []\n",
    "for ind, row in df.iterrows():\n",
    "    genes = row['Genes']\n",
    "    genecount = float(row['Gene_Count'])\n",
    "    assert genecount == len(genes.split(\" \")) # make sure the gene count is correct\n",
    "    \n",
    "    analysis = row['gpt_4_default Analysis']\n",
    "    overlap_counts = get_LLM_gene_count(genes, analysis)\n",
    "    counts.append(overlap_counts)\n",
    "    perc_gene_mentioned.append(overlap_counts/genecount)\n",
    "\n",
    "df['num_gene_mentioned'] = counts\n",
    "df['perc_gene_mentioned'] = perc_gene_mentioned\n",
    "\n",
    "df.to_csv('./data/GO_term_analysis/simrank_pval_LLM_processed_selected_1000_go_terms.tsv', sep='\\t', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    1000.000000\n",
       "mean        0.811926\n",
       "std         0.302239\n",
       "min         0.015487\n",
       "25%         0.638369\n",
       "50%         1.000000\n",
       "75%         1.000000\n",
       "max         1.000000\n",
       "Name: perc_gene_mentioned, dtype: float64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['perc_gene_mentioned'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of GO terms that cover at least 90% of the genes: 797\n",
      "Number of GO terms that cover at least 50% of the genes: 942\n"
     ]
    }
   ],
   "source": [
    "# find the number of times that covers at least 90% of the genes\n",
    "\n",
    "print(\"Number of GO terms that cover at least 90% of the genes: {}\".format(sum(1 for i in perc_gene_mentioned if i >= 0.9)))\n",
    "print(\"Number of GO terms that cover at least 50% of the genes: {}\".format(sum(1 for i in perc_gene_mentioned if i >= 0.5)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lower bound: 0.8652777777777778\n",
      "Number of outliers: 186\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "count    186.000000\n",
       "mean      55.854839\n",
       "std       20.657710\n",
       "min        6.000000\n",
       "25%       40.000000\n",
       "50%       53.000000\n",
       "75%       72.000000\n",
       "max      100.000000\n",
       "Name: Gene_Count, dtype: float64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check the size of the GO terms for outliers\n",
    "import numpy as np\n",
    "\n",
    "# Calculate quartiles and IQR\n",
    "q1 = np.percentile(df['perc_gene_mentioned'], 25)\n",
    "q3 = np.percentile(df['perc_gene_mentioned'], 75)\n",
    "iqr = q3 - q1\n",
    "\n",
    "# Calculate the upper and lower bounds for outliers\n",
    "lower_bound = q1 - 1.5 * iqr\n",
    "upper_bound = q3 + 1.5 * iqr\n",
    "\n",
    "# Extract the outliers from the DataFrame\n",
    "outliers = df[(df['perc_gene_mentioned'] < lower_bound) | (df['perc_gene_mentioned'] > upper_bound)]\n",
    "print('lower bound: {}'.format(lower_bound))\n",
    "print(\"Number of outliers: {}\".format(len(outliers)))\n",
    "# outliers size distribution\n",
    "outliers['Gene_Count'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 300x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## plot a box plot \n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "df = pd.read_csv('./data/GO_term_analysis/simrank_pval_LLM_processed_selected_1000_go_terms.tsv', sep='\\t')\n",
    "perc_gene_mentioned = df['perc_gene_mentioned'].tolist()\n",
    "plt.rcParams['svg.fonttype'] = 'none'\n",
    "plt.rcParams.update({'font.size': 7, 'font.family': 'sans-serif'})\n",
    "fig,ax = plt.subplots(figsize=(3,3))\n",
    "\n",
    "box_plot = sns.boxplot(y=[frac *100 for frac in perc_gene_mentioned], ax=ax, color='white', linewidth=1, fliersize=1.5)\n",
    "\n",
    "# changing color and linewidth of the median line\n",
    "median_line = box_plot.get_lines()[4] # get the median line\n",
    "median_line.set_color('black') # change the color\n",
    "median_line.set_linewidth(1.5) # change the linewidth\n",
    "\n",
    "ax.set_ylabel(\"Percentage of genes mentioned (%)\")\n",
    "ax.set_xlabel(\"GPT-4 analysis of GO terms\")\n",
    "\n",
    "plt.savefig('./figures/LLM_GO_gene_mention_percentage.svg', dpi = 300,bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "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</th>\n",
       "      <th>Genes</th>\n",
       "      <th>Gene_Count</th>\n",
       "      <th>Term_Description</th>\n",
       "      <th>gpt_4_default Name</th>\n",
       "      <th>gpt_4_default Analysis</th>\n",
       "      <th>gpt_4_default Score</th>\n",
       "      <th>gpt_4_default Score Bin</th>\n",
       "      <th>LLM_name_GO_term_sim</th>\n",
       "      <th>sim_rank</th>\n",
       "      <th>...</th>\n",
       "      <th>top_3_hits</th>\n",
       "      <th>top_3_sim</th>\n",
       "      <th>pvals</th>\n",
       "      <th>adj_pvals</th>\n",
       "      <th>random_pvals</th>\n",
       "      <th>random_adj_pvals</th>\n",
       "      <th>JI</th>\n",
       "      <th>random JI</th>\n",
       "      <th>num_gene_mentioned</th>\n",
       "      <th>perc_gene_mentioned</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>GO:0009081</td>\n",
       "      <td>HIBCH BCAT1 HMGCLL1 ACADSB BCKDHA SLC25A44 SDS...</td>\n",
       "      <td>27</td>\n",
       "      <td>branched-chain amino acid metabolic process</td>\n",
       "      <td>Branched-Chain Amino Acid Metabolism and Catab...</td>\n",
       "      <td>1. The proteins HIBCH, BCAT1, BCAT2, BCKDHA, B...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>High Confidence</td>\n",
       "      <td>0.946014</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>branched-chain amino acid metabolic process|br...</td>\n",
       "      <td>0.9460139|0.87900156|0.83314335</td>\n",
       "      <td>6.426397022410732e-87|2.154030905169728e-64|6....</td>\n",
       "      <td>1.691157111160719e-85|5.523156167101867e-63|3....</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0|0.7777777777777778|0.18518518518518517</td>\n",
       "      <td>0.0</td>\n",
       "      <td>26</td>\n",
       "      <td>0.962963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>GO:0048240</td>\n",
       "      <td>EFCAB9 ADAM7 TCP11 CATSPER4 SPINK1 BSPH1 ROPN1...</td>\n",
       "      <td>33</td>\n",
       "      <td>sperm capacitation</td>\n",
       "      <td>Sperm Capacitation and Fertilization</td>\n",
       "      <td>1. The majority of the proteins listed are rel...</td>\n",
       "      <td>0.92</td>\n",
       "      <td>High Confidence</td>\n",
       "      <td>0.936600</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>sperm capacitation|regulation of sperm capacit...</td>\n",
       "      <td>0.9366003|0.7948351|0.7883392</td>\n",
       "      <td>2.1280089332582335e-103|1.9935150264533243e-14...</td>\n",
       "      <td>7.60003190449369e-102|8.8554274588964e-14|1.01...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0|0.15151515151515152|0.06060606060606061</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20</td>\n",
       "      <td>0.606061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>GO:0006768</td>\n",
       "      <td>BTD MCCC1 HLCS SLC5A6</td>\n",
       "      <td>4</td>\n",
       "      <td>biotin metabolic process</td>\n",
       "      <td>Biotin Metabolism and Catabolism</td>\n",
       "      <td>1. BTD, or biotinidase, is an enzyme that recy...</td>\n",
       "      <td>0.89</td>\n",
       "      <td>High Confidence</td>\n",
       "      <td>0.933146</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>biotin metabolic process|biotin transport|coba...</td>\n",
       "      <td>0.93314576|0.76734674|0.7521958</td>\n",
       "      <td>2.85903800815798e-16|0.0007048043924525059|1.0</td>\n",
       "      <td>1.181420664528091e-15|0.0013149335680084065|1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0|0.16666666666666666|0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>GO:0090183</td>\n",
       "      <td>PAX8 WNT4 BASP1 GDNF SOX9 TACSTD2 GREM1 HNF1B ...</td>\n",
       "      <td>34</td>\n",
       "      <td>regulation of kidney development</td>\n",
       "      <td>Kidney Development and Function Regulation</td>\n",
       "      <td>1. Several proteins in this system, including ...</td>\n",
       "      <td>0.92</td>\n",
       "      <td>High Confidence</td>\n",
       "      <td>0.912659</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>regulation of kidney development|kidney develo...</td>\n",
       "      <td>0.91265875|0.8040259|0.7702517</td>\n",
       "      <td>4.2585228799752766e-106|1.991676329704935e-60|...</td>\n",
       "      <td>1.6378934153751065e-104|4.064645570826398e-59|...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0|0.10793650793650794|0.16279069767441862</td>\n",
       "      <td>0.0</td>\n",
       "      <td>33</td>\n",
       "      <td>0.970588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GO:0001732</td>\n",
       "      <td>EIF3C EIF3E EIF3M EIF3D EIF5 EIF3K EIF3F EIF3B...</td>\n",
       "      <td>15</td>\n",
       "      <td>formation of cytoplasmic translation initiatio...</td>\n",
       "      <td>Eukaryotic Translation Initiation Complex Asse...</td>\n",
       "      <td>1. The eukaryotic initiation factors (eIFs) ar...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>High Confidence</td>\n",
       "      <td>0.909025</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>formation of cytoplasmic translation initiatio...</td>\n",
       "      <td>0.9090252|0.77489007|0.7727462</td>\n",
       "      <td>4.504407445815161e-52|3.593179735087371e-33|6....</td>\n",
       "      <td>5.701781576981217e-51|3.0194787689809845e-32|1...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0|0.12096774193548387|0.39473684210526316</td>\n",
       "      <td>0.0</td>\n",
       "      <td>15</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>GO:2000065</td>\n",
       "      <td>REST BMP5 BMP2 DKK3</td>\n",
       "      <td>4</td>\n",
       "      <td>negative regulation of cortisol biosynthetic p...</td>\n",
       "      <td>Regulation of Neurogenesis and Bone Morphogenesis</td>\n",
       "      <td>1. REST, also known as RE1-Silencing Transcrip...</td>\n",
       "      <td>0.85</td>\n",
       "      <td>Medium Confidence</td>\n",
       "      <td>0.170712</td>\n",
       "      <td>11207</td>\n",
       "      <td>...</td>\n",
       "      <td>regulation of neurogenesis|regulation of bone ...</td>\n",
       "      <td>0.7791767|0.7555887|0.7188432</td>\n",
       "      <td>0.0029398113167312862|1.0|0.02375509025866538</td>\n",
       "      <td>0.004726384753587277|1.0|0.04152987807458983</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.005194805194805195|0.0|0.009523809523809525</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>GO:0001542</td>\n",
       "      <td>SIRT1 MMP2 FOXO3 NOS3 NRIP1 PGR ADAMTS1 AFP LEP</td>\n",
       "      <td>9</td>\n",
       "      <td>ovulation from ovarian follicle</td>\n",
       "      <td>Regulation of Metabolic and Cardiovascular Pro...</td>\n",
       "      <td>1. SIRT1, or sirtuin 1, is a NAD-dependent dea...</td>\n",
       "      <td>0.85</td>\n",
       "      <td>Medium Confidence</td>\n",
       "      <td>0.131500</td>\n",
       "      <td>11242</td>\n",
       "      <td>...</td>\n",
       "      <td>regulation of metabolic process|regulation of ...</td>\n",
       "      <td>0.72351325|0.7129016|0.6734365</td>\n",
       "      <td>0.07530245980152894|1.0|0.008954116541057816</td>\n",
       "      <td>0.10176008081287695|1.0|0.017055460078205364</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0009358914365933552|0.0|0.04</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>GO:1902966</td>\n",
       "      <td>RDX ROCK2 DTX3L EZR EGF MSN MGAT3 NF2 SORL1</td>\n",
       "      <td>9</td>\n",
       "      <td>positive regulation of protein localization to...</td>\n",
       "      <td>Actin Cytoskeleton Regulation and Signal Trans...</td>\n",
       "      <td>1. RDX (radixin) and EZR (ezrin) are members o...</td>\n",
       "      <td>0.85</td>\n",
       "      <td>Medium Confidence</td>\n",
       "      <td>0.149689</td>\n",
       "      <td>11325</td>\n",
       "      <td>...</td>\n",
       "      <td>actin cytoskeleton organization|regulation of ...</td>\n",
       "      <td>0.78401446|0.75558466|0.73912597</td>\n",
       "      <td>0.00031448591713256275|2.5111480956844267e-05|...</td>\n",
       "      <td>0.0005566122427125004|5.543373279656571e-05|4....</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0056179775280898875|0.0106951871657754|0.011...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>GO:0045599</td>\n",
       "      <td>ZFPM2 WNT3A LRP3 SIRT1 YAP1 CCDC85B TGFB1 DDIT...</td>\n",
       "      <td>52</td>\n",
       "      <td>negative regulation of fat cell differentiation</td>\n",
       "      <td>Regulation of Wnt Signaling and Cellular Stres...</td>\n",
       "      <td>1. The Wnt signaling pathway is represented by...</td>\n",
       "      <td>0.85</td>\n",
       "      <td>Medium Confidence</td>\n",
       "      <td>0.125487</td>\n",
       "      <td>11554</td>\n",
       "      <td>...</td>\n",
       "      <td>regulation of cellular response to stress|regu...</td>\n",
       "      <td>0.7262345|0.7084471|0.6966753</td>\n",
       "      <td>0.00013166826369465795|1.0206922127068983e-11|...</td>\n",
       "      <td>0.0002424038513729637|3.837188769574805e-11|0....</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.013062409288824383|0.036011080332409975|0.01...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>18</td>\n",
       "      <td>0.346154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>GO:1904339</td>\n",
       "      <td>WNT3A GSK3B SHH</td>\n",
       "      <td>3</td>\n",
       "      <td>negative regulation of dopaminergic neuron dif...</td>\n",
       "      <td>Canonical Wnt and Hedgehog Signaling Pathway I...</td>\n",
       "      <td>1. WNT3A is a member of the WNT protein family...</td>\n",
       "      <td>0.92</td>\n",
       "      <td>High Confidence</td>\n",
       "      <td>0.099194</td>\n",
       "      <td>11662</td>\n",
       "      <td>...</td>\n",
       "      <td>canonical Wnt signaling pathway|smoothened sig...</td>\n",
       "      <td>0.67585087|0.64332056|0.6393231</td>\n",
       "      <td>5.5845001895334345e-06|1.0|1.7507850439351638e-05</td>\n",
       "      <td>1.178164596947982e-05|1.0|4.489192420346574e-05</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.009900990099009901|0.0|0.006772009029345372</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           GO                                              Genes  Gene_Count  \\\n",
       "0  GO:0009081  HIBCH BCAT1 HMGCLL1 ACADSB BCKDHA SLC25A44 SDS...          27   \n",
       "1  GO:0048240  EFCAB9 ADAM7 TCP11 CATSPER4 SPINK1 BSPH1 ROPN1...          33   \n",
       "2  GO:0006768                              BTD MCCC1 HLCS SLC5A6           4   \n",
       "3  GO:0090183  PAX8 WNT4 BASP1 GDNF SOX9 TACSTD2 GREM1 HNF1B ...          34   \n",
       "4  GO:0001732  EIF3C EIF3E EIF3M EIF3D EIF5 EIF3K EIF3F EIF3B...          15   \n",
       "5  GO:2000065                                REST BMP5 BMP2 DKK3           4   \n",
       "6  GO:0001542    SIRT1 MMP2 FOXO3 NOS3 NRIP1 PGR ADAMTS1 AFP LEP           9   \n",
       "7  GO:1902966        RDX ROCK2 DTX3L EZR EGF MSN MGAT3 NF2 SORL1           9   \n",
       "8  GO:0045599  ZFPM2 WNT3A LRP3 SIRT1 YAP1 CCDC85B TGFB1 DDIT...          52   \n",
       "9  GO:1904339                                    WNT3A GSK3B SHH           3   \n",
       "\n",
       "                                    Term_Description  \\\n",
       "0        branched-chain amino acid metabolic process   \n",
       "1                                 sperm capacitation   \n",
       "2                           biotin metabolic process   \n",
       "3                   regulation of kidney development   \n",
       "4  formation of cytoplasmic translation initiatio...   \n",
       "5  negative regulation of cortisol biosynthetic p...   \n",
       "6                    ovulation from ovarian follicle   \n",
       "7  positive regulation of protein localization to...   \n",
       "8    negative regulation of fat cell differentiation   \n",
       "9  negative regulation of dopaminergic neuron dif...   \n",
       "\n",
       "                                  gpt_4_default Name  \\\n",
       "0  Branched-Chain Amino Acid Metabolism and Catab...   \n",
       "1               Sperm Capacitation and Fertilization   \n",
       "2                   Biotin Metabolism and Catabolism   \n",
       "3         Kidney Development and Function Regulation   \n",
       "4  Eukaryotic Translation Initiation Complex Asse...   \n",
       "5  Regulation of Neurogenesis and Bone Morphogenesis   \n",
       "6  Regulation of Metabolic and Cardiovascular Pro...   \n",
       "7  Actin Cytoskeleton Regulation and Signal Trans...   \n",
       "8  Regulation of Wnt Signaling and Cellular Stres...   \n",
       "9  Canonical Wnt and Hedgehog Signaling Pathway I...   \n",
       "\n",
       "                              gpt_4_default Analysis  gpt_4_default Score  \\\n",
       "0  1. The proteins HIBCH, BCAT1, BCAT2, BCKDHA, B...                 0.95   \n",
       "1  1. The majority of the proteins listed are rel...                 0.92   \n",
       "2  1. BTD, or biotinidase, is an enzyme that recy...                 0.89   \n",
       "3  1. Several proteins in this system, including ...                 0.92   \n",
       "4  1. The eukaryotic initiation factors (eIFs) ar...                 0.98   \n",
       "5  1. REST, also known as RE1-Silencing Transcrip...                 0.85   \n",
       "6  1. SIRT1, or sirtuin 1, is a NAD-dependent dea...                 0.85   \n",
       "7  1. RDX (radixin) and EZR (ezrin) are members o...                 0.85   \n",
       "8  1. The Wnt signaling pathway is represented by...                 0.85   \n",
       "9  1. WNT3A is a member of the WNT protein family...                 0.92   \n",
       "\n",
       "  gpt_4_default Score Bin  LLM_name_GO_term_sim  sim_rank  ...  \\\n",
       "0         High Confidence              0.946014         1  ...   \n",
       "1         High Confidence              0.936600         1  ...   \n",
       "2         High Confidence              0.933146         1  ...   \n",
       "3         High Confidence              0.912659         1  ...   \n",
       "4         High Confidence              0.909025         1  ...   \n",
       "5       Medium Confidence              0.170712     11207  ...   \n",
       "6       Medium Confidence              0.131500     11242  ...   \n",
       "7       Medium Confidence              0.149689     11325  ...   \n",
       "8       Medium Confidence              0.125487     11554  ...   \n",
       "9         High Confidence              0.099194     11662  ...   \n",
       "\n",
       "                                          top_3_hits  \\\n",
       "0  branched-chain amino acid metabolic process|br...   \n",
       "1  sperm capacitation|regulation of sperm capacit...   \n",
       "2  biotin metabolic process|biotin transport|coba...   \n",
       "3  regulation of kidney development|kidney develo...   \n",
       "4  formation of cytoplasmic translation initiatio...   \n",
       "5  regulation of neurogenesis|regulation of bone ...   \n",
       "6  regulation of metabolic process|regulation of ...   \n",
       "7  actin cytoskeleton organization|regulation of ...   \n",
       "8  regulation of cellular response to stress|regu...   \n",
       "9  canonical Wnt signaling pathway|smoothened sig...   \n",
       "\n",
       "                          top_3_sim  \\\n",
       "0   0.9460139|0.87900156|0.83314335   \n",
       "1     0.9366003|0.7948351|0.7883392   \n",
       "2   0.93314576|0.76734674|0.7521958   \n",
       "3    0.91265875|0.8040259|0.7702517   \n",
       "4    0.9090252|0.77489007|0.7727462   \n",
       "5     0.7791767|0.7555887|0.7188432   \n",
       "6    0.72351325|0.7129016|0.6734365   \n",
       "7  0.78401446|0.75558466|0.73912597   \n",
       "8     0.7262345|0.7084471|0.6966753   \n",
       "9   0.67585087|0.64332056|0.6393231   \n",
       "\n",
       "                                               pvals  \\\n",
       "0  6.426397022410732e-87|2.154030905169728e-64|6....   \n",
       "1  2.1280089332582335e-103|1.9935150264533243e-14...   \n",
       "2     2.85903800815798e-16|0.0007048043924525059|1.0   \n",
       "3  4.2585228799752766e-106|1.991676329704935e-60|...   \n",
       "4  4.504407445815161e-52|3.593179735087371e-33|6....   \n",
       "5      0.0029398113167312862|1.0|0.02375509025866538   \n",
       "6       0.07530245980152894|1.0|0.008954116541057816   \n",
       "7  0.00031448591713256275|2.5111480956844267e-05|...   \n",
       "8  0.00013166826369465795|1.0206922127068983e-11|...   \n",
       "9  5.5845001895334345e-06|1.0|1.7507850439351638e-05   \n",
       "\n",
       "                                           adj_pvals  random_pvals  \\\n",
       "0  1.691157111160719e-85|5.523156167101867e-63|3....           1.0   \n",
       "1  7.60003190449369e-102|8.8554274588964e-14|1.01...           1.0   \n",
       "2    1.181420664528091e-15|0.0013149335680084065|1.0           1.0   \n",
       "3  1.6378934153751065e-104|4.064645570826398e-59|...           1.0   \n",
       "4  5.701781576981217e-51|3.0194787689809845e-32|1...           1.0   \n",
       "5       0.004726384753587277|1.0|0.04152987807458983           1.0   \n",
       "6       0.10176008081287695|1.0|0.017055460078205364           1.0   \n",
       "7  0.0005566122427125004|5.543373279656571e-05|4....           1.0   \n",
       "8  0.0002424038513729637|3.837188769574805e-11|0....           1.0   \n",
       "9    1.178164596947982e-05|1.0|4.489192420346574e-05           1.0   \n",
       "\n",
       "  random_adj_pvals                                                 JI  \\\n",
       "0              1.0         1.0|0.7777777777777778|0.18518518518518517   \n",
       "1              1.0        1.0|0.15151515151515152|0.06060606060606061   \n",
       "2              1.0                        1.0|0.16666666666666666|0.0   \n",
       "3              1.0        1.0|0.10793650793650794|0.16279069767441862   \n",
       "4              1.0        1.0|0.12096774193548387|0.39473684210526316   \n",
       "5              1.0      0.005194805194805195|0.0|0.009523809523809525   \n",
       "6              1.0                     0.0009358914365933552|0.0|0.04   \n",
       "7              1.0  0.0056179775280898875|0.0106951871657754|0.011...   \n",
       "8              1.0  0.013062409288824383|0.036011080332409975|0.01...   \n",
       "9              1.0      0.009900990099009901|0.0|0.006772009029345372   \n",
       "\n",
       "  random JI num_gene_mentioned  perc_gene_mentioned  \n",
       "0       0.0                 26             0.962963  \n",
       "1       0.0                 20             0.606061  \n",
       "2       0.0                  4             1.000000  \n",
       "3       0.0                 33             0.970588  \n",
       "4       0.0                 15             1.000000  \n",
       "5       0.0                  4             1.000000  \n",
       "6       0.0                  9             1.000000  \n",
       "7       0.0                  9             1.000000  \n",
       "8       0.0                 18             0.346154  \n",
       "9       0.0                  3             1.000000  \n",
       "\n",
       "[10 rows x 25 columns]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## collect top 10 and bottom 10 \n",
    "import pandas as pd \n",
    "df = pd.read_csv('./data/GO_term_analysis/simrank_pval_LLM_processed_selected_1000_go_terms.tsv', sep='\\t')\n",
    "# not include the 'System of unrelated proteins'\n",
    "df.drop(df[df['gpt_4_default Name'] == 'System of unrelated proteins'].index, inplace = True)\n",
    "df = df.sort_values(by=['true_GO_term_sim_percentile', 'LLM_name_GO_term_sim'], ascending=False)\n",
    "top = df.head(5)\n",
    "bottom = df.tail(5)\n",
    "combine_df = pd.concat([top,bottom], ignore_index=True)\n",
    "combine_df.to_csv('./data/GO_term_analysis/LLM_processed_GO_representative_top_bottom_5.tsv', sep='\\t', index=False)\n",
    "combine_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}