2B.[Omics_revampled]Calculate_LLM_JI.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "11943\n"
     ]
    }
   ],
   "source": [
    "## create embeddings for all GO Terms and save the embeddings \n",
    "from semanticSimFunctions import getSentenceEmbedding\n",
    "from transformers import AutoTokenizer, AutoModel\n",
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "import pandas as pd\n",
    "from tqdm import tqdm\n",
    "\n",
    "SapBERT_tokenizer = AutoTokenizer.from_pretrained('cambridgeltl/SapBERT-from-PubMedBERT-fulltext')\n",
    "SapBERT_model = AutoModel.from_pretrained('cambridgeltl/SapBERT-from-PubMedBERT-fulltext')\n",
    "\n",
    "import pickle\n",
    "# get the embeddings for the go terms we made in TASK1\n",
    "with open('data/all_go_terms_embeddings_dict.pkl', 'rb') as handle:\n",
    "    all_go_terms_embeddings_dict = pickle.load(handle)\n",
    "print(len(all_go_terms_embeddings_dict))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "           GO                                              Genes  Gene_Count  \\\n",
      "0  GO:0006311                                 TEX11 PRDM9 RNF212           3   \n",
      "1  GO:0006855  ABCB11 ABCC4 ABCC3 SLC29A2 SLC37A3 SLC19A1 SLC...          16   \n",
      "2  GO:0006890  GOLPH3L ARCN1 GBF1 KDELR1 BET1L PICK1 COPZ2 SC...          52   \n",
      "3  GO:1900372                                PARP1 FLCN RD3 PID1           4   \n",
      "4  GO:0031645  LPIN1 HCRT GPR35 EIF2AK3 GLRA1 MTNR1B CCN3 MTM...          16   \n",
      "\n",
      "                                    Term_Description  \n",
      "0                            meiotic gene conversion  \n",
      "1                 xenobiotic transmembrane transport  \n",
      "2  retrograde vesicle-mediated transport, Golgi t...  \n",
      "3  negative regulation of purine nucleotide biosy...  \n",
      "4      negative regulation of nervous system process  \n"
     ]
    }
   ],
   "source": [
    "# load our GO term table \n",
    "go_terms = pd.read_csv('data/go_terms.csv', index_col=0)\n",
    "print(go_terms.head())"
   ]
  },
  {
   "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>Source</th>\n",
       "      <th>GeneSetID</th>\n",
       "      <th>GeneSetName</th>\n",
       "      <th>GeneList</th>\n",
       "      <th>n_Genes</th>\n",
       "      <th>Supporting Genes</th>\n",
       "      <th>Supporting Count</th>\n",
       "      <th>LLM Support Analysis</th>\n",
       "      <th>GenesMentionedInText</th>\n",
       "      <th>LLM_coverage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>L1000</td>\n",
       "      <td>BRD-A00546892_-666_MCF7_6.0_h_10.0_um</td>\n",
       "      <td>BRD-A00546892 -666 MCF7 6.0 h 10.0 um</td>\n",
       "      <td>CITED2 COL5A1 CRABP2 KCTD12 MDFIC MMP2 NRP1 OR...</td>\n",
       "      <td>24</td>\n",
       "      <td>CITED2 TWIST1 COL5A1 MMP2 NRP1</td>\n",
       "      <td>5</td>\n",
       "      <td>-- Summary --||The selected genes support the ...</td>\n",
       "      <td>24</td>\n",
       "      <td>0.21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>L1000</td>\n",
       "      <td>BRD-A00993607_ALPRENOLOL_MCF7_6.0_h_10.0_um</td>\n",
       "      <td>BRD-A00993607 ALPRENOLOL MCF7 6.0 h 10.0 um</td>\n",
       "      <td>1060P11.3 ADM AHR AMIGO2 ARL4C ATP10D CAV2 CD4...</td>\n",
       "      <td>47</td>\n",
       "      <td>CD44 EMP1 AMIGO2 POSTN COMP EXT1</td>\n",
       "      <td>6</td>\n",
       "      <td>-- Summary --||The selected genes support the ...</td>\n",
       "      <td>15</td>\n",
       "      <td>0.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>L1000</td>\n",
       "      <td>BRD-A01593789_CHLORMADINONE ACETATE_MCF7_6.0_h...</td>\n",
       "      <td>BRD-A01593789 CHLORMADINONE ACETATE MCF7 6.0 h...</td>\n",
       "      <td>1060P11.3 ADH5 ADM ATF3 CASP1 CLEC2B CPA3 CRIM...</td>\n",
       "      <td>38</td>\n",
       "      <td>ATF3 CASP1 DKK3 PIK3R1 TNC</td>\n",
       "      <td>5</td>\n",
       "      <td>-- Summary --||The selected genes support the ...</td>\n",
       "      <td>10</td>\n",
       "      <td>0.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>L1000</td>\n",
       "      <td>BRD-A02006392_NITRENDIPINE_MCF7_6.0_h_10.0_um</td>\n",
       "      <td>BRD-A02006392 NITRENDIPINE MCF7 6.0 h 10.0 um</td>\n",
       "      <td>1060P11.3 ADAM17 AKAP12 ALCAM AMFR BIK CASP2 C...</td>\n",
       "      <td>76</td>\n",
       "      <td>TP53 CASP2 BIK HSPA1A GPX3 CDKN2A TLR4</td>\n",
       "      <td>7</td>\n",
       "      <td>-- Summary --||The selected genes support the ...</td>\n",
       "      <td>11</td>\n",
       "      <td>0.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>L1000</td>\n",
       "      <td>BRD-A02176148_TUBAIC ACID_MCF7_6.0_h_10.0_um</td>\n",
       "      <td>BRD-A02176148 TUBAIC ACID MCF7 6.0 h 10.0 um</td>\n",
       "      <td>CASP1 CAST CD52 CLC COL1A2 COL3A1 COL5A2 CPA3 ...</td>\n",
       "      <td>34</td>\n",
       "      <td>COL1A2 COL3A1 COL5A2 FN1 LUM SPON1</td>\n",
       "      <td>6</td>\n",
       "      <td>-- Summary --||The selected genes support the ...</td>\n",
       "      <td>15</td>\n",
       "      <td>0.18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Source                                          GeneSetID  \\\n",
       "0  L1000              BRD-A00546892_-666_MCF7_6.0_h_10.0_um   \n",
       "1  L1000        BRD-A00993607_ALPRENOLOL_MCF7_6.0_h_10.0_um   \n",
       "2  L1000  BRD-A01593789_CHLORMADINONE ACETATE_MCF7_6.0_h...   \n",
       "3  L1000      BRD-A02006392_NITRENDIPINE_MCF7_6.0_h_10.0_um   \n",
       "4  L1000       BRD-A02176148_TUBAIC ACID_MCF7_6.0_h_10.0_um   \n",
       "\n",
       "                                         GeneSetName  \\\n",
       "0              BRD-A00546892 -666 MCF7 6.0 h 10.0 um   \n",
       "1        BRD-A00993607 ALPRENOLOL MCF7 6.0 h 10.0 um   \n",
       "2  BRD-A01593789 CHLORMADINONE ACETATE MCF7 6.0 h...   \n",
       "3      BRD-A02006392 NITRENDIPINE MCF7 6.0 h 10.0 um   \n",
       "4       BRD-A02176148 TUBAIC ACID MCF7 6.0 h 10.0 um   \n",
       "\n",
       "                                            GeneList  n_Genes  \\\n",
       "0  CITED2 COL5A1 CRABP2 KCTD12 MDFIC MMP2 NRP1 OR...       24   \n",
       "1  1060P11.3 ADM AHR AMIGO2 ARL4C ATP10D CAV2 CD4...       47   \n",
       "2  1060P11.3 ADH5 ADM ATF3 CASP1 CLEC2B CPA3 CRIM...       38   \n",
       "3  1060P11.3 ADAM17 AKAP12 ALCAM AMFR BIK CASP2 C...       76   \n",
       "4  CASP1 CAST CD52 CLC COL1A2 COL3A1 COL5A2 CPA3 ...       34   \n",
       "\n",
       "                         Supporting Genes  Supporting Count  \\\n",
       "0          CITED2 TWIST1 COL5A1 MMP2 NRP1                 5   \n",
       "1        CD44 EMP1 AMIGO2 POSTN COMP EXT1                 6   \n",
       "2              ATF3 CASP1 DKK3 PIK3R1 TNC                 5   \n",
       "3  TP53 CASP2 BIK HSPA1A GPX3 CDKN2A TLR4                 7   \n",
       "4      COL1A2 COL3A1 COL5A2 FN1 LUM SPON1                 6   \n",
       "\n",
       "                                LLM Support Analysis  GenesMentionedInText  \\\n",
       "0  -- Summary --||The selected genes support the ...                    24   \n",
       "1  -- Summary --||The selected genes support the ...                    15   \n",
       "2  -- Summary --||The selected genes support the ...                    10   \n",
       "3  -- Summary --||The selected genes support the ...                    11   \n",
       "4  -- Summary --||The selected genes support the ...                    15   \n",
       "\n",
       "   LLM_coverage  \n",
       "0          0.21  \n",
       "1          0.13  \n",
       "2          0.13  \n",
       "3          0.09  \n",
       "4          0.18  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inputFile = \"data/omics_revamped_LLM_genecounts_DF.tsv\"\n",
    "data = pd.read_csv(inputFile, sep='\\t')\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# def calc_LLM_JI(supporting_count, query_gene_list, GO_gene_list):\n",
    "    \n",
    "#     union = len(set(query_gene_list).union(set(GO_gene_list)))\n",
    "#     return supporting_count/union if union != 0 else 0\n",
    "def calc_LLM_JI(supporting_count, query_size, GO_size):\n",
    "    # if the best matching GO term size is smaller than the supporting count, then make the size be the same as the supporting count\n",
    "    if GO_size < supporting_count:\n",
    "         GO_size = supporting_count\n",
    "    union = query_size + GO_size - supporting_count\n",
    "    return supporting_count/union if union != 0 else 0\n",
    "\n",
    "\n",
    "def get_closest_GO(LLM_name, all_go_terms_embeddings_dict):\n",
    "        # get llm name embedding\n",
    "        LLM_name_emb = getSentenceEmbedding(LLM_name, SapBERT_tokenizer, SapBERT_model)\n",
    "        # get the cosine similarity between the LLM name and all GO terms\n",
    "        LLM_GO_sim = {}\n",
    "        for term, emb in all_go_terms_embeddings_dict.items():\n",
    "            GO_emb = emb\n",
    "            GO_llm_sim = cosine_similarity(LLM_name_emb, GO_emb)[0][0]\n",
    "            LLM_GO_sim[term] = GO_llm_sim\n",
    "        # get the best matching GO term\n",
    "        best_matching_GO = max(LLM_GO_sim, key=LLM_GO_sim.get)\n",
    "        return best_matching_GO"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/300 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 300/300 [12:54<00:00,  2.58s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DONE\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "col_keep = ['Source','GeneSetID','GeneList','n_Genes', 'LLM Name', 'LLM_best_matching_GO','best_matching_GO_ID', 'best_matching_GO_term_genes', 'best_matching_GO_term_size', 'Supporting Count', 'LLM_JI']\n",
    "\n",
    "# initialize the data frame \n",
    "df = data.copy()\n",
    "df['LLM_best_matching_GO'] = None\n",
    "df['best_matching_GO_ID'] = None\n",
    "df['best_matching_GO_term_genes'] = None\n",
    "df['best_matching_GO_term_size'] = None\n",
    "df['LLM_JI'] = None\n",
    "\n",
    "for ind, row in tqdm(df.iterrows(), total=df.shape[0]):\n",
    "    # if not pd.isna(row['LLM_best_matching_GO']):\n",
    "    #     # no need to calculate the embeddings again\n",
    "    #     continue\n",
    "    LLM_name = row['LLM Name'] \n",
    "    if LLM_name == 'System of unrelated proteins':\n",
    "        # dont find best matching GO term for the one does not named\n",
    "        continue\n",
    "    else:\n",
    "        best_matching_GO = get_closest_GO(LLM_name, all_go_terms_embeddings_dict)\n",
    "        df.loc[ind, 'LLM_best_matching_GO'] = best_matching_GO\n",
    "        df.loc[ind, 'best_matching_GO_ID'] = go_terms.loc[go_terms['Term_Description'] == best_matching_GO, 'GO'].values[0]\n",
    "        # get GO gene list \n",
    "        gene_list = go_terms.loc[go_terms['Term_Description'] == best_matching_GO, 'Genes'].values[0].split(' ')\n",
    "        df.loc[ind, 'best_matching_GO_term_genes'] = ' '.join(gene_list)\n",
    "        # get the GO term size \n",
    "        term_size = go_terms.loc[go_terms['Term_Description'] == best_matching_GO, 'Gene_Count'].values[0]\n",
    "        assert len(gene_list) == term_size, f'{best_matching_GO}: GO gene list and term size does not match'\n",
    "        df.loc[ind, 'best_matching_GO_term_size'] = term_size\n",
    "        # # get intersections \n",
    "       \n",
    "        # intersection = list(set(gene_list).intersection(LLM_genes))\n",
    "        # df.loc[ind, 'best_matching_GO_term_intersection'] = ' '.join(intersection)\n",
    "\n",
    "        # calculate JI \n",
    "        LLM_genes = row['updated GeneList'].split(' ')\n",
    "        supporting_count = row['Supporting Count']\n",
    "        JI = calc_LLM_JI(supporting_count,LLM_genes, gene_list)\n",
    "        df.loc[ind, 'LLM_JI'] = JI\n",
    "\n",
    "    df.loc[:, col_keep].to_csv('data/omics_revamped_LLM_w_best_matching_GO_terms_for_JI.tsv', sep='\\t', index=False)\n",
    "\n",
    "print('DONE')\n",
    "df.loc[:, col_keep].to_csv('data/omics_revamped_LLM_w_best_matching_GO_terms_for_JI.tsv', sep='\\t', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Source</th>\n",
       "      <th>GeneSetID</th>\n",
       "      <th>GeneList</th>\n",
       "      <th>n_Genes</th>\n",
       "      <th>LLM Name</th>\n",
       "      <th>LLM_best_matching_GO</th>\n",
       "      <th>best_matching_GO_ID</th>\n",
       "      <th>best_matching_GO_term_genes</th>\n",
       "      <th>best_matching_GO_term_size</th>\n",
       "      <th>Supporting Count</th>\n",
       "      <th>LLM_JI</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>L1000</td>\n",
       "      <td>BRD-A00546892_-666_MCF7_6.0_h_10.0_um</td>\n",
       "      <td>CITED2 COL5A1 CRABP2 KCTD12 MDFIC MMP2 NRP1 OR...</td>\n",
       "      <td>24</td>\n",
       "      <td>Cellular Matrix Remodeling and Tissue Development</td>\n",
       "      <td>connective tissue development</td>\n",
       "      <td>GO:0061448</td>\n",
       "      <td>CRIP1 GDF5 IL17F MAF NOTCH1 AMELX VPS13B NFATC...</td>\n",
       "      <td>286.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.016181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>L1000</td>\n",
       "      <td>BRD-A00993607_ALPRENOLOL_MCF7_6.0_h_10.0_um</td>\n",
       "      <td>1060P11.3 ADM AHR AMIGO2 ARL4C ATP10D CAV2 CD4...</td>\n",
       "      <td>47</td>\n",
       "      <td>Cellular Adhesion and Extracellular Matrix Int...</td>\n",
       "      <td>cell-matrix adhesion</td>\n",
       "      <td>GO:0007160</td>\n",
       "      <td>CX3CL1 BCR ARHGEF7 PLET1 ITGA7 TNXB ADAMTS13 N...</td>\n",
       "      <td>235.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.021429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>L1000</td>\n",
       "      <td>BRD-A00993607_Alprenolol hydrochloride_MCF7_6....</td>\n",
       "      <td>ABAT ASS1 CHI3L1 CHST2 CLDN3 EIF5B FRZB GAL HE...</td>\n",
       "      <td>59</td>\n",
       "      <td>System of unrelated proteins</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>L1000</td>\n",
       "      <td>BRD-A01320529_Salmeterol_MCF7_6.0_h_10.0_um</td>\n",
       "      <td>AMIGO2 AREG GAS6 GPR37 IFT57 PELI1 SQLE AKAP12...</td>\n",
       "      <td>9</td>\n",
       "      <td>System of unrelated proteins</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>L1000</td>\n",
       "      <td>BRD-A01346607_FLUMETHASONE_MCF7_6.0_h_10.0_um</td>\n",
       "      <td>1060P11.3 CPE EFNB2 HIST1H2AC IL1R2 INHBB LYPD...</td>\n",
       "      <td>15</td>\n",
       "      <td>System of unrelated proteins</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>295</th>\n",
       "      <td>Viral_Infections</td>\n",
       "      <td>hMPV_72Hour</td>\n",
       "      <td>RNMT SAMD9L CEBPB CEBPG PSMD12 RSRC2 XBP1 GADD...</td>\n",
       "      <td>22</td>\n",
       "      <td>Cellular Stress Response and Protein Homeostasis</td>\n",
       "      <td>cellular response to stress</td>\n",
       "      <td>GO:0033554</td>\n",
       "      <td>COMT ALKBH8 SESN3 STRADB ATP5IF1 MID1 CASP9 ST...</td>\n",
       "      <td>1920.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.003101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>296</th>\n",
       "      <td>Viral_Infections</td>\n",
       "      <td>icSARA deltaORF6_48Hour</td>\n",
       "      <td>FGF19 ADM2 LRIT1 UCHL1 SLC19A1 TSSC1 INPP5J HY...</td>\n",
       "      <td>49</td>\n",
       "      <td>System of unrelated proteins</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>297</th>\n",
       "      <td>Viral_Infections</td>\n",
       "      <td>icSARS CoV_12Hour...321</td>\n",
       "      <td>TSPY3 LGALS14 PPBPP2 LARS MRPS10 PAX9 RBM22 LO...</td>\n",
       "      <td>16</td>\n",
       "      <td>System of unrelated proteins</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>298</th>\n",
       "      <td>Viral_Infections</td>\n",
       "      <td>icSARS CoV_3Hour</td>\n",
       "      <td>ANK2 GAB3 TGFB3 CECR3 MUC19 LOC100507053 ZNF36...</td>\n",
       "      <td>99</td>\n",
       "      <td>System of unrelated proteins</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>299</th>\n",
       "      <td>Viral_Infections</td>\n",
       "      <td>icSARS CoV_72Hour...15</td>\n",
       "      <td>KRT75 ZDBF2 HIF3A TFAP2D F12 CASP14 FAM151B NG...</td>\n",
       "      <td>72</td>\n",
       "      <td>System of unrelated proteins</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>300 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               Source                                          GeneSetID  \\\n",
       "0               L1000              BRD-A00546892_-666_MCF7_6.0_h_10.0_um   \n",
       "1               L1000        BRD-A00993607_ALPRENOLOL_MCF7_6.0_h_10.0_um   \n",
       "2               L1000  BRD-A00993607_Alprenolol hydrochloride_MCF7_6....   \n",
       "3               L1000        BRD-A01320529_Salmeterol_MCF7_6.0_h_10.0_um   \n",
       "4               L1000      BRD-A01346607_FLUMETHASONE_MCF7_6.0_h_10.0_um   \n",
       "..                ...                                                ...   \n",
       "295  Viral_Infections                                        hMPV_72Hour   \n",
       "296  Viral_Infections                            icSARA deltaORF6_48Hour   \n",
       "297  Viral_Infections                            icSARS CoV_12Hour...321   \n",
       "298  Viral_Infections                                   icSARS CoV_3Hour   \n",
       "299  Viral_Infections                             icSARS CoV_72Hour...15   \n",
       "\n",
       "                                              GeneList  n_Genes  \\\n",
       "0    CITED2 COL5A1 CRABP2 KCTD12 MDFIC MMP2 NRP1 OR...       24   \n",
       "1    1060P11.3 ADM AHR AMIGO2 ARL4C ATP10D CAV2 CD4...       47   \n",
       "2    ABAT ASS1 CHI3L1 CHST2 CLDN3 EIF5B FRZB GAL HE...       59   \n",
       "3    AMIGO2 AREG GAS6 GPR37 IFT57 PELI1 SQLE AKAP12...        9   \n",
       "4    1060P11.3 CPE EFNB2 HIST1H2AC IL1R2 INHBB LYPD...       15   \n",
       "..                                                 ...      ...   \n",
       "295  RNMT SAMD9L CEBPB CEBPG PSMD12 RSRC2 XBP1 GADD...       22   \n",
       "296  FGF19 ADM2 LRIT1 UCHL1 SLC19A1 TSSC1 INPP5J HY...       49   \n",
       "297  TSPY3 LGALS14 PPBPP2 LARS MRPS10 PAX9 RBM22 LO...       16   \n",
       "298  ANK2 GAB3 TGFB3 CECR3 MUC19 LOC100507053 ZNF36...       99   \n",
       "299  KRT75 ZDBF2 HIF3A TFAP2D F12 CASP14 FAM151B NG...       72   \n",
       "\n",
       "                                              LLM Name  \\\n",
       "0    Cellular Matrix Remodeling and Tissue Development   \n",
       "1    Cellular Adhesion and Extracellular Matrix Int...   \n",
       "2                         System of unrelated proteins   \n",
       "3                         System of unrelated proteins   \n",
       "4                         System of unrelated proteins   \n",
       "..                                                 ...   \n",
       "295   Cellular Stress Response and Protein Homeostasis   \n",
       "296                       System of unrelated proteins   \n",
       "297                       System of unrelated proteins   \n",
       "298                       System of unrelated proteins   \n",
       "299                       System of unrelated proteins   \n",
       "\n",
       "              LLM_best_matching_GO best_matching_GO_ID  \\\n",
       "0    connective tissue development          GO:0061448   \n",
       "1             cell-matrix adhesion          GO:0007160   \n",
       "2                              NaN                 NaN   \n",
       "3                              NaN                 NaN   \n",
       "4                              NaN                 NaN   \n",
       "..                             ...                 ...   \n",
       "295    cellular response to stress          GO:0033554   \n",
       "296                            NaN                 NaN   \n",
       "297                            NaN                 NaN   \n",
       "298                            NaN                 NaN   \n",
       "299                            NaN                 NaN   \n",
       "\n",
       "                           best_matching_GO_term_genes  \\\n",
       "0    CRIP1 GDF5 IL17F MAF NOTCH1 AMELX VPS13B NFATC...   \n",
       "1    CX3CL1 BCR ARHGEF7 PLET1 ITGA7 TNXB ADAMTS13 N...   \n",
       "2                                                  NaN   \n",
       "3                                                  NaN   \n",
       "4                                                  NaN   \n",
       "..                                                 ...   \n",
       "295  COMT ALKBH8 SESN3 STRADB ATP5IF1 MID1 CASP9 ST...   \n",
       "296                                                NaN   \n",
       "297                                                NaN   \n",
       "298                                                NaN   \n",
       "299                                                NaN   \n",
       "\n",
       "     best_matching_GO_term_size  Supporting Count    LLM_JI  \n",
       "0                         286.0               5.0  0.016181  \n",
       "1                         235.0               6.0  0.021429  \n",
       "2                           NaN               NaN       NaN  \n",
       "3                           NaN               NaN       NaN  \n",
       "4                           NaN               NaN       NaN  \n",
       "..                          ...               ...       ...  \n",
       "295                      1920.0               6.0  0.003101  \n",
       "296                         NaN               NaN       NaN  \n",
       "297                         NaN               NaN       NaN  \n",
       "298                         NaN               NaN       NaN  \n",
       "299                         NaN               NaN       NaN  \n",
       "\n",
       "[300 rows x 11 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "llm_JI_df = pd.read_csv('data/omics_revamped_LLM_w_best_matching_GO_terms_for_JI.tsv', sep='\\t')\n",
    "llm_JI_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "llm_eval",
   "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": 2
}