{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3ee9fe07",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0f2a51dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np \n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "64cf9afb",
   "metadata": {},
   "outputs": [],
   "source": [
    "import faiss\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6af84fe1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# print(\"Loading FAISS index...\")\n",
    "# index = faiss.read_index(str(\"data/data/faiss.index\"))\n",
    "with open('data/data/metadata.pkl', \"rb\") as f:\n",
    "    metadata = pickle.load(f)\n",
    "# print(f\"Index loaded with {index.ntotal} documents\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "790819dd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "100000"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "index.ntotal"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "e25e1ea9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Species                           Host                  \n",
       "TTV-like mini virus               Homo sapiens              668\n",
       "Torque teno midi virus            Homo sapiens              651\n",
       "Orthoflavivirus denguei           Homo sapiens              624\n",
       "Norovirus norwalkense             Homo sapiens              617\n",
       "Enterovirus alphacoxsackie        Homo sapiens              607\n",
       "                                                           ... \n",
       "Alphainfluenzavirus influenzae    Gallus gallus             152\n",
       "Rotavirus alphagastroenteritidis  Sus scrofa                151\n",
       "Morbillivirus canis               Canis lupus familiaris    150\n",
       "Gammainfluenzavirus influenzae    Homo sapiens              147\n",
       "Mastadenovirus blackbeardi        Homo sapiens              145\n",
       "Length: 100, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metadata.groupby(['Species', 'Host']).size().sort_values(ascending=False).head(100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7ffd8faf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'AGCAAAAGCAGGTAGATATTGAAAGATGAGTCTTCTAACTGAGGTCGAAACGTACGTMOTIF_ID_TC_REPEAT_3_REPEAT_GROUP_2TATCGTCCCGTCAGGCCCCCTCAAAGCCGAGATCGCGCAGAGACTTGAAGATGTCTTTGCAGGGAAGAACACCGATCTCGAGGCTCTCATGGAATGGCTAAAGACAAGACCAATCCTGTCACCTCTGACTAAGGGGATTTTAGGGTTTGTGTTCACGCTCACCGTGCCCAGTGAGCGAGGACTGCAGCGTAGACGCTTTGTCCAAAATGCCCTAAATGGAAATGGAGACCCAAACAACATGGACAGGGCAGTCAAACTGTACAGGAAGTTGAAAMOTIF_ID_AG_REPEAT_3_REPEAT_GROUP_2ATAACATTCCATGGGGCTAAAGAAGTTGCACTCAGCTACTCAACCGGTGCACTTGCCAGTTGTATGGGTCTCATATACAACAGGATGGGAACAGTGACCACAGAAGTGGCTTTTGGCCTAMOTIF_ID_GT_REPEAT_3_REPEAT_GROUP_2GCCACCTGTGAGCAGATTGCTGATTCACAGCACCGGTCCCACAGGCAGATGGTAACTACCACCAACCCACTAATCAGGCATGAAAACAGGATGGTGCTGGCCAGCACTACGGCTAAGGCTATGGAGCASATGGCTGGGTCGAGTGAGCAGGCAGCGGAAGCCATGGAGGTTGCTAGTCAGGCTAGGCAGATGGTGCAGGCGATGAGGACAATTGGAACTCACCCTAGCTCCAGTGCCGGTCTGAAAGATGATCTTCTTGAAAATTTGCAGGCCTACCAGAAACGGATGGGAGTGCAAATGCAGCGATTCAAGTGATCCTCTCGTTATTGCCGCAAGTATCATTGGGATCTTGCACTTGATATTGTGGATTCTTGATCGTCTTTTCTTCAAATGTATTTATCGTCGCCTTAAATACGGTTTGAAAAGAGGGCCTTCTACGGAAGGAGTGCCTGAGTCTATGAGGGAAGAGTATCGGCAGGAACAGCAGAGTGCTGTGGATGTTGACGATGGTCATTTTGTCAACATAGAGCTGGAGTMOTIF_ID_A_REPEAT_6_REPEAT_GROUP_3CTACCTTGTTTCTACTAATACGAGACGATATA'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metadata[metadata['Species'].str.contains('Alphainfluenzavirus')].iloc[0]['seq_with_repeat_tokens']\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f09ac404",
   "metadata": {},
   "outputs": [],
   "source": [
    "metadata['embeddings'] = embeddings.tolist()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "36a1ded3",
   "metadata": {},
   "outputs": [],
   "source": [
    "no_alphainfluenze = metadata[~metadata['Species'].str.contains('Alphainfluenzavirus')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "ca22a8b3",
   "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>index</th>\n",
       "      <th>id</th>\n",
       "      <th>sequence_x</th>\n",
       "      <th>Species</th>\n",
       "      <th>Host</th>\n",
       "      <th>quality_percentage</th>\n",
       "      <th>Molecule_type</th>\n",
       "      <th>Nuc_Completeness</th>\n",
       "      <th>Genus</th>\n",
       "      <th>Family</th>\n",
       "      <th>Sequence_Type</th>\n",
       "      <th>sequence_y</th>\n",
       "      <th>seq_with_repeat_tokens</th>\n",
       "      <th>__sequence__</th>\n",
       "      <th>embeddings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>745225</td>\n",
       "      <td>MN779000.1</td>\n",
       "      <td>GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...</td>\n",
       "      <td>Torque teno midi virus</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssDNA(-)</td>\n",
       "      <td>partial</td>\n",
       "      <td>None</td>\n",
       "      <td>Anelloviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...</td>\n",
       "      <td>GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...</td>\n",
       "      <td>GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...</td>\n",
       "      <td>[1.5784817934036255, -0.7151118516921997, -0.7...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>333935</td>\n",
       "      <td>OP074769.1</td>\n",
       "      <td>TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...</td>\n",
       "      <td>Bacteriophage sp.</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>complete</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...</td>\n",
       "      <td>TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...</td>\n",
       "      <td>TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...</td>\n",
       "      <td>[0.35540127754211426, -0.16885411739349365, -0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>587805</td>\n",
       "      <td>MG904773.1</td>\n",
       "      <td>CCTATGTACAGGAAGCCCAGGATATACAGGATGTACAGAAGTCCAG...</td>\n",
       "      <td>Cotton leaf curl Rajasthan virus</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssDNA(+/-)</td>\n",
       "      <td>partial</td>\n",
       "      <td>Begomovirus</td>\n",
       "      <td>Geminiviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>CCTATGTACAGGAAGCCCAGGATATACAGGATGTACAGAAGTCCAG...</td>\n",
       "      <td>CCTATGTACAGGAAGCCCAGGATATACAGGATGTACAGAAGTCCAG...</td>\n",
       "      <td>CCTATGTACAGGAAGCCCAGGATATACAGGATGTACAGAAGTCCAG...</td>\n",
       "      <td>[1.380758285522461, -0.6527102589607239, -0.54...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>71069</td>\n",
       "      <td>PQ609851.1</td>\n",
       "      <td>GGTGGGGTGACAATCACACTGTTCTCAGCCAACATCGATGCCATCA...</td>\n",
       "      <td>Avibirnavirus gumboroense</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0</td>\n",
       "      <td>dsRNA</td>\n",
       "      <td>partial</td>\n",
       "      <td>Avibirnavirus</td>\n",
       "      <td>Birnaviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>GGTGGGGTGACAATCACACTGTTCTCAGCCAACATCGATGCCATCA...</td>\n",
       "      <td>GGTGGGGTGACAATCACACTGTTCTCAGCCAACATCGATGCCATCA...</td>\n",
       "      <td>GGTGGGGTGACAATCACACTGTTCTCAGCCAACATCGATGCCATCA...</td>\n",
       "      <td>[1.3650176525115967, -1.3555208444595337, -0.5...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>743588</td>\n",
       "      <td>MN777688.1</td>\n",
       "      <td>GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...</td>\n",
       "      <td>Torque teno midi virus</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssDNA(-)</td>\n",
       "      <td>partial</td>\n",
       "      <td>None</td>\n",
       "      <td>Anelloviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...</td>\n",
       "      <td>GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...</td>\n",
       "      <td>GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...</td>\n",
       "      <td>[1.4490833282470703, -0.6404349207878113, -0.5...</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99995</th>\n",
       "      <td>543599</td>\n",
       "      <td>BK071906.1</td>\n",
       "      <td>CTAAGATCAAAGTGTAGTATCTGTGAAAGCACCCTTTGAATGTCAA...</td>\n",
       "      <td>Potato cyst nematode orthomyxo-like virus 1</td>\n",
       "      <td>Globodera rostochiensis</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssRNA(-)</td>\n",
       "      <td>partial</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>CTAAGATCAAAGTGTAGTATCTGTGAAAGCACCCTTTGAATGTCAA...</td>\n",
       "      <td>CTAAGATCAAAGTGTAGTATCTGTGAAAGCACCCTTTGAATGTCAA...</td>\n",
       "      <td>CTAAGATCAAAGTGTAGTATCTGTGAAAGCACCCTTTGAATGTCAA...</td>\n",
       "      <td>[1.44771409034729, -1.0718882083892822, -0.585...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99996</th>\n",
       "      <td>515678</td>\n",
       "      <td>PP747384.1</td>\n",
       "      <td>TATTTCATCGTCAGCTGCATCAATACGGCAATTCAACGATTGGGCA...</td>\n",
       "      <td>Poacevirus sacchari</td>\n",
       "      <td>Saccharum officinarum</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssRNA(+)</td>\n",
       "      <td>partial</td>\n",
       "      <td>Poacevirus</td>\n",
       "      <td>Potyviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>TATTTCATCGTCAGCTGCATCAATACGGCAATTCAACGATTGGGCA...</td>\n",
       "      <td>TATTTCATCGTCAGCTGCATCAATACGGCAATTCAACGATTGGGCA...</td>\n",
       "      <td>TATTTCATCGTCAGCTGCATCAATACGGCAATTCAACGATTGGGCA...</td>\n",
       "      <td>[0.8059760332107544, -0.9930003881454468, -0.0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99997</th>\n",
       "      <td>766809</td>\n",
       "      <td>KJ201611.1</td>\n",
       "      <td>GAAAAATTGTGGGTCACAGTCTATTATGGGGTACCTGTGTGGAAAG...</td>\n",
       "      <td>Lentivirus humimdef1</td>\n",
       "      <td>Macaca nemestrina</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssRNA-RT</td>\n",
       "      <td>partial</td>\n",
       "      <td>Lentivirus</td>\n",
       "      <td>Retroviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>GAAAAATTGTGGGTCACAGTCTATTATGGGGTACCMOTIF_ID_TG...</td>\n",
       "      <td>GAAAAATTGTGGGTCACAGTCTATTATGGGGTACCMOTIF_ID_TG...</td>\n",
       "      <td>GAAAAATTGTGGGTCACAGTCTATTATGGGGTACCMOTIF_ID_TG...</td>\n",
       "      <td>[1.1333006620407104, -0.10236950218677521, -0....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99998</th>\n",
       "      <td>523672</td>\n",
       "      <td>KF008044.1</td>\n",
       "      <td>GGTGGTCAGCGTATTGACAAGCAGTATGGTGACTGGATGCAAATCT...</td>\n",
       "      <td>uncultured phycodnavirus</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0</td>\n",
       "      <td>dsDNA</td>\n",
       "      <td>partial</td>\n",
       "      <td>None</td>\n",
       "      <td>Phycodnaviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>None</td>\n",
       "      <td>GGTGGTCAGCGTATTGACAAGCAGTATGGTGACTGGATGCAAATCT...</td>\n",
       "      <td>GGTGGTCAGCGTATTGACAAGCAGTATGGTGACTGGATGCAAATCT...</td>\n",
       "      <td>[0.6877675652503967, -0.7062620520591736, -0.6...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99999</th>\n",
       "      <td>407458</td>\n",
       "      <td>DQ792459.1</td>\n",
       "      <td>ATGGCCGGCCGTGGAATGCAACTACGGTCCGGCCGTACCACCGGCG...</td>\n",
       "      <td>Rhadinovirus macacinegamma8</td>\n",
       "      <td>Macaca nemestrina</td>\n",
       "      <td>0.0</td>\n",
       "      <td>dsDNA</td>\n",
       "      <td>partial</td>\n",
       "      <td>Rhadinovirus</td>\n",
       "      <td>Orthoherpesviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>ATGGCCGGCCGTGGAATGCAACTACGGTCCGGCCGTACCACCGGCG...</td>\n",
       "      <td>ATGGCCGGCCGTGGAATGCAACTACGGTCCGGCCGTACCACCGGCG...</td>\n",
       "      <td>ATGGCCGGCCGTGGAATGCAACTACGGTCCGGCCGTACCACCGGCG...</td>\n",
       "      <td>[0.02847452461719513, -0.1403609663248062, -0....</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>98910 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        index          id                                         sequence_x  \\\n",
       "0      745225  MN779000.1  GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...   \n",
       "1      333935  OP074769.1  TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...   \n",
       "2      587805  MG904773.1  CCTATGTACAGGAAGCCCAGGATATACAGGATGTACAGAAGTCCAG...   \n",
       "3       71069  PQ609851.1  GGTGGGGTGACAATCACACTGTTCTCAGCCAACATCGATGCCATCA...   \n",
       "4      743588  MN777688.1  GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...   \n",
       "...       ...         ...                                                ...   \n",
       "99995  543599  BK071906.1  CTAAGATCAAAGTGTAGTATCTGTGAAAGCACCCTTTGAATGTCAA...   \n",
       "99996  515678  PP747384.1  TATTTCATCGTCAGCTGCATCAATACGGCAATTCAACGATTGGGCA...   \n",
       "99997  766809  KJ201611.1  GAAAAATTGTGGGTCACAGTCTATTATGGGGTACCTGTGTGGAAAG...   \n",
       "99998  523672  KF008044.1  GGTGGTCAGCGTATTGACAAGCAGTATGGTGACTGGATGCAAATCT...   \n",
       "99999  407458  DQ792459.1  ATGGCCGGCCGTGGAATGCAACTACGGTCCGGCCGTACCACCGGCG...   \n",
       "\n",
       "                                           Species                     Host  \\\n",
       "0                           Torque teno midi virus             Homo sapiens   \n",
       "1                                Bacteriophage sp.             Homo sapiens   \n",
       "2                 Cotton leaf curl Rajasthan virus                     None   \n",
       "3                        Avibirnavirus gumboroense                     None   \n",
       "4                           Torque teno midi virus             Homo sapiens   \n",
       "...                                            ...                      ...   \n",
       "99995  Potato cyst nematode orthomyxo-like virus 1  Globodera rostochiensis   \n",
       "99996                          Poacevirus sacchari    Saccharum officinarum   \n",
       "99997                         Lentivirus humimdef1        Macaca nemestrina   \n",
       "99998                     uncultured phycodnavirus                     None   \n",
       "99999                  Rhadinovirus macacinegamma8        Macaca nemestrina   \n",
       "\n",
       "       quality_percentage Molecule_type Nuc_Completeness          Genus  \\\n",
       "0                     0.0      ssDNA(-)          partial           None   \n",
       "1                     0.0       unknown         complete           None   \n",
       "2                     0.0    ssDNA(+/-)          partial    Begomovirus   \n",
       "3                     0.0         dsRNA          partial  Avibirnavirus   \n",
       "4                     0.0      ssDNA(-)          partial           None   \n",
       "...                   ...           ...              ...            ...   \n",
       "99995                 0.0      ssRNA(-)          partial           None   \n",
       "99996                 0.0      ssRNA(+)          partial     Poacevirus   \n",
       "99997                 0.0      ssRNA-RT          partial     Lentivirus   \n",
       "99998                 0.0         dsDNA          partial           None   \n",
       "99999                 0.0         dsDNA          partial   Rhadinovirus   \n",
       "\n",
       "                   Family Sequence_Type  \\\n",
       "0           Anelloviridae       GenBank   \n",
       "1                    None       GenBank   \n",
       "2           Geminiviridae       GenBank   \n",
       "3            Birnaviridae       GenBank   \n",
       "4           Anelloviridae       GenBank   \n",
       "...                   ...           ...   \n",
       "99995                None       GenBank   \n",
       "99996         Potyviridae       GenBank   \n",
       "99997        Retroviridae       GenBank   \n",
       "99998     Phycodnaviridae       GenBank   \n",
       "99999  Orthoherpesviridae       GenBank   \n",
       "\n",
       "                                              sequence_y  \\\n",
       "0      GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...   \n",
       "1      TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...   \n",
       "2      CCTATGTACAGGAAGCCCAGGATATACAGGATGTACAGAAGTCCAG...   \n",
       "3      GGTGGGGTGACAATCACACTGTTCTCAGCCAACATCGATGCCATCA...   \n",
       "4      GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...   \n",
       "...                                                  ...   \n",
       "99995  CTAAGATCAAAGTGTAGTATCTGTGAAAGCACCCTTTGAATGTCAA...   \n",
       "99996  TATTTCATCGTCAGCTGCATCAATACGGCAATTCAACGATTGGGCA...   \n",
       "99997  GAAAAATTGTGGGTCACAGTCTATTATGGGGTACCMOTIF_ID_TG...   \n",
       "99998                                               None   \n",
       "99999  ATGGCCGGCCGTGGAATGCAACTACGGTCCGGCCGTACCACCGGCG...   \n",
       "\n",
       "                                  seq_with_repeat_tokens  \\\n",
       "0      GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...   \n",
       "1      TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...   \n",
       "2      CCTATGTACAGGAAGCCCAGGATATACAGGATGTACAGAAGTCCAG...   \n",
       "3      GGTGGGGTGACAATCACACTGTTCTCAGCCAACATCGATGCCATCA...   \n",
       "4      GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...   \n",
       "...                                                  ...   \n",
       "99995  CTAAGATCAAAGTGTAGTATCTGTGAAAGCACCCTTTGAATGTCAA...   \n",
       "99996  TATTTCATCGTCAGCTGCATCAATACGGCAATTCAACGATTGGGCA...   \n",
       "99997  GAAAAATTGTGGGTCACAGTCTATTATGGGGTACCMOTIF_ID_TG...   \n",
       "99998  GGTGGTCAGCGTATTGACAAGCAGTATGGTGACTGGATGCAAATCT...   \n",
       "99999  ATGGCCGGCCGTGGAATGCAACTACGGTCCGGCCGTACCACCGGCG...   \n",
       "\n",
       "                                            __sequence__  \\\n",
       "0      GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...   \n",
       "1      TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...   \n",
       "2      CCTATGTACAGGAAGCCCAGGATATACAGGATGTACAGAAGTCCAG...   \n",
       "3      GGTGGGGTGACAATCACACTGTTCTCAGCCAACATCGATGCCATCA...   \n",
       "4      GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...   \n",
       "...                                                  ...   \n",
       "99995  CTAAGATCAAAGTGTAGTATCTGTGAAAGCACCCTTTGAATGTCAA...   \n",
       "99996  TATTTCATCGTCAGCTGCATCAATACGGCAATTCAACGATTGGGCA...   \n",
       "99997  GAAAAATTGTGGGTCACAGTCTATTATGGGGTACCMOTIF_ID_TG...   \n",
       "99998  GGTGGTCAGCGTATTGACAAGCAGTATGGTGACTGGATGCAAATCT...   \n",
       "99999  ATGGCCGGCCGTGGAATGCAACTACGGTCCGGCCGTACCACCGGCG...   \n",
       "\n",
       "                                              embeddings  \n",
       "0      [1.5784817934036255, -0.7151118516921997, -0.7...  \n",
       "1      [0.35540127754211426, -0.16885411739349365, -0...  \n",
       "2      [1.380758285522461, -0.6527102589607239, -0.54...  \n",
       "3      [1.3650176525115967, -1.3555208444595337, -0.5...  \n",
       "4      [1.4490833282470703, -0.6404349207878113, -0.5...  \n",
       "...                                                  ...  \n",
       "99995  [1.44771409034729, -1.0718882083892822, -0.585...  \n",
       "99996  [0.8059760332107544, -0.9930003881454468, -0.0...  \n",
       "99997  [1.1333006620407104, -0.10236950218677521, -0....  \n",
       "99998  [0.6877675652503967, -0.7062620520591736, -0.6...  \n",
       "99999  [0.02847452461719513, -0.1403609663248062, -0....  \n",
       "\n",
       "[98910 rows x 15 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "no_alphainfluenze"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "adace419",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/440 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 440/440 [00:01<00:00, 434.52it/s]\n"
     ]
    }
   ],
   "source": [
    "# select only one specie for each human host\n",
    "\n",
    "onlyhumans = no_alphainfluenze[no_alphainfluenze['Host'] == 'Homo sapiens']\n",
    "species = onlyhumans['Species'].unique()\n",
    "embedds = []\n",
    "for specie in tqdm(species):\n",
    "    embeds = onlyhumans[onlyhumans['Species'] == specie]['embeddings'].tolist()\n",
    "    embeds = np.array(embeds)\n",
    "    embeds_mean = embeds.mean(axis=0).flatten()\n",
    "    # print(embeds_mean.shape)\n",
    "    embedds.append(embeds_mean.tolist())\n",
    "    # break\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2db912ed",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 18%|█▊        | 78/441 [00:00<00:00, 772.15it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 441/441 [00:00<00:00, 801.06it/s]\n",
      "100%|██████████| 9839/9839 [00:38<00:00, 254.81it/s]\n",
      "100%|██████████| 20/20 [00:00<00:00, 46.91it/s]\n"
     ]
    }
   ],
   "source": [
    "onlyhumans = metadata[metadata['Host'] == 'Homo sapiens']\n",
    "species = onlyhumans['Species'].unique()\n",
    "atleast_100samples = []\n",
    "for specie in tqdm(species):\n",
    "    if len(onlyhumans[onlyhumans['Species'] == specie]) >= 100:\n",
    "        atleast_100samples.append(specie)\n",
    "\n",
    "nonhuman = metadata[metadata['Host'] != 'Homo sapiens']\n",
    "species = nonhuman['Species'].unique()\n",
    "atleast_100samples_non = []\n",
    "for specie in tqdm(species):\n",
    "    if len(nonhuman[nonhuman['Species'] == specie]) >= 100:\n",
    "        atleast_100samples_non.append(specie)\n",
    "\n",
    "# only keep species with at least 100 samples in both human and non-human datasets\n",
    "common_species = list(set(atleast_100samples) & set(atleast_100samples_non))\n",
    "# get human_embeds & nonhuman avg embeds\n",
    "final_embds = []\n",
    "labels = []\n",
    "for specie in tqdm(common_species):\n",
    "    labels.append('human')\n",
    "    labels.append('non_human')\n",
    "    human_embeds = onlyhumans[onlyhumans['Species'] == specie]['embeddings'].tolist()\n",
    "    human_embeds = np.array(human_embeds).mean(axis=0)\n",
    "    final_embds.append(human_embeds)\n",
    "    nonhuman_embeds = nonhuman[nonhuman['Species'] == specie]['embeddings'].tolist()\n",
    "    nonhuman_embeds = np.array(nonhuman_embeds).mean(axis=0)\n",
    "    final_embds.append(nonhuman_embeds)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "cf541b6c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 441/441 [00:00<00:00, 795.12it/s]\n",
      "100%|██████████| 9839/9839 [00:37<00:00, 261.78it/s]\n",
      "100%|██████████| 20/20 [00:00<00:00, 49.29it/s]\n"
     ]
    }
   ],
   "source": [
    "onlyhumans = metadata[metadata['Host'] == 'Homo sapiens']\n",
    "species = onlyhumans['Species'].unique()\n",
    "atleast_100samples = []\n",
    "for specie in tqdm(species):\n",
    "    if len(onlyhumans[onlyhumans['Species'] == specie]) >= 100:\n",
    "        atleast_100samples.append(specie)\n",
    "\n",
    "nonhuman = metadata[metadata['Host'] != 'Homo sapiens']\n",
    "species = nonhuman['Species'].unique()\n",
    "atleast_100samples_non = []\n",
    "for specie in tqdm(species):\n",
    "    if len(nonhuman[nonhuman['Species'] == specie]) >= 100:\n",
    "        atleast_100samples_non.append(specie)\n",
    "\n",
    "# only keep species with at least 100 samples in both human and non-human datasets\n",
    "common_species = list(set(atleast_100samples) & set(atleast_100samples_non))\n",
    "# get human_embeds & nonhuman avg embeds\n",
    "final_embds = []\n",
    "labels = []\n",
    "for specie in tqdm(common_species):\n",
    "    labels.append('human')\n",
    "    labels.append('non_human')\n",
    "    human_embeds = onlyhumans[onlyhumans['Species'] == specie]['embeddings'].tolist()\n",
    "    human_embeds = np.array(human_embeds).mean(axis=0)\n",
    "    final_embds.append(human_embeds)\n",
    "    nonhuman_embeds = nonhuman[nonhuman['Species'] == specie]['embeddings'].tolist()\n",
    "    nonhuman_embeds = np.array(nonhuman_embeds).mean(axis=0)\n",
    "    final_embds.append(nonhuman_embeds)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "id": "656819b4",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20/20 [00:00<00:00, 37.19it/s]\n"
     ]
    }
   ],
   "source": [
    "\n",
    "from scipy.spatial import distance\n",
    "# only keep species with at least 100 samples in both human and non-human datasets\n",
    "common_species = list(set(atleast_100samples) & set(atleast_100samples_non))\n",
    "# get human_embeds & nonhuman avg embeds\n",
    "final_embds = []\n",
    "labels = []\n",
    "difference = []\n",
    "counts_non_and_human = []\n",
    "Molecule_type = []\n",
    "for specie in tqdm(common_species):\n",
    "    labels.append('human')\n",
    "    labels.append('non_human')\n",
    "    Molecule_type.append(onlyhumans[onlyhumans['Species'] == specie]['Molecule_type'].iloc[0])\n",
    "    human_embeds = onlyhumans[onlyhumans['Species'] == specie]['embeddings'].tolist()\n",
    "    counts = (len(human_embeds), len(nonhuman[nonhuman['Species'] == specie]))\n",
    "    counts_non_and_human.append(counts)\n",
    "    human_embeds = np.array(human_embeds).mean(axis=0)\n",
    "    final_embds.append(human_embeds)\n",
    "    nonhuman_embeds = nonhuman[nonhuman['Species'] == specie]['embeddings'].tolist()\n",
    "    nonhuman_embeds = np.array(nonhuman_embeds).mean(axis=0)\n",
    "    final_embds.append(nonhuman_embeds)\n",
    "    difference.append(distance.euclidean(human_embeds, nonhuman_embeds))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "619d4d3a",
   "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>index</th>\n",
       "      <th>id</th>\n",
       "      <th>sequence_x</th>\n",
       "      <th>Species</th>\n",
       "      <th>Host</th>\n",
       "      <th>quality_percentage</th>\n",
       "      <th>Molecule_type</th>\n",
       "      <th>Nuc_Completeness</th>\n",
       "      <th>Genus</th>\n",
       "      <th>Family</th>\n",
       "      <th>Sequence_Type</th>\n",
       "      <th>sequence_y</th>\n",
       "      <th>seq_with_repeat_tokens</th>\n",
       "      <th>__sequence__</th>\n",
       "      <th>embeddings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>745225</td>\n",
       "      <td>MN779000.1</td>\n",
       "      <td>GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...</td>\n",
       "      <td>Torque teno midi virus</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssDNA(-)</td>\n",
       "      <td>partial</td>\n",
       "      <td>None</td>\n",
       "      <td>Anelloviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...</td>\n",
       "      <td>GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...</td>\n",
       "      <td>GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...</td>\n",
       "      <td>[1.5784817934036255, -0.7151118516921997, -0.7...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>333935</td>\n",
       "      <td>OP074769.1</td>\n",
       "      <td>TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...</td>\n",
       "      <td>Bacteriophage sp.</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>complete</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...</td>\n",
       "      <td>TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...</td>\n",
       "      <td>TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...</td>\n",
       "      <td>[0.35540127754211426, -0.16885411739349365, -0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>743588</td>\n",
       "      <td>MN777688.1</td>\n",
       "      <td>GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...</td>\n",
       "      <td>Torque teno midi virus</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssDNA(-)</td>\n",
       "      <td>partial</td>\n",
       "      <td>None</td>\n",
       "      <td>Anelloviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...</td>\n",
       "      <td>GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...</td>\n",
       "      <td>GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...</td>\n",
       "      <td>[1.4490833282470703, -0.6404349207878113, -0.5...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>555077</td>\n",
       "      <td>MN728852.1</td>\n",
       "      <td>GTCATAGTAGCTTAGCTGAACTGGGCCGTGGGGGTCGTCATCATCT...</td>\n",
       "      <td>Lymphocryptovirus humangamma4</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>dsDNA</td>\n",
       "      <td>partial</td>\n",
       "      <td>Lymphocryptovirus</td>\n",
       "      <td>Orthoherpesviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>None</td>\n",
       "      <td>GTCATAGTAGCTTAGCTGAACTGGGCCGTGGGGGTCGTCATCATCT...</td>\n",
       "      <td>GTCATAGTAGCTTAGCTGAACTGGGCCGTGGGGGTCGTCATCATCT...</td>\n",
       "      <td>[1.3683521747589111, -1.4705158472061157, -0.1...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>226539</td>\n",
       "      <td>MH834996.1</td>\n",
       "      <td>TCGCCCACAGGACGTCAAGTTCCCGGGTGGCGGTCAGATCGTTGGT...</td>\n",
       "      <td>Hepacivirus hominis</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssRNA(+)</td>\n",
       "      <td>partial</td>\n",
       "      <td>Hepacivirus</td>\n",
       "      <td>Flaviviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>TCGCCCACAGGACGTCAAGTTCCCGGGTGGCGGTCAGATCGTTGGT...</td>\n",
       "      <td>TCGCCCACAGGACGTCAAGTTCCCGGGTGGCGGTCAGATCGTTGGT...</td>\n",
       "      <td>TCGCCCACAGGACGTCAAGTTCCCGGGTGGCGGTCAGATCGTTGGT...</td>\n",
       "      <td>[1.127502202987671, -0.5482251644134521, -0.71...</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99984</th>\n",
       "      <td>426600</td>\n",
       "      <td>MH322371.1</td>\n",
       "      <td>ATGGCTACCCCTTCGATGATGCCGCAGTGGTCTTACATGCACATCT...</td>\n",
       "      <td>Mastadenovirus caesari</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>dsDNA</td>\n",
       "      <td>partial</td>\n",
       "      <td>Mastadenovirus</td>\n",
       "      <td>Adenoviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>ATGGCTACCCCTTCGATGATGCCGCAGTGGTCTTACATGCACATCT...</td>\n",
       "      <td>ATGGCTACCCCTTCGATGATGCCGCAGTGGTCTTACATGCACATCT...</td>\n",
       "      <td>ATGGCTACCCCTTCGATGATGCCGCAGTGGTCTTACATGCACATCT...</td>\n",
       "      <td>[1.163127064704895, -1.039842963218689, -0.194...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99989</th>\n",
       "      <td>749839</td>\n",
       "      <td>PX122279.1</td>\n",
       "      <td>ATTTTCAAGCCGGCTATATATTCTGCTATATTGTTACTAGTATCAA...</td>\n",
       "      <td>Orthopoxvirus monkeypox</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>dsDNA</td>\n",
       "      <td>partial</td>\n",
       "      <td>Orthopoxvirus</td>\n",
       "      <td>Poxviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>ATTTTCAAGCCGGCMOTIF_ID_TA_REPEAT_3_REPEAT_GROU...</td>\n",
       "      <td>ATTTTCAAGCCGGCMOTIF_ID_TA_REPEAT_3_REPEAT_GROU...</td>\n",
       "      <td>ATTTTCAAGCCGGCMOTIF_ID_TA_REPEAT_3_REPEAT_GROU...</td>\n",
       "      <td>[0.5123036503791809, -0.3153320550918579, -0.3...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99990</th>\n",
       "      <td>703430</td>\n",
       "      <td>FJ807891.1</td>\n",
       "      <td>ATGGAGTTTATCCCAACCCAAACTTTCTACAATAGGAGGTACCAGC...</td>\n",
       "      <td>Alphavirus chikungunya</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssRNA(+)</td>\n",
       "      <td>partial</td>\n",
       "      <td>Alphavirus</td>\n",
       "      <td>Togaviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>ATGGAGTTTATCCCAACCCAAACTTTCTACAATAGGAGGTACCAGC...</td>\n",
       "      <td>ATGGAGTTTATCCCAACCCAAACTTTCTACAATAGGAGGTACCAGC...</td>\n",
       "      <td>ATGGAGTTTATCCCAACCCAAACTTTCTACAATAGGAGGTACCAGC...</td>\n",
       "      <td>[1.582106113433838, -0.9431564807891846, -0.52...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99992</th>\n",
       "      <td>413749</td>\n",
       "      <td>LC337740.1</td>\n",
       "      <td>ATGTCTTGGAAAGTGATGATTATCATTTCGTTACTCATAACACCTC...</td>\n",
       "      <td>Metapneumovirus hominis</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssRNA(-)</td>\n",
       "      <td>partial</td>\n",
       "      <td>Metapneumovirus</td>\n",
       "      <td>Pneumoviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>ATGTCTTGGAAAGTGATGATTATCATTTCGTTACTCATAACACCTC...</td>\n",
       "      <td>ATGTCTTGGAAAGTGATGATTATCATTTCGTTACTCATAACACCTC...</td>\n",
       "      <td>ATGTCTTGGAAAGTGATGATTATCATTTCGTTACTCATAACACCTC...</td>\n",
       "      <td>[1.430707335472107, -1.3506540060043335, -0.34...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99993</th>\n",
       "      <td>354856</td>\n",
       "      <td>MN090470.1</td>\n",
       "      <td>GGACCTGAAAGCGAAAGAGAAACCAGAGGAGCTCTCTCGACGCAGG...</td>\n",
       "      <td>Lentivirus humimdef1</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ssRNA-RT</td>\n",
       "      <td>complete</td>\n",
       "      <td>Lentivirus</td>\n",
       "      <td>Retroviridae</td>\n",
       "      <td>GenBank</td>\n",
       "      <td>GGACCTGAAAGCGAAAGAGAAACCAGAGGAGMOTIF_ID_CT_REP...</td>\n",
       "      <td>GGACCTGAAAGCGAAAGAGAAACCAGAGGAGMOTIF_ID_CT_REP...</td>\n",
       "      <td>GGACCTGAAAGCGAAAGAGAAACCAGAGGAGMOTIF_ID_CT_REP...</td>\n",
       "      <td>[1.3395997285842896, -0.37854230403900146, -0....</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>22707 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        index          id                                         sequence_x  \\\n",
       "0      745225  MN779000.1  GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...   \n",
       "1      333935  OP074769.1  TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...   \n",
       "4      743588  MN777688.1  GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...   \n",
       "6      555077  MN728852.1  GTCATAGTAGCTTAGCTGAACTGGGCCGTGGGGGTCGTCATCATCT...   \n",
       "8      226539  MH834996.1  TCGCCCACAGGACGTCAAGTTCCCGGGTGGCGGTCAGATCGTTGGT...   \n",
       "...       ...         ...                                                ...   \n",
       "99984  426600  MH322371.1  ATGGCTACCCCTTCGATGATGCCGCAGTGGTCTTACATGCACATCT...   \n",
       "99989  749839  PX122279.1  ATTTTCAAGCCGGCTATATATTCTGCTATATTGTTACTAGTATCAA...   \n",
       "99990  703430  FJ807891.1  ATGGAGTTTATCCCAACCCAAACTTTCTACAATAGGAGGTACCAGC...   \n",
       "99992  413749  LC337740.1  ATGTCTTGGAAAGTGATGATTATCATTTCGTTACTCATAACACCTC...   \n",
       "99993  354856  MN090470.1  GGACCTGAAAGCGAAAGAGAAACCAGAGGAGCTCTCTCGACGCAGG...   \n",
       "\n",
       "                             Species          Host  quality_percentage  \\\n",
       "0             Torque teno midi virus  Homo sapiens                 0.0   \n",
       "1                  Bacteriophage sp.  Homo sapiens                 0.0   \n",
       "4             Torque teno midi virus  Homo sapiens                 0.0   \n",
       "6      Lymphocryptovirus humangamma4  Homo sapiens                 0.0   \n",
       "8                Hepacivirus hominis  Homo sapiens                 0.0   \n",
       "...                              ...           ...                 ...   \n",
       "99984         Mastadenovirus caesari  Homo sapiens                 0.0   \n",
       "99989        Orthopoxvirus monkeypox  Homo sapiens                 0.0   \n",
       "99990         Alphavirus chikungunya  Homo sapiens                 0.0   \n",
       "99992        Metapneumovirus hominis  Homo sapiens                 0.0   \n",
       "99993           Lentivirus humimdef1  Homo sapiens                 0.0   \n",
       "\n",
       "      Molecule_type Nuc_Completeness              Genus              Family  \\\n",
       "0          ssDNA(-)          partial               None       Anelloviridae   \n",
       "1           unknown         complete               None                None   \n",
       "4          ssDNA(-)          partial               None       Anelloviridae   \n",
       "6             dsDNA          partial  Lymphocryptovirus  Orthoherpesviridae   \n",
       "8          ssRNA(+)          partial        Hepacivirus        Flaviviridae   \n",
       "...             ...              ...                ...                 ...   \n",
       "99984         dsDNA          partial     Mastadenovirus        Adenoviridae   \n",
       "99989         dsDNA          partial      Orthopoxvirus          Poxviridae   \n",
       "99990      ssRNA(+)          partial         Alphavirus         Togaviridae   \n",
       "99992      ssRNA(-)          partial    Metapneumovirus       Pneumoviridae   \n",
       "99993      ssRNA-RT         complete         Lentivirus        Retroviridae   \n",
       "\n",
       "      Sequence_Type                                         sequence_y  \\\n",
       "0           GenBank  GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...   \n",
       "1           GenBank  TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...   \n",
       "4           GenBank  GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...   \n",
       "6           GenBank                                               None   \n",
       "8           GenBank  TCGCCCACAGGACGTCAAGTTCCCGGGTGGCGGTCAGATCGTTGGT...   \n",
       "...             ...                                                ...   \n",
       "99984       GenBank  ATGGCTACCCCTTCGATGATGCCGCAGTGGTCTTACATGCACATCT...   \n",
       "99989       GenBank  ATTTTCAAGCCGGCMOTIF_ID_TA_REPEAT_3_REPEAT_GROU...   \n",
       "99990       GenBank  ATGGAGTTTATCCCAACCCAAACTTTCTACAATAGGAGGTACCAGC...   \n",
       "99992       GenBank  ATGTCTTGGAAAGTGATGATTATCATTTCGTTACTCATAACACCTC...   \n",
       "99993       GenBank  GGACCTGAAAGCGAAAGAGAAACCAGAGGAGMOTIF_ID_CT_REP...   \n",
       "\n",
       "                                  seq_with_repeat_tokens  \\\n",
       "0      GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...   \n",
       "1      TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...   \n",
       "4      GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...   \n",
       "6      GTCATAGTAGCTTAGCTGAACTGGGCCGTGGGGGTCGTCATCATCT...   \n",
       "8      TCGCCCACAGGACGTCAAGTTCCCGGGTGGCGGTCAGATCGTTGGT...   \n",
       "...                                                  ...   \n",
       "99984  ATGGCTACCCCTTCGATGATGCCGCAGTGGTCTTACATGCACATCT...   \n",
       "99989  ATTTTCAAGCCGGCMOTIF_ID_TA_REPEAT_3_REPEAT_GROU...   \n",
       "99990  ATGGAGTTTATCCCAACCCAAACTTTCTACAATAGGAGGTACCAGC...   \n",
       "99992  ATGTCTTGGAAAGTGATGATTATCATTTCGTTACTCATAACACCTC...   \n",
       "99993  GGACCTGAAAGCGAAAGAGAAACCAGAGGAGMOTIF_ID_CT_REP...   \n",
       "\n",
       "                                            __sequence__  \\\n",
       "0      GGAGGTCCCCGGCTGCCCAAGGGCGGGAGCCCGAGGTGAGTGAAAC...   \n",
       "1      TGTGCGCCCGTGAAATTGTTTCACGTGGAACACTGCAACACCGATG...   \n",
       "4      GTGCAGGGACCGGATCGAGCGCAGCGAGGAGGTCCCCGGCTGCCCA...   \n",
       "6      GTCATAGTAGCTTAGCTGAACTGGGCCGTGGGGGTCGTCATCATCT...   \n",
       "8      TCGCCCACAGGACGTCAAGTTCCCGGGTGGCGGTCAGATCGTTGGT...   \n",
       "...                                                  ...   \n",
       "99984  ATGGCTACCCCTTCGATGATGCCGCAGTGGTCTTACATGCACATCT...   \n",
       "99989  ATTTTCAAGCCGGCMOTIF_ID_TA_REPEAT_3_REPEAT_GROU...   \n",
       "99990  ATGGAGTTTATCCCAACCCAAACTTTCTACAATAGGAGGTACCAGC...   \n",
       "99992  ATGTCTTGGAAAGTGATGATTATCATTTCGTTACTCATAACACCTC...   \n",
       "99993  GGACCTGAAAGCGAAAGAGAAACCAGAGGAGMOTIF_ID_CT_REP...   \n",
       "\n",
       "                                              embeddings  \n",
       "0      [1.5784817934036255, -0.7151118516921997, -0.7...  \n",
       "1      [0.35540127754211426, -0.16885411739349365, -0...  \n",
       "4      [1.4490833282470703, -0.6404349207878113, -0.5...  \n",
       "6      [1.3683521747589111, -1.4705158472061157, -0.1...  \n",
       "8      [1.127502202987671, -0.5482251644134521, -0.71...  \n",
       "...                                                  ...  \n",
       "99984  [1.163127064704895, -1.039842963218689, -0.194...  \n",
       "99989  [0.5123036503791809, -0.3153320550918579, -0.3...  \n",
       "99990  [1.582106113433838, -0.9431564807891846, -0.52...  \n",
       "99992  [1.430707335472107, -1.3506540060043335, -0.34...  \n",
       "99993  [1.3395997285842896, -0.37854230403900146, -0....  \n",
       "\n",
       "[22707 rows x 15 columns]"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "onlyhumans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "1f84236d",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "# Your original array\n",
    "original_labels = np.array(['ssRNA(+)', 'dsDNA', 'ssDNA(-)', 'ssDNA(+/-)', 'unknown', \n",
    "                           'dsRNA', 'ssRNA(-)', 'ssRNA(+/-)', 'ssRNA-RT', 'dsDNA-RT'])\n",
    "\n",
    "# Mapping dictionary to simplify\n",
    "clean_map = {\n",
    "    'ssRNA(+)': 'ssRNA', 'ssRNA(-)': 'ssRNA', 'ssRNA(+/-)': 'ssRNA', 'ssRNA-RT': 'ssRNA',\n",
    "    'dsRNA': 'dsRNA',\n",
    "    'ssDNA(-)': 'ssDNA', 'ssDNA(+)': 'ssDNA', 'ssDNA(+/-)': 'ssDNA',\n",
    "    'dsDNA': 'dsDNA', 'dsDNA-RT': 'dsDNA',\n",
    "    'unknown': 'Unknown'\n",
    "}\n",
    "\n",
    "# Apply the mapping\n",
    "simple_labels = np.array([clean_map.get(label, 'Other') for label in Molecule_type])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "69a902b5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Create the DataFrame\n",
    "df = pd.DataFrame({\n",
    "    'difference': difference,\n",
    "    'type': simple_labels,\n",
    "    'species': common_species\n",
    "})\n",
    "\n",
    "# Define \"Outliers\" (e.g., the top 5 largest differences)\n",
    "# You can also use: df[df['difference'] > threshold]\n",
    "outliers = df.nlargest(5, 'difference')\n",
    "\n",
    "# Create the plot\n",
    "plt.figure(figsize=(12, 6))\n",
    "ax = sns.histplot(data=df, x='difference', hue='type', bins=30, multiple=\"stack\")\n",
    "\n",
    "# Add the labels for the outliers\n",
    "for i, row in outliers.iterrows():\n",
    "    # We place the text slightly above the x-axis or at a specific height\n",
    "    plt.annotate(\n",
    "        row['species'], \n",
    "        xy=(row['difference'], 0),      # Point to the data location\n",
    "        xytext=(row['difference'], 5),  # Text location (shifted up)\n",
    "        arrowprops=dict(facecolor='black', shrink=0.05, width=1, headwidth=5),\n",
    "        rotation=45,\n",
    "        ha='left',\n",
    "        fontsize=9,\n",
    "        fontweight='bold'\n",
    "    )\n",
    "\n",
    "plt.title(\"Genomic Differences with Outlier Species Labeled\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "id": "a6643403",
   "metadata": {},
   "outputs": [],
   "source": [
    "outliers = df.nlargest(5, 'difference')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "id": "0b39132f",
   "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>difference</th>\n",
       "      <th>type</th>\n",
       "      <th>species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>17.358904</td>\n",
       "      <td>ssRNA</td>\n",
       "      <td>Lentivirus humimdef1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>15.446966</td>\n",
       "      <td>ssRNA</td>\n",
       "      <td>Orthoflavivirus denguei</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>9.390351</td>\n",
       "      <td>dsDNA</td>\n",
       "      <td>Caudoviricetes sp.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>8.546765</td>\n",
       "      <td>ssRNA</td>\n",
       "      <td>Orthopneumovirus bovis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>7.898471</td>\n",
       "      <td>ssRNA</td>\n",
       "      <td>Enterovirus alphacoxsackie</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    difference   type                     species\n",
       "43   17.358904  ssRNA        Lentivirus humimdef1\n",
       "24   15.446966  ssRNA     Orthoflavivirus denguei\n",
       "30    9.390351  dsDNA          Caudoviricetes sp.\n",
       "16    8.546765  ssRNA      Orthopneumovirus bovis\n",
       "39    7.898471  ssRNA  Enterovirus alphacoxsackie"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "79b1502d",
   "metadata": {},
   "outputs": [
    {
     "ename": "ModuleNotFoundError",
     "evalue": "No module named 'umap'",
     "output_type": "error",
     "traceback": [
      "\u001b[31m---------------------------------------------------------------------------\u001b[39m",
      "\u001b[31mModuleNotFoundError\u001b[39m                       Traceback (most recent call last)",
      "\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[66]\u001b[39m\u001b[32m, line 4\u001b[39m\n\u001b[32m      2\u001b[39m \u001b[38;5;28;01mimport\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mpandas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mpd\u001b[39;00m\n\u001b[32m      3\u001b[39m \u001b[38;5;28;01mimport\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mnumpy\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mnp\u001b[39;00m\n\u001b[32m----> \u001b[39m\u001b[32m4\u001b[39m \u001b[38;5;28;01mimport\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mumap\u001b[39;00m\n",
      "\u001b[31mModuleNotFoundError\u001b[39m: No module named 'umap'"
     ]
    }
   ],
   "source": [
    "import sklearn.datasets\n",
    "import pandas as pd\n",
    "import numpy as np\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "11801b9c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.52528826, -0.48906102, -0.77509444, ..., -0.09142731,\n",
       "        -0.14950409,  0.03409395],\n",
       "       [ 0.77453233, -1.8310526 ,  0.08784599, ...,  2.18337157,\n",
       "         0.56579514,  0.98507481],\n",
       "       [ 0.8738077 , -1.0557334 ,  1.03922419, ..., -0.13426347,\n",
       "         0.86687653,  0.50680748],\n",
       "       ...,\n",
       "       [ 0.8837494 , -1.2456221 , -0.77899884, ...,  1.56792927,\n",
       "         1.00026099,  1.88152304],\n",
       "       [-1.30289476,  1.02449966,  0.81268137, ..., -0.41357554,\n",
       "        -0.54531597,  0.14805078],\n",
       "       [-3.70826866,  3.08256337,  1.26085462, ..., -0.41230168,\n",
       "        -3.30523349, -0.13255314]], shape=(44, 512))"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import preprocessing\n",
    "import numpy as np\n",
    "X_train = final_embds\n",
    "scaler = preprocessing.StandardScaler().fit(difference)\n",
    "scaler\n",
    "\n",
    "scaler.mean_\n",
    "\n",
    "scaler.scale_\n",
    "\n",
    "X_scaled = scaler.transform(difference)\n",
    "X_scaled"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "7eedaa59",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.11826702,  0.10656931,  0.00952668,  0.01155277, -0.0036464 ,\n",
       "        0.01362973,  0.07826762, -0.01556595,  0.05214115,  0.08853014,\n",
       "        0.01649623, -0.00280654, -0.17468103,  0.1250343 , -0.05893999,\n",
       "       -0.10565191,  0.07742578,  0.09836444, -0.06842282, -0.06210458,\n",
       "        0.07331916,  0.05028733,  0.08316718,  0.03289016,  0.14819595,\n",
       "        0.11273334, -0.01589784, -0.0538015 ,  0.03887626,  0.01019054,\n",
       "        0.02267047, -0.10979362,  0.04818909,  0.04108151,  0.07898751,\n",
       "       -0.03626374,  0.09759948,  0.06671937,  0.13176716,  0.01880216,\n",
       "       -0.05136631,  0.03389503, -0.2297648 ,  0.03876104, -0.17353243,\n",
       "       -0.05602483, -0.03419147,  0.12235103, -0.00573563, -0.01299198,\n",
       "        0.04694373, -0.0897054 ,  0.02344495,  0.04163683,  0.26078797,\n",
       "       -0.05831128,  0.16174395, -0.00196099, -0.06443227,  0.10904721,\n",
       "        0.085366  ,  0.20751294, -0.17647894, -0.01191628,  0.0803137 ,\n",
       "        0.02465544,  0.07531016, -0.04449111,  0.05630648, -0.03130906,\n",
       "       -0.05708081, -0.01838893, -0.05445285, -0.07288948,  0.19048141,\n",
       "        0.11604944,  0.13488638,  0.18135961, -0.06520843, -0.11111529,\n",
       "       -0.12663792, -0.07024514,  0.13291787, -0.1575718 ,  0.1181291 ,\n",
       "        0.10864834, -0.0163    , -0.12504926,  0.01930129,  0.04993767,\n",
       "       -0.04659841, -0.00483647,  0.02875254, -0.11738016,  0.01855658,\n",
       "        0.0334962 ,  0.04709809,  0.18724501, -0.1248958 ,  0.01850903,\n",
       "       -0.03153642, -0.07873458,  0.01978654, -0.07381321,  0.22381832,\n",
       "        0.00765181,  0.11387284, -0.07890469,  0.06395411,  0.0791205 ,\n",
       "       -0.02382313,  0.08185035, -0.03350754,  0.0951895 , -0.13408777,\n",
       "       -0.06143904, -0.16334297, -0.01908186,  0.17097759,  0.10388974,\n",
       "        0.05878084, -0.12518224,  0.01041812,  0.05029419, -0.04436305,\n",
       "        0.07192371, -0.00386673, -0.11520742,  0.18235991, -0.18332942,\n",
       "       -0.23196216,  0.03158349, -0.01783191, -0.00801877, -0.0060655 ,\n",
       "       -0.06892787,  0.00802493, -0.04122208,  0.04140389, -0.02685665,\n",
       "       -0.07533424, -0.00111427,  0.0236892 , -0.23346744,  0.06413496,\n",
       "       -0.08877175,  0.0351316 ,  0.05796962,  0.04908024, -0.10735556,\n",
       "        0.0526904 , -0.06762235, -0.04297338, -0.01457921, -0.13527181,\n",
       "       -0.1284612 , -0.04163372, -0.0434812 ,  0.13677484,  0.19534383,\n",
       "        0.1302731 ,  0.16686016, -0.11836864,  0.03229045, -0.03273393,\n",
       "       -0.14129137,  0.03381272, -0.03232398,  0.09681888, -0.26095487,\n",
       "       -0.01026824, -0.05824349,  0.00361409,  0.10489386,  0.05132231,\n",
       "        0.15965703, -0.12531371, -0.06266663,  0.00441202,  0.04260402,\n",
       "       -0.0715869 , -0.00879758, -0.15860874,  0.14773851, -0.1399162 ,\n",
       "        0.02779156,  0.10008079, -0.20459295,  0.09231173, -0.23922916,\n",
       "       -0.17156948,  0.2162625 ,  0.12497747,  0.12191243, -0.05968926,\n",
       "        0.00048894, -0.05703721, -0.07392601, -0.14404233,  0.06814013,\n",
       "        0.04588825, -0.0631024 , -0.03346177, -0.01280804, -0.05696043,\n",
       "        0.10696201, -0.15595433, -0.02148754,  0.10141454,  0.01092494,\n",
       "        0.03155835, -0.00888372, -0.06398281,  0.07678424, -0.00491971,\n",
       "        0.09932077, -0.096598  ,  0.01122419,  0.05614117, -0.05123619,\n",
       "        0.06027807, -0.03081487,  0.02732255,  0.01394182, -0.04775955,\n",
       "       -0.10435779, -0.02565282,  0.09192015, -0.19686402,  0.11398341,\n",
       "        0.17017441, -0.01229679, -0.01919833, -0.00434892,  0.04277632,\n",
       "        0.04126479,  0.1992878 ,  0.13232032,  0.05522215, -0.17099563,\n",
       "       -0.01010378,  0.0059948 , -0.08830738, -0.05975782, -0.07900338,\n",
       "        0.02702376,  0.08608351,  0.06003513, -0.09250419,  0.03289928,\n",
       "        0.15513919,  0.01255525, -0.0612108 , -0.03846161,  0.224105  ,\n",
       "        0.07132966,  0.1241037 , -0.18896003, -0.10660256, -0.07896445,\n",
       "       -0.06298888, -0.10712486,  0.08345618, -0.08992796, -0.20951154,\n",
       "       -0.02034598, -0.10836408, -0.22086548,  0.11647697,  0.04267843,\n",
       "       -0.03660746,  0.01250195,  0.04294764,  0.20806573, -0.01910027,\n",
       "       -0.14677994,  0.1339223 , -0.01420352,  0.17458138, -0.11456056,\n",
       "        0.050146  ,  0.15725795, -0.12256332, -0.17665508, -0.2134696 ,\n",
       "        0.02940404, -0.04099337,  0.05117583,  0.04062918, -0.03134876,\n",
       "        0.07693368,  0.11463797,  0.05954959, -0.16719716,  0.03970544,\n",
       "       -0.0565487 , -0.08307503, -0.03568541, -0.06486399, -0.25134527,\n",
       "       -0.00067374,  0.11648348, -0.07194415, -0.18222426, -0.06580557,\n",
       "        0.00890155,  0.13092345, -0.04842759,  0.07011641, -0.08313514,\n",
       "       -0.02065196,  0.05844559,  0.03671276,  0.06171165,  0.05737772,\n",
       "        0.00102745,  0.02043169,  0.16353483, -0.05880475, -0.03119979,\n",
       "       -0.00226952, -0.07439121, -0.0051804 ,  0.10897107,  0.10796846,\n",
       "       -0.13286459,  0.05597976,  0.03434891,  0.17481112,  0.1072033 ,\n",
       "       -0.22956484,  0.07433442, -0.10598633, -0.03731839, -0.14822525,\n",
       "        0.04299132, -0.06252825, -0.14666029, -0.0855995 ,  0.09805925,\n",
       "        0.12678799, -0.14138369,  0.05369006, -0.05359324, -0.11293268,\n",
       "        0.02416382,  0.10424335, -0.15637583, -0.07210166,  0.01075929,\n",
       "       -0.03189686,  0.09350748, -0.12760643,  0.22676434,  0.03092151,\n",
       "       -0.04785775, -0.14135823, -0.1648976 ,  0.07378951, -0.01961578,\n",
       "        0.14664125, -0.08949427, -0.07260371,  0.1880499 , -0.00237543,\n",
       "       -0.00031977, -0.0609338 , -0.08470597, -0.08201055, -0.07593412,\n",
       "        0.00307599,  0.03699099,  0.2111746 , -0.17727648,  0.06848525,\n",
       "        0.11014923,  0.01665131, -0.02306221, -0.01964788,  0.28250917,\n",
       "        0.08374891,  0.02787239, -0.04126069, -0.02282465, -0.08801089,\n",
       "        0.04092935,  0.13861325, -0.10168947,  0.08285918,  0.07083781,\n",
       "        0.0006006 ,  0.07303856,  0.13063113, -0.02895838,  0.04228141,\n",
       "        0.07991643, -0.03915888,  0.20110737,  0.01942172, -0.24066451,\n",
       "        0.06493962,  0.13520116,  0.16754821,  0.07081199,  0.1467688 ,\n",
       "        0.05541789,  0.05770754,  0.08775903,  0.10379817,  0.1044952 ,\n",
       "       -0.09803296, -0.12804764, -0.02462201,  0.04753012, -0.08497261,\n",
       "        0.14366919, -0.2364507 , -0.00501923, -0.05423873, -0.27086696,\n",
       "       -0.07385492, -0.24852252,  0.00985616, -0.05155023, -0.01161322,\n",
       "        0.07150506, -0.13082604, -0.14501548, -0.1246951 ,  0.06661557,\n",
       "        0.06908679,  0.04159047,  0.14080869, -0.04270218, -0.02916402,\n",
       "        0.00646315,  0.02520579,  0.08339035, -0.02305597,  0.11166978,\n",
       "        0.13785175,  0.07499945,  0.03826173, -0.01492224, -0.18436202,\n",
       "       -0.06509631, -0.1576496 ,  0.01895446, -0.00677837, -0.13552884,\n",
       "       -0.03297255,  0.05106766, -0.10111326, -0.16643592,  0.1178231 ,\n",
       "       -0.10059402,  0.00878905, -0.07251859, -0.0192677 , -0.04561202,\n",
       "        0.01700338, -0.04428936,  0.09935351, -0.00584767, -0.29345229,\n",
       "       -0.13424619,  0.17506197,  0.01934148, -0.07580524, -0.26642706,\n",
       "        0.03427641,  0.05536025, -0.1515776 ,  0.00834645,  0.10610453,\n",
       "        0.09720968,  0.00091738,  0.16817837, -0.00298761, -0.06955121,\n",
       "        0.03127913,  0.04862075, -0.00788778,  0.01195447, -0.06127647,\n",
       "       -0.17824713,  0.08920697, -0.03245315,  0.14442241, -0.20210036,\n",
       "       -0.05471003,  0.06884111,  0.23491878,  0.00520145, -0.04704316,\n",
       "        0.03317961,  0.02421927,  0.216112  ,  0.02396454, -0.1484807 ,\n",
       "        0.02773508,  0.04987883,  0.19059188,  0.14839748,  0.11957869,\n",
       "       -0.02974497, -0.06076054,  0.10290011,  0.0029877 , -0.0523704 ,\n",
       "       -0.03882024, -0.06073262])"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(difference, axis=0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "2e0a84c8",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/user/hassanahmed.hassan/u21055/.project/dir.project/towards_better_genomic_models/.venv/lib/python3.12/site-packages/sklearn/manifold/_mds.py:677: FutureWarning: The default value of `n_init` will change from 4 to 1 in 1.9.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(44, 2)"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.datasets import load_digits\n",
    "from sklearn.manifold import MDS\n",
    "\n",
    "embedding = MDS(n_components=2)\n",
    "X_transformed = embedding.fit_transform(difference)\n",
    "X_transformed.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "ff81750e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<Axes: >,)"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "plt.figure(figsize=(10,10))\n",
    "# c is speci\n",
    "sns.scatterplot(x=X_transformed[:,0], y=X_transformed[:,1]),# hue=common_species)\n",
    "# c=['red' if label=='human' else 'blue' for label in labels])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "d482599c",
   "metadata": {},
   "outputs": [],
   "source": [
    "dff = pd.DataFrame()\n",
    "dff['embed_diff'] = X_transformed.tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "46bab820",
   "metadata": {},
   "outputs": [],
   "source": [
    "dff['specie'] = common_species\n",
    "dff['counts_non_and_human'] = counts_non_and_human"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "cceae970",
   "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>embed_diff</th>\n",
       "      <th>specie</th>\n",
       "      <th>counts_non_and_human</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[-0.13763951244769218, 0.1766350106969233]</td>\n",
       "      <td>Rhadinovirus</td>\n",
       "      <td>(117, 128)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[0.1219625014355477, -0.044148872976712175]</td>\n",
       "      <td>Orthohepadnavirus</td>\n",
       "      <td>(567, 640)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[-0.10162890466443956, -0.14123188263183853]</td>\n",
       "      <td>Betainfluenzavirus</td>\n",
       "      <td>(547, 228)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[-0.11268897025277129, -0.08703998350455902]</td>\n",
       "      <td>Orthonairovirus</td>\n",
       "      <td>(188, 218)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[-0.10768941791741027, -0.10277596455244477]</td>\n",
       "      <td>Orthohantavirus</td>\n",
       "      <td>(124, 757)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[0.16683898140282974, 0.031563617205617164]</td>\n",
       "      <td>Metapneumovirus</td>\n",
       "      <td>(560, 106)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>[0.15586293058363865, 0.011449311539129549]</td>\n",
       "      <td>Morbillivirus</td>\n",
       "      <td>(367, 498)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>[0.09841072506185206, -0.06619201367868136]</td>\n",
       "      <td>Bandavirus</td>\n",
       "      <td>(581, 110)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>[-0.08698346028508039, -0.12699047485256862]</td>\n",
       "      <td>Rotavirus</td>\n",
       "      <td>(762, 1644)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>[-0.1392519729227442, 0.17346662156734716]</td>\n",
       "      <td>Betapolyomavirus</td>\n",
       "      <td>(211, 212)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>[-0.08534263559917774, -0.15043296092463446]</td>\n",
       "      <td>Cytomegalovirus</td>\n",
       "      <td>(222, 200)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>[0.18740675123873005, 0.10024623455411354]</td>\n",
       "      <td>Lentivirus</td>\n",
       "      <td>(1777, 2973)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>[0.1912645073455752, 0.11096619552536519]</td>\n",
       "      <td>Orthoflavivirus</td>\n",
       "      <td>(929, 2119)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>[-0.14747839975198213, 0.2502481027076783]</td>\n",
       "      <td>Simplexvirus</td>\n",
       "      <td>(319, 159)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>[0.06356809867134204, -0.09343125178386913]</td>\n",
       "      <td>Orthobunyavirus</td>\n",
       "      <td>(269, 424)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>[0.17108483320811155, 0.04557947562961259]</td>\n",
       "      <td>Norovirus</td>\n",
       "      <td>(619, 528)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>[0.06462390466597107, -0.07140312391625978]</td>\n",
       "      <td>Deltaretrovirus</td>\n",
       "      <td>(301, 422)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>[-0.14631634961961554, 0.2345126036404797]</td>\n",
       "      <td>Mamastrovirus</td>\n",
       "      <td>(105, 263)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>[0.18740675123873027, 0.1002462345541134]</td>\n",
       "      <td>Mammarenavirus</td>\n",
       "      <td>(135, 240)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>[0.1366045250314483, -0.008178589279885998]</td>\n",
       "      <td>Enterovirus</td>\n",
       "      <td>(2880, 1510)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>[-0.1131427072750879, -0.0544868948377319]</td>\n",
       "      <td>Mastadenovirus</td>\n",
       "      <td>(686, 286)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>[-0.08025913272665164, -0.14481010034284833]</td>\n",
       "      <td>Phlebovirus</td>\n",
       "      <td>(111, 265)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>[-0.08419884436327682, -0.15372049262950502]</td>\n",
       "      <td>Betacoronavirus</td>\n",
       "      <td>(138, 206)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>[-0.0914687334765774, -0.15240020627687756]</td>\n",
       "      <td>Alphapapillomavirus</td>\n",
       "      <td>(793, 189)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>[-0.15095973134914736, 0.2835972227463461]</td>\n",
       "      <td>Alphavirus</td>\n",
       "      <td>(539, 453)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>[0.14059437300046487, -0.013575671511743789]</td>\n",
       "      <td>Paslahepevirus</td>\n",
       "      <td>(109, 110)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>[-0.09998302340018402, -0.06147814581121143]</td>\n",
       "      <td>Alphainfluenzavirus</td>\n",
       "      <td>(234, 856)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>[-0.009963610897503824, -0.12083176876865355]</td>\n",
       "      <td>Orthopneumovirus</td>\n",
       "      <td>(809, 629)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>[0.16842921051342608, 0.05078886470451921]</td>\n",
       "      <td>Hepacivirus</td>\n",
       "      <td>(581, 688)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>[-0.14835142474231863, 0.26215215713229767]</td>\n",
       "      <td>Bocaparvovirus</td>\n",
       "      <td>(229, 230)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                       embed_diff               specie  \\\n",
       "0      [-0.13763951244769218, 0.1766350106969233]         Rhadinovirus   \n",
       "1     [0.1219625014355477, -0.044148872976712175]    Orthohepadnavirus   \n",
       "2    [-0.10162890466443956, -0.14123188263183853]   Betainfluenzavirus   \n",
       "3    [-0.11268897025277129, -0.08703998350455902]      Orthonairovirus   \n",
       "4    [-0.10768941791741027, -0.10277596455244477]      Orthohantavirus   \n",
       "5     [0.16683898140282974, 0.031563617205617164]      Metapneumovirus   \n",
       "6     [0.15586293058363865, 0.011449311539129549]        Morbillivirus   \n",
       "7     [0.09841072506185206, -0.06619201367868136]           Bandavirus   \n",
       "8    [-0.08698346028508039, -0.12699047485256862]            Rotavirus   \n",
       "9      [-0.1392519729227442, 0.17346662156734716]     Betapolyomavirus   \n",
       "10   [-0.08534263559917774, -0.15043296092463446]      Cytomegalovirus   \n",
       "11     [0.18740675123873005, 0.10024623455411354]           Lentivirus   \n",
       "12      [0.1912645073455752, 0.11096619552536519]      Orthoflavivirus   \n",
       "13     [-0.14747839975198213, 0.2502481027076783]         Simplexvirus   \n",
       "14    [0.06356809867134204, -0.09343125178386913]      Orthobunyavirus   \n",
       "15     [0.17108483320811155, 0.04557947562961259]            Norovirus   \n",
       "16    [0.06462390466597107, -0.07140312391625978]      Deltaretrovirus   \n",
       "17     [-0.14631634961961554, 0.2345126036404797]        Mamastrovirus   \n",
       "18      [0.18740675123873027, 0.1002462345541134]       Mammarenavirus   \n",
       "19    [0.1366045250314483, -0.008178589279885998]          Enterovirus   \n",
       "20     [-0.1131427072750879, -0.0544868948377319]       Mastadenovirus   \n",
       "21   [-0.08025913272665164, -0.14481010034284833]          Phlebovirus   \n",
       "22   [-0.08419884436327682, -0.15372049262950502]      Betacoronavirus   \n",
       "23    [-0.0914687334765774, -0.15240020627687756]  Alphapapillomavirus   \n",
       "24     [-0.15095973134914736, 0.2835972227463461]           Alphavirus   \n",
       "25   [0.14059437300046487, -0.013575671511743789]       Paslahepevirus   \n",
       "26   [-0.09998302340018402, -0.06147814581121143]  Alphainfluenzavirus   \n",
       "27  [-0.009963610897503824, -0.12083176876865355]     Orthopneumovirus   \n",
       "28     [0.16842921051342608, 0.05078886470451921]          Hepacivirus   \n",
       "29    [-0.14835142474231863, 0.26215215713229767]       Bocaparvovirus   \n",
       "\n",
       "   counts_non_and_human  \n",
       "0            (117, 128)  \n",
       "1            (567, 640)  \n",
       "2            (547, 228)  \n",
       "3            (188, 218)  \n",
       "4            (124, 757)  \n",
       "5            (560, 106)  \n",
       "6            (367, 498)  \n",
       "7            (581, 110)  \n",
       "8           (762, 1644)  \n",
       "9            (211, 212)  \n",
       "10           (222, 200)  \n",
       "11         (1777, 2973)  \n",
       "12          (929, 2119)  \n",
       "13           (319, 159)  \n",
       "14           (269, 424)  \n",
       "15           (619, 528)  \n",
       "16           (301, 422)  \n",
       "17           (105, 263)  \n",
       "18           (135, 240)  \n",
       "19         (2880, 1510)  \n",
       "20           (686, 286)  \n",
       "21           (111, 265)  \n",
       "22           (138, 206)  \n",
       "23           (793, 189)  \n",
       "24           (539, 453)  \n",
       "25           (109, 110)  \n",
       "26           (234, 856)  \n",
       "27           (809, 629)  \n",
       "28           (581, 688)  \n",
       "29           (229, 230)  "
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "52ccdb5d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Hepacivirus hominis',\n",
       " 'Bacteriophage sp.',\n",
       " 'Enterovirus betacoxsackie',\n",
       " 'Alphainfluenzavirus influenzae',\n",
       " 'Betainfluenzavirus influenzae',\n",
       " 'Rotavirus alphagastroenteritidis',\n",
       " 'Enterovirus coxsackiepol',\n",
       " 'Paslahepevirus balayani',\n",
       " 'Orthoflavivirus denguei',\n",
       " 'Norovirus norwalkense',\n",
       " 'Cytomegalovirus humanbeta5',\n",
       " 'Caudoviricetes sp.',\n",
       " 'Simian-Human immunodeficiency virus',\n",
       " 'Orthopneumovirus hominis',\n",
       " 'Deltaretrovirus priTlym1',\n",
       " 'Mammarenavirus lassaense',\n",
       " 'Alphapapillomavirus 9',\n",
       " 'Orthohepadnavirus hominoidei',\n",
       " 'Enterovirus alphacoxsackie',\n",
       " 'Lentivirus humimdef1']"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "common_species"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "818ab6a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "avg_human = np.array(embedds).mean(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7c514c8f",
   "metadata": {},
   "outputs": [],
   "source": [
    "alphainfluenze = metadata[metadata['Species'].str.contains('Alphainfluenzavirus')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "517dfeae",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/57 [00:00<?, ?it/s]/tmp/ipykernel_3368205/245567597.py:12: RuntimeWarning: Mean of empty slice.\n",
      "  embeds_mean = embeds.mean(axis=0).flatten()\n",
      "/user/hassanahmed.hassan/u21055/.project/dir.project/towards_better_genomic_models/.venv/lib/python3.12/site-packages/numpy/_core/_methods.py:144: RuntimeWarning: invalid value encountered in scalar divide\n",
      "  ret = ret.dtype.type(ret / rcount)\n",
      "100%|██████████| 57/57 [00:00<00:00, 1547.71it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(512,)\n",
      "Unexpected shape: (1,), (array([nan]), array([], dtype=float64), None)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n",
      "(512,)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "# select only one specie for each human host\n",
    "\n",
    "onlyhumans = alphainfluenze[~(alphainfluenze['Host'] == 'Homo sapiens')]\n",
    "species = onlyhumans['Host'].unique()\n",
    "embedds = []\n",
    "for specie in tqdm(species):\n",
    "    embeds = onlyhumans[onlyhumans['Host'] == specie]['embeddings'].tolist()\n",
    "    embeds = np.array(embeds)\n",
    "    if embeds.shape[0] == 1:\n",
    "        embeds_mean = embeds.flatten()\n",
    "    else:\n",
    "        embeds_mean = embeds.mean(axis=0).flatten()\n",
    "    if embeds_mean.shape[0] != 512:\n",
    "        print(f\"Unexpected shape: {embeds_mean.shape}, {embeds_mean, embeds, specie}\")\n",
    "        continue\n",
    "\n",
    "    print(embeds_mean.shape)\n",
    "    embedds.append(embeds_mean.tolist())\n",
    "    # break\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "e48a3958",
   "metadata": {},
   "outputs": [],
   "source": [
    "avg_influnze_no_human = np.array(embedds).mean(axis=0).tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2bd9174e",
   "metadata": {},
   "outputs": [],
   "source": [
    "avg_influnze_human = embedds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "b3b184ad",
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'list' object has no attribute 'shape'",
     "output_type": "error",
     "traceback": [
      "\u001b[31m---------------------------------------------------------------------------\u001b[39m",
      "\u001b[31mAttributeError\u001b[39m                            Traceback (most recent call last)",
      "\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[40]\u001b[39m\u001b[32m, line 1\u001b[39m\n\u001b[32m----> \u001b[39m\u001b[32m1\u001b[39m \u001b[43mavg_influnze_human\u001b[49m\u001b[43m.\u001b[49m\u001b[43mshape\u001b[49m\n",
      "\u001b[31mAttributeError\u001b[39m: 'list' object has no attribute 'shape'"
     ]
    }
   ],
   "source": [
    "avg_influnze_human.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d13070d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.99944964]])"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "\n",
    "cosine_similarity(np.array(avg_influnze_human[0]).reshape(1,512), np.array(avg_influnze_no_human).reshape(1,512))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "2aa19c7b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.09784692]])"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cosine_similarity(humans, avg_human.reshape(1,512))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "e3a6f928",
   "metadata": {},
   "outputs": [],
   "source": [
    "humans = np.array(avg_influnze_human[0]).reshape(1,512) - np.array(avg_influnze_no_human).reshape(1,512)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "3ed5a35b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "512"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(avg_influnze_human[0])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "1858ebef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.        , 0.        ],\n",
       "       [0.57735027, 0.81649658]])"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "X = [[0, 0, 0], [1, 1, 1]]\n",
    "Y = [[1, 0, 0], [1, 1, 0]]\n",
    "cosine_similarity(X, Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "a69e16f3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9774370059258591"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "distance.euclidean(avg_influnze_human[0], nnn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "42092821",
   "metadata": {},
   "outputs": [],
   "source": [
    "nnn = avg_influnze_no_human + (0.01 *avg_human)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "903c0ed7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1.3627573605786023,\n",
       " -0.9745412524948772,\n",
       " -0.4468247321330839,\n",
       " 1.1175651794824846,\n",
       " 0.2975374684454157,\n",
       " -0.7778587546358761,\n",
       " -0.6518397906269783,\n",
       " 0.1737956221897128,\n",
       " -1.2056968540717394,\n",
       " 0.6302159123667954,\n",
       " -0.1529835437334433,\n",
       " -0.4460651825508501,\n",
       " 2.516385886913691,\n",
       " -0.87882979417968,\n",
       " -0.0351385485752024,\n",
       " 1.197892635041832,\n",
       " -0.9733002715640597,\n",
       " -1.998334291144314,\n",
       " -0.22094950834254368,\n",
       " 0.3893189242420097,\n",
       " -0.9293508761458926,\n",
       " -2.1001845037835274,\n",
       " -0.2651567118586654,\n",
       " -0.7607102122826453,\n",
       " -0.8007213259513817,\n",
       " -1.035428555857422,\n",
       " 0.029184411384954326,\n",
       " 1.2200952398980784,\n",
       " -0.7879050489928987,\n",
       " 0.04038690966673386,\n",
       " -0.08236495739607236,\n",
       " 2.484253864512484,\n",
       " -0.5935717369628768,\n",
       " -0.0618778196711722,\n",
       " -0.8206673348561312,\n",
       " 1.110784141299052,\n",
       " -0.5124204857354491,\n",
       " 0.3460729581017334,\n",
       " -1.2130683861736558,\n",
       " -1.9757098000273745,\n",
       " 0.8437188117422609,\n",
       " -0.7776825226779677,\n",
       " 2.926715990417024,\n",
       " 0.23537885382946613,\n",
       " 1.6096562880736132,\n",
       " 1.1843590051190467,\n",
       " -0.15501332788786137,\n",
       " -1.6140859654316535,\n",
       " -0.04085378125237806,\n",
       " 0.6090058832087069,\n",
       " -1.0765866479303083,\n",
       " 0.5878119785816242,\n",
       " -1.652558918692108,\n",
       " -0.6952533569091406,\n",
       " -2.749964021719419,\n",
       " 0.729868560965754,\n",
       " -2.215478027987684,\n",
       " -0.3236756168308859,\n",
       " 0.3656961973866002,\n",
       " -1.1456358004838993,\n",
       " -0.9729086456135807,\n",
       " -2.501770613030491,\n",
       " 1.6528889261759245,\n",
       " 0.5043957489582463,\n",
       " -0.35744560736333,\n",
       " -0.14706401354773815,\n",
       " -1.4780659125401423,\n",
       " 0.17336075979436183,\n",
       " -0.4857029653677685,\n",
       " 0.28474916086301333,\n",
       " -0.20850159052130568,\n",
       " -1.958379978298122,\n",
       " -0.19379714726690306,\n",
       " 0.991924596647931,\n",
       " -2.0764713908872032,\n",
       " -2.0574892265164952,\n",
       " -1.0668887888264453,\n",
       " -1.98441589388073,\n",
       " -0.4461051930124179,\n",
       " 0.8410063079024992,\n",
       " 1.7706944534921238,\n",
       " 1.233976136160712,\n",
       " -0.9071188771571869,\n",
       " 0.6921407698820798,\n",
       " -0.808091981925516,\n",
       " -1.374319730915575,\n",
       " -0.02942128892755136,\n",
       " 1.9011858059809759,\n",
       " 0.2569271052518907,\n",
       " -0.1277515574606657,\n",
       " 1.5400205015117288,\n",
       " -0.03965220031971271,\n",
       " -0.41580372603021115,\n",
       " 0.38095796217696154,\n",
       " 0.2567562628607067,\n",
       " -1.4532162908814912,\n",
       " 0.3226626563626222,\n",
       " -2.08609499839636,\n",
       " 2.131095840380742,\n",
       " -0.49250237681926823,\n",
       " 0.4156186787936932,\n",
       " 0.9898105526072347,\n",
       " 0.7115367969386598,\n",
       " 1.0522745283495667,\n",
       " -1.5458993267299783,\n",
       " 0.17390043494932386,\n",
       " -1.2736639107903864,\n",
       " 1.8960224109837132,\n",
       " -1.8012487776259072,\n",
       " -0.26783868800808924,\n",
       " -0.265486659301031,\n",
       " -1.6106861921457143,\n",
       " 0.6803993369078535,\n",
       " -0.47644006283744833,\n",
       " 1.4445448189209669,\n",
       " 0.6184715647728015,\n",
       " 1.7613357977989392,\n",
       " 0.9556965079063025,\n",
       " -2.7792981047915597,\n",
       " -1.2657244870295892,\n",
       " -0.8832192583980724,\n",
       " 2.113070318332085,\n",
       " 0.29635316758758873,\n",
       " 0.37624540319069266,\n",
       " 0.29532876109274536,\n",
       " -0.30916038539129287,\n",
       " 0.4886771662112994,\n",
       " 1.6856776317979536,\n",
       " -2.7828762617885556,\n",
       " 1.8722323000940502,\n",
       " 3.0132913344945664,\n",
       " 0.2003845483558173,\n",
       " 2.775234478151696,\n",
       " 0.43294958954947627,\n",
       " -0.8604102111779727,\n",
       " 0.4941705067315672,\n",
       " -0.37533931182617813,\n",
       " 0.21532632669823992,\n",
       " -0.513388719314184,\n",
       " -0.09244508898576212,\n",
       " 1.1422439635309398,\n",
       " -0.16899774758019345,\n",
       " -0.6924542597471139,\n",
       " 1.6988479632597704,\n",
       " -0.7599479678349618,\n",
       " 1.7566104268925822,\n",
       " 0.23766125040526828,\n",
       " 0.02981456918187605,\n",
       " -1.5156208383731353,\n",
       " 0.9674736330142388,\n",
       " -0.6874241406082088,\n",
       " 0.2410408611230017,\n",
       " 0.8945180559769655,\n",
       " 0.3884571067785892,\n",
       " 1.3368654006566756,\n",
       " 2.000280041470487,\n",
       " 0.4457839592877362,\n",
       " 1.3285671897933014,\n",
       " -1.4463525572393694,\n",
       " -2.2593776975941453,\n",
       " -1.9399715361432133,\n",
       " -1.4779185232443688,\n",
       " 2.202137020408598,\n",
       " 1.4451393892622402,\n",
       " -0.0442905338795987,\n",
       " 0.9238732108193585,\n",
       " -1.0780605248406403,\n",
       " 0.5167138125651922,\n",
       " -0.9824605014843818,\n",
       " 2.415674808188381,\n",
       " -0.2787562138659673,\n",
       " 0.9093983211578467,\n",
       " -0.5277982890351206,\n",
       " -1.0524461549571438,\n",
       " -0.47938556771757257,\n",
       " -2.198902201958192,\n",
       " 1.699965486669133,\n",
       " 0.49801525349418324,\n",
       " 0.8142799544028747,\n",
       " -1.30174119808735,\n",
       " 0.3288768952469636,\n",
       " -0.5393306008643575,\n",
       " 1.7312577911931226,\n",
       " -1.6787108840086522,\n",
       " 0.750037835067154,\n",
       " -0.23523496283600345,\n",
       " -0.5418383320156708,\n",
       " 2.3080511251066484,\n",
       " -1.4450848573802884,\n",
       " 1.776263519994214,\n",
       " 1.9687592438143542,\n",
       " -2.158936334980859,\n",
       " -1.5739310750594506,\n",
       " -2.030275168072464,\n",
       " 0.9062683346689256,\n",
       " 0.22817894699386296,\n",
       " 0.8834686410159637,\n",
       " 0.1622313321023308,\n",
       " 1.8562170608430846,\n",
       " -0.7373334357881138,\n",
       " 0.16418390077999973,\n",
       " 1.2780632170347066,\n",
       " 0.6165337536452163,\n",
       " 0.29082584065227457,\n",
       " 0.6197150713867612,\n",
       " -1.2634146998071263,\n",
       " 1.3672949343155592,\n",
       " 0.5614612941813265,\n",
       " -1.642670705787137,\n",
       " 0.31132268385949713,\n",
       " -0.6932299415477448,\n",
       " -0.04264322888557839,\n",
       " 1.2583827949487245,\n",
       " -0.900110768711465,\n",
       " 0.2386623015696915,\n",
       " -1.5526978388810768,\n",
       " 1.2449179345216506,\n",
       " 0.370233847727824,\n",
       " 0.45272338681239394,\n",
       " 0.8082948165953668,\n",
       " -0.06853985340602131,\n",
       " 0.04330588209554235,\n",
       " -0.4998553621017525,\n",
       " 0.3955826247190563,\n",
       " 0.6473818223000082,\n",
       " 1.0824575253531463,\n",
       " 0.4726707006398684,\n",
       " -1.3011661571824653,\n",
       " 2.8607510847923083,\n",
       " -0.4585846721425525,\n",
       " -2.202077627182007,\n",
       " 0.41538409949240523,\n",
       " 0.9819123189673464,\n",
       " 0.3539121926276602,\n",
       " -1.3078424703871083,\n",
       " 0.10160898184528153,\n",
       " -3.1148358257407818,\n",
       " -1.512262585071417,\n",
       " -0.5321600473462007,\n",
       " 1.4481977231991596,\n",
       " -0.47698458265831584,\n",
       " -0.8438973491772627,\n",
       " 0.5107290798909644,\n",
       " 1.8689966038760977,\n",
       " 0.06059733594361788,\n",
       " 0.3495596916749118,\n",
       " -1.4753885460205567,\n",
       " -1.3651550565010462,\n",
       " 1.6744828939947307,\n",
       " -0.6465085604442999,\n",
       " -1.1658631612857182,\n",
       " 0.49267339655476755,\n",
       " 1.0277826470744789,\n",
       " 1.46519733290387,\n",
       " -2.161928645056537,\n",
       " -0.0016848365500127561,\n",
       " -1.2978617938665242,\n",
       " 2.033262483075134,\n",
       " 0.7658507639908383,\n",
       " 0.41878896813967037,\n",
       " 1.1559490421388903,\n",
       " 1.3209740297916608,\n",
       " -1.576380881998274,\n",
       " 0.327681554433627,\n",
       " 1.3836237156493032,\n",
       " -0.34306330708627647,\n",
       " 0.6114512453349228,\n",
       " 2.1185314283411727,\n",
       " -1.030139688243214,\n",
       " -1.9033369686868455,\n",
       " -0.6065800761183103,\n",
       " -0.47680766000133806,\n",
       " -0.628791354278214,\n",
       " -2.0884603542140407,\n",
       " 0.5642455012625099,\n",
       " 1.7093813383681142,\n",
       " -0.9195322106536638,\n",
       " 0.16848230806356454,\n",
       " -1.8949475502356505,\n",
       " 1.7586180954916864,\n",
       " -0.014917487327335801,\n",
       " -1.0060581734292529,\n",
       " 3.726681635929988,\n",
       " 1.7264997551583836,\n",
       " 1.2852643162776263,\n",
       " -1.325273780995964,\n",
       " 0.8114002825867417,\n",
       " -0.23764681563022125,\n",
       " -0.4087429160538774,\n",
       " 1.0198221117512793,\n",
       " -1.1288801574299479,\n",
       " -1.1893825933464572,\n",
       " -0.15431532900159559,\n",
       " 0.873893888714986,\n",
       " 0.22455554461090738,\n",
       " 0.7103287817066551,\n",
       " 0.9883729550573561,\n",
       " 0.8137837124940677,\n",
       " 3.011957804361979,\n",
       " 2.623529715415759,\n",
       " 0.45570650862322915,\n",
       " -0.40538285007206804,\n",
       " 0.6625649777487812,\n",
       " 2.981163956161238,\n",
       " 0.9668534577657015,\n",
       " -0.6278867112775134,\n",
       " -1.5220091747931945,\n",
       " 0.2684403389153015,\n",
       " -0.5473947875971834,\n",
       " 1.0931975268400633,\n",
       " 0.943502191040251,\n",
       " -1.1486690757621048,\n",
       " -0.29932659658895405,\n",
       " -0.06033799674848998,\n",
       " -1.0010653297361145,\n",
       " -0.3878619592700504,\n",
       " -0.7823589796948637,\n",
       " -2.237660378981859,\n",
       " 0.4031511309266122,\n",
       " 1.8180760537457263,\n",
       " 0.7667759411896651,\n",
       " 1.501807747488348,\n",
       " -0.020471969043784257,\n",
       " -0.8839007589297417,\n",
       " -1.241484182003217,\n",
       " 0.9908623973019103,\n",
       " -0.8103834013016815,\n",
       " -0.46586314164118,\n",
       " -2.5743398564493556,\n",
       " -0.72135897531596,\n",
       " 2.7488120585425286,\n",
       " -1.0953830200382786,\n",
       " 1.4812416948823848,\n",
       " -1.894158435682965,\n",
       " 1.3382933718017025,\n",
       " -0.9633357722280372,\n",
       " -0.035489595152199384,\n",
       " 2.6483734713660345,\n",
       " 0.9332997717409053,\n",
       " -0.7930425358378989,\n",
       " -1.7430142969147773,\n",
       " 2.357573406309144,\n",
       " -0.09533807756100646,\n",
       " 0.3497626791493251,\n",
       " 1.4573625216626713,\n",
       " -0.39844233657305056,\n",
       " -0.7912093157696928,\n",
       " 2.286486305232741,\n",
       " 0.6243517686668624,\n",
       " 0.10884455323883961,\n",
       " 0.6503572730649995,\n",
       " -0.9941733445112522,\n",
       " 2.2997496026193995,\n",
       " -2.2075220804948072,\n",
       " -0.43393356413043177,\n",
       " 0.44055224705137247,\n",
       " 1.714541886606787,\n",
       " 1.5382257592983735,\n",
       " -0.20859961868669244,\n",
       " 1.6397563422846997,\n",
       " -1.8552136563847208,\n",
       " 0.6270430751590648,\n",
       " 0.4244950498925904,\n",
       " -2.284899719759949,\n",
       " 1.425712739300524,\n",
       " -0.20067195746546182,\n",
       " 0.9112770224993045,\n",
       " 1.3486824359139826,\n",
       " 1.0019614803485382,\n",
       " 1.0122569595646655,\n",
       " -1.433178243983505,\n",
       " -1.3544248309400346,\n",
       " -1.464428922559461,\n",
       " 2.473477246924343,\n",
       " -0.4488412929682905,\n",
       " -1.3324663394536727,\n",
       " -1.0225025611555474,\n",
       " 0.4837332077515431,\n",
       " -0.100630749306546,\n",
       " -3.689639560177795,\n",
       " -0.7582542638047638,\n",
       " -1.056179077961506,\n",
       " 0.07523621652329054,\n",
       " 0.24622889569936654,\n",
       " 1.0899118877883651,\n",
       " -0.4883857925032449,\n",
       " -1.0091816621203709,\n",
       " 1.2092781107649844,\n",
       " -1.6637767996543493,\n",
       " -0.5425305487806948,\n",
       " -0.5573253840462774,\n",
       " -0.5737289358408023,\n",
       " -2.0878806623638186,\n",
       " -0.27147358398223853,\n",
       " -0.10539957610233568,\n",
       " -0.7847973911935447,\n",
       " 0.25737896396278037,\n",
       " -2.2154030754015994,\n",
       " 0.710455367835159,\n",
       " 2.892098610217755,\n",
       " -0.9023872252203461,\n",
       " -1.3715867397622166,\n",
       " -2.3040411991950793,\n",
       " -0.9975669873066437,\n",
       " -1.8401759745728257,\n",
       " -0.7759538544421523,\n",
       " -1.185163624266274,\n",
       " -1.2273436955407135,\n",
       " -1.2656523687207801,\n",
       " -1.6248663419332259,\n",
       " 1.0709720144414494,\n",
       " 1.345171080949979,\n",
       " 0.16558542999470782,\n",
       " -0.6427128296504673,\n",
       " 0.7951949674349564,\n",
       " -1.2565240032143064,\n",
       " 2.5396370882661934,\n",
       " -0.1295853482659321,\n",
       " 0.7556541724464833,\n",
       " 3.163482532541976,\n",
       " 1.5891329381201003,\n",
       " 3.1873963969385524,\n",
       " 1.000617087396801,\n",
       " 0.33256880381208265,\n",
       " -0.9531904890751227,\n",
       " -0.30196333589414376,\n",
       " 1.46397203207016,\n",
       " 1.887232993912493,\n",
       " 0.6134385593935975,\n",
       " -1.7801752304419494,\n",
       " 0.06030666211626938,\n",
       " -1.2902167872485952,\n",
       " -1.4634509280196621,\n",
       " -0.621709981162109,\n",
       " 0.7728689941458213,\n",
       " 0.4673213458651852,\n",
       " -1.630576649792174,\n",
       " -0.9953014013858942,\n",
       " 0.9244751604194315,\n",
       " -0.43482020726570714,\n",
       " -0.6948951430555083,\n",
       " -1.2109204212315061,\n",
       " 0.4986119890330821,\n",
       " 0.16036274132265188,\n",
       " 2.056106370738429,\n",
       " 0.963291039578935,\n",
       " 1.4056170052952237,\n",
       " 0.1214276632186101,\n",
       " -0.10291075607976662,\n",
       " 1.293353547143121,\n",
       " 1.331257279102619,\n",
       " -1.0382056725330842,\n",
       " 1.1329310569498274,\n",
       " 2.191652073819413,\n",
       " -1.283078462904335,\n",
       " -0.21862288282857809,\n",
       " 0.18135949942924115,\n",
       " 0.13651835048047858,\n",
       " -0.29662502475349534,\n",
       " 1.090536742256238,\n",
       " 0.0017656288222745848,\n",
       " 1.3166355556911893,\n",
       " -1.3266946540938482,\n",
       " -1.5758376042557578,\n",
       " 4.30120840948871,\n",
       " 1.2409261093180404,\n",
       " -2.147565381649213,\n",
       " 0.32085175352553147,\n",
       " 0.9256364307724513,\n",
       " 3.32480963682517,\n",
       " -0.821211685354893,\n",
       " -1.4562621710137424,\n",
       " 1.8513060186663244,\n",
       " -0.041930621621720336,\n",
       " -1.3373341002525427,\n",
       " -1.4066898038244655,\n",
       " -0.009225590247810125,\n",
       " -1.9509349353293068,\n",
       " 0.6769769139524199,\n",
       " -0.28537431478890407,\n",
       " 0.4012915896068717,\n",
       " -1.1392147961335304,\n",
       " -0.4267871764739259,\n",
       " -0.019926783873532444,\n",
       " 0.7214261201075,\n",
       " 2.226633189070938,\n",
       " -1.520251724964533,\n",
       " -1.2652143828379803,\n",
       " -1.0714683630782316,\n",
       " 2.489250980890714,\n",
       " 0.42759139341119135,\n",
       " -0.9578790576794208,\n",
       " -2.8229558177483387,\n",
       " 0.39440190504726946,\n",
       " -0.15157083187124923,\n",
       " -0.6042534725533591,\n",
       " -0.29010283587993974,\n",
       " -2.3057845902239156,\n",
       " -0.5319382952064531,\n",
       " 1.3255272533139613,\n",
       " -1.3705512919487097,\n",
       " -0.041927855920905255,\n",
       " -1.6102203589219313,\n",
       " -1.2200027510651157,\n",
       " -1.1359950805831158,\n",
       " 0.3415090963622539,\n",
       " -0.025909008098065726,\n",
       " -2.538314772467328,\n",
       " -0.11031572007709461,\n",
       " 0.48399652853910613,\n",
       " 0.0789614228077002,\n",
       " 0.4043652080332773]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embedds[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d0514257",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1.09755653e+00, -7.20708990e-01, -3.93439434e-01,  1.09143486e+00,\n",
       "        2.92203880e-01, -7.41890458e-01, -4.46404766e-01,  1.35269066e-01,\n",
       "       -1.09095233e+00,  8.37994218e-01, -8.45538119e-02, -3.54715221e-01,\n",
       "        2.03748072e+00, -5.22993809e-01, -2.44071744e-01,  8.72565978e-01,\n",
       "       -7.90654944e-01, -1.65591921e+00, -3.90442703e-01,  2.24182824e-01,\n",
       "       -7.54146944e-01, -1.85286775e+00, -5.55226512e-02, -6.61903947e-01,\n",
       "       -3.18032436e-01, -7.80328580e-01, -1.12754988e-01,  1.06230692e+00,\n",
       "       -7.34309688e-01,  4.81425986e-02,  1.27005090e-01,  2.06635441e+00,\n",
       "       -3.81654707e-01,  9.91543972e-02, -6.34157406e-01,  9.44231430e-01,\n",
       "       -3.08989436e-01,  6.00587956e-01, -7.80740764e-01, -1.87759096e+00,\n",
       "        7.92722588e-01, -5.57848550e-01,  2.36547505e+00,  3.46443905e-01,\n",
       "        1.15331592e+00,  1.02738223e+00, -2.64705078e-01, -1.37379119e+00,\n",
       "       -3.48525138e-02,  6.01083066e-01, -9.22682729e-01,  3.25868898e-01,\n",
       "       -1.55350253e+00, -5.41420615e-01, -2.03260427e+00,  5.74776408e-01,\n",
       "       -1.73308350e+00, -4.01593507e-01,  2.71712314e-01, -8.33397091e-01,\n",
       "       -7.07733862e-01, -1.93923707e+00,  1.15415843e+00,  4.54818528e-01,\n",
       "       -2.00305938e-01, -2.40714388e-02, -1.24388114e+00, -4.15786892e-02,\n",
       "       -2.79586472e-01,  1.80489753e-01, -3.92091331e-01, -1.94055777e+00,\n",
       "       -3.11508432e-01,  8.81046500e-01, -1.52036135e+00, -1.73766760e+00,\n",
       "       -8.15432545e-01, -1.55469019e+00, -6.11579478e-01,  6.21991547e-01,\n",
       "        1.50877517e+00,  9.94181722e-01, -5.28416292e-01,  3.17427497e-01,\n",
       "       -5.14723580e-01, -1.12717293e+00, -8.16914507e-02,  1.51904440e+00,\n",
       "        2.74125110e-01, -1.86718739e-02,  1.38986861e+00, -7.24230072e-02,\n",
       "       -2.99767613e-01,  7.41698717e-02,  3.73695345e-01, -1.37074064e+00,\n",
       "        4.56435332e-01, -1.62704098e+00,  1.85736933e+00, -4.89388735e-01,\n",
       "        3.88896002e-01,  7.68534837e-01,  8.09072748e-01,  8.55763159e-01,\n",
       "       -9.19230913e-01,  1.57898449e-01, -8.63410719e-01,  1.67866075e+00,\n",
       "       -1.65898036e+00, -1.57607899e-01, -3.23502149e-01, -1.31682482e+00,\n",
       "        5.11495502e-01, -2.18902704e-01,  1.01503548e+00,  4.44365702e-01,\n",
       "        1.29273240e+00,  9.97329214e-01, -2.30355862e+00, -9.96084548e-01,\n",
       "       -6.22479764e-01,  1.81920109e+00,  3.32218816e-01,  6.15531298e-01,\n",
       "        1.85993156e-01, -1.39403675e-01,  4.71466569e-01,  1.42166216e+00,\n",
       "       -2.22047979e+00,  1.33941688e+00,  2.43303407e+00,  2.65177341e-01,\n",
       "        2.69142436e+00,  4.60717494e-01, -8.57712632e-01,  2.99065281e-01,\n",
       "       -3.52411353e-01,  5.49260055e-02, -4.77399105e-01, -4.39961614e-02,\n",
       "        8.39183691e-01, -1.21547607e-01, -4.42751812e-01,  1.20539731e+00,\n",
       "       -5.54875649e-01,  1.54588253e+00,  2.91707888e-01,  1.24716359e-01,\n",
       "       -1.39564657e+00,  7.50813866e-01, -5.20021034e-01, -8.10581647e-04,\n",
       "        8.28555301e-01,  3.37669002e-01,  9.05413055e-01,  1.62429891e+00,\n",
       "        2.69546036e-01,  1.20706019e+00, -1.11369732e+00, -1.60834362e+00,\n",
       "       -1.58350729e+00, -1.08797830e+00,  1.84992474e+00,  1.48338169e+00,\n",
       "       -1.10106838e-01,  5.42756384e-01, -1.00477998e+00,  4.51719949e-01,\n",
       "       -6.28736749e-01,  1.61036682e+00, -2.58702147e-01,  6.82872164e-01,\n",
       "       -5.93441236e-01, -9.37178271e-01, -3.26875061e-01, -1.71119116e+00,\n",
       "        1.38936187e+00,  3.81198540e-01,  7.88974079e-01, -1.22461791e+00,\n",
       "        1.79597787e-01, -4.85830275e-01,  1.25148436e+00, -1.36382723e+00,\n",
       "        4.74615322e-01, -1.67446583e-01, -2.29437830e-01,  1.69399595e+00,\n",
       "       -1.16448625e+00,  1.06467683e+00,  1.62527168e+00, -1.54389489e+00,\n",
       "       -1.25830394e+00, -1.74808814e+00,  7.81471427e-01,  3.46699155e-01,\n",
       "        8.38331348e-01, -7.63469161e-03,  1.33108107e+00, -5.40332682e-01,\n",
       "        3.68325705e-01,  1.04661738e+00,  5.63086386e-01,  2.80384756e-01,\n",
       "        4.45917057e-01, -9.60741659e-01,  9.57805854e-01,  4.60413276e-01,\n",
       "       -1.33750370e+00,  2.95359432e-01, -6.59865848e-01, -1.12071685e-01,\n",
       "        1.14577533e+00, -5.93939231e-01,  2.10911973e-01, -1.22278084e+00,\n",
       "        1.01069515e+00,  3.59149110e-01,  5.32375939e-01,  6.65446202e-01,\n",
       "       -1.03854895e-02, -7.99142653e-02, -4.59193289e-01,  4.27373157e-01,\n",
       "        4.89571587e-01,  7.97020208e-01,  3.54197603e-01, -1.02714413e+00,\n",
       "        2.21476560e+00, -2.15791166e-01, -1.74503050e+00,  3.78190094e-01,\n",
       "        8.52960140e-01,  3.52229299e-01, -1.14729974e+00,  1.98529534e-01,\n",
       "       -2.52460778e+00, -1.13254462e+00, -4.42678304e-01,  1.01767039e+00,\n",
       "       -4.78551259e-01, -8.18259603e-01,  2.96626872e-01,  1.73846540e+00,\n",
       "       -2.01481166e-01,  3.36005627e-01, -1.20739294e+00, -1.18297510e+00,\n",
       "        1.49807624e+00, -5.89299161e-01, -7.65577979e-01,  4.42035782e-01,\n",
       "        8.90957520e-01,  1.29448520e+00, -1.56056036e+00,  1.58621077e-01,\n",
       "       -1.05474814e+00,  1.45377865e+00,  4.62747541e-01,  1.14770427e-01,\n",
       "        9.61422143e-01,  1.00341620e+00, -1.33091751e+00,  1.67409967e-01,\n",
       "        8.03328081e-01, -4.20650243e-01,  3.04672587e-01,  1.40884769e+00,\n",
       "       -6.96871735e-01, -1.77186017e+00, -7.28445801e-01, -5.35905407e-01,\n",
       "       -4.79916483e-01, -1.55846428e+00,  5.30135441e-01,  1.30629818e+00,\n",
       "       -7.02884732e-01,  1.27520943e-01, -1.42561832e+00,  1.41101085e+00,\n",
       "        1.36588464e-01, -6.00956439e-01,  3.36505397e+00,  1.34635114e+00,\n",
       "        7.37145291e-01, -1.24267434e+00,  7.10853392e-01, -2.13074149e-01,\n",
       "       -2.96981404e-01,  8.40677545e-01, -9.80949709e-01, -8.76908000e-01,\n",
       "       -8.63526670e-02,  5.25560625e-01,  3.79063522e-01,  5.61986305e-01,\n",
       "        7.84582628e-01,  7.78921413e-01,  2.76859430e+00,  2.06864779e+00,\n",
       "        4.20129344e-01, -1.64871532e-01,  4.64628177e-01,  2.37093241e+00,\n",
       "        8.02165590e-01, -5.62769513e-01, -1.16214403e+00,  1.18089118e-01,\n",
       "       -3.63832709e-01,  8.11280233e-01,  8.44567521e-01, -8.51256673e-01,\n",
       "       -1.65358747e-01,  1.10774470e-01, -7.88561773e-01, -4.57825822e-01,\n",
       "       -7.02589743e-01, -1.87158468e+00,  2.13114000e-01,  1.67480127e+00,\n",
       "        8.53636856e-01,  1.28289857e+00, -2.98033599e-02, -6.19535262e-01,\n",
       "       -9.20349670e-01,  6.18862222e-01, -6.77288214e-01, -3.25415967e-01,\n",
       "       -2.11918645e+00, -4.41249889e-01,  2.22486239e+00, -8.67064483e-01,\n",
       "        1.21030352e+00, -1.91076570e+00,  8.59964646e-01, -8.07147217e-01,\n",
       "       -1.74745846e-01,  2.18703172e+00,  6.69839003e-01, -4.92733654e-01,\n",
       "       -1.35664839e+00,  1.89357394e+00,  6.92011174e-02,  2.35936984e-01,\n",
       "        1.12535810e+00, -3.47107435e-01, -4.94164763e-01,  1.81984282e+00,\n",
       "        4.36407932e-01,  1.12495144e-01,  5.93240274e-01, -8.28367739e-01,\n",
       "        1.90361740e+00, -1.52733823e+00, -3.57425204e-01,  4.33640878e-01,\n",
       "        1.33852337e+00,  1.15466368e+00,  2.27539474e-02,  1.54506173e+00,\n",
       "       -1.46765812e+00,  4.35340418e-01,  2.65575086e-01, -1.77108247e+00,\n",
       "        1.41813064e+00, -1.98171880e-01,  6.92095274e-01,  1.15197952e+00,\n",
       "        8.28151847e-01,  8.91841266e-01, -1.38797833e+00, -1.30439081e+00,\n",
       "       -8.22638778e-01,  1.94928146e+00, -3.29158058e-01, -9.68316415e-01,\n",
       "       -9.18548272e-01,  4.29291921e-01, -1.77105711e-01, -2.99868884e+00,\n",
       "       -4.75448036e-01, -9.26219983e-01, -1.33327197e-01,  1.83764098e-01,\n",
       "        8.99351920e-01, -3.57002241e-01, -7.13535482e-01,  8.16167261e-01,\n",
       "       -1.35236332e+00, -4.60192640e-01, -6.23824424e-01, -3.86573907e-01,\n",
       "       -1.65560977e+00, -4.10818248e-01,  6.67105027e-02, -7.05411156e-01,\n",
       "        1.51875667e-01, -1.65965019e+00,  6.50095383e-01,  2.31382682e+00,\n",
       "       -6.92839482e-01, -1.00863042e+00, -1.87151270e+00, -7.80752525e-01,\n",
       "       -1.50121707e+00, -5.93006701e-01, -1.07355653e+00, -1.03607119e+00,\n",
       "       -9.49449450e-01, -1.30035993e+00,  7.65090877e-01,  9.84387337e-01,\n",
       "        9.31530268e-02, -4.99935039e-01,  5.97424216e-01, -8.23266324e-01,\n",
       "        1.89464289e+00, -2.53164058e-01,  6.53536433e-01,  2.41480266e+00,\n",
       "        1.35481286e+00,  2.52268740e+00,  1.07986182e+00,  8.85925332e-02,\n",
       "       -9.45830966e-01, -1.43253623e-01,  1.00522136e+00,  1.60666639e+00,\n",
       "        2.75150315e-01, -1.62077485e+00,  2.52524925e-01, -1.16894831e+00,\n",
       "       -1.04894076e+00, -7.03620381e-01,  6.53793147e-01,  4.57517195e-01,\n",
       "       -1.50388881e+00, -6.38072597e-01,  8.50202872e-01, -1.62361052e-01,\n",
       "       -2.87724810e-01, -9.80838824e-01,  6.36486358e-01,  3.87623849e-02,\n",
       "        1.56400278e+00,  7.80265715e-01,  9.57815063e-01,  2.33295426e-02,\n",
       "       -5.12010919e-02,  9.16566681e-01,  1.21787145e+00, -9.31773235e-01,\n",
       "        8.55028557e-01,  1.81220231e+00, -1.01288619e+00, -4.98423098e-01,\n",
       "        2.17470901e-01, -1.40531805e-01, -3.59688667e-01,  8.78738758e-01,\n",
       "        1.62037840e-02,  1.19846591e+00, -1.02795739e+00, -1.56559994e+00,\n",
       "        3.67852632e+00,  8.82889925e-01, -1.69670736e+00,  3.15879594e-01,\n",
       "        8.05004465e-01,  2.74164855e+00, -7.08151303e-01, -1.23166343e+00,\n",
       "        1.48913103e+00, -6.92814829e-02, -1.09670702e+00, -1.21950781e+00,\n",
       "       -5.29306923e-02, -1.54066534e+00,  6.25857068e-01, -4.54476523e-01,\n",
       "        4.55351600e-01, -1.00524724e+00, -4.20399909e-01, -1.15677287e-02,\n",
       "        5.42396641e-01,  1.76856266e+00, -1.25680907e+00, -1.31512349e+00,\n",
       "       -6.35010488e-01,  1.95997631e+00,  2.38650896e-01, -7.03030122e-01,\n",
       "       -2.10248371e+00,  3.41805995e-01, -2.74037735e-01, -5.02486517e-01,\n",
       "       -2.66302417e-01, -1.71700966e+00, -3.71733217e-01,  9.52813499e-01,\n",
       "       -1.23031948e+00, -1.82940031e-02, -1.16243802e+00, -8.26715382e-01,\n",
       "       -8.14251097e-01,  2.38086818e-01, -2.02126086e-01, -2.21812409e+00,\n",
       "       -1.27318924e-01,  2.48984179e-01,  4.78491195e-03,  2.03663613e-01])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "avg_human"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5484df31",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "9d7a9614",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100000, 512)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embedddings = \"/user/hassanahmed.hassan/u21055/.project/dir.project/towards_better_genomic_models/data/ideal_tokens/index_search/data/data/embeddings.npy\"\n",
    "embeddings = np.load(embedddings)\n",
    "embeddings.shape"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}