6_AE_deep_metaclonotypes.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluate the performance and relevance of the deep AutoEncoder in detecting some essential structures for predicting similar sequences"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly as in the notebook *3_AE_relevance_metaclonotypes*, this notebook explores if the deep AutoEncoder is able to accurately detect some essential structures and patterns in the TCR sequences. This allows us to evaluate and determine if our deep model can cluster well and better TCR sequences with similar structures based on TCRdist."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-06-10 12:36:40.198975: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
     ]
    }
   ],
   "source": [
    "# Import all the necessary modules and functions from automate.py\n",
    "from automate import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load pre-processed data\n",
    "with (open(\"df_comparison.pkl\", \"rb\")) as openfile:\n",
    "    df = pickle.load(openfile)\n",
    "\n",
    "with (open(\"data.pkl\", \"rb\")) as openfile:\n",
    "    data = pickle.load(openfile)\n",
    "\n",
    "data[\"v_gene_one_hot\"]  = data[\"v_gene_one_hot\"].apply(lambda x: x.reshape(54))\n",
    "data[\"j_gene_one_hot\"]  = data[\"j_gene_one_hot\"].apply(lambda x: x.reshape(13))\n",
    "data['CDR3_al_one_hot'] = df['CDR3_al_integer']\n",
    "data = [np.stack(col.values) for _, col in data.items()]\n",
    "X_train = data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>CDR3</th>\n",
       "      <th>CDR3_al</th>\n",
       "      <th>tokenized_smiles</th>\n",
       "      <th>TCR BioIdentity</th>\n",
       "      <th>TCR Nucleotide Sequence</th>\n",
       "      <th>Experiment</th>\n",
       "      <th>ORF Coverage</th>\n",
       "      <th>Amino Acids</th>\n",
       "      <th>Start Index in Genome</th>\n",
       "      <th>End Index in Genome</th>\n",
       "      <th>v_gene</th>\n",
       "      <th>j_gene</th>\n",
       "      <th>Amino Acids 1</th>\n",
       "      <th>CDR3_al_one_hot</th>\n",
       "      <th>CDR3_al_integer</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CASSAQGTGDRGYTF</td>\n",
       "      <td>CASSA-----QGTGDRGYTF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 31, 27, 13]</td>\n",
       "      <td>CASSAQGTGDRGYTF+TCRBV27-01+TCRBJ01-02</td>\n",
       "      <td>GAGTCGCCCAGCCCCAACCAGACCTCTCTGTACTTCTGTGCCAGCA...</td>\n",
       "      <td>eAV93</td>\n",
       "      <td>ORF1ab,surface glycoprotein</td>\n",
       "      <td>ADAGFIKQY,AELEGIQY,LADAGFIKQY,TLADAGFIK</td>\n",
       "      <td>533</td>\n",
       "      <td>24073</td>\n",
       "      <td>TCRBV27-01</td>\n",
       "      <td>TCRBJ01-02</td>\n",
       "      <td>ADAGFIKQY</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 1, 0, 0, 0, 0, 0, 14, 6, 17, 6,...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CASSLVATGNTGELFF</td>\n",
       "      <td>CASSL----VATGNTGELFF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 23, 27, 27, 13]</td>\n",
       "      <td>CASSLVATGNTGELFF+TCRBV07-09+TCRBJ02-02</td>\n",
       "      <td>CGCACAGAGCAGGGGGACTCGGCCATGTATCTCTGTGCCAGCAGCT...</td>\n",
       "      <td>eOX56</td>\n",
       "      <td>ORF1ab,surface glycoprotein</td>\n",
       "      <td>ADAGFIKQY,AELEGIQY,LADAGFIKQY,TLADAGFIK</td>\n",
       "      <td>533</td>\n",
       "      <td>24073</td>\n",
       "      <td>TCRBV07-09</td>\n",
       "      <td>TCRBJ02-02</td>\n",
       "      <td>ADAGFIKQY</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 10, 0, 0, 0, 0, 18, 1, 17, 6, 1...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CALKVGADTQYF</td>\n",
       "      <td>CALKV--------GADTQYF</td>\n",
       "      <td>[12, 16, 31, 31, 31, 31, 31, 31, 31, 31, 27, 13]</td>\n",
       "      <td>CALKVGADTQYF+TCRBV30-01+TCRBJ02-03</td>\n",
       "      <td>CTGAGTTCTAAGAAGCTCCTTCTCAGTGACTCTGGCTTCTATCTCT...</td>\n",
       "      <td>eQD124</td>\n",
       "      <td>ORF1ab,surface glycoprotein</td>\n",
       "      <td>ADAGFIKQY,AELEGIQY,LADAGFIKQY,TLADAGFIK</td>\n",
       "      <td>533</td>\n",
       "      <td>24073</td>\n",
       "      <td>TCRBV30-01</td>\n",
       "      <td>TCRBJ02-03</td>\n",
       "      <td>ADAGFIKQY</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 10, 9, 18, 0, 0, 0, 0, 0, 0, 0, 0, 6, 1...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CASSLWASGRGGTGELFF</td>\n",
       "      <td>CASSLW--ASGRGGTGELFF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 34, 27, 27, 13]</td>\n",
       "      <td>CASSLWASGRGGTGELFF+TCRBV27-01+TCRBJ02-02</td>\n",
       "      <td>AGCCCCAACCAGACCTCTCTGTACTTCTGTGCCAGCAGTTTATGGG...</td>\n",
       "      <td>eAV93</td>\n",
       "      <td>ORF1ab,surface glycoprotein</td>\n",
       "      <td>ADAGFIKQY,AELEGIQY,LADAGFIKQY,TLADAGFIK</td>\n",
       "      <td>533</td>\n",
       "      <td>24073</td>\n",
       "      <td>TCRBV27-01</td>\n",
       "      <td>TCRBJ02-02</td>\n",
       "      <td>ADAGFIKQY</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 10, 19, 0, 0, 1, 16, 6, 15, 6, ...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CASSLLGWEQLDEQFF</td>\n",
       "      <td>CASSL----LGWEQLDEQFF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 27, 27, 13]</td>\n",
       "      <td>CASSLLGWEQLDEQFF+TCRBV27-01+TCRBJ02-01</td>\n",
       "      <td>TCGCCCAGCCCCAACCAGACCTCTCTGTACTTCTGTGCCAGCAGTT...</td>\n",
       "      <td>eMR16</td>\n",
       "      <td>ORF1ab,surface glycoprotein</td>\n",
       "      <td>ADAGFIKQY,AELEGIQY,LADAGFIKQY,TLADAGFIK</td>\n",
       "      <td>533</td>\n",
       "      <td>24073</td>\n",
       "      <td>TCRBV27-01</td>\n",
       "      <td>TCRBJ02-01</td>\n",
       "      <td>ADAGFIKQY</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 10, 0, 0, 0, 0, 10, 6, 19, 4, 1...</td>\n",
       "      <td>0</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",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236340</th>\n",
       "      <td>CASSSGPQETQYF</td>\n",
       "      <td>CASSS-------GPQETQYF</td>\n",
       "      <td>[12, 16, 34, 34, 34, 31, 31, 31, 31, 31, 31, 3...</td>\n",
       "      <td>CASSSGPQETQYF+TCRBV07-09+TCRBJ02-05</td>\n",
       "      <td>GAGATCCAGCGCACAGAGCAGGGGGACTCGGCCATGTATCTCTGTG...</td>\n",
       "      <td>eAV88</td>\n",
       "      <td>ORF10</td>\n",
       "      <td>AQVDVVNFNL,NYIAQVDVV</td>\n",
       "      <td>29630</td>\n",
       "      <td>29668</td>\n",
       "      <td>TCRBV07-09</td>\n",
       "      <td>TCRBJ02-05</td>\n",
       "      <td>AQVDVVNFNL</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 16, 0, 0, 0, 0, 0, 0, 0, 6, 13,...</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236341</th>\n",
       "      <td>CASSKRDSQETQYF</td>\n",
       "      <td>CASSK------RDSQETQYF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...</td>\n",
       "      <td>CASSKRDSQETQYF+TCRBV07-09+TCRBJ02-05</td>\n",
       "      <td>ATCCAGCGCACAGAGCAGGGGGACTCGGCCATGTATCTCTGTGCCA...</td>\n",
       "      <td>eDH105</td>\n",
       "      <td>ORF10</td>\n",
       "      <td>AQVDVVNFNL,NYIAQVDVV</td>\n",
       "      <td>29630</td>\n",
       "      <td>29668</td>\n",
       "      <td>TCRBV07-09</td>\n",
       "      <td>TCRBJ02-05</td>\n",
       "      <td>AQVDVVNFNL</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 9, 0, 0, 0, 0, 0, 0, 15, 3, 16,...</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236342</th>\n",
       "      <td>CASSQLPGQGKVQYF</td>\n",
       "      <td>CASSQ-----LPGQGKVQYF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 31, 45, 27, 13]</td>\n",
       "      <td>CASSQLPGQGKVQYF+TCRBV04-01+TCRBJ02-07</td>\n",
       "      <td>CACGCCCTGCAGCCAGAAGACTCAGCCCTGTATCTCTGCGCCAGCA...</td>\n",
       "      <td>eAV93</td>\n",
       "      <td>ORF10</td>\n",
       "      <td>AQVDVVNFNL,NYIAQVDVV</td>\n",
       "      <td>29630</td>\n",
       "      <td>29668</td>\n",
       "      <td>TCRBV04-01</td>\n",
       "      <td>TCRBJ02-07</td>\n",
       "      <td>AQVDVVNFNL</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 14, 0, 0, 0, 0, 0, 10, 13, 6, 1...</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236343</th>\n",
       "      <td>CASSPGDNEQFF</td>\n",
       "      <td>CASSP--------GDNEQFF</td>\n",
       "      <td>[12, 16, 34, 34, 45, 31, 31, 31, 31, 31, 31, 3...</td>\n",
       "      <td>CASSPGDNEQFF+TCRBV09-01+TCRBJ02-01</td>\n",
       "      <td>CTAAACCTGAGCTCTCTGGAGCTGGGGGACTCAGCTTTGTATTTCT...</td>\n",
       "      <td>eEE224</td>\n",
       "      <td>ORF10</td>\n",
       "      <td>AQVDVVNFNL,NYIAQVDVV</td>\n",
       "      <td>29630</td>\n",
       "      <td>29668</td>\n",
       "      <td>TCRBV09-01</td>\n",
       "      <td>TCRBJ02-01</td>\n",
       "      <td>AQVDVVNFNL</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 13, 0, 0, 0, 0, 0, 0, 0, 0, 6, ...</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236344</th>\n",
       "      <td>CASSVGDPDTQYF</td>\n",
       "      <td>CASSV-------GDPDTQYF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 31, 4...</td>\n",
       "      <td>CASSVGDPDTQYF+TCRBV09-01+TCRBJ02-03</td>\n",
       "      <td>AACCTGAGCTCTCTGGAGCTGGGGGACTCAGCTTTGTATTTCTGTG...</td>\n",
       "      <td>eAV88</td>\n",
       "      <td>ORF10</td>\n",
       "      <td>AQVDVVNFNL,NYIAQVDVV</td>\n",
       "      <td>29630</td>\n",
       "      <td>29668</td>\n",
       "      <td>TCRBV09-01</td>\n",
       "      <td>TCRBJ02-03</td>\n",
       "      <td>AQVDVVNFNL</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 18, 0, 0, 0, 0, 0, 0, 0, 6, 3, ...</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>236345 rows × 16 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      CDR3               CDR3_al  \\\n",
       "0          CASSAQGTGDRGYTF  CASSA-----QGTGDRGYTF   \n",
       "1         CASSLVATGNTGELFF  CASSL----VATGNTGELFF   \n",
       "2             CALKVGADTQYF  CALKV--------GADTQYF   \n",
       "3       CASSLWASGRGGTGELFF  CASSLW--ASGRGGTGELFF   \n",
       "4         CASSLLGWEQLDEQFF  CASSL----LGWEQLDEQFF   \n",
       "...                    ...                   ...   \n",
       "236340       CASSSGPQETQYF  CASSS-------GPQETQYF   \n",
       "236341      CASSKRDSQETQYF  CASSK------RDSQETQYF   \n",
       "236342     CASSQLPGQGKVQYF  CASSQ-----LPGQGKVQYF   \n",
       "236343        CASSPGDNEQFF  CASSP--------GDNEQFF   \n",
       "236344       CASSVGDPDTQYF  CASSV-------GDPDTQYF   \n",
       "\n",
       "                                         tokenized_smiles  \\\n",
       "0            [12, 16, 34, 34, 31, 31, 31, 31, 31, 27, 13]   \n",
       "1        [12, 16, 34, 34, 31, 31, 31, 31, 23, 27, 27, 13]   \n",
       "2        [12, 16, 31, 31, 31, 31, 31, 31, 31, 31, 27, 13]   \n",
       "3                [12, 16, 34, 34, 31, 31, 34, 27, 27, 13]   \n",
       "4            [12, 16, 34, 34, 31, 31, 31, 31, 27, 27, 13]   \n",
       "...                                                   ...   \n",
       "236340  [12, 16, 34, 34, 34, 31, 31, 31, 31, 31, 31, 3...   \n",
       "236341  [12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...   \n",
       "236342   [12, 16, 34, 34, 31, 31, 31, 31, 31, 45, 27, 13]   \n",
       "236343  [12, 16, 34, 34, 45, 31, 31, 31, 31, 31, 31, 3...   \n",
       "236344  [12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 31, 4...   \n",
       "\n",
       "                                 TCR BioIdentity  \\\n",
       "0          CASSAQGTGDRGYTF+TCRBV27-01+TCRBJ01-02   \n",
       "1         CASSLVATGNTGELFF+TCRBV07-09+TCRBJ02-02   \n",
       "2             CALKVGADTQYF+TCRBV30-01+TCRBJ02-03   \n",
       "3       CASSLWASGRGGTGELFF+TCRBV27-01+TCRBJ02-02   \n",
       "4         CASSLLGWEQLDEQFF+TCRBV27-01+TCRBJ02-01   \n",
       "...                                          ...   \n",
       "236340       CASSSGPQETQYF+TCRBV07-09+TCRBJ02-05   \n",
       "236341      CASSKRDSQETQYF+TCRBV07-09+TCRBJ02-05   \n",
       "236342     CASSQLPGQGKVQYF+TCRBV04-01+TCRBJ02-07   \n",
       "236343        CASSPGDNEQFF+TCRBV09-01+TCRBJ02-01   \n",
       "236344       CASSVGDPDTQYF+TCRBV09-01+TCRBJ02-03   \n",
       "\n",
       "                                  TCR Nucleotide Sequence Experiment  \\\n",
       "0       GAGTCGCCCAGCCCCAACCAGACCTCTCTGTACTTCTGTGCCAGCA...      eAV93   \n",
       "1       CGCACAGAGCAGGGGGACTCGGCCATGTATCTCTGTGCCAGCAGCT...      eOX56   \n",
       "2       CTGAGTTCTAAGAAGCTCCTTCTCAGTGACTCTGGCTTCTATCTCT...     eQD124   \n",
       "3       AGCCCCAACCAGACCTCTCTGTACTTCTGTGCCAGCAGTTTATGGG...      eAV93   \n",
       "4       TCGCCCAGCCCCAACCAGACCTCTCTGTACTTCTGTGCCAGCAGTT...      eMR16   \n",
       "...                                                   ...        ...   \n",
       "236340  GAGATCCAGCGCACAGAGCAGGGGGACTCGGCCATGTATCTCTGTG...      eAV88   \n",
       "236341  ATCCAGCGCACAGAGCAGGGGGACTCGGCCATGTATCTCTGTGCCA...     eDH105   \n",
       "236342  CACGCCCTGCAGCCAGAAGACTCAGCCCTGTATCTCTGCGCCAGCA...      eAV93   \n",
       "236343  CTAAACCTGAGCTCTCTGGAGCTGGGGGACTCAGCTTTGTATTTCT...     eEE224   \n",
       "236344  AACCTGAGCTCTCTGGAGCTGGGGGACTCAGCTTTGTATTTCTGTG...      eAV88   \n",
       "\n",
       "                       ORF Coverage                              Amino Acids  \\\n",
       "0       ORF1ab,surface glycoprotein  ADAGFIKQY,AELEGIQY,LADAGFIKQY,TLADAGFIK   \n",
       "1       ORF1ab,surface glycoprotein  ADAGFIKQY,AELEGIQY,LADAGFIKQY,TLADAGFIK   \n",
       "2       ORF1ab,surface glycoprotein  ADAGFIKQY,AELEGIQY,LADAGFIKQY,TLADAGFIK   \n",
       "3       ORF1ab,surface glycoprotein  ADAGFIKQY,AELEGIQY,LADAGFIKQY,TLADAGFIK   \n",
       "4       ORF1ab,surface glycoprotein  ADAGFIKQY,AELEGIQY,LADAGFIKQY,TLADAGFIK   \n",
       "...                             ...                                      ...   \n",
       "236340                        ORF10                     AQVDVVNFNL,NYIAQVDVV   \n",
       "236341                        ORF10                     AQVDVVNFNL,NYIAQVDVV   \n",
       "236342                        ORF10                     AQVDVVNFNL,NYIAQVDVV   \n",
       "236343                        ORF10                     AQVDVVNFNL,NYIAQVDVV   \n",
       "236344                        ORF10                     AQVDVVNFNL,NYIAQVDVV   \n",
       "\n",
       "        Start Index in Genome  End Index in Genome      v_gene      j_gene  \\\n",
       "0                         533                24073  TCRBV27-01  TCRBJ01-02   \n",
       "1                         533                24073  TCRBV07-09  TCRBJ02-02   \n",
       "2                         533                24073  TCRBV30-01  TCRBJ02-03   \n",
       "3                         533                24073  TCRBV27-01  TCRBJ02-02   \n",
       "4                         533                24073  TCRBV27-01  TCRBJ02-01   \n",
       "...                       ...                  ...         ...         ...   \n",
       "236340                  29630                29668  TCRBV07-09  TCRBJ02-05   \n",
       "236341                  29630                29668  TCRBV07-09  TCRBJ02-05   \n",
       "236342                  29630                29668  TCRBV04-01  TCRBJ02-07   \n",
       "236343                  29630                29668  TCRBV09-01  TCRBJ02-01   \n",
       "236344                  29630                29668  TCRBV09-01  TCRBJ02-03   \n",
       "\n",
       "       Amino Acids 1                                    CDR3_al_one_hot  \\\n",
       "0          ADAGFIKQY  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "1          ADAGFIKQY  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "2          ADAGFIKQY  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "3          ADAGFIKQY  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "4          ADAGFIKQY  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "...              ...                                                ...   \n",
       "236340    AQVDVVNFNL  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "236341    AQVDVVNFNL  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "236342    AQVDVVNFNL  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "236343    AQVDVVNFNL  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "236344    AQVDVVNFNL  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "\n",
       "                                          CDR3_al_integer  label  \n",
       "0       [2, 1, 16, 16, 1, 0, 0, 0, 0, 0, 14, 6, 17, 6,...      0  \n",
       "1       [2, 1, 16, 16, 10, 0, 0, 0, 0, 18, 1, 17, 6, 1...      0  \n",
       "2       [2, 1, 10, 9, 18, 0, 0, 0, 0, 0, 0, 0, 0, 6, 1...      0  \n",
       "3       [2, 1, 16, 16, 10, 19, 0, 0, 1, 16, 6, 15, 6, ...      0  \n",
       "4       [2, 1, 16, 16, 10, 0, 0, 0, 0, 10, 6, 19, 4, 1...      0  \n",
       "...                                                   ...    ...  \n",
       "236340  [2, 1, 16, 16, 16, 0, 0, 0, 0, 0, 0, 0, 6, 13,...     28  \n",
       "236341  [2, 1, 16, 16, 9, 0, 0, 0, 0, 0, 0, 15, 3, 16,...     28  \n",
       "236342  [2, 1, 16, 16, 14, 0, 0, 0, 0, 0, 10, 13, 6, 1...     28  \n",
       "236343  [2, 1, 16, 16, 13, 0, 0, 0, 0, 0, 0, 0, 0, 6, ...     28  \n",
       "236344  [2, 1, 16, 16, 18, 0, 0, 0, 0, 0, 0, 0, 6, 3, ...     28  \n",
       "\n",
       "[236345 rows x 16 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_8\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " cdr3_input (InputLayer)        [(None, 20)]         0           []                               \n",
      "                                                                                                  \n",
      " dense_19 (Dense)               (None, 40)           840         ['cdr3_input[0][0]']             \n",
      "                                                                                                  \n",
      " batch_normalization_12 (BatchN  (None, 40)          160         ['dense_19[0][0]']               \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " v_gene_input (InputLayer)      [(None, 54)]         0           []                               \n",
      "                                                                                                  \n",
      " j_gene_input (InputLayer)      [(None, 13)]         0           []                               \n",
      "                                                                                                  \n",
      " re_lu_12 (ReLU)                (None, 40)           0           ['batch_normalization_12[0][0]'] \n",
      "                                                                                                  \n",
      " v_gene_embedding (Dense)       (None, 54)           2970        ['v_gene_input[0][0]']           \n",
      "                                                                                                  \n",
      " j_gene_embedding (Dense)       (None, 13)           182         ['j_gene_input[0][0]']           \n",
      "                                                                                                  \n",
      " merged_embedding (Concatenate)  (None, 107)         0           ['re_lu_12[0][0]',               \n",
      "                                                                  'v_gene_embedding[0][0]',       \n",
      "                                                                  'j_gene_embedding[0][0]']       \n",
      "                                                                                                  \n",
      " encoder_dense_1 (Dense)        (None, 75)           8100        ['merged_embedding[0][0]']       \n",
      "                                                                                                  \n",
      " encoder_dense_2 (Dense)        (None, 75)           5700        ['encoder_dense_1[0][0]']        \n",
      "                                                                                                  \n",
      " bottleneck (Dense)             (None, 20)           1520        ['encoder_dense_2[0][0]']        \n",
      "                                                                                                  \n",
      " dense_20 (Dense)               (None, 40)           840         ['bottleneck[0][0]']             \n",
      "                                                                                                  \n",
      " batch_normalization_13 (BatchN  (None, 40)          160         ['dense_20[0][0]']               \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " re_lu_13 (ReLU)                (None, 40)           0           ['batch_normalization_13[0][0]'] \n",
      "                                                                                                  \n",
      " dense_21 (Dense)               (None, 20)           820         ['re_lu_13[0][0]']               \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 21,292\n",
      "Trainable params: 21,132\n",
      "Non-trainable params: 160\n",
      "__________________________________________________________________________________________________\n",
      "None\n",
      "Epoch 1/10\n",
      "3693/3693 [==============================] - 6s 1ms/step - loss: 6.4020\n",
      "Epoch 2/10\n",
      "3693/3693 [==============================] - 5s 1ms/step - loss: 0.7217\n",
      "Epoch 3/10\n",
      "3693/3693 [==============================] - 5s 1ms/step - loss: 0.5908\n",
      "Epoch 4/10\n",
      "3693/3693 [==============================] - 5s 1ms/step - loss: 0.5236\n",
      "Epoch 5/10\n",
      "3693/3693 [==============================] - 5s 1ms/step - loss: 0.4862\n",
      "Epoch 6/10\n",
      "3693/3693 [==============================] - 5s 1ms/step - loss: 0.4507\n",
      "Epoch 7/10\n",
      "3693/3693 [==============================] - 5s 1ms/step - loss: 0.4309\n",
      "Epoch 8/10\n",
      "3693/3693 [==============================] - 5s 1ms/step - loss: 0.3930\n",
      "Epoch 9/10\n",
      "3693/3693 [==============================] - 5s 1ms/step - loss: 0.3648\n",
      "Epoch 10/10\n",
      "3693/3693 [==============================] - 5s 1ms/step - loss: 0.3504\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Setting the parameters for model building and training of the deep AE\n",
    "params = {\n",
    "            \"latent_dim\": 20,\n",
    "            \"dense_nodes\": 75,\n",
    "            \"aa_embedding_dim\": 21,\n",
    "            \"v_gene_embedding_dim\": 54,\n",
    "            \"j_gene_embedding_dim\": 13,\n",
    "            \"beta\": 0.2,\n",
    "            \"max_cdr3_len\": 20,\n",
    "            \"n_aas\": 21,\n",
    "            \"n_v_genes\": 54,\n",
    "            \"n_j_genes\": 13,\n",
    "            \"stopping_monitor\": \"val_loss\",\n",
    "            \"batch_size\": 100,\n",
    "            \"pretrains\": 2,\n",
    "            \"warmup_period\": 3,\n",
    "            \"epochs\": 10,\n",
    "            \"patience\": 20,\n",
    "            \"n_inputs\" : 20,\n",
    "            \"v_inputs\" : 54,\n",
    "            \"j_inputs\" : 13\n",
    "        }\n",
    "model, encoder, decoder = train_model_complex(df, X_train, params=params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reload data\n",
    "with (open(\"df_comparison.pkl\", \"rb\")) as openfile:\n",
    "    df = pickle.load(openfile)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "58\n",
      "1183\n",
      "[ 20 217  75 246]\n"
     ]
    }
   ],
   "source": [
    "# Inspect metaclonotypes \n",
    "metaclonotypes_filenames =[\n",
    "    \"mira_epitope_55_524_ALRKVPTDNYITTY_KVPTDNYITTY.tcrdist3.csv.ranked_centers_bkgd_ctlr_1E5.tsv\",\n",
    "    \"mira_epitope_70_345_DTDFVNEFYAY_NRDVDTDFVNEFY.tcrdist3.csv.ranked_centers_bkgd_ctlr_1E5.tsv\",\n",
    "    \"mira_epitope_90_216_GYQPYRVVVL_PYRVVVLSF_QPYRVVVL_QPYRVVVLSF.tcrdist3.csv.ranked_centers_bkgd_ctlr_1E5.tsv\",\n",
    "    \"mira_epitope_53_532_NYLYRLFRK_NYNYLYRLF.tcrdist3.csv.ranked_centers_bkgd_ctlr_1E5.tsv\"\n",
    "]\n",
    "f = pd.read_csv('./covid19/hla_restricted_meta_clonotypes/mira_epitope_55_524_ALRKVPTDNYITTY_KVPTDNYITTY.tcrdist3.csv.ranked_centers_bkgd_ctlr_1E5.tsv', sep='\\t')\n",
    "index = list(f['cdr3_b_aa'].values).index(f['cdr3_b_aa'][0])\n",
    "\n",
    "print(len(f['target_seqs'][index].strip('][').split(', ')))\n",
    "f['target_seqs'][index]\n",
    "group = f['target_seqs'][index].strip(\"][\").replace(\"'\", \"\").split(', ')\n",
    "sample = df[df[\"CDR3\"].isin(group)]\n",
    "print(len(sample))\n",
    "print(sample[\"label\"].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "20     1142\n",
       "217      33\n",
       "75        5\n",
       "246       3\n",
       "Name: label, dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Count values label\n",
    "sample[\"label\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Label metaclonotype group\n",
    "sample = sample[sample[\"label\"] == 20]\n",
    "sample[\"file\"] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "30\n",
      "1856\n",
      "[219  25  71]\n"
     ]
    }
   ],
   "source": [
    "# Repeat these steps for a different metaclonotype group\n",
    "f = pd.read_csv('./covid19/hla_restricted_meta_clonotypes/mira_epitope_67_382_APHGVVFL_APHGVVFLHV_GVVFLHVTY_VVFLHVTYV.tcrdist3.csv.ranked_centers_bkgd_ctlr_1E5.tsv', sep='\\t')\n",
    "index = list(f['cdr3_b_aa'].values).index(f['cdr3_b_aa'][0])\n",
    "print(len(f['target_seqs'][index].strip('][').split(', ')))\n",
    "f['target_seqs'][index]\n",
    "group = f['target_seqs'][index].strip(\"][\").replace(\"'\", \"\").split(', ')\n",
    "sample2 = df[df[\"CDR3\"].isin(group)]\n",
    "print(len(sample2))\n",
    "print(sample2[\"label\"].unique())\n",
    "sample2 = sample2[sample2[\"label\"] == 25]\n",
    "sample2[\"file\"] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25\n",
      "687\n",
      "[111  70  50 121   7   8  64 250]\n"
     ]
    }
   ],
   "source": [
    "# Repeat these steps for a different metaclonotype group\n",
    "f = pd.read_csv('./covid19/hla_restricted_meta_clonotypes/mira_epitope_70_345_DTDFVNEFYAY_NRDVDTDFVNEFY.tcrdist3.csv.ranked_centers_bkgd_ctlr_1E5.tsv', sep='\\t')\n",
    "index = list(f['cdr3_b_aa'].values).index(f['cdr3_b_aa'][0])\n",
    "print(len(f['target_seqs'][index].strip('][').split(', ')))\n",
    "f['target_seqs'][index]\n",
    "group = f['target_seqs'][index].strip(\"][\").replace(\"'\", \"\").split(', ')\n",
    "sample3 = df[df[\"CDR3\"].isin(group)]\n",
    "print(len(sample3))\n",
    "print(sample3[\"label\"].unique())\n",
    "sample3 = sample3[sample3[\"label\"] == 50]\n",
    "sample3[\"file\"] = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/m4r/lib/python3.9/site-packages/logomaker/src/matrix.py:584: FutureWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`\n",
      "  counts_df.loc[:, c] = tmp_mat.sum(axis=0).T\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([<AxesSubplot: >, <AxesSubplot: >, <AxesSubplot: >], dtype=object)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x200 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot logos of sequences of each metaclonotype group\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import pyrepseq as rs\n",
    "import pyrepseq.plotting as rsp\n",
    "import matplotlib.pyplot as plt\n",
    "fig, axes = plt.subplots(ncols=3, nrows=1, sharey=True, figsize=(20, 2),\n",
    "                         gridspec_kw=dict(width_ratios=(1, 3, 1),\n",
    "                                          wspace=0.1, hspace=0.5)\n",
    "                         )\n",
    "rsp.seqlogos_vj(sample, \"CDR3_al\",'v_gene','j_gene',axes=axes[:3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/m4r/lib/python3.9/site-packages/logomaker/src/matrix.py:584: FutureWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`\n",
      "  counts_df.loc[:, c] = tmp_mat.sum(axis=0).T\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([<AxesSubplot: >, <AxesSubplot: >, <AxesSubplot: >], dtype=object)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x200 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(ncols=3, nrows=1, sharey=True, figsize=(20, 2),\n",
    "                         gridspec_kw=dict(width_ratios=(1, 3, 1),\n",
    "                                          wspace=0.1, hspace=0.5)\n",
    "                         )\n",
    "rsp.seqlogos_vj(sample2, \"CDR3_al\",'v_gene','j_gene',axes=axes[:3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/m4r/lib/python3.9/site-packages/logomaker/src/matrix.py:584: FutureWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`\n",
      "  counts_df.loc[:, c] = tmp_mat.sum(axis=0).T\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([<AxesSubplot: >, <AxesSubplot: >, <AxesSubplot: >], dtype=object)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x200 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(ncols=3, nrows=1, sharey=True, figsize=(20, 2),\n",
    "                         gridspec_kw=dict(width_ratios=(1, 3, 1),\n",
    "                                          wspace=0.1, hspace=0.5)\n",
    "                         )\n",
    "rsp.seqlogos_vj(sample3, \"CDR3_al\",'v_gene','j_gene',axes=axes[:3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3503, 20)\n"
     ]
    }
   ],
   "source": [
    "# Processing of the labels of the 3 different metaclonotypes groups\n",
    "groups_index = list(sample.index) + list(sample2.index) + list(sample3.index)\n",
    "X_test = X_train.copy()\n",
    "for i in range(len(X_test)):\n",
    "    X_test[i] = X_test[i][groups_index]\n",
    "print(X_test[0].shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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>CDR3</th>\n",
       "      <th>CDR3_al</th>\n",
       "      <th>tokenized_smiles</th>\n",
       "      <th>TCR BioIdentity</th>\n",
       "      <th>TCR Nucleotide Sequence</th>\n",
       "      <th>Experiment</th>\n",
       "      <th>ORF Coverage</th>\n",
       "      <th>Amino Acids</th>\n",
       "      <th>Start Index in Genome</th>\n",
       "      <th>End Index in Genome</th>\n",
       "      <th>v_gene</th>\n",
       "      <th>j_gene</th>\n",
       "      <th>Amino Acids 1</th>\n",
       "      <th>CDR3_al_one_hot</th>\n",
       "      <th>CDR3_al_integer</th>\n",
       "      <th>label</th>\n",
       "      <th>file</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CASSRKTDSYEQYF</td>\n",
       "      <td>CASSR------KTDSYEQYF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...</td>\n",
       "      <td>CASSRKTDSYEQYF+TCRBV28-01+TCRBJ02-07</td>\n",
       "      <td>CTGGAGTCCGCCAGCACCAACCAGACATCTATGTACCTCTGTGCCA...</td>\n",
       "      <td>eQD110</td>\n",
       "      <td>ORF1ab</td>\n",
       "      <td>ALRKVPTDNYITTY,KVPTDNYITTY</td>\n",
       "      <td>4211</td>\n",
       "      <td>4252</td>\n",
       "      <td>TCRBV28-01</td>\n",
       "      <td>TCRBJ02-07</td>\n",
       "      <td>ALRKVPTDNYITTY</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 15, 0, 0, 0, 0, 0, 0, 9, 17, 3,...</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CASSRKTDSYEQYF</td>\n",
       "      <td>CASSR------KTDSYEQYF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...</td>\n",
       "      <td>CASSRKTDSYEQYF+TCRBV28-01+TCRBJ02-07</td>\n",
       "      <td>CTGGAGTCCGCCAGCACCAACCAGACATCTATGTACCTCTGTGCCA...</td>\n",
       "      <td>eQD126</td>\n",
       "      <td>ORF1ab</td>\n",
       "      <td>ALRKVPTDNYITTY,KVPTDNYITTY</td>\n",
       "      <td>4211</td>\n",
       "      <td>4252</td>\n",
       "      <td>TCRBV28-01</td>\n",
       "      <td>TCRBJ02-07</td>\n",
       "      <td>ALRKVPTDNYITTY</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 15, 0, 0, 0, 0, 0, 0, 9, 17, 3,...</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CASSRKTDSYEQYF</td>\n",
       "      <td>CASSR------KTDSYEQYF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...</td>\n",
       "      <td>CASSRKTDSYEQYF+TCRBV28-01+TCRBJ02-07</td>\n",
       "      <td>CTGGAGTCCGCCAGCACCAACCAGACATCTATGTACCTCTGTGCCA...</td>\n",
       "      <td>eQD110</td>\n",
       "      <td>ORF1ab</td>\n",
       "      <td>ALRKVPTDNYITTY,KVPTDNYITTY</td>\n",
       "      <td>4211</td>\n",
       "      <td>4252</td>\n",
       "      <td>TCRBV28-01</td>\n",
       "      <td>TCRBJ02-07</td>\n",
       "      <td>ALRKVPTDNYITTY</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 15, 0, 0, 0, 0, 0, 0, 9, 17, 3,...</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CASSRKTDSYEQYF</td>\n",
       "      <td>CASSR------KTDSYEQYF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...</td>\n",
       "      <td>CASSRKTDSYEQYF+TCRBV28-01+TCRBJ02-07</td>\n",
       "      <td>CTGGAGTCCGCCAGCACCAACCAGACATCTATGTACCTCTGTGCCA...</td>\n",
       "      <td>eQD126</td>\n",
       "      <td>ORF1ab</td>\n",
       "      <td>ALRKVPTDNYITTY,KVPTDNYITTY</td>\n",
       "      <td>4211</td>\n",
       "      <td>4252</td>\n",
       "      <td>TCRBV28-01</td>\n",
       "      <td>TCRBJ02-07</td>\n",
       "      <td>ALRKVPTDNYITTY</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 15, 0, 0, 0, 0, 0, 0, 9, 17, 3,...</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CASSLRTDSYEQYF</td>\n",
       "      <td>CASSL------RTDSYEQYF</td>\n",
       "      <td>[12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...</td>\n",
       "      <td>CASSLRTDSYEQYF+TCRBV28-01+TCRBJ02-07</td>\n",
       "      <td>CTGGAGTCCGCCAGCACCAACCAGACATCTATGTACCTCTGTGCCA...</td>\n",
       "      <td>eQD110</td>\n",
       "      <td>ORF1ab</td>\n",
       "      <td>ALRKVPTDNYITTY,KVPTDNYITTY</td>\n",
       "      <td>4211</td>\n",
       "      <td>4252</td>\n",
       "      <td>TCRBV28-01</td>\n",
       "      <td>TCRBJ02-07</td>\n",
       "      <td>ALRKVPTDNYITTY</td>\n",
       "      <td>[[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...</td>\n",
       "      <td>[2, 1, 16, 16, 10, 0, 0, 0, 0, 0, 0, 15, 17, 3...</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             CDR3               CDR3_al  \\\n",
       "0  CASSRKTDSYEQYF  CASSR------KTDSYEQYF   \n",
       "1  CASSRKTDSYEQYF  CASSR------KTDSYEQYF   \n",
       "2  CASSRKTDSYEQYF  CASSR------KTDSYEQYF   \n",
       "3  CASSRKTDSYEQYF  CASSR------KTDSYEQYF   \n",
       "4  CASSLRTDSYEQYF  CASSL------RTDSYEQYF   \n",
       "\n",
       "                                    tokenized_smiles  \\\n",
       "0  [12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...   \n",
       "1  [12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...   \n",
       "2  [12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...   \n",
       "3  [12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...   \n",
       "4  [12, 16, 34, 34, 31, 31, 31, 31, 31, 31, 34, 2...   \n",
       "\n",
       "                        TCR BioIdentity  \\\n",
       "0  CASSRKTDSYEQYF+TCRBV28-01+TCRBJ02-07   \n",
       "1  CASSRKTDSYEQYF+TCRBV28-01+TCRBJ02-07   \n",
       "2  CASSRKTDSYEQYF+TCRBV28-01+TCRBJ02-07   \n",
       "3  CASSRKTDSYEQYF+TCRBV28-01+TCRBJ02-07   \n",
       "4  CASSLRTDSYEQYF+TCRBV28-01+TCRBJ02-07   \n",
       "\n",
       "                             TCR Nucleotide Sequence Experiment ORF Coverage  \\\n",
       "0  CTGGAGTCCGCCAGCACCAACCAGACATCTATGTACCTCTGTGCCA...     eQD110       ORF1ab   \n",
       "1  CTGGAGTCCGCCAGCACCAACCAGACATCTATGTACCTCTGTGCCA...     eQD126       ORF1ab   \n",
       "2  CTGGAGTCCGCCAGCACCAACCAGACATCTATGTACCTCTGTGCCA...     eQD110       ORF1ab   \n",
       "3  CTGGAGTCCGCCAGCACCAACCAGACATCTATGTACCTCTGTGCCA...     eQD126       ORF1ab   \n",
       "4  CTGGAGTCCGCCAGCACCAACCAGACATCTATGTACCTCTGTGCCA...     eQD110       ORF1ab   \n",
       "\n",
       "                  Amino Acids  Start Index in Genome  End Index in Genome  \\\n",
       "0  ALRKVPTDNYITTY,KVPTDNYITTY                   4211                 4252   \n",
       "1  ALRKVPTDNYITTY,KVPTDNYITTY                   4211                 4252   \n",
       "2  ALRKVPTDNYITTY,KVPTDNYITTY                   4211                 4252   \n",
       "3  ALRKVPTDNYITTY,KVPTDNYITTY                   4211                 4252   \n",
       "4  ALRKVPTDNYITTY,KVPTDNYITTY                   4211                 4252   \n",
       "\n",
       "       v_gene      j_gene   Amino Acids 1  \\\n",
       "0  TCRBV28-01  TCRBJ02-07  ALRKVPTDNYITTY   \n",
       "1  TCRBV28-01  TCRBJ02-07  ALRKVPTDNYITTY   \n",
       "2  TCRBV28-01  TCRBJ02-07  ALRKVPTDNYITTY   \n",
       "3  TCRBV28-01  TCRBJ02-07  ALRKVPTDNYITTY   \n",
       "4  TCRBV28-01  TCRBJ02-07  ALRKVPTDNYITTY   \n",
       "\n",
       "                                     CDR3_al_one_hot  \\\n",
       "0  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "1  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "2  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "3  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "4  [[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...   \n",
       "\n",
       "                                     CDR3_al_integer  label  file  \n",
       "0  [2, 1, 16, 16, 15, 0, 0, 0, 0, 0, 0, 9, 17, 3,...     20     0  \n",
       "1  [2, 1, 16, 16, 15, 0, 0, 0, 0, 0, 0, 9, 17, 3,...     20     0  \n",
       "2  [2, 1, 16, 16, 15, 0, 0, 0, 0, 0, 0, 9, 17, 3,...     20     0  \n",
       "3  [2, 1, 16, 16, 15, 0, 0, 0, 0, 0, 0, 9, 17, 3,...     20     0  \n",
       "4  [2, 1, 16, 16, 10, 0, 0, 0, 0, 0, 0, 15, 17, 3...     20     0  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Assemble the dataframe with all the sequences from the 3 groups\n",
    "df = pd.concat([sample, sample2, sample3])\n",
    "df.reset_index(inplace=True, drop=True)\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save dataset\n",
    "df.to_csv(\"metaclonotypes_tcrdist_3groups.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_11\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " cdr3_input (InputLayer)        [(None, 20)]         0           []                               \n",
      "                                                                                                  \n",
      " dense_22 (Dense)               (None, 40)           840         ['cdr3_input[0][0]']             \n",
      "                                                                                                  \n",
      " batch_normalization_14 (BatchN  (None, 40)          160         ['dense_22[0][0]']               \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " v_gene_input (InputLayer)      [(None, 54)]         0           []                               \n",
      "                                                                                                  \n",
      " j_gene_input (InputLayer)      [(None, 13)]         0           []                               \n",
      "                                                                                                  \n",
      " re_lu_14 (ReLU)                (None, 40)           0           ['batch_normalization_14[0][0]'] \n",
      "                                                                                                  \n",
      " v_gene_embedding (Dense)       (None, 54)           2970        ['v_gene_input[0][0]']           \n",
      "                                                                                                  \n",
      " j_gene_embedding (Dense)       (None, 13)           182         ['j_gene_input[0][0]']           \n",
      "                                                                                                  \n",
      " merged_embedding (Concatenate)  (None, 107)         0           ['re_lu_14[0][0]',               \n",
      "                                                                  'v_gene_embedding[0][0]',       \n",
      "                                                                  'j_gene_embedding[0][0]']       \n",
      "                                                                                                  \n",
      " encoder_dense_1 (Dense)        (None, 75)           8100        ['merged_embedding[0][0]']       \n",
      "                                                                                                  \n",
      " encoder_dense_2 (Dense)        (None, 75)           5700        ['encoder_dense_1[0][0]']        \n",
      "                                                                                                  \n",
      " bottleneck (Dense)             (None, 20)           1520        ['encoder_dense_2[0][0]']        \n",
      "                                                                                                  \n",
      " dense_23 (Dense)               (None, 40)           840         ['bottleneck[0][0]']             \n",
      "                                                                                                  \n",
      " batch_normalization_15 (BatchN  (None, 40)          160         ['dense_23[0][0]']               \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " re_lu_15 (ReLU)                (None, 40)           0           ['batch_normalization_15[0][0]'] \n",
      "                                                                                                  \n",
      " dense_24 (Dense)               (None, 20)           820         ['re_lu_15[0][0]']               \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 21,292\n",
      "Trainable params: 21,132\n",
      "Non-trainable params: 160\n",
      "__________________________________________________________________________________________________\n",
      "None\n",
      "Epoch 1/10\n",
      "55/55 [==============================] - 1s 2ms/step - loss: 70.6810\n",
      "Epoch 2/10\n",
      "55/55 [==============================] - 0s 2ms/step - loss: 55.1679\n",
      "Epoch 3/10\n",
      "55/55 [==============================] - 0s 2ms/step - loss: 40.8693\n",
      "Epoch 4/10\n",
      "55/55 [==============================] - 0s 2ms/step - loss: 28.8135\n",
      "Epoch 5/10\n",
      "55/55 [==============================] - 0s 1ms/step - loss: 19.6701\n",
      "Epoch 6/10\n",
      "55/55 [==============================] - 0s 2ms/step - loss: 13.2108\n",
      "Epoch 7/10\n",
      "55/55 [==============================] - 0s 2ms/step - loss: 8.7511\n",
      "Epoch 8/10\n",
      "55/55 [==============================] - 0s 1ms/step - loss: 5.6337\n",
      "Epoch 9/10\n",
      "55/55 [==============================] - 0s 1ms/step - loss: 3.5499\n",
      "Epoch 10/10\n",
      "55/55 [==============================] - 0s 1ms/step - loss: 2.1465\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Setting the parameters for model building and training of the deep AE\n",
    "params = {\n",
    "            \"latent_dim\": 20,\n",
    "            \"dense_nodes\": 75,\n",
    "            \"aa_embedding_dim\": 21,\n",
    "            \"v_gene_embedding_dim\": 54,\n",
    "            \"j_gene_embedding_dim\": 13,\n",
    "            \"beta\": 0.2,\n",
    "            \"max_cdr3_len\": 20,\n",
    "            \"n_aas\": 21,\n",
    "            \"n_v_genes\": 54,\n",
    "            \"n_j_genes\": 13,\n",
    "            \"stopping_monitor\": \"val_loss\",\n",
    "            \"batch_size\": 100,\n",
    "            \"pretrains\": 2,\n",
    "            \"warmup_period\": 3,\n",
    "            \"epochs\": 10,\n",
    "            \"patience\": 20,\n",
    "            \"n_inputs\" : 20,\n",
    "            \"v_inputs\" : 54,\n",
    "            \"j_inputs\" : 13\n",
    "        }\n",
    "model, encoder, decoder = train_model_complex(df, X_test, params=params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "110/110 [==============================] - 0s 658us/step\n"
     ]
    }
   ],
   "source": [
    "# Get embeddings in the latent space and perform PCA\n",
    "X_test_encode = encoder.predict(X_test)\n",
    "# Umap embedding\n",
    "import umap.umap_ as umap\n",
    "reducer = umap.UMAP(random_state=42)\n",
    "principalComponents = reducer.fit_transform(X_test_encode)\n",
    "plot_X_test = principalComponents"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d898d3f0baf943c38e1d9f3fa302de73",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3503 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3503, 20)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the latent space with data points coloured by their metaclonotype group\n",
    "label = \"file\"\n",
    "labels = []\n",
    "labels_encoder = LabelEncoder()\n",
    "labels_encoder = labels_encoder.fit(df[label].unique())\n",
    "for k in tqdm(df.index):\n",
    "    labels.append(labels_encoder.transform([df.loc[k,label]]))\n",
    "labels = [int(y) for y in labels]\n",
    "df[label] = labels\n",
    "rgb_values = sns.color_palette(\"Spectral\", df[label].nunique())\n",
    "df[str(label+\"_color\")] = df[label].apply(lambda x: rgb_values[x])\n",
    "print(X_test_encode.shape)\n",
    "plt.scatter(plot_X_test[:,0], plot_X_test[:,1], color=df[str(label+\"_color\")], alpha=0.5)\n",
    "plt.title(\"UMAP\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  0.9980017128175849\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/m4r/lib/python3.9/site-packages/sklearn/cluster/_kmeans.py:870: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
      "  warnings.warn(\n",
      "/opt/anaconda3/envs/m4r/lib/python3.9/site-packages/matplotlib/text.py:1241: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
      "  if s != self._text:\n",
      "/opt/anaconda3/envs/m4r/lib/python3.9/site-packages/matplotlib/text.py:1241: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
      "  if s != self._text:\n",
      "/opt/anaconda3/envs/m4r/lib/python3.9/site-packages/matplotlib/text.py:1241: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
      "  if s != self._text:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Perform K-means on the latent space embeddings\n",
    "k = 3\n",
    "kmeans = KMeans(n_clusters=k, random_state=0).fit(X_test_encode)\n",
    "labels = pd.Series(kmeans.labels_)\n",
    "rgb_values = sns.color_palette(\"Spectral\", k)\n",
    "col_kmeans = labels.apply(lambda x: rgb_values[x])\n",
    "df[str(label+\"_kmeans_label\")] = kmeans.labels_\n",
    "label_dict = {}\n",
    "for cluster in range(k):\n",
    "    label_ind = df[df[str(label+\"_kmeans_label\")] == cluster][label].value_counts().index[0]\n",
    "    label_dict[cluster] = label_ind\n",
    "\n",
    "print(\"Accuracy: \", sum([label_dict[x] == y for x,y in zip(df[str(label+\"_kmeans_label\")], df[label])])/len(df))\n",
    "plt.scatter(plot_X_test[:,0], plot_X_test[:,1], color=col_kmeans, alpha=0.5)\n",
    "centroids = kmeans.cluster_centers_\n",
    "for cluster in range(k):\n",
    "    plt.text(centroids[cluster,0], centroids[cluster,1], labels_encoder.inverse_transform([label_dict[cluster]]), fontsize=10)\n",
    "plt.title(\"Kmeans clustering for \"+label)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/m4r/lib/python3.9/site-packages/logomaker/src/matrix.py:584: FutureWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`\n",
      "  counts_df.loc[:, c] = tmp_mat.sum(axis=0).T\n",
      "/opt/anaconda3/envs/m4r/lib/python3.9/site-packages/logomaker/src/matrix.py:584: FutureWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`\n",
      "  counts_df.loc[:, c] = tmp_mat.sum(axis=0).T\n",
      "/opt/anaconda3/envs/m4r/lib/python3.9/site-packages/logomaker/src/matrix.py:584: FutureWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`\n",
      "  counts_df.loc[:, c] = tmp_mat.sum(axis=0).T\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x200 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x200 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x200 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot logos for each K-means group \n",
    "for cluster in range(k):\n",
    "    fig, axes = plt.subplots(ncols=3, nrows=1, sharey=True, figsize=(20, 2),gridspec_kw=dict(width_ratios=(1, 3, 1),wspace=0.1, hspace=0.5))\n",
    "    cluster_df = df[df[str(label+\"_kmeans_label\")] == cluster]\n",
    "    rsp.seqlogos_vj(cluster_df, \"CDR3_al\",'v_gene','j_gene',axes=axes[:3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjEAAAGdCAYAAADjWSL8AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAmRUlEQVR4nO3df1CU94HH8c+WHxvjwVMRYXcvK6E9dUwgnkJPob36MygVaaJXNWY4nXikdzWmDDqtpNMLuekELzeJ6cVpLs1ZjUpH5+bUZk6PBmvUOoQE8WjUeh5psOKEldSDXTBmofjcHz2fZAVUUnD54vs188ywz/Pdh+8+eSLvefZZcNm2bQsAAMAwn4v2BAAAAD4LIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARiJiAACAkWKjPYGhcvXqVX3wwQdKSEiQy+WK9nQAAMAtsG1bHR0d8vl8+tznbnytZcRGzAcffCC/3x/taQAAgM+gublZ99xzzw3HjNiISUhIkPSHg5CYmBjl2QAAgFsRCoXk9/udn+M3MmIj5tpbSImJiUQMAACGuZVbQbixFwAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARoqN9gRw+9y7YX+0pzBg5zYujPYUAADDFFdiAACAkYgYAABgJCIGAAAYiYgBAABGGnDEHD16VIsWLZLP55PL5dK+ffsitrtcrj6Xf/qnf3LGzJo1q9f25cuXR+ynra1NRUVFsixLlmWpqKhI7e3tn+lFAgCAkWfAEXP58mVNmTJFmzdv7nN7S0tLxPKTn/xELpdLS5YsiRhXXFwcMe6VV16J2L5ixQo1NDSoqqpKVVVVamhoUFFR0UCnCwAARqgBf8Q6Pz9f+fn5/W73eDwRj3/2s59p9uzZ+sIXvhCx/u677+419pozZ86oqqpKtbW1mj59uiTp1VdfVU5Ojs6ePatJkyYNdNoAAGCEGdJ7Yi5evKj9+/dr9erVvbZVVlYqOTlZ999/v9avX6+Ojg5n21tvvSXLspyAkaQZM2bIsizV1NT0+b3C4bBCoVDEAgAARq4h/WV3r732mhISErR48eKI9Y8++qjS09Pl8Xh06tQplZWV6Ve/+pWqq6slSYFAQCkpKb32l5KSokAg0Of3qqio0DPPPDP4LwIAAAxLQxoxP/nJT/Too4/qrrvuilhfXFzsfJ2RkaEJEyYoOztbJ06c0LRp0yT94Qbh69m23ed6SSorK1NpaanzOBQKye/3D8bLAAAAw9CQRcwvf/lLnT17Vrt3777p2GnTpikuLk6NjY2aNm2aPB6PLl682Gvchx9+qNTU1D734Xa75Xa7/+h5AwAAMwzZPTFbtmxRVlaWpkyZctOxp0+fVnd3t7xeryQpJydHwWBQ77zzjjPm7bffVjAYVG5u7lBNGQAAGGTAV2I6Ozv13nvvOY+bmprU0NCgpKQkjR8/XtIf3sr5t3/7Nz3//PO9nv+b3/xGlZWV+trXvqbk5GT9+te/1rp16zR16lR9+ctfliRNnjxZCxYsUHFxsfPR68cff1wFBQV8MgkAAEj6DFdijh8/rqlTp2rq1KmSpNLSUk2dOlV///d/74zZtWuXbNvWI4880uv58fHx+sUvfqH58+dr0qRJevLJJ5WXl6eDBw8qJibGGVdZWanMzEzl5eUpLy9PDzzwgHbs2PFZXiMAABiBXLZt29GexFAIhUKyLEvBYFCJiYnRns6wcO+G/dGewoCd27gw2lMAANxGA/n5zd9OAgAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARiJiAACAkQYcMUePHtWiRYvk8/nkcrm0b9++iO2rVq2Sy+WKWGbMmBExJhwOa+3atUpOTtbo0aNVWFioCxcuRIxpa2tTUVGRLMuSZVkqKipSe3v7gF8gAAAYmQYcMZcvX9aUKVO0efPmfscsWLBALS0tznLgwIGI7SUlJdq7d6927dqlY8eOqbOzUwUFBerp6XHGrFixQg0NDaqqqlJVVZUaGhpUVFQ00OkCAIARKnagT8jPz1d+fv4Nx7jdbnk8nj63BYNBbdmyRTt27NC8efMkSTt37pTf79fBgwc1f/58nTlzRlVVVaqtrdX06dMlSa+++qpycnJ09uxZTZo0aaDTBgAAI8yQ3BNz+PBhpaSkaOLEiSouLlZra6uzrb6+Xt3d3crLy3PW+Xw+ZWRkqKamRpL01ltvybIsJ2AkacaMGbIsyxlzvXA4rFAoFLEAAICRa9AjJj8/X5WVlTp06JCef/551dXVac6cOQqHw5KkQCCg+Ph4jRkzJuJ5qampCgQCzpiUlJRe+05JSXHGXK+iosK5f8ayLPn9/kF+ZQAAYDgZ8NtJN7Ns2TLn64yMDGVnZystLU379+/X4sWL+32ebdtyuVzO409/3d+YTysrK1NpaanzOBQKETIAAIxgQ/4Ra6/Xq7S0NDU2NkqSPB6Purq61NbWFjGutbVVqampzpiLFy/22teHH37ojLme2+1WYmJixAIAAEauIY+YS5cuqbm5WV6vV5KUlZWluLg4VVdXO2NaWlp06tQp5ebmSpJycnIUDAb1zjvvOGPefvttBYNBZwwAALizDfjtpM7OTr333nvO46amJjU0NCgpKUlJSUkqLy/XkiVL5PV6de7cOT311FNKTk7Www8/LEmyLEurV6/WunXrNHbsWCUlJWn9+vXKzMx0Pq00efJkLViwQMXFxXrllVckSY8//rgKCgr4ZBIAAJD0GSLm+PHjmj17tvP42n0oK1eu1Msvv6yTJ09q+/btam9vl9fr1ezZs7V7924lJCQ4z9m0aZNiY2O1dOlSXblyRXPnztW2bdsUExPjjKmsrNSTTz7pfIqpsLDwhr+bBgAA3Flctm3b0Z7EUAiFQrIsS8FgkPtj/t+9G/ZHewoDdm7jwmhPAQBwGw3k5zd/OwkAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYacMQcPXpUixYtks/nk8vl0r59+5xt3d3d+u53v6vMzEyNHj1aPp9Pf/3Xf60PPvggYh+zZs2Sy+WKWJYvXx4xpq2tTUVFRbIsS5ZlqaioSO3t7Z/pRQIAgJFnwBFz+fJlTZkyRZs3b+617aOPPtKJEyf0/e9/XydOnNCePXv0P//zPyosLOw1tri4WC0tLc7yyiuvRGxfsWKFGhoaVFVVpaqqKjU0NKioqGig0wUAACNU7ECfkJ+fr/z8/D63WZal6urqiHUvvfSS/uIv/kLnz5/X+PHjnfV33323PB5Pn/s5c+aMqqqqVFtbq+nTp0uSXn31VeXk5Ojs2bOaNGnSQKcNAABGmCG/JyYYDMrlcunzn/98xPrKykolJyfr/vvv1/r169XR0eFse+utt2RZlhMwkjRjxgxZlqWampo+v084HFYoFIpYAADAyDXgKzED8fHHH2vDhg1asWKFEhMTnfWPPvqo0tPT5fF4dOrUKZWVlelXv/qVcxUnEAgoJSWl1/5SUlIUCAT6/F4VFRV65plnhuaFAACAYWfIIqa7u1vLly/X1atX9aMf/ShiW3FxsfN1RkaGJkyYoOzsbJ04cULTpk2TJLlcrl77tG27z/WSVFZWptLSUudxKBSS3+8fjJcCAACGoSGJmO7ubi1dulRNTU06dOhQxFWYvkybNk1xcXFqbGzUtGnT5PF4dPHixV7jPvzwQ6Wmpva5D7fbLbfbPSjzBwAAw9+g3xNzLWAaGxt18OBBjR079qbPOX36tLq7u+X1eiVJOTk5CgaDeuedd5wxb7/9toLBoHJzcwd7ygAAwEADvhLT2dmp9957z3nc1NSkhoYGJSUlyefz6a/+6q904sQJ/cd//Id6enqce1iSkpIUHx+v3/zmN6qsrNTXvvY1JScn69e//rXWrVunqVOn6stf/rIkafLkyVqwYIGKi4udj14//vjjKigo4JNJAABA0meImOPHj2v27NnO42v3oaxcuVLl5eV6/fXXJUl//ud/HvG8N998U7NmzVJ8fLx+8Ytf6Ic//KE6Ozvl9/u1cOFCPf3004qJiXHGV1ZW6sknn1ReXp4kqbCwsM/fTQMAAO5MA46YWbNmybbtfrffaJsk+f1+HTly5KbfJykpSTt37hzo9AAAwB2Cv50EAACMRMQAAAAjETEAAMBIRAwAADASEQMAAIxExAAAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAIxExAADASEQMAAAwEhEDAACMRMQAAAAjxUZ7Aqa6d8P+aE8BAIA7GldiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGGnAEXP06FEtWrRIPp9PLpdL+/bti9hu27bKy8vl8/k0atQozZo1S6dPn44YEw6HtXbtWiUnJ2v06NEqLCzUhQsXIsa0tbWpqKhIlmXJsiwVFRWpvb19wC8QAACMTAOOmMuXL2vKlCnavHlzn9ufe+45vfDCC9q8ebPq6urk8Xj04IMPqqOjwxlTUlKivXv3ateuXTp27Jg6OztVUFCgnp4eZ8yKFSvU0NCgqqoqVVVVqaGhQUVFRZ/hJQIAgJHIZdu2/Zmf7HJp7969euihhyT94SqMz+dTSUmJvvvd70r6w1WX1NRU/eM//qO++c1vKhgMaty4cdqxY4eWLVsmSfrggw/k9/t14MABzZ8/X2fOnNF9992n2tpaTZ8+XZJUW1urnJwc/fd//7cmTZp007mFQiFZlqVgMKjExMTP+hL7de+G/YO+T/R2buPCaE8BAHAbDeTn96DeE9PU1KRAIKC8vDxnndvt1syZM1VTUyNJqq+vV3d3d8QYn8+njIwMZ8xbb70ly7KcgJGkGTNmyLIsZ8z1wuGwQqFQxAIAAEauQY2YQCAgSUpNTY1Yn5qa6mwLBAKKj4/XmDFjbjgmJSWl1/5TUlKcMderqKhw7p+xLEt+v/+Pfj0AAGD4GpJPJ7lcrojHtm33Wne968f0Nf5G+ykrK1MwGHSW5ubmzzBzAABgikGNGI/HI0m9rpa0trY6V2c8Ho+6urrU1tZ2wzEXL17stf8PP/yw11Wea9xutxITEyMWAAAwcg1qxKSnp8vj8ai6utpZ19XVpSNHjig3N1eSlJWVpbi4uIgxLS0tOnXqlDMmJydHwWBQ77zzjjPm7bffVjAYdMYAAIA7W+xAn9DZ2an33nvPedzU1KSGhgYlJSVp/PjxKikp0bPPPqsJEyZowoQJevbZZ3X33XdrxYoVkiTLsrR69WqtW7dOY8eOVVJSktavX6/MzEzNmzdPkjR58mQtWLBAxcXFeuWVVyRJjz/+uAoKCm7pk0kAAGDkG3DEHD9+XLNnz3Yel5aWSpJWrlypbdu26Tvf+Y6uXLmib33rW2pra9P06dP1xhtvKCEhwXnOpk2bFBsbq6VLl+rKlSuaO3eutm3bppiYGGdMZWWlnnzySedTTIWFhf3+bhoAAHDn+aN+T8xwxu+JGRn4PTEAcGeJ2u+JAQAAuF2IGAAAYCQiBgAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYCQiBgAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRiBgAAGAkIgYAABiJiAEAAEYiYgAAgJGIGAAAYKRBj5h7771XLper17JmzRpJ0qpVq3ptmzFjRsQ+wuGw1q5dq+TkZI0ePVqFhYW6cOHCYE8VAAAYbNAjpq6uTi0tLc5SXV0tSfrGN77hjFmwYEHEmAMHDkTso6SkRHv37tWuXbt07NgxdXZ2qqCgQD09PYM9XQAAYKjYwd7huHHjIh5v3LhRX/ziFzVz5kxnndvtlsfj6fP5wWBQW7Zs0Y4dOzRv3jxJ0s6dO+X3+3Xw4EHNnz9/sKcMAAAMNKT3xHR1dWnnzp167LHH5HK5nPWHDx9WSkqKJk6cqOLiYrW2tjrb6uvr1d3drby8PGedz+dTRkaGampq+v1e4XBYoVAoYgEAACPXkEbMvn371N7erlWrVjnr8vPzVVlZqUOHDun5559XXV2d5syZo3A4LEkKBAKKj4/XmDFjIvaVmpqqQCDQ7/eqqKiQZVnO4vf7h+Q1AQCA4WHQ3076tC1btig/P18+n89Zt2zZMufrjIwMZWdnKy0tTfv379fixYv73Zdt2xFXc65XVlam0tJS53EoFCJkAAAYwYYsYn7729/q4MGD2rNnzw3Heb1epaWlqbGxUZLk8XjU1dWltra2iKsxra2tys3N7Xc/brdbbrd7cCYPAACGvSF7O2nr1q1KSUnRwoULbzju0qVLam5ultfrlSRlZWUpLi7O+VSTJLW0tOjUqVM3jBgAAHBnGZIrMVevXtXWrVu1cuVKxcZ+8i06OztVXl6uJUuWyOv16ty5c3rqqaeUnJyshx9+WJJkWZZWr16tdevWaezYsUpKStL69euVmZnpfFoJAABgSCLm4MGDOn/+vB577LGI9TExMTp58qS2b9+u9vZ2eb1ezZ49W7t371ZCQoIzbtOmTYqNjdXSpUt15coVzZ07V9u2bVNMTMxQTBcAABjIZdu2He1JDIVQKCTLshQMBpWYmDjo+793w/5B3yd6O7fxxm9HAgBGloH8/OZvJwEAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAIxExAADASEQMAAAwEhEDAACMRMQAAAAjETEAAMBIRAwAADASEQMAAIxExAAAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAIxExAADASEQMAAAwEhEDAACMRMQAAAAjETEAAMBIRAwAADASEQMAAIxExAAAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAI8VGewIAAEC6d8P+aE9hwM5tXBjV78+VGAAAYCQiBgAAGImIAQAARiJiAACAkYgYAABgJCIGAAAYiYgBAABGImIAAICRBj1iysvL5XK5IhaPx+Nst21b5eXl8vl8GjVqlGbNmqXTp09H7CMcDmvt2rVKTk7W6NGjVVhYqAsXLgz2VAEAgMGG5ErM/fffr5aWFmc5efKks+25557TCy+8oM2bN6uurk4ej0cPPvigOjo6nDElJSXau3evdu3apWPHjqmzs1MFBQXq6ekZiukCAAADDcmfHYiNjY24+nKNbdt68cUX9b3vfU+LFy+WJL322mtKTU3VT3/6U33zm99UMBjUli1btGPHDs2bN0+StHPnTvn9fh08eFDz588fiikDAADDDMmVmMbGRvl8PqWnp2v58uV6//33JUlNTU0KBALKy8tzxrrdbs2cOVM1NTWSpPr6enV3d0eM8fl8ysjIcMb0JRwOKxQKRSwAAGDkGvSImT59urZv366f//znevXVVxUIBJSbm6tLly4pEAhIklJTUyOek5qa6mwLBAKKj4/XmDFj+h3Tl4qKClmW5Sx+v3+QXxkAABhOBj1i8vPztWTJEmVmZmrevHnav/8Pf5Xztddec8a4XK6I59i23Wvd9W42pqysTMFg0Fmam5v/iFcBAACGuyH/iPXo0aOVmZmpxsZG5z6Z66+otLa2OldnPB6Purq61NbW1u+YvrjdbiUmJkYsAABg5BryiAmHwzpz5oy8Xq/S09Pl8XhUXV3tbO/q6tKRI0eUm5srScrKylJcXFzEmJaWFp06dcoZAwAAMOifTlq/fr0WLVqk8ePHq7W1VT/4wQ8UCoW0cuVKuVwulZSU6Nlnn9WECRM0YcIEPfvss7r77ru1YsUKSZJlWVq9erXWrVunsWPHKikpSevXr3fengIAAJCGIGIuXLigRx55RL/73e80btw4zZgxQ7W1tUpLS5Mkfec739GVK1f0rW99S21tbZo+fbreeOMNJSQkOPvYtGmTYmNjtXTpUl25ckVz587Vtm3bFBMTM9jTBQAAhnLZtm1HexJDIRQKybIsBYPBIbk/5t4N+wd9n+jt3MaF0Z4CANwWJv5cGYp/owfy85u/nQQAAIxExAAAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAIxExAADASEQMAAAwEhEDAACMRMQAAAAjETEAAMBIRAwAADASEQMAAIxExAAAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAIxExAADASEQMAAAwEhEDAACMRMQAAAAjETEAAMBIRAwAADASEQMAAIxExAAAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAIxExAADASEQMAAAwEhEDAACMRMQAAAAjETEAAMBIRAwAADASEQMAAIxExAAAACMNesRUVFToS1/6khISEpSSkqKHHnpIZ8+ejRizatUquVyuiGXGjBkRY8LhsNauXavk5GSNHj1ahYWFunDhwmBPFwAAGGrQI+bIkSNas2aNamtrVV1drd///vfKy8vT5cuXI8YtWLBALS0tznLgwIGI7SUlJdq7d6927dqlY8eOqbOzUwUFBerp6RnsKQMAAAPFDvYOq6qqIh5v3bpVKSkpqq+v11e/+lVnvdvtlsfj6XMfwWBQW7Zs0Y4dOzRv3jxJ0s6dO+X3+3Xw4EHNnz9/sKcNAAAMM+T3xASDQUlSUlJSxPrDhw8rJSVFEydOVHFxsVpbW51t9fX16u7uVl5enrPO5/MpIyNDNTU1fX6fcDisUCgUsQAAgJFrSCPGtm2VlpbqK1/5ijIyMpz1+fn5qqys1KFDh/T888+rrq5Oc+bMUTgcliQFAgHFx8drzJgxEftLTU1VIBDo83tVVFTIsixn8fv9Q/fCAABA1A3620mf9sQTT+jdd9/VsWPHItYvW7bM+TojI0PZ2dlKS0vT/v37tXjx4n73Z9u2XC5Xn9vKyspUWlrqPA6FQoQMAAAj2JBdiVm7dq1ef/11vfnmm7rnnntuONbr9SotLU2NjY2SJI/Ho66uLrW1tUWMa21tVWpqap/7cLvdSkxMjFgAAMDINegRY9u2nnjiCe3Zs0eHDh1Senr6TZ9z6dIlNTc3y+v1SpKysrIUFxen6upqZ0xLS4tOnTql3NzcwZ4yAAAw0KC/nbRmzRr99Kc/1c9+9jMlJCQ497BYlqVRo0aps7NT5eXlWrJkibxer86dO6ennnpKycnJevjhh52xq1ev1rp16zR27FglJSVp/fr1yszMdD6tBAAA7myDHjEvv/yyJGnWrFkR67du3apVq1YpJiZGJ0+e1Pbt29Xe3i6v16vZs2dr9+7dSkhIcMZv2rRJsbGxWrp0qa5cuaK5c+dq27ZtiomJGewpAwAAAw16xNi2fcPto0aN0s9//vOb7ueuu+7SSy+9pJdeemmwpgYAAEYQ/nYSAAAwEhEDAACMRMQAAAAjETEAAMBIRAwAADASEQMAAIxExAAAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAIxExAADASEQMAAAwEhEDAACMRMQAAAAjETEAAMBIRAwAADASEQMAAIxExAAAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAIxExAADASEQMAAAwEhEDAACMRMQAAAAjETEAAMBIRAwAADASEQMAAIxExAAAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAIxExAADASEQMAAAwEhEDAACMNOwj5kc/+pHS09N11113KSsrS7/85S+jPSUAADAMDOuI2b17t0pKSvS9731P//Vf/6W//Mu/VH5+vs6fPx/tqQEAgCgb1hHzwgsvaPXq1fqbv/kbTZ48WS+++KL8fr9efvnlaE8NAABEWWy0J9Cfrq4u1dfXa8OGDRHr8/LyVFNT02t8OBxWOBx2HgeDQUlSKBQakvldDX80JPtFpKH67wcAw42JP1eG4t/oa/u0bfumY4dtxPzud79TT0+PUlNTI9anpqYqEAj0Gl9RUaFnnnmm13q/3z9kc8TQs16M9gwAAP0Zyn+jOzo6ZFnWDccM24i5xuVyRTy2bbvXOkkqKytTaWmp8/jq1av63//9X40dO7bP8X+MUCgkv9+v5uZmJSYmDuq+RxqO1a3jWN06jtWt41gNDMfr1g3VsbJtWx0dHfL5fDcdO2wjJjk5WTExMb2uurS2tva6OiNJbrdbbrc7Yt3nP//5oZyiEhMTOclvEcfq1nGsbh3H6tZxrAaG43XrhuJY3ewKzDXD9sbe+Ph4ZWVlqbq6OmJ9dXW1cnNzozQrAAAwXAzbKzGSVFpaqqKiImVnZysnJ0c//vGPdf78ef3t3/5ttKcGAACibFhHzLJly3Tp0iX9wz/8g1paWpSRkaEDBw4oLS0tqvNyu916+umne719hd44VreOY3XrOFa3jmM1MByvWzccjpXLvpXPMAEAAAwzw/aeGAAAgBshYgAAgJGIGAAAYCQiBgAAGImI6UdFRYW+9KUvKSEhQSkpKXrooYd09uzZiDG2bau8vFw+n0+jRo3SrFmzdPr06SjNOHpu5VitWrVKLpcrYpkxY0aUZhw9L7/8sh544AHnl0Pl5OToP//zP53tnFOfuNmx4pzqX0VFhVwul0pKSpx1nFt96+tYcW59ory8vNex8Hg8zvZon1dETD+OHDmiNWvWqLa2VtXV1fr973+vvLw8Xb582Rnz3HPP6YUXXtDmzZtVV1cnj8ejBx98UB0dHVGc+e13K8dKkhYsWKCWlhZnOXDgQJRmHD333HOPNm7cqOPHj+v48eOaM2eOvv71rzv/03NOfeJmx0rinOpLXV2dfvzjH+uBBx6IWM+51Vt/x0ri3Pq0+++/P+JYnDx50tkW9fPKxi1pbW21JdlHjhyxbdu2r169ans8Hnvjxo3OmI8//ti2LMv+l3/5l2hNc1i4/ljZtm2vXLnS/vrXvx69SQ1jY8aMsf/1X/+Vc+oWXDtWts051ZeOjg57woQJdnV1tT1z5kz729/+tm3b/HvVl/6OlW1zbn3a008/bU+ZMqXPbcPhvOJKzC0KBoOSpKSkJElSU1OTAoGA8vLynDFut1szZ85UTU1NVOY4XFx/rK45fPiwUlJSNHHiRBUXF6u1tTUa0xs2enp6tGvXLl2+fFk5OTmcUzdw/bG6hnMq0po1a7Rw4ULNmzcvYj3nVm/9HatrOLc+0djYKJ/Pp/T0dC1fvlzvv/++pOFxXg3r39g7XNi2rdLSUn3lK19RRkaGJDl/mPL6P0aZmpqq3/72t7d9jsNFX8dKkvLz8/WNb3xDaWlpampq0ve//33NmTNH9fX1d9xvxjx58qRycnL08ccf60/+5E+0d+9e3Xfffc7/9JxTn+jvWEmcU9fbtWuXTpw4obq6ul7b+Pcq0o2OlcS59WnTp0/X9u3bNXHiRF28eFE/+MEPlJubq9OnTw+L84qIuQVPPPGE3n33XR07dqzXNpfLFfHYtu1e6+4k/R2rZcuWOV9nZGQoOztbaWlp2r9/vxYvXny7pxlVkyZNUkNDg9rb2/Xv//7vWrlypY4cOeJs55z6RH/H6r777uOc+pTm5mZ9+9vf1htvvKG77rqr33GcW7d2rDi3PpGfn+98nZmZqZycHH3xi1/Ua6+95tzsHM3zireTbmLt2rV6/fXX9eabb+qee+5x1l+7O/taiV7T2traq0rvFP0dq754vV6lpaWpsbHxNs1u+IiPj9ef/dmfKTs7WxUVFZoyZYp++MMfck71ob9j1Zc7+Zyqr69Xa2ursrKyFBsbq9jYWB05ckT//M//rNjYWOf84dy6+bHq6enp9Zw7+dy63ujRo5WZmanGxsZh8W8WEdMP27b1xBNPaM+ePTp06JDS09Mjtqenp8vj8ai6utpZ19XVpSNHjig3N/d2Tzeqbnas+nLp0iU1NzfL6/XehhkOb7ZtKxwOc07dgmvHqi938jk1d+5cnTx5Ug0NDc6SnZ2tRx99VA0NDfrCF77AufX/bnasYmJiej3nTj63rhcOh3XmzBl5vd7h8W/Wbbl92EB/93d/Z1uWZR8+fNhuaWlxlo8++sgZs3HjRtuyLHvPnj32yZMn7UceecT2er12KBSK4sxvv5sdq46ODnvdunV2TU2N3dTUZL/55pt2Tk6O/ad/+qd33LEqKyuzjx49ajc1Ndnvvvuu/dRTT9mf+9zn7DfeeMO2bc6pT7vRseKcurnrP3HDudW/Tx8rzq1I69atsw8fPmy///77dm1trV1QUGAnJCTY586ds207+ucVEdMPSX0uW7dudcZcvXrVfvrpp22Px2O73W77q1/9qn3y5MnoTTpKbnasPvroIzsvL88eN26cHRcXZ48fP95euXKlff78+ehOPAoee+wxOy0tzY6Pj7fHjRtnz5071wkY2+ac+rQbHSvOqZu7PmI4t/r36WPFuRVp2bJlttfrtePi4myfz2cvXrzYPn36tLM92ueVy7Zt+/Zc8wEAABg83BMDAACMRMQAAAAjETEAAMBIRAwAADASEQMAAIxExAAAACMRMQAAwEhEDAAAMBIRAwAAjETEAAAAIxExAADASEQMAAAw0v8Bz3pNh+Ee8KMAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Check distribution of the 3 metaclonotypes groups in our dataset\n",
    "plt.hist(df[\"label\"], bins=10)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "m4r",
   "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": "3.9.16"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "527563dfeefebc70ac19b49ae61825616aae9a571fceec61e8d81d36990e36c5"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}