{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c3f43948",
   "metadata": {},
   "source": [
    "# Length dependency of conformational change"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23364e75",
   "metadata": {},
   "source": [
    "## Introduction\n",
    "\n",
    "In this notebook, we set out to determine if there was a correlation between conformational change and the length of CDR loops or peptides.\n",
    "We looked at both the correlation of bulk movements when loops where aligned on the framework regions and also the deformation effects when loops are aligned to one another.\n",
    "All peptide conformations are measure from alignment on the antigen binding groove floor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "765a694e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import scipy\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a29a7531",
   "metadata": {},
   "source": [
    "## Loading Meta data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "08d3997f",
   "metadata": {},
   "outputs": [],
   "source": [
    "DATA_DIR = '../data/processed/apo-holo-tcr-pmhc-class-I-comparisons'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ab6eb5e3",
   "metadata": {},
   "outputs": [],
   "source": [
    "apo_holo_summary_df = pd.read_csv('../data/processed/apo-holo-tcr-pmhc-class-I/apo_holo_summary.csv')\n",
    "\n",
    "apo_holo_summary_df['id'] = apo_holo_summary_df['file_name'].str.replace('.pdb$', '', regex=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8d90718a",
   "metadata": {},
   "outputs": [],
   "source": [
    "cdr_types = ['CDR-A1', 'CDR-A2', 'CDR-A3','CDR-B1', 'CDR-B2', 'CDR-B3']\n",
    "apo_holo_summary_df[cdr_types] = apo_holo_summary_df['cdr_sequences_collated'].str.split('-').apply(pd.Series)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "244f1c48",
   "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>file_name</th>\n",
       "      <th>pdb_id</th>\n",
       "      <th>structure_type</th>\n",
       "      <th>state</th>\n",
       "      <th>alpha_chain</th>\n",
       "      <th>beta_chain</th>\n",
       "      <th>antigen_chain</th>\n",
       "      <th>mhc_chain1</th>\n",
       "      <th>mhc_chain2</th>\n",
       "      <th>cdr_sequences_collated</th>\n",
       "      <th>peptide_sequence</th>\n",
       "      <th>mhc_slug</th>\n",
       "      <th>id</th>\n",
       "      <th>CDR-A1</th>\n",
       "      <th>CDR-A2</th>\n",
       "      <th>CDR-A3</th>\n",
       "      <th>CDR-B1</th>\n",
       "      <th>CDR-B2</th>\n",
       "      <th>CDR-B3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1ao7_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>1ao7</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>D</td>\n",
       "      <td>E</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>DRGSQS-IYSNGD-AVTTDSWGKLQ-MNHEY-SVGAGI-ASRPGLA...</td>\n",
       "      <td>LLFGYPVYV</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>1ao7_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS</td>\n",
       "      <td>IYSNGD</td>\n",
       "      <td>AVTTDSWGKLQ</td>\n",
       "      <td>MNHEY</td>\n",
       "      <td>SVGAGI</td>\n",
       "      <td>ASRPGLAGGRPEQY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1b0g_C-A-B_pmhc.pdb</td>\n",
       "      <td>1b0g</td>\n",
       "      <td>pmhc</td>\n",
       "      <td>apo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>NaN</td>\n",
       "      <td>ALWGFFPVL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>1b0g_C-A-B_pmhc</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1b0g_F-D-E_pmhc.pdb</td>\n",
       "      <td>1b0g</td>\n",
       "      <td>pmhc</td>\n",
       "      <td>apo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>F</td>\n",
       "      <td>D</td>\n",
       "      <td>E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>ALWGFFPVL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>1b0g_F-D-E_pmhc</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1bd2_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>1bd2</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>D</td>\n",
       "      <td>E</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>NSMFDY-ISSIKDK-AAMEGAQKLV-MNHEY-SVGAGI-ASSYPGG...</td>\n",
       "      <td>LLFGYPVYV</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>1bd2_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>NSMFDY</td>\n",
       "      <td>ISSIKDK</td>\n",
       "      <td>AAMEGAQKLV</td>\n",
       "      <td>MNHEY</td>\n",
       "      <td>SVGAGI</td>\n",
       "      <td>ASSYPGGGFYEQY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1bii_P-A-B_pmhc.pdb</td>\n",
       "      <td>1bii</td>\n",
       "      <td>pmhc</td>\n",
       "      <td>apo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>P</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>NaN</td>\n",
       "      <td>RGPGRAFVTI</td>\n",
       "      <td>h2_dd</td>\n",
       "      <td>1bii_P-A-B_pmhc</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <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>386</th>\n",
       "      <td>7rtd_C-A-B_pmhc.pdb</td>\n",
       "      <td>7rtd</td>\n",
       "      <td>pmhc</td>\n",
       "      <td>apo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>NaN</td>\n",
       "      <td>YLQPRTFLL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>7rtd_C-A-B_pmhc</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>387</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>7rtr</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>D</td>\n",
       "      <td>E</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY</td>\n",
       "      <td>YLQPRTFLL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS</td>\n",
       "      <td>IYSNGD</td>\n",
       "      <td>AVNRDDKII</td>\n",
       "      <td>SEHNR</td>\n",
       "      <td>FQNEAQ</td>\n",
       "      <td>ASSPDIEQY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>388</th>\n",
       "      <td>8gvb_A-B-P-H-L_tcr_pmhc.pdb</td>\n",
       "      <td>8gvb</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>P</td>\n",
       "      <td>H</td>\n",
       "      <td>L</td>\n",
       "      <td>YGATPY-YFSGDTLV-AVGFTGGGNKLT-SEHNR-FQNEAQ-ASSD...</td>\n",
       "      <td>RYPLTFGW</td>\n",
       "      <td>hla_a_24_02</td>\n",
       "      <td>8gvb_A-B-P-H-L_tcr_pmhc</td>\n",
       "      <td>YGATPY</td>\n",
       "      <td>YFSGDTLV</td>\n",
       "      <td>AVGFTGGGNKLT</td>\n",
       "      <td>SEHNR</td>\n",
       "      <td>FQNEAQ</td>\n",
       "      <td>ASSDRDRVPETQY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>389</th>\n",
       "      <td>8gvg_A-B-P-H-L_tcr_pmhc.pdb</td>\n",
       "      <td>8gvg</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>P</td>\n",
       "      <td>H</td>\n",
       "      <td>L</td>\n",
       "      <td>YGATPY-YFSGDTLV-AVGFTGGGNKLT-SEHNR-FQNEAQ-ASSD...</td>\n",
       "      <td>RFPLTFGW</td>\n",
       "      <td>hla_a_24_02</td>\n",
       "      <td>8gvg_A-B-P-H-L_tcr_pmhc</td>\n",
       "      <td>YGATPY</td>\n",
       "      <td>YFSGDTLV</td>\n",
       "      <td>AVGFTGGGNKLT</td>\n",
       "      <td>SEHNR</td>\n",
       "      <td>FQNEAQ</td>\n",
       "      <td>ASSDRDRVPETQY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>390</th>\n",
       "      <td>8gvi_A-B-P-H-L_tcr_pmhc.pdb</td>\n",
       "      <td>8gvi</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>P</td>\n",
       "      <td>H</td>\n",
       "      <td>L</td>\n",
       "      <td>YGATPY-YFSGDTLV-AVVFTGGGNKLT-SEHNR-FQNEAQ-ASSL...</td>\n",
       "      <td>RYPLTFGW</td>\n",
       "      <td>hla_a_24_02</td>\n",
       "      <td>8gvi_A-B-P-H-L_tcr_pmhc</td>\n",
       "      <td>YGATPY</td>\n",
       "      <td>YFSGDTLV</td>\n",
       "      <td>AVVFTGGGNKLT</td>\n",
       "      <td>SEHNR</td>\n",
       "      <td>FQNEAQ</td>\n",
       "      <td>ASSLRDRVPETQY</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>391 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       file_name pdb_id structure_type state alpha_chain  \\\n",
       "0    1ao7_D-E-C-A-B_tcr_pmhc.pdb   1ao7       tcr_pmhc  holo           D   \n",
       "1            1b0g_C-A-B_pmhc.pdb   1b0g           pmhc   apo         NaN   \n",
       "2            1b0g_F-D-E_pmhc.pdb   1b0g           pmhc   apo         NaN   \n",
       "3    1bd2_D-E-C-A-B_tcr_pmhc.pdb   1bd2       tcr_pmhc  holo           D   \n",
       "4            1bii_P-A-B_pmhc.pdb   1bii           pmhc   apo         NaN   \n",
       "..                           ...    ...            ...   ...         ...   \n",
       "386          7rtd_C-A-B_pmhc.pdb   7rtd           pmhc   apo         NaN   \n",
       "387  7rtr_D-E-C-A-B_tcr_pmhc.pdb   7rtr       tcr_pmhc  holo           D   \n",
       "388  8gvb_A-B-P-H-L_tcr_pmhc.pdb   8gvb       tcr_pmhc  holo           A   \n",
       "389  8gvg_A-B-P-H-L_tcr_pmhc.pdb   8gvg       tcr_pmhc  holo           A   \n",
       "390  8gvi_A-B-P-H-L_tcr_pmhc.pdb   8gvi       tcr_pmhc  holo           A   \n",
       "\n",
       "    beta_chain antigen_chain mhc_chain1 mhc_chain2  \\\n",
       "0            E             C          A          B   \n",
       "1          NaN             C          A          B   \n",
       "2          NaN             F          D          E   \n",
       "3            E             C          A          B   \n",
       "4          NaN             P          A          B   \n",
       "..         ...           ...        ...        ...   \n",
       "386        NaN             C          A          B   \n",
       "387          E             C          A          B   \n",
       "388          B             P          H          L   \n",
       "389          B             P          H          L   \n",
       "390          B             P          H          L   \n",
       "\n",
       "                                cdr_sequences_collated peptide_sequence  \\\n",
       "0    DRGSQS-IYSNGD-AVTTDSWGKLQ-MNHEY-SVGAGI-ASRPGLA...        LLFGYPVYV   \n",
       "1                                                  NaN        ALWGFFPVL   \n",
       "2                                                  NaN        ALWGFFPVL   \n",
       "3    NSMFDY-ISSIKDK-AAMEGAQKLV-MNHEY-SVGAGI-ASSYPGG...        LLFGYPVYV   \n",
       "4                                                  NaN       RGPGRAFVTI   \n",
       "..                                                 ...              ...   \n",
       "386                                                NaN        YLQPRTFLL   \n",
       "387     DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY        YLQPRTFLL   \n",
       "388  YGATPY-YFSGDTLV-AVGFTGGGNKLT-SEHNR-FQNEAQ-ASSD...         RYPLTFGW   \n",
       "389  YGATPY-YFSGDTLV-AVGFTGGGNKLT-SEHNR-FQNEAQ-ASSD...         RFPLTFGW   \n",
       "390  YGATPY-YFSGDTLV-AVVFTGGGNKLT-SEHNR-FQNEAQ-ASSL...         RYPLTFGW   \n",
       "\n",
       "        mhc_slug                       id  CDR-A1    CDR-A2        CDR-A3  \\\n",
       "0    hla_a_02_01  1ao7_D-E-C-A-B_tcr_pmhc  DRGSQS    IYSNGD   AVTTDSWGKLQ   \n",
       "1    hla_a_02_01          1b0g_C-A-B_pmhc     NaN       NaN           NaN   \n",
       "2    hla_a_02_01          1b0g_F-D-E_pmhc     NaN       NaN           NaN   \n",
       "3    hla_a_02_01  1bd2_D-E-C-A-B_tcr_pmhc  NSMFDY   ISSIKDK    AAMEGAQKLV   \n",
       "4          h2_dd          1bii_P-A-B_pmhc     NaN       NaN           NaN   \n",
       "..           ...                      ...     ...       ...           ...   \n",
       "386  hla_a_02_01          7rtd_C-A-B_pmhc     NaN       NaN           NaN   \n",
       "387  hla_a_02_01  7rtr_D-E-C-A-B_tcr_pmhc  DRGSQS    IYSNGD     AVNRDDKII   \n",
       "388  hla_a_24_02  8gvb_A-B-P-H-L_tcr_pmhc  YGATPY  YFSGDTLV  AVGFTGGGNKLT   \n",
       "389  hla_a_24_02  8gvg_A-B-P-H-L_tcr_pmhc  YGATPY  YFSGDTLV  AVGFTGGGNKLT   \n",
       "390  hla_a_24_02  8gvi_A-B-P-H-L_tcr_pmhc  YGATPY  YFSGDTLV  AVVFTGGGNKLT   \n",
       "\n",
       "    CDR-B1  CDR-B2          CDR-B3  \n",
       "0    MNHEY  SVGAGI  ASRPGLAGGRPEQY  \n",
       "1      NaN     NaN             NaN  \n",
       "2      NaN     NaN             NaN  \n",
       "3    MNHEY  SVGAGI   ASSYPGGGFYEQY  \n",
       "4      NaN     NaN             NaN  \n",
       "..     ...     ...             ...  \n",
       "386    NaN     NaN             NaN  \n",
       "387  SEHNR  FQNEAQ       ASSPDIEQY  \n",
       "388  SEHNR  FQNEAQ   ASSDRDRVPETQY  \n",
       "389  SEHNR  FQNEAQ   ASSDRDRVPETQY  \n",
       "390  SEHNR  FQNEAQ   ASSLRDRVPETQY  \n",
       "\n",
       "[391 rows x 19 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "apo_holo_summary_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0a44559a",
   "metadata": {},
   "source": [
    "## TCR Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "37decc9c",
   "metadata": {},
   "source": [
    "### Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "80d21b03",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_tcr_fw_align = pd.read_csv(os.path.join(DATA_DIR, 'rmsd_cdr_fw_align_results.csv'))\n",
    "results_tcr_fw_align['alignment'] = 'framework'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "03e06544",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_tcr_loop_align = pd.read_csv(os.path.join(DATA_DIR, 'rmsd_cdr_loop_align_results.csv'))\n",
    "results_tcr_loop_align['alignment'] = 'loop'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "46562e1a",
   "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>complex_id</th>\n",
       "      <th>structure_x_name</th>\n",
       "      <th>structure_y_name</th>\n",
       "      <th>chain_type</th>\n",
       "      <th>cdr</th>\n",
       "      <th>rmsd</th>\n",
       "      <th>alignment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>1.932806</td>\n",
       "      <td>framework</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>2</td>\n",
       "      <td>1.308598</td>\n",
       "      <td>framework</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>1.244062</td>\n",
       "      <td>framework</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>0.809066</td>\n",
       "      <td>framework</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>2</td>\n",
       "      <td>0.688597</td>\n",
       "      <td>framework</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>809</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>2</td>\n",
       "      <td>0.206414</td>\n",
       "      <td>loop</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>810</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>0.459090</td>\n",
       "      <td>loop</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>811</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>0.255668</td>\n",
       "      <td>loop</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>812</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>2</td>\n",
       "      <td>0.175123</td>\n",
       "      <td>loop</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>813</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>0.223801</td>\n",
       "      <td>loop</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1628 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  complex_id             structure_x_name  \\\n",
       "0    3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "1    3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "2    3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "3    3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "4    3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "..                       ...                          ...   \n",
       "809  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "810  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "811  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "812  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "813  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "\n",
       "                structure_y_name   chain_type  cdr      rmsd  alignment  \n",
       "0               3qeu_A-B_tcr.pdb  alpha_chain    1  1.932806  framework  \n",
       "1               3qeu_A-B_tcr.pdb  alpha_chain    2  1.308598  framework  \n",
       "2               3qeu_A-B_tcr.pdb  alpha_chain    3  1.244062  framework  \n",
       "3               3qeu_A-B_tcr.pdb   beta_chain    1  0.809066  framework  \n",
       "4               3qeu_A-B_tcr.pdb   beta_chain    2  0.688597  framework  \n",
       "..                           ...          ...  ...       ...        ...  \n",
       "809  7rtr_D-E-C-A-B_tcr_pmhc.pdb  alpha_chain    2  0.206414       loop  \n",
       "810  7rtr_D-E-C-A-B_tcr_pmhc.pdb  alpha_chain    3  0.459090       loop  \n",
       "811  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    1  0.255668       loop  \n",
       "812  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    2  0.175123       loop  \n",
       "813  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3  0.223801       loop  \n",
       "\n",
       "[1628 rows x 7 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_tcr = pd.concat([results_tcr_fw_align, results_tcr_loop_align])\n",
    "results_tcr"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a26611ef",
   "metadata": {},
   "source": [
    "### Merge with metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f3c4807f",
   "metadata": {},
   "outputs": [],
   "source": [
    "apo_holo_summary_cdrs_df = apo_holo_summary_df.melt(\n",
    "    id_vars=[col for col in apo_holo_summary_df.columns if col not in cdr_types],\n",
    "    var_name='cdr_type',\n",
    "    value_name='cdr_sequence',\n",
    "    value_vars=cdr_types,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3eaa16bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "apo_holo_summary_cdrs_df[['chain_type', 'cdr']] = apo_holo_summary_cdrs_df['cdr_type'].map(\n",
    "    lambda cdr_type: ('alpha_chain' if cdr_type[-2] == 'A' else 'beta_chain', int(cdr_type[-1]))\n",
    ").apply(pd.Series)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e2bf4834",
   "metadata": {},
   "outputs": [],
   "source": [
    "apo_holo_summary_cdrs_df['cdr_length'] = apo_holo_summary_cdrs_df['cdr_sequence'].str.len()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d22070df",
   "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>file_name</th>\n",
       "      <th>pdb_id</th>\n",
       "      <th>structure_type</th>\n",
       "      <th>state</th>\n",
       "      <th>alpha_chain</th>\n",
       "      <th>beta_chain</th>\n",
       "      <th>antigen_chain</th>\n",
       "      <th>mhc_chain1</th>\n",
       "      <th>mhc_chain2</th>\n",
       "      <th>cdr_sequences_collated</th>\n",
       "      <th>peptide_sequence</th>\n",
       "      <th>mhc_slug</th>\n",
       "      <th>id</th>\n",
       "      <th>cdr_type</th>\n",
       "      <th>cdr_sequence</th>\n",
       "      <th>chain_type</th>\n",
       "      <th>cdr</th>\n",
       "      <th>cdr_length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1ao7_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>1ao7</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>D</td>\n",
       "      <td>E</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>DRGSQS-IYSNGD-AVTTDSWGKLQ-MNHEY-SVGAGI-ASRPGLA...</td>\n",
       "      <td>LLFGYPVYV</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>1ao7_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>DRGSQS</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1b0g_C-A-B_pmhc.pdb</td>\n",
       "      <td>1b0g</td>\n",
       "      <td>pmhc</td>\n",
       "      <td>apo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>NaN</td>\n",
       "      <td>ALWGFFPVL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>1b0g_C-A-B_pmhc</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1b0g_F-D-E_pmhc.pdb</td>\n",
       "      <td>1b0g</td>\n",
       "      <td>pmhc</td>\n",
       "      <td>apo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>F</td>\n",
       "      <td>D</td>\n",
       "      <td>E</td>\n",
       "      <td>NaN</td>\n",
       "      <td>ALWGFFPVL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>1b0g_F-D-E_pmhc</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1bd2_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>1bd2</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>D</td>\n",
       "      <td>E</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>NSMFDY-ISSIKDK-AAMEGAQKLV-MNHEY-SVGAGI-ASSYPGG...</td>\n",
       "      <td>LLFGYPVYV</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>1bd2_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>NSMFDY</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1bii_P-A-B_pmhc.pdb</td>\n",
       "      <td>1bii</td>\n",
       "      <td>pmhc</td>\n",
       "      <td>apo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>P</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>NaN</td>\n",
       "      <td>RGPGRAFVTI</td>\n",
       "      <td>h2_dd</td>\n",
       "      <td>1bii_P-A-B_pmhc</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <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>2341</th>\n",
       "      <td>7rtd_C-A-B_pmhc.pdb</td>\n",
       "      <td>7rtd</td>\n",
       "      <td>pmhc</td>\n",
       "      <td>apo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>NaN</td>\n",
       "      <td>YLQPRTFLL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>7rtd_C-A-B_pmhc</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2342</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>7rtr</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>D</td>\n",
       "      <td>E</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY</td>\n",
       "      <td>YLQPRTFLL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>ASSPDIEQY</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2343</th>\n",
       "      <td>8gvb_A-B-P-H-L_tcr_pmhc.pdb</td>\n",
       "      <td>8gvb</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>P</td>\n",
       "      <td>H</td>\n",
       "      <td>L</td>\n",
       "      <td>YGATPY-YFSGDTLV-AVGFTGGGNKLT-SEHNR-FQNEAQ-ASSD...</td>\n",
       "      <td>RYPLTFGW</td>\n",
       "      <td>hla_a_24_02</td>\n",
       "      <td>8gvb_A-B-P-H-L_tcr_pmhc</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>ASSDRDRVPETQY</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2344</th>\n",
       "      <td>8gvg_A-B-P-H-L_tcr_pmhc.pdb</td>\n",
       "      <td>8gvg</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>P</td>\n",
       "      <td>H</td>\n",
       "      <td>L</td>\n",
       "      <td>YGATPY-YFSGDTLV-AVGFTGGGNKLT-SEHNR-FQNEAQ-ASSD...</td>\n",
       "      <td>RFPLTFGW</td>\n",
       "      <td>hla_a_24_02</td>\n",
       "      <td>8gvg_A-B-P-H-L_tcr_pmhc</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>ASSDRDRVPETQY</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2345</th>\n",
       "      <td>8gvi_A-B-P-H-L_tcr_pmhc.pdb</td>\n",
       "      <td>8gvi</td>\n",
       "      <td>tcr_pmhc</td>\n",
       "      <td>holo</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>P</td>\n",
       "      <td>H</td>\n",
       "      <td>L</td>\n",
       "      <td>YGATPY-YFSGDTLV-AVVFTGGGNKLT-SEHNR-FQNEAQ-ASSL...</td>\n",
       "      <td>RYPLTFGW</td>\n",
       "      <td>hla_a_24_02</td>\n",
       "      <td>8gvi_A-B-P-H-L_tcr_pmhc</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>ASSLRDRVPETQY</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2346 rows × 18 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                        file_name pdb_id structure_type state alpha_chain  \\\n",
       "0     1ao7_D-E-C-A-B_tcr_pmhc.pdb   1ao7       tcr_pmhc  holo           D   \n",
       "1             1b0g_C-A-B_pmhc.pdb   1b0g           pmhc   apo         NaN   \n",
       "2             1b0g_F-D-E_pmhc.pdb   1b0g           pmhc   apo         NaN   \n",
       "3     1bd2_D-E-C-A-B_tcr_pmhc.pdb   1bd2       tcr_pmhc  holo           D   \n",
       "4             1bii_P-A-B_pmhc.pdb   1bii           pmhc   apo         NaN   \n",
       "...                           ...    ...            ...   ...         ...   \n",
       "2341          7rtd_C-A-B_pmhc.pdb   7rtd           pmhc   apo         NaN   \n",
       "2342  7rtr_D-E-C-A-B_tcr_pmhc.pdb   7rtr       tcr_pmhc  holo           D   \n",
       "2343  8gvb_A-B-P-H-L_tcr_pmhc.pdb   8gvb       tcr_pmhc  holo           A   \n",
       "2344  8gvg_A-B-P-H-L_tcr_pmhc.pdb   8gvg       tcr_pmhc  holo           A   \n",
       "2345  8gvi_A-B-P-H-L_tcr_pmhc.pdb   8gvi       tcr_pmhc  holo           A   \n",
       "\n",
       "     beta_chain antigen_chain mhc_chain1 mhc_chain2  \\\n",
       "0             E             C          A          B   \n",
       "1           NaN             C          A          B   \n",
       "2           NaN             F          D          E   \n",
       "3             E             C          A          B   \n",
       "4           NaN             P          A          B   \n",
       "...         ...           ...        ...        ...   \n",
       "2341        NaN             C          A          B   \n",
       "2342          E             C          A          B   \n",
       "2343          B             P          H          L   \n",
       "2344          B             P          H          L   \n",
       "2345          B             P          H          L   \n",
       "\n",
       "                                 cdr_sequences_collated peptide_sequence  \\\n",
       "0     DRGSQS-IYSNGD-AVTTDSWGKLQ-MNHEY-SVGAGI-ASRPGLA...        LLFGYPVYV   \n",
       "1                                                   NaN        ALWGFFPVL   \n",
       "2                                                   NaN        ALWGFFPVL   \n",
       "3     NSMFDY-ISSIKDK-AAMEGAQKLV-MNHEY-SVGAGI-ASSYPGG...        LLFGYPVYV   \n",
       "4                                                   NaN       RGPGRAFVTI   \n",
       "...                                                 ...              ...   \n",
       "2341                                                NaN        YLQPRTFLL   \n",
       "2342     DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY        YLQPRTFLL   \n",
       "2343  YGATPY-YFSGDTLV-AVGFTGGGNKLT-SEHNR-FQNEAQ-ASSD...         RYPLTFGW   \n",
       "2344  YGATPY-YFSGDTLV-AVGFTGGGNKLT-SEHNR-FQNEAQ-ASSD...         RFPLTFGW   \n",
       "2345  YGATPY-YFSGDTLV-AVVFTGGGNKLT-SEHNR-FQNEAQ-ASSL...         RYPLTFGW   \n",
       "\n",
       "         mhc_slug                       id cdr_type   cdr_sequence  \\\n",
       "0     hla_a_02_01  1ao7_D-E-C-A-B_tcr_pmhc   CDR-A1         DRGSQS   \n",
       "1     hla_a_02_01          1b0g_C-A-B_pmhc   CDR-A1            NaN   \n",
       "2     hla_a_02_01          1b0g_F-D-E_pmhc   CDR-A1            NaN   \n",
       "3     hla_a_02_01  1bd2_D-E-C-A-B_tcr_pmhc   CDR-A1         NSMFDY   \n",
       "4           h2_dd          1bii_P-A-B_pmhc   CDR-A1            NaN   \n",
       "...           ...                      ...      ...            ...   \n",
       "2341  hla_a_02_01          7rtd_C-A-B_pmhc   CDR-B3            NaN   \n",
       "2342  hla_a_02_01  7rtr_D-E-C-A-B_tcr_pmhc   CDR-B3      ASSPDIEQY   \n",
       "2343  hla_a_24_02  8gvb_A-B-P-H-L_tcr_pmhc   CDR-B3  ASSDRDRVPETQY   \n",
       "2344  hla_a_24_02  8gvg_A-B-P-H-L_tcr_pmhc   CDR-B3  ASSDRDRVPETQY   \n",
       "2345  hla_a_24_02  8gvi_A-B-P-H-L_tcr_pmhc   CDR-B3  ASSLRDRVPETQY   \n",
       "\n",
       "       chain_type  cdr  cdr_length  \n",
       "0     alpha_chain    1         6.0  \n",
       "1     alpha_chain    1         NaN  \n",
       "2     alpha_chain    1         NaN  \n",
       "3     alpha_chain    1         6.0  \n",
       "4     alpha_chain    1         NaN  \n",
       "...           ...  ...         ...  \n",
       "2341   beta_chain    3         NaN  \n",
       "2342   beta_chain    3         9.0  \n",
       "2343   beta_chain    3        13.0  \n",
       "2344   beta_chain    3        13.0  \n",
       "2345   beta_chain    3        13.0  \n",
       "\n",
       "[2346 rows x 18 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "apo_holo_summary_cdrs_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c43a3b30",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_tcr = results_tcr.merge(\n",
    "    apo_holo_summary_cdrs_df[['file_name', 'pdb_id', 'structure_type', 'state']],\n",
    "    how='left',\n",
    "    left_on='structure_x_name',\n",
    "    right_on='file_name',\n",
    ").merge(\n",
    "    apo_holo_summary_cdrs_df[['file_name', 'pdb_id', 'structure_type', 'state']],\n",
    "    how='left',\n",
    "    left_on='structure_y_name',\n",
    "    right_on='file_name',\n",
    ").merge(\n",
    "    apo_holo_summary_cdrs_df[['id',\n",
    "                              'cdr_sequences_collated',\n",
    "                              'peptide_sequence',\n",
    "                              'mhc_slug',\n",
    "                              'cdr_sequence',\n",
    "                              'cdr_length',\n",
    "                              'cdr_type',\n",
    "                              'chain_type',\n",
    "                              'cdr']],\n",
    "    how='left',\n",
    "    left_on=['complex_id', 'chain_type', 'cdr'],\n",
    "    right_on=['id', 'chain_type', 'cdr'],\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "303ff63d",
   "metadata": {},
   "source": [
    "### Normalise data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "740898cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_tcr['comparison'] = results_tcr['state_x'] + '-' + results_tcr['state_y']\n",
    "results_tcr['comparison'] = results_tcr['comparison'].map(lambda entry: 'apo-holo' if entry == 'holo-apo' else entry)\n",
    "results_tcr = results_tcr.query(\"comparison == 'apo-holo'\").reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "0f94c8ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_tcr['structure_comparison'] = results_tcr.apply(\n",
    "    lambda row: '-'.join(sorted([row.structure_x_name, row.structure_y_name])),\n",
    "    axis='columns',\n",
    ")\n",
    "results_tcr = results_tcr.drop_duplicates(['structure_comparison', 'chain_type', 'cdr', 'alignment'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "9d61dd03",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_tcr = results_tcr.groupby(['cdr_sequences_collated',\n",
    "                                   'alignment',\n",
    "                                   'comparison',\n",
    "                                   'cdr_length',\n",
    "                                   'cdr_sequence',\n",
    "                                   'cdr_type',\n",
    "                                   'chain_type',\n",
    "                                   'cdr'])['rmsd'].mean().reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb297fed",
   "metadata": {},
   "source": [
    "### Visualise Results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6dd5d049",
   "metadata": {},
   "source": [
    "#### Alignment on Framework region"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "3df358a7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = results_tcr.query(\"alignment == 'framework'\")\n",
    "\n",
    "r, p_val = scipy.stats.pearsonr(data['cdr_length'], data['rmsd'])\n",
    "\n",
    "sns.lmplot(data.sort_values('cdr_type'), x='cdr_length', y='rmsd', scatter=False)\n",
    "sns.scatterplot(data.sort_values('cdr_type'), x='cdr_length', y='rmsd', hue='cdr_type')\n",
    "\n",
    "plt.text(4, 8, f'$r^2$ = {r**2: .2f}, p-value = {p_val: .2e}')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "654c1cb1",
   "metadata": {},
   "source": [
    "#### Alignment on loops"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "553a81dd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = results_tcr.query(\"alignment == 'loop'\")\n",
    "\n",
    "r, p_val = scipy.stats.pearsonr(data['cdr_length'], data['rmsd'])\n",
    "\n",
    "sns.lmplot(data.sort_values('cdr_type'), x='cdr_length', y='rmsd', scatter=False)\n",
    "sns.scatterplot(data.sort_values('cdr_type'), x='cdr_length', y='rmsd', hue='cdr_type')\n",
    "\n",
    "plt.text(7, 2.5, f'$r^2$ = {r**2: .2f}, p-value = {p_val: .2e}')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9cd6f7b7",
   "metadata": {},
   "source": [
    "#### TCR apex calculations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "c2c33135",
   "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>complex_id</th>\n",
       "      <th>structure_x_name</th>\n",
       "      <th>structure_y_name</th>\n",
       "      <th>chain_type</th>\n",
       "      <th>cdr</th>\n",
       "      <th>residue_name</th>\n",
       "      <th>residue_seq_id</th>\n",
       "      <th>residue_insert_code</th>\n",
       "      <th>rmsd</th>\n",
       "      <th>ca_distance</th>\n",
       "      <th>chi_angle_change</th>\n",
       "      <th>com_distance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>ASP</td>\n",
       "      <td>27</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.922807</td>\n",
       "      <td>2.215234</td>\n",
       "      <td>-1.001709</td>\n",
       "      <td>3.836500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>ARG</td>\n",
       "      <td>28</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.683418</td>\n",
       "      <td>2.322292</td>\n",
       "      <td>-1.010462</td>\n",
       "      <td>6.119157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>GLY</td>\n",
       "      <td>29</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.657793</td>\n",
       "      <td>0.718576</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.452200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>SER</td>\n",
       "      <td>36</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.224430</td>\n",
       "      <td>0.404912</td>\n",
       "      <td>-2.505061</td>\n",
       "      <td>0.866544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>GLN</td>\n",
       "      <td>37</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.133408</td>\n",
       "      <td>0.467132</td>\n",
       "      <td>0.667185</td>\n",
       "      <td>0.798590</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6228</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>ASP</td>\n",
       "      <td>109</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.507077</td>\n",
       "      <td>0.180564</td>\n",
       "      <td>0.198956</td>\n",
       "      <td>0.208659</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6229</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>ILE</td>\n",
       "      <td>114</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.164965</td>\n",
       "      <td>0.158682</td>\n",
       "      <td>3.679442</td>\n",
       "      <td>0.829175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6230</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>GLU</td>\n",
       "      <td>115</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.577728</td>\n",
       "      <td>0.195887</td>\n",
       "      <td>3.091039</td>\n",
       "      <td>0.995351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6231</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>GLN</td>\n",
       "      <td>116</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.204783</td>\n",
       "      <td>0.197683</td>\n",
       "      <td>-0.014126</td>\n",
       "      <td>0.166472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6232</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>TYR</td>\n",
       "      <td>117</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.180118</td>\n",
       "      <td>0.164981</td>\n",
       "      <td>-0.024244</td>\n",
       "      <td>0.170852</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6233 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                   complex_id             structure_x_name  \\\n",
       "0     3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "1     3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "2     3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "3     3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "4     3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "...                       ...                          ...   \n",
       "6228  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "6229  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "6230  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "6231  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "6232  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "\n",
       "                 structure_y_name   chain_type  cdr residue_name  \\\n",
       "0                3qeu_A-B_tcr.pdb  alpha_chain    1          ASP   \n",
       "1                3qeu_A-B_tcr.pdb  alpha_chain    1          ARG   \n",
       "2                3qeu_A-B_tcr.pdb  alpha_chain    1          GLY   \n",
       "3                3qeu_A-B_tcr.pdb  alpha_chain    1          SER   \n",
       "4                3qeu_A-B_tcr.pdb  alpha_chain    1          GLN   \n",
       "...                           ...          ...  ...          ...   \n",
       "6228  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3          ASP   \n",
       "6229  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3          ILE   \n",
       "6230  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3          GLU   \n",
       "6231  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3          GLN   \n",
       "6232  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3          TYR   \n",
       "\n",
       "      residue_seq_id residue_insert_code      rmsd  ca_distance  \\\n",
       "0                 27                 NaN  4.922807     2.215234   \n",
       "1                 28                 NaN  7.683418     2.322292   \n",
       "2                 29                 NaN  0.657793     0.718576   \n",
       "3                 36                 NaN  1.224430     0.404912   \n",
       "4                 37                 NaN  1.133408     0.467132   \n",
       "...              ...                 ...       ...          ...   \n",
       "6228             109                 NaN  0.507077     0.180564   \n",
       "6229             114                 NaN  2.164965     0.158682   \n",
       "6230             115                 NaN  1.577728     0.195887   \n",
       "6231             116                 NaN  0.204783     0.197683   \n",
       "6232             117                 NaN  0.180118     0.164981   \n",
       "\n",
       "      chi_angle_change  com_distance  \n",
       "0            -1.001709      3.836500  \n",
       "1            -1.010462      6.119157  \n",
       "2                  NaN      0.452200  \n",
       "3            -2.505061      0.866544  \n",
       "4             0.667185      0.798590  \n",
       "...                ...           ...  \n",
       "6228          0.198956      0.208659  \n",
       "6229          3.679442      0.829175  \n",
       "6230          3.091039      0.995351  \n",
       "6231         -0.014126      0.166472  \n",
       "6232         -0.024244      0.170852  \n",
       "\n",
       "[6233 rows x 12 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_per_res_tcr = pd.read_csv(os.path.join(DATA_DIR, 'tcr_per_res_apo_holo_loop_align.csv'))\n",
    "results_per_res_tcr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "d4cafbb3",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_per_res_tcr = results_per_res_tcr.merge(\n",
    "    apo_holo_summary_cdrs_df[['file_name', 'pdb_id', 'structure_type', 'state']],\n",
    "    how='left',\n",
    "    left_on='structure_x_name',\n",
    "    right_on='file_name',\n",
    ").merge(\n",
    "    apo_holo_summary_cdrs_df[['file_name', 'pdb_id', 'structure_type', 'state']],\n",
    "    how='left',\n",
    "    left_on='structure_y_name',\n",
    "    right_on='file_name',\n",
    ").merge(\n",
    "    apo_holo_summary_cdrs_df[['id',\n",
    "                              'cdr_sequences_collated',\n",
    "                              'peptide_sequence',\n",
    "                              'mhc_slug',\n",
    "                              'cdr_sequence',\n",
    "                              'cdr_length',\n",
    "                              'cdr_type',\n",
    "                              'chain_type',\n",
    "                              'cdr']],\n",
    "    how='left',\n",
    "    left_on=['complex_id', 'chain_type', 'cdr'],\n",
    "    right_on=['id', 'chain_type', 'cdr'],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "3298f58c",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_per_res_tcr['comparison'] = results_per_res_tcr['state_x'] + '-' + results_per_res_tcr['state_y']\n",
    "results_per_res_tcr['comparison'] = results_per_res_tcr['comparison'].map(\n",
    "    lambda entry: 'apo-holo' if entry == 'holo-apo' else entry\n",
    ")\n",
    "results_per_res_tcr = results_per_res_tcr.query(\"comparison == 'apo-holo'\").reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "8915cad1",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_per_res_tcr['structure_comparison'] = results_per_res_tcr.apply(\n",
    "    lambda row: '-'.join(sorted([row.structure_x_name, row.structure_y_name])),\n",
    "    axis='columns',\n",
    ")\n",
    "results_per_res_tcr = results_per_res_tcr.drop_duplicates(['structure_comparison', 'chain_type', 'cdr',\n",
    "                                                           'residue_name', 'residue_seq_id', 'residue_insert_code'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "6d649918",
   "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>complex_id</th>\n",
       "      <th>structure_x_name</th>\n",
       "      <th>structure_y_name</th>\n",
       "      <th>chain_type</th>\n",
       "      <th>cdr</th>\n",
       "      <th>residue_name</th>\n",
       "      <th>residue_seq_id</th>\n",
       "      <th>residue_insert_code</th>\n",
       "      <th>rmsd</th>\n",
       "      <th>ca_distance</th>\n",
       "      <th>...</th>\n",
       "      <th>state_y</th>\n",
       "      <th>id</th>\n",
       "      <th>cdr_sequences_collated</th>\n",
       "      <th>peptide_sequence</th>\n",
       "      <th>mhc_slug</th>\n",
       "      <th>cdr_sequence</th>\n",
       "      <th>cdr_length</th>\n",
       "      <th>cdr_type</th>\n",
       "      <th>comparison</th>\n",
       "      <th>structure_comparison</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>ASP</td>\n",
       "      <td>27</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.922807</td>\n",
       "      <td>2.215234</td>\n",
       "      <td>...</td>\n",
       "      <td>apo</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNFGGGKLI-MRHNA-SNTAGT-ASSLSFGTEAF</td>\n",
       "      <td>ELAGIGILTV</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>DRGSQS</td>\n",
       "      <td>6.0</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb-3qeu_A-B_tcr.pdb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>ARG</td>\n",
       "      <td>28</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.683418</td>\n",
       "      <td>2.322292</td>\n",
       "      <td>...</td>\n",
       "      <td>apo</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNFGGGKLI-MRHNA-SNTAGT-ASSLSFGTEAF</td>\n",
       "      <td>ELAGIGILTV</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>DRGSQS</td>\n",
       "      <td>6.0</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb-3qeu_A-B_tcr.pdb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>GLY</td>\n",
       "      <td>29</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.657793</td>\n",
       "      <td>0.718576</td>\n",
       "      <td>...</td>\n",
       "      <td>apo</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNFGGGKLI-MRHNA-SNTAGT-ASSLSFGTEAF</td>\n",
       "      <td>ELAGIGILTV</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>DRGSQS</td>\n",
       "      <td>6.0</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb-3qeu_A-B_tcr.pdb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>108</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>SER</td>\n",
       "      <td>36</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.224430</td>\n",
       "      <td>0.404912</td>\n",
       "      <td>...</td>\n",
       "      <td>apo</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNFGGGKLI-MRHNA-SNTAGT-ASSLSFGTEAF</td>\n",
       "      <td>ELAGIGILTV</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>DRGSQS</td>\n",
       "      <td>6.0</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb-3qeu_A-B_tcr.pdb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>144</th>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qeu_A-B_tcr.pdb</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>1</td>\n",
       "      <td>GLN</td>\n",
       "      <td>37</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.133408</td>\n",
       "      <td>0.467132</td>\n",
       "      <td>...</td>\n",
       "      <td>apo</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNFGGGKLI-MRHNA-SNTAGT-ASSLSFGTEAF</td>\n",
       "      <td>ELAGIGILTV</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>DRGSQS</td>\n",
       "      <td>6.0</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>3qdg_D-E-C-A-B_tcr_pmhc.pdb-3qeu_A-B_tcr.pdb</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>158220</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>ASP</td>\n",
       "      <td>109</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.507077</td>\n",
       "      <td>0.180564</td>\n",
       "      <td>...</td>\n",
       "      <td>holo</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY</td>\n",
       "      <td>YLQPRTFLL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>ASSPDIEQY</td>\n",
       "      <td>9.0</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>7n1d_A-B_tcr.pdb-7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>158256</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>ILE</td>\n",
       "      <td>114</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.164965</td>\n",
       "      <td>0.158682</td>\n",
       "      <td>...</td>\n",
       "      <td>holo</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY</td>\n",
       "      <td>YLQPRTFLL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>ASSPDIEQY</td>\n",
       "      <td>9.0</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>7n1d_A-B_tcr.pdb-7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>158292</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>GLU</td>\n",
       "      <td>115</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.577728</td>\n",
       "      <td>0.195887</td>\n",
       "      <td>...</td>\n",
       "      <td>holo</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY</td>\n",
       "      <td>YLQPRTFLL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>ASSPDIEQY</td>\n",
       "      <td>9.0</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>7n1d_A-B_tcr.pdb-7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>158328</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>GLN</td>\n",
       "      <td>116</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.204783</td>\n",
       "      <td>0.197683</td>\n",
       "      <td>...</td>\n",
       "      <td>holo</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY</td>\n",
       "      <td>YLQPRTFLL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>ASSPDIEQY</td>\n",
       "      <td>9.0</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>7n1d_A-B_tcr.pdb-7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>158364</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n1d_A-B_tcr.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>3</td>\n",
       "      <td>TYR</td>\n",
       "      <td>117</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.180118</td>\n",
       "      <td>0.164981</td>\n",
       "      <td>...</td>\n",
       "      <td>holo</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY</td>\n",
       "      <td>YLQPRTFLL</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>ASSPDIEQY</td>\n",
       "      <td>9.0</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>7n1d_A-B_tcr.pdb-7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4400 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     complex_id             structure_x_name  \\\n",
       "0       3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "36      3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "72      3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "108     3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "144     3qdg_D-E-C-A-B_tcr_pmhc  3qdg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "...                         ...                          ...   \n",
       "158220  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "158256  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "158292  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "158328  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "158364  7rtr_D-E-C-A-B_tcr_pmhc             7n1d_A-B_tcr.pdb   \n",
       "\n",
       "                   structure_y_name   chain_type  cdr residue_name  \\\n",
       "0                  3qeu_A-B_tcr.pdb  alpha_chain    1          ASP   \n",
       "36                 3qeu_A-B_tcr.pdb  alpha_chain    1          ARG   \n",
       "72                 3qeu_A-B_tcr.pdb  alpha_chain    1          GLY   \n",
       "108                3qeu_A-B_tcr.pdb  alpha_chain    1          SER   \n",
       "144                3qeu_A-B_tcr.pdb  alpha_chain    1          GLN   \n",
       "...                             ...          ...  ...          ...   \n",
       "158220  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3          ASP   \n",
       "158256  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3          ILE   \n",
       "158292  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3          GLU   \n",
       "158328  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3          GLN   \n",
       "158364  7rtr_D-E-C-A-B_tcr_pmhc.pdb   beta_chain    3          TYR   \n",
       "\n",
       "        residue_seq_id residue_insert_code      rmsd  ca_distance  ...  \\\n",
       "0                   27                 NaN  4.922807     2.215234  ...   \n",
       "36                  28                 NaN  7.683418     2.322292  ...   \n",
       "72                  29                 NaN  0.657793     0.718576  ...   \n",
       "108                 36                 NaN  1.224430     0.404912  ...   \n",
       "144                 37                 NaN  1.133408     0.467132  ...   \n",
       "...                ...                 ...       ...          ...  ...   \n",
       "158220             109                 NaN  0.507077     0.180564  ...   \n",
       "158256             114                 NaN  2.164965     0.158682  ...   \n",
       "158292             115                 NaN  1.577728     0.195887  ...   \n",
       "158328             116                 NaN  0.204783     0.197683  ...   \n",
       "158364             117                 NaN  0.180118     0.164981  ...   \n",
       "\n",
       "        state_y                       id  \\\n",
       "0           apo  3qdg_D-E-C-A-B_tcr_pmhc   \n",
       "36          apo  3qdg_D-E-C-A-B_tcr_pmhc   \n",
       "72          apo  3qdg_D-E-C-A-B_tcr_pmhc   \n",
       "108         apo  3qdg_D-E-C-A-B_tcr_pmhc   \n",
       "144         apo  3qdg_D-E-C-A-B_tcr_pmhc   \n",
       "...         ...                      ...   \n",
       "158220     holo  7rtr_D-E-C-A-B_tcr_pmhc   \n",
       "158256     holo  7rtr_D-E-C-A-B_tcr_pmhc   \n",
       "158292     holo  7rtr_D-E-C-A-B_tcr_pmhc   \n",
       "158328     holo  7rtr_D-E-C-A-B_tcr_pmhc   \n",
       "158364     holo  7rtr_D-E-C-A-B_tcr_pmhc   \n",
       "\n",
       "                                   cdr_sequences_collated peptide_sequence  \\\n",
       "0       DRGSQS-IYSNGD-AVNFGGGKLI-MRHNA-SNTAGT-ASSLSFGTEAF       ELAGIGILTV   \n",
       "36      DRGSQS-IYSNGD-AVNFGGGKLI-MRHNA-SNTAGT-ASSLSFGTEAF       ELAGIGILTV   \n",
       "72      DRGSQS-IYSNGD-AVNFGGGKLI-MRHNA-SNTAGT-ASSLSFGTEAF       ELAGIGILTV   \n",
       "108     DRGSQS-IYSNGD-AVNFGGGKLI-MRHNA-SNTAGT-ASSLSFGTEAF       ELAGIGILTV   \n",
       "144     DRGSQS-IYSNGD-AVNFGGGKLI-MRHNA-SNTAGT-ASSLSFGTEAF       ELAGIGILTV   \n",
       "...                                                   ...              ...   \n",
       "158220     DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY        YLQPRTFLL   \n",
       "158256     DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY        YLQPRTFLL   \n",
       "158292     DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY        YLQPRTFLL   \n",
       "158328     DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY        YLQPRTFLL   \n",
       "158364     DRGSQS-IYSNGD-AVNRDDKII-SEHNR-FQNEAQ-ASSPDIEQY        YLQPRTFLL   \n",
       "\n",
       "           mhc_slug cdr_sequence cdr_length cdr_type comparison  \\\n",
       "0       hla_a_02_01       DRGSQS        6.0   CDR-A1   apo-holo   \n",
       "36      hla_a_02_01       DRGSQS        6.0   CDR-A1   apo-holo   \n",
       "72      hla_a_02_01       DRGSQS        6.0   CDR-A1   apo-holo   \n",
       "108     hla_a_02_01       DRGSQS        6.0   CDR-A1   apo-holo   \n",
       "144     hla_a_02_01       DRGSQS        6.0   CDR-A1   apo-holo   \n",
       "...             ...          ...        ...      ...        ...   \n",
       "158220  hla_a_02_01    ASSPDIEQY        9.0   CDR-B3   apo-holo   \n",
       "158256  hla_a_02_01    ASSPDIEQY        9.0   CDR-B3   apo-holo   \n",
       "158292  hla_a_02_01    ASSPDIEQY        9.0   CDR-B3   apo-holo   \n",
       "158328  hla_a_02_01    ASSPDIEQY        9.0   CDR-B3   apo-holo   \n",
       "158364  hla_a_02_01    ASSPDIEQY        9.0   CDR-B3   apo-holo   \n",
       "\n",
       "                                structure_comparison  \n",
       "0       3qdg_D-E-C-A-B_tcr_pmhc.pdb-3qeu_A-B_tcr.pdb  \n",
       "36      3qdg_D-E-C-A-B_tcr_pmhc.pdb-3qeu_A-B_tcr.pdb  \n",
       "72      3qdg_D-E-C-A-B_tcr_pmhc.pdb-3qeu_A-B_tcr.pdb  \n",
       "108     3qdg_D-E-C-A-B_tcr_pmhc.pdb-3qeu_A-B_tcr.pdb  \n",
       "144     3qdg_D-E-C-A-B_tcr_pmhc.pdb-3qeu_A-B_tcr.pdb  \n",
       "...                                              ...  \n",
       "158220  7n1d_A-B_tcr.pdb-7rtr_D-E-C-A-B_tcr_pmhc.pdb  \n",
       "158256  7n1d_A-B_tcr.pdb-7rtr_D-E-C-A-B_tcr_pmhc.pdb  \n",
       "158292  7n1d_A-B_tcr.pdb-7rtr_D-E-C-A-B_tcr_pmhc.pdb  \n",
       "158328  7n1d_A-B_tcr.pdb-7rtr_D-E-C-A-B_tcr_pmhc.pdb  \n",
       "158364  7n1d_A-B_tcr.pdb-7rtr_D-E-C-A-B_tcr_pmhc.pdb  \n",
       "\n",
       "[4400 rows x 29 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_per_res_tcr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "9d50c214",
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_apex_movement(loop: pd.DataFrame) -> float:\n",
    "    loop_len = len(loop)\n",
    "\n",
    "    if loop_len % 2 == 0:\n",
    "        return np.mean([loop.iloc[loop_len // 2]['ca_distance'], loop.iloc[(loop_len // 2) - 1]['ca_distance']])\n",
    "\n",
    "    return loop.iloc[loop_len // 2]['ca_distance']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "b8d4f41e",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_tcr_apex = results_per_res_tcr.groupby(['structure_x_name',\n",
    "                                                'structure_y_name',\n",
    "                                                'cdr_type',\n",
    "                                                'cdr',\n",
    "                                                'chain_type',\n",
    "                                                'cdr_length']).apply(compute_apex_movement)\n",
    "results_tcr_apex.name = 'apex_ca_distance'\n",
    "results_tcr_apex = results_tcr_apex.reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "61b24c7b",
   "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>structure_x_name</th>\n",
       "      <th>structure_y_name</th>\n",
       "      <th>cdr_type</th>\n",
       "      <th>cdr</th>\n",
       "      <th>chain_type</th>\n",
       "      <th>cdr_length</th>\n",
       "      <th>apex_ca_distance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1ao7_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qh3_A-B_tcr.pdb</td>\n",
       "      <td>CDR-A1</td>\n",
       "      <td>1</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.153049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1ao7_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qh3_A-B_tcr.pdb</td>\n",
       "      <td>CDR-A2</td>\n",
       "      <td>2</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.447122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1ao7_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qh3_A-B_tcr.pdb</td>\n",
       "      <td>CDR-A3</td>\n",
       "      <td>3</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>11.0</td>\n",
       "      <td>2.461146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1ao7_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qh3_A-B_tcr.pdb</td>\n",
       "      <td>CDR-B1</td>\n",
       "      <td>1</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.319232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1ao7_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>3qh3_A-B_tcr.pdb</td>\n",
       "      <td>CDR-B2</td>\n",
       "      <td>2</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.167444</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>569</th>\n",
       "      <td>7r7z_A-B_tcr.pdb</td>\n",
       "      <td>7r80_A-B-E-C-D_tcr_pmhc.pdb</td>\n",
       "      <td>CDR-A2</td>\n",
       "      <td>2</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.173771</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>570</th>\n",
       "      <td>7r7z_A-B_tcr.pdb</td>\n",
       "      <td>7r80_A-B-E-C-D_tcr_pmhc.pdb</td>\n",
       "      <td>CDR-A3</td>\n",
       "      <td>3</td>\n",
       "      <td>alpha_chain</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0.508615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>571</th>\n",
       "      <td>7r7z_A-B_tcr.pdb</td>\n",
       "      <td>7r80_A-B-E-C-D_tcr_pmhc.pdb</td>\n",
       "      <td>CDR-B1</td>\n",
       "      <td>1</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.108211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>572</th>\n",
       "      <td>7r7z_A-B_tcr.pdb</td>\n",
       "      <td>7r80_A-B-E-C-D_tcr_pmhc.pdb</td>\n",
       "      <td>CDR-B2</td>\n",
       "      <td>2</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.219891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>573</th>\n",
       "      <td>7r7z_A-B_tcr.pdb</td>\n",
       "      <td>7r80_A-B-E-C-D_tcr_pmhc.pdb</td>\n",
       "      <td>CDR-B3</td>\n",
       "      <td>3</td>\n",
       "      <td>beta_chain</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.789117</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>574 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                structure_x_name             structure_y_name cdr_type  cdr  \\\n",
       "0    1ao7_D-E-C-A-B_tcr_pmhc.pdb             3qh3_A-B_tcr.pdb   CDR-A1    1   \n",
       "1    1ao7_D-E-C-A-B_tcr_pmhc.pdb             3qh3_A-B_tcr.pdb   CDR-A2    2   \n",
       "2    1ao7_D-E-C-A-B_tcr_pmhc.pdb             3qh3_A-B_tcr.pdb   CDR-A3    3   \n",
       "3    1ao7_D-E-C-A-B_tcr_pmhc.pdb             3qh3_A-B_tcr.pdb   CDR-B1    1   \n",
       "4    1ao7_D-E-C-A-B_tcr_pmhc.pdb             3qh3_A-B_tcr.pdb   CDR-B2    2   \n",
       "..                           ...                          ...      ...  ...   \n",
       "569             7r7z_A-B_tcr.pdb  7r80_A-B-E-C-D_tcr_pmhc.pdb   CDR-A2    2   \n",
       "570             7r7z_A-B_tcr.pdb  7r80_A-B-E-C-D_tcr_pmhc.pdb   CDR-A3    3   \n",
       "571             7r7z_A-B_tcr.pdb  7r80_A-B-E-C-D_tcr_pmhc.pdb   CDR-B1    1   \n",
       "572             7r7z_A-B_tcr.pdb  7r80_A-B-E-C-D_tcr_pmhc.pdb   CDR-B2    2   \n",
       "573             7r7z_A-B_tcr.pdb  7r80_A-B-E-C-D_tcr_pmhc.pdb   CDR-B3    3   \n",
       "\n",
       "      chain_type  cdr_length  apex_ca_distance  \n",
       "0    alpha_chain         6.0          0.153049  \n",
       "1    alpha_chain         6.0          0.447122  \n",
       "2    alpha_chain        11.0          2.461146  \n",
       "3     beta_chain         5.0          0.319232  \n",
       "4     beta_chain         6.0          0.167444  \n",
       "..           ...         ...               ...  \n",
       "569  alpha_chain         7.0          0.173771  \n",
       "570  alpha_chain        11.0          0.508615  \n",
       "571   beta_chain         5.0          0.108211  \n",
       "572   beta_chain         6.0          0.219891  \n",
       "573   beta_chain        13.0          0.789117  \n",
       "\n",
       "[574 rows x 7 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_tcr_apex"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "c9e8e137",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "r, p_val = scipy.stats.pearsonr(results_tcr_apex['cdr_length'], results_tcr_apex['apex_ca_distance'])\n",
    "\n",
    "sns.lmplot(results_tcr_apex.sort_values('cdr_type'), x='cdr_length', y='apex_ca_distance', scatter=False)\n",
    "sns.scatterplot(results_tcr_apex.sort_values('cdr_type'), x='cdr_length', y='apex_ca_distance', hue='cdr_type')\n",
    "\n",
    "plt.text(6.75, 4, f'$r^2$ = {r**2: .2f}, p-value = {p_val: .2e}')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e44d6c69",
   "metadata": {},
   "source": [
    "## Peptide Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e8ec65a",
   "metadata": {},
   "source": [
    "### Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "47abcbee",
   "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>complex_id</th>\n",
       "      <th>structure_x_name</th>\n",
       "      <th>structure_y_name</th>\n",
       "      <th>rmsd</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5c0a_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>5c0a_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>5n1y_C-A-B_pmhc.pdb</td>\n",
       "      <td>0.448858</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5wlg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>5wlg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>5wli_C-A-B_pmhc.pdb</td>\n",
       "      <td>0.498148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>5wlg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>5wlg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>5wli_F-D-E_pmhc.pdb</td>\n",
       "      <td>0.519507</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>5wlg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>5wlg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>5wli_I-G-H_pmhc.pdb</td>\n",
       "      <td>0.459025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>5wlg_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>5wlg_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>5wli_L-J-K_pmhc.pdb</td>\n",
       "      <td>0.494705</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3301</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n6d_O-M-N_pmhc.pdb</td>\n",
       "      <td>7rtd_C-A-B_pmhc.pdb</td>\n",
       "      <td>0.486296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3304</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7n6d_O-M-N_pmhc.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>0.448905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3307</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7p3d_C-A-B_pmhc.pdb</td>\n",
       "      <td>7rtd_C-A-B_pmhc.pdb</td>\n",
       "      <td>0.773692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3310</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7p3d_C-A-B_pmhc.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>0.765295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3313</th>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc</td>\n",
       "      <td>7rtd_C-A-B_pmhc.pdb</td>\n",
       "      <td>7rtr_D-E-C-A-B_tcr_pmhc.pdb</td>\n",
       "      <td>0.455078</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1096 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                   complex_id             structure_x_name  \\\n",
       "2     5c0a_D-E-C-A-B_tcr_pmhc  5c0a_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "5     5wlg_D-E-C-A-B_tcr_pmhc  5wlg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "8     5wlg_D-E-C-A-B_tcr_pmhc  5wlg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "11    5wlg_D-E-C-A-B_tcr_pmhc  5wlg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "14    5wlg_D-E-C-A-B_tcr_pmhc  5wlg_D-E-C-A-B_tcr_pmhc.pdb   \n",
       "...                       ...                          ...   \n",
       "3301  7rtr_D-E-C-A-B_tcr_pmhc          7n6d_O-M-N_pmhc.pdb   \n",
       "3304  7rtr_D-E-C-A-B_tcr_pmhc          7n6d_O-M-N_pmhc.pdb   \n",
       "3307  7rtr_D-E-C-A-B_tcr_pmhc          7p3d_C-A-B_pmhc.pdb   \n",
       "3310  7rtr_D-E-C-A-B_tcr_pmhc          7p3d_C-A-B_pmhc.pdb   \n",
       "3313  7rtr_D-E-C-A-B_tcr_pmhc          7rtd_C-A-B_pmhc.pdb   \n",
       "\n",
       "                 structure_y_name      rmsd  \n",
       "2             5n1y_C-A-B_pmhc.pdb  0.448858  \n",
       "5             5wli_C-A-B_pmhc.pdb  0.498148  \n",
       "8             5wli_F-D-E_pmhc.pdb  0.519507  \n",
       "11            5wli_I-G-H_pmhc.pdb  0.459025  \n",
       "14            5wli_L-J-K_pmhc.pdb  0.494705  \n",
       "...                           ...       ...  \n",
       "3301          7rtd_C-A-B_pmhc.pdb  0.486296  \n",
       "3304  7rtr_D-E-C-A-B_tcr_pmhc.pdb  0.448905  \n",
       "3307          7rtd_C-A-B_pmhc.pdb  0.773692  \n",
       "3310  7rtr_D-E-C-A-B_tcr_pmhc.pdb  0.765295  \n",
       "3313  7rtr_D-E-C-A-B_tcr_pmhc.pdb  0.455078  \n",
       "\n",
       "[1096 rows x 4 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_peptide = pd.read_csv(os.path.join(DATA_DIR, 'pmhc_tcr_contact_apo_holo.csv'))\n",
    "\n",
    "results_peptide = results_peptide.query(\"chain_type == 'antigen_chain'\")\n",
    "results_peptide = results_peptide.drop(columns=['chain_type', 'tcr_contact'])\n",
    "\n",
    "results_peptide"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "00bce7a6",
   "metadata": {},
   "source": [
    "### Merge with metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "e2e323cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "apo_holo_summary_df['peptide_length'] = apo_holo_summary_df['peptide_sequence'].str.len()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "3333ba7c",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_peptide = results_peptide.merge(\n",
    "    apo_holo_summary_df[['file_name', 'pdb_id', 'state']],\n",
    "    how='left',\n",
    "    left_on='structure_x_name',\n",
    "    right_on='file_name',\n",
    ").merge(\n",
    "    apo_holo_summary_df[['file_name', 'pdb_id', 'state']],\n",
    "    how='left',\n",
    "    left_on='structure_y_name',\n",
    "    right_on='file_name',\n",
    ").merge(\n",
    "    apo_holo_summary_df[['id', 'peptide_sequence', 'peptide_length', 'mhc_slug']],\n",
    "    how='left',\n",
    "    left_on='complex_id',\n",
    "    right_on='id',\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24d6d9af",
   "metadata": {},
   "source": [
    "### Normalise data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "202e703c",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_peptide['comparison'] = results_peptide['state_x'] + '-' + results_peptide['state_y']\n",
    "results_peptide['comparison'] = results_peptide['comparison'].map(\n",
    "    lambda entry: 'apo-holo' if entry == 'holo-apo' else entry,\n",
    ")\n",
    "results_peptide = results_peptide.query(\"comparison == 'apo-holo'\").reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "fdb6dd82",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_peptide['structure_comparison'] = results_peptide.apply(\n",
    "    lambda row: '-'.join(sorted([row.structure_x_name, row.structure_y_name])),\n",
    "    axis='columns',\n",
    ")\n",
    "results_peptide = results_peptide.drop_duplicates('structure_comparison')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "08e770cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_peptide = results_peptide.groupby(['peptide_sequence',\n",
    "                                           'comparison',\n",
    "                                           'peptide_length',\n",
    "                                           'mhc_slug'])['rmsd'].mean().reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "eb267508",
   "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>peptide_sequence</th>\n",
       "      <th>comparison</th>\n",
       "      <th>peptide_length</th>\n",
       "      <th>mhc_slug</th>\n",
       "      <th>rmsd</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AAGIGILTV</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>9.0</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>1.025622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ALGIGILTV</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>9.0</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>1.274516</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ALWGFFPVL</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>9.0</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>0.351748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ALWGPDPAAA</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>10.0</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>0.703094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>APRGPHGGAASGL</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>13.0</td>\n",
       "      <td>hla_b_07_02</td>\n",
       "      <td>3.523593</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>VVVGAGGVGK</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>10.0</td>\n",
       "      <td>hla_a_11_01</td>\n",
       "      <td>1.483028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>YGFRNVVHI</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>9.0</td>\n",
       "      <td>h2_db</td>\n",
       "      <td>0.275006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>77</th>\n",
       "      <td>YLGGPDFPTI</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>10.0</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>0.868938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>78</th>\n",
       "      <td>YLQPRTFLL</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>9.0</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>0.584577</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>YQFGPDFPIA</td>\n",
       "      <td>apo-holo</td>\n",
       "      <td>10.0</td>\n",
       "      <td>hla_a_02_01</td>\n",
       "      <td>0.408370</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>80 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   peptide_sequence comparison  peptide_length     mhc_slug      rmsd\n",
       "0         AAGIGILTV   apo-holo             9.0  hla_a_02_01  1.025622\n",
       "1         ALGIGILTV   apo-holo             9.0  hla_a_02_01  1.274516\n",
       "2         ALWGFFPVL   apo-holo             9.0  hla_a_02_01  0.351748\n",
       "3        ALWGPDPAAA   apo-holo            10.0  hla_a_02_01  0.703094\n",
       "4     APRGPHGGAASGL   apo-holo            13.0  hla_b_07_02  3.523593\n",
       "..              ...        ...             ...          ...       ...\n",
       "75       VVVGAGGVGK   apo-holo            10.0  hla_a_11_01  1.483028\n",
       "76        YGFRNVVHI   apo-holo             9.0        h2_db  0.275006\n",
       "77       YLGGPDFPTI   apo-holo            10.0  hla_a_02_01  0.868938\n",
       "78        YLQPRTFLL   apo-holo             9.0  hla_a_02_01  0.584577\n",
       "79       YQFGPDFPIA   apo-holo            10.0  hla_a_02_01  0.408370\n",
       "\n",
       "[80 rows x 5 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_peptide"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84557444",
   "metadata": {},
   "source": [
    "### Visualise results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "957172f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "r, p_val = scipy.stats.pearsonr(results_peptide['peptide_length'], results_peptide['rmsd'])\n",
    "\n",
    "sns.lmplot(results_peptide.sort_values('mhc_slug'), x='peptide_length', y='rmsd', scatter=False)\n",
    "ax = sns.scatterplot(results_peptide.sort_values('mhc_slug'), x='peptide_length', y='rmsd', hue='mhc_slug')\n",
    "sns.move_legend(ax, \"upper left\", bbox_to_anchor=(1, 1))\n",
    "\n",
    "plt.text(8.5, 4, f'$r^2$ = {r**2: .2f}, p-value = {p_val: .2e}')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21c88b9c",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "\n",
    "Both TCR CDR loops and peptides show a correlation between length and amount of conformational change.\n",
    "For the TCRs, there is a correlation in both paradigms, framework alignment and loop alignment, but the correlation is much stronger looking at the loop alignments.\n",
    "The increased correlation for loop alignments make sense since the conformational changes from the framework regions can be driven by other parts of the protein, but when the loops are aligned together, the only differences can be driven by changes in the loops themselves, implying the loop length has more of an effect."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}