tutorials/halluc_SH3_binder/halluc_sh3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Tutorial: scaffolding a linear motif to create SH3 binders\n",
    "\n",
    "This demonstrates scaffolding a linear peptide from 2w0z with the goal of binding an SH3 domain. \n",
    "\n",
    "Here we will do monomer hallucination followed by 2-chain refinement, and score final outputs using AlphaFold2 interface predictions. For hallucination here, we will use the `rf_Nov05_2021` version of RoseTTAFold, which is currently the best model (as of Jan 2022).\n",
    "\n",
    "If you haven't already, look through the PD-L1 binder hallucination tutorial to get the basics. This has less explanation but more steps, which is a more realistic design scenario.\n",
    "\n",
    "Working commands below call scripts using relative paths from this folder (`<rfdesign repo>/tutorials/halluc_SH3_binder/`). Some paths instead have a placeholder `<rfdesign repo>`, which should be replaced with the path to your clone of the RFDesign repository."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import glob, os, shutil"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append(\"util/\") # some functions to make plotting/analysis easier\n",
    "from plotting import *\n",
    "from functions import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "figdir = 'fig/'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run 1: Monomer hallucination\n",
    "\n",
    "This will use gradient descent to generate scaffolds holding the linear motif, with a repulsive loss to avoid clashes with the binding target.\n",
    "\n",
    "The goal is to generate good-looking scaffolds. We will fine-tune its sequence and interactions with the binding target in a second step (see run 2)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate commands"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [],
   "source": [
    "ref_pdb = 'input/SH3_2w0z.pdb'\n",
    "receptor = 'input/SH3_2w0z_rec.pdb'\n",
    "run = 'sh3_r1'\n",
    "contigs = 'B7-14'\n",
    "force_aa = 'B7-14'\n",
    "\n",
    "batch = 2\n",
    "start_num = 0\n",
    "total_num = 200\n",
    "outdir = f'output/20220103_{run}/'\n",
    "\n",
    "with open('r1.list','w') as outf:\n",
    "    for istart in np.arange(start_num, start_num+total_num, batch):\n",
    "        os.makedirs(outdir,exist_ok=True)\n",
    "        if all([os.path.exists(f'{outdir}{run}_{j}.pdb') for j in range(istart, istart+batch)]):\n",
    "            continue\n",
    "\n",
    "        run_name = f'{run}'\n",
    "        logfile = f'{outdir}/{run_name}_{istart}.log'\n",
    "        trd_cmd = f'source activate SE3-nvidia; python ../../hallucination/hallucinate.py '\\\n",
    "                  f'--network_name rf_Nov05_2021 '\\\n",
    "                  f'--pdb={ref_pdb} --out={outdir}/{run_name} --contigs={contigs} --len=55-100 '\\\n",
    "                  f'--steps=g600 --num={batch} --start_num={istart} '\\\n",
    "                  f'--w_rep=2 --rep_sigma=3.5 --rep_pdb={receptor} '\\\n",
    "                  f'--w_rog=1 --rog_thresh=16 '\\\n",
    "                  f'--force_aa {force_aa} --exclude_aa C --save_pdb=True --track_step=10 --cautious=True '\\\n",
    "                  f'&>> {logfile}'\n",
    "\n",
    "        print(trd_cmd, file=outf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Run using:\n",
    "\n",
    "    ./array_submit.sh r1.list"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scoring\n",
    "\n",
    "Generate relaxed design models with side chains:\n",
    "\n",
    "    cd output/20220103_sh3_r1/\n",
    "    <rfdesign repo>/scripts/trfold_relax.sh .\n",
    "    \n",
    "wait until finished, then:\n",
    "\n",
    "    cd trf_relax\n",
    "    sbatch -p gpu -c 2 --mem 12g --gres gpu:rtx2080:1 --wrap \"<rfdesign repo>/scripts/af2_metrics.py .\"\n",
    "    sbatch --mem 5g --wrap \"<rfdesign repo>/scripts/pyrosetta_metrics.py .\"\n",
    "    <rfdesign repo>/scripts/compile_metrics.py .\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Analyze"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "datadir = 'output/20220103_sh3_r1/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(datadir+'trf_relax/combined_metrics.csv')\n",
    "df['folder'] = datadir+'trf_relax/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(200, 24)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x144 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig,axes = plt.subplots(1,2,figsize=(6,2))\n",
    "for ax, metric in zip(axes.flat,['af2_lddt','contig_rmsd_af2']):\n",
    "    sns.distplot(df[metric],ax=ax)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Pick hits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "import shutil"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "hits = df[(df['af2_lddt']>80) & (df['contig_rmsd_af2']<0.6) & (df['contig_rmsd']<1) & (df['rog']<16)\n",
    "          & (df['topology']!='HH')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(13, 24)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hits.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>name</th>\n",
       "      <th>af2_lddt</th>\n",
       "      <th>contig_rmsd_af2</th>\n",
       "      <th>rog</th>\n",
       "      <th>topology</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>sh3_r1_120</td>\n",
       "      <td>82.917949</td>\n",
       "      <td>0.558620</td>\n",
       "      <td>13.438005</td>\n",
       "      <td>HHHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>sh3_r1_112</td>\n",
       "      <td>81.830960</td>\n",
       "      <td>0.472451</td>\n",
       "      <td>11.803300</td>\n",
       "      <td>HHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>sh3_r1_87</td>\n",
       "      <td>81.388544</td>\n",
       "      <td>0.539688</td>\n",
       "      <td>13.514336</td>\n",
       "      <td>HHHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>sh3_r1_192</td>\n",
       "      <td>86.161309</td>\n",
       "      <td>0.536761</td>\n",
       "      <td>11.573707</td>\n",
       "      <td>HHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>sh3_r1_33</td>\n",
       "      <td>80.818507</td>\n",
       "      <td>0.569302</td>\n",
       "      <td>12.369105</td>\n",
       "      <td>HHHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86</th>\n",
       "      <td>sh3_r1_22</td>\n",
       "      <td>81.911010</td>\n",
       "      <td>0.419346</td>\n",
       "      <td>13.836364</td>\n",
       "      <td>HHHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>sh3_r1_85</td>\n",
       "      <td>85.730110</td>\n",
       "      <td>0.410799</td>\n",
       "      <td>12.039831</td>\n",
       "      <td>HHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102</th>\n",
       "      <td>sh3_r1_146</td>\n",
       "      <td>85.665627</td>\n",
       "      <td>0.465928</td>\n",
       "      <td>12.251873</td>\n",
       "      <td>HHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>sh3_r1_66</td>\n",
       "      <td>88.249932</td>\n",
       "      <td>0.532329</td>\n",
       "      <td>11.887709</td>\n",
       "      <td>HHHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>sh3_r1_110</td>\n",
       "      <td>86.046411</td>\n",
       "      <td>0.305571</td>\n",
       "      <td>13.929003</td>\n",
       "      <td>HHHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>156</th>\n",
       "      <td>sh3_r1_69</td>\n",
       "      <td>89.168160</td>\n",
       "      <td>0.428238</td>\n",
       "      <td>11.787690</td>\n",
       "      <td>HHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185</th>\n",
       "      <td>sh3_r1_43</td>\n",
       "      <td>81.494475</td>\n",
       "      <td>0.478631</td>\n",
       "      <td>11.965088</td>\n",
       "      <td>HHHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>sh3_r1_78</td>\n",
       "      <td>85.511798</td>\n",
       "      <td>0.350242</td>\n",
       "      <td>12.471108</td>\n",
       "      <td>HHHH</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           name   af2_lddt  contig_rmsd_af2        rog topology\n",
       "14   sh3_r1_120  82.917949         0.558620  13.438005     HHHH\n",
       "17   sh3_r1_112  81.830960         0.472451  11.803300      HHH\n",
       "40    sh3_r1_87  81.388544         0.539688  13.514336     HHHH\n",
       "45   sh3_r1_192  86.161309         0.536761  11.573707      HHH\n",
       "56    sh3_r1_33  80.818507         0.569302  12.369105     HHHH\n",
       "86    sh3_r1_22  81.911010         0.419346  13.836364     HHHH\n",
       "99    sh3_r1_85  85.730110         0.410799  12.039831      HHH\n",
       "102  sh3_r1_146  85.665627         0.465928  12.251873      HHH\n",
       "123   sh3_r1_66  88.249932         0.532329  11.887709     HHHH\n",
       "133  sh3_r1_110  86.046411         0.305571  13.929003     HHHH\n",
       "156   sh3_r1_69  89.168160         0.428238  11.787690      HHH\n",
       "185   sh3_r1_43  81.494475         0.478631  11.965088     HHHH\n",
       "194   sh3_r1_78  85.511798         0.350242  12.471108     HHHH"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hits[['name','af2_lddt','contig_rmsd_af2','rog','topology']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "outdir = 'output/hits_sh3_r1/'\n",
    "os.makedirs(outdir, exist_ok=True)\n",
    "os.makedirs(outdir+'af2/', exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i,row in hits.iterrows():\n",
    "    shutil.copyfile(row['folder']+'/'+row['name']+'.pdb', outdir+row['name']+'.pdb')  \n",
    "    shutil.copyfile(row['folder']+'af2/'+row['name']+'.pdb', outdir+'af2/'+row['name']+'_af2pred.pdb')    # copy af2 model\n",
    "    shutil.copyfile(row['folder']+'../'+row['name']+'.trb', outdir+row['name']+'.trb')    \n",
    "    shutil.copyfile(row['folder']+'../'+row['name']+'.fas', outdir+row['name']+'.fas')    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "hits.to_csv(outdir+'metrics.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Align hits and save to pymol session for visual inspection\n",
    "\n",
    "    cd output/hits_sh3_r1/\n",
    "    <rfdesign repo>/scripts/pymol_align.py --template ../../input/SH3_2w0z.pdb -o aligned.pse *pdb \n",
    "    \n",
    "Most of these look good by eye. Some don't look amazing, or have weird sequences (lots of surface M), but I will just do 2-chain refinement on all of these hits (which might rescue some minor defects) and do more stringent filtering after that."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run 2: 2-chain MCMC refinement from run 1 hits\n",
    "2021-1-4\n",
    "\n",
    "This run will take hits from run 1 and continue to mutate them, but in the context of the receptor (both are being predicted by RoseTTAFold). This will allow us to build interface interactions, and also include (non-differentiable) losses to avoid surface hydrophobics and decrease the surface charge (to a target of -7)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [],
   "source": [
    "ref_pdb = 'input/SH3_2w0z.pdb'\n",
    "receptor = 'input/SH3_2w0z_rec.pdb'\n",
    "seeds = glob.glob('output/hits_sh3_r1/*pdb')\n",
    "run = 'r2'\n",
    "contigs = 'B7-14'\n",
    "force_aa = 'B7-14'\n",
    "\n",
    "batch = 5\n",
    "start_num = 0\n",
    "total_num = 50 # per seed design\n",
    "outdir = f'output/20220104_sh3_{run}/'\n",
    "os.makedirs(outdir,exist_ok=True)\n",
    "\n",
    "with open(run+'.list','w') as outf:\n",
    "    for istart in np.arange(start_num, start_num+total_num, batch):\n",
    "        for fn in seeds:\n",
    "            name = os.path.basename(fn.replace('.pdb',''))\n",
    "            if all([os.path.exists(f'{outdir}{run}_{j}.pdb') for j in range(istart, istart+batch)]):\n",
    "                continue\n",
    "\n",
    "            trb = np.load(fn.replace('.pdb','.trb'),allow_pickle=True)\n",
    "\n",
    "            run_name = f'{name}_{run}'\n",
    "            logfile = f'{outdir}/{run_name}_{istart}.log'\n",
    "            trd_cmd = f'source activate SE3-nvidia; python ../../hallucination/hallucinate.py '\\\n",
    "                      f'--network_name rf_Nov05_2021 '\\\n",
    "                      f'--pdb={ref_pdb} --out={outdir}/{run_name} '\\\n",
    "                      f'--mask {trb[\"sampled_mask\"]} --use_template {contigs} '\\\n",
    "                      f'--spike_fas {fn.replace(\".pdb\",\".fas\")} --spike 0.999 '\\\n",
    "                      f'--force_aa {force_aa} --exclude_aa C '\\\n",
    "                      f'--receptor {receptor} --rec_placement second --w_surfnp 1 --w_nc 0.02 '\\\n",
    "                      f'--steps=m300 --num={batch} --start_num={istart} '\\\n",
    "                      f'--w_rog=1 --rog_thresh=16 '\\\n",
    "                      f'--save_pdb=True --track_step=1 --cautious=True '\\\n",
    "                      f'&>> {logfile}'\n",
    "\n",
    "            print(trd_cmd, file=outf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Run commands:\n",
    "\n",
    "    ./array_submit r2.list"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scoring\n",
    "Generate relaxed models:\n",
    "\n",
    "    cd output/20220104_sh3_r2/\n",
    "    <rfdesign repo>/scripts/trfold_relax.sh ."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then split up the files so scoring jobs can run faster."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "split_folder('output/20220104_sh3_r2/trf_relax/', batch_size=50, trb_dir='../..')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now run scoring jobs, looping over the subfolders. For this section you will need to install [`silent_tools`](https://github.com/bcov77/silent_tools).\n",
    "\n",
    "AF2 prediction-based metrics, i.e. `pae_interaction`:\n",
    "\n",
    "    for i in */; do sbatch -p gpu --mem 12g --gres gpu:rtx2080:1 -J af2int --wrap \"cd $i; silentfrompdbs *pdb > in.silent; <rfdesign repo>/scripts/af2_interface_metrics.py -silent in.silent\"; done\n",
    "    \n",
    "Relax AF2 predicted interfaces, then calculate interface metrics (ddg, contact MS, etc):\n",
    "\n",
    "    for i in 0* 1*; do sbatch --mem 5g -J rel.$i --wrap \"mkdir $i/relax; cd $i/relax; ln -s ../out.silent in.silent; <path to rosetta>/main/source/cmake/build_release/rosetta_scripts @<rfdesign repo>/scripts/relax.flags -parser:protocol <rfdesign repo>/scripts/minterface.xml; silentextract out.silent; <rfdesign repo>/scripts/get_interface_metrics.py .\"; done\n",
    "    \n",
    "Calculate RMSD of binder after superimposing on receptor:\n",
    "\n",
    "    for i in 0* 1*; do sbatch --mem 2g --wrap \"<rfdesign repo>/scripts/get_binder_rmsd.py $i\"; done\n",
    "    \n",
    "The scripts above are written by Nate Bennett and come from [this preprint](https://www.biorxiv.org/content/10.1101/2022.06.15.495993v1)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Analyze"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "filenames = sorted(glob.glob(f'output/20220104_sh3_r2/trf_relax/*/'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame()\n",
    "for folder in filenames:\n",
    "    if 'logs' in folder: continue\n",
    "    tmp = load_af2_int_metrics(folder)\n",
    "    df = df.append(tmp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(650, 31)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['SCORE:', 'pae_binder', 'pae_interaction', 'pae_interaction1',\n",
       "       'pae_interaction2', 'pae_target', 'plddt_binder', 'plddt_target',\n",
       "       'plddt_total', 'time', 'name', 'folder', 'Unnamed: 0_x',\n",
       "       'binder_blocked_sap', 'buns_heavy_ball_1.1D',\n",
       "       'contact_molec_sq5_apap_target', 'contact_molecular_surface',\n",
       "       'contact_molecular_surface_ap_target',\n",
       "       'contact_molecular_surface_apap_target', 'ddg', 'interface_buried_sasa',\n",
       "       'interface_sc', 'mismatch_probability', 'sap_score', 'sap_score_target',\n",
       "       'target_blocked_sap', 'Unnamed: 0_y', 'binder_rmsd_af2_des',\n",
       "       'complex_rmsd_af2_des', 'binder_rmsd', 'complex_rmsd'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "seqs  = load_seqs('output/20220104_sh3_r2/')\n",
    "df = df.merge(seqs,on='name')\n",
    "df['seed'] = df['name'].apply(lambda x: '_'.join(x.split('_')[:-2]))\n",
    "df['net_charge'] = df['seq'].apply(lambda x: x.count('K')+x.count('R')-x.count('D')-x.count('E'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(650, 34)"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[df['ddg']<20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x600 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig,axes = plt.subplots(1,4,figsize=(12,6))\n",
    "plt.subplots_adjust(wspace=0.1,hspace=0.2,top=0.92)\n",
    "for ax,metric in zip(axes.flat,['pae_interaction','plddt_binder','ddg','sap_score']):\n",
    "    sns.violinplot(data=df,y='seed',x=metric,ax=ax,color=[.8]*3,zorder=-1,inner=None)\n",
    "    sns.stripplot(data=df,y='seed',x=metric,ax=ax,color=[.3]*3,s=3,jitter=False,\n",
    "                  label='Halluc -> 2-chain MCMC',zorder=1)\n",
    "#     sns.stripplot(data=seed,y='seed',x=metric,ax=ax,color='r',s=5,jitter=False,\n",
    "#                   label='Seed',zorder=1)\n",
    "    yl = ax.get_ylim()\n",
    "    ax.set_ylim(yl)\n",
    "#     ax.plot([native[metric].values[0]]*2,yl,'-',color='C0',linewidth=1,label='Native')\n",
    "    if ax!=axes.flat[0]:\n",
    "        ax.set_yticklabels('')\n",
    "        ax.set_ylabel('')\n",
    "outside_legend(ax)\n",
    "sns.despine()\n",
    "# plt.suptitle('Improving EpcamBind1_cb4 using inpainting + MPNN');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These results look pretty good! We need `pae_interaction < 10` to expect binding activity, and `plddt_binder` should also be ideally >80."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "hits = df[(df['pae_interaction']<10) & (df['plddt_binder']>80) & (df['sap_score']<35) & (df['binder_rmsd_af2_des']<5) \n",
    "          & (df['net_charge']<=-5)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(45, 34)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hits.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "outdir = 'output/hits_sh3_r2/'\n",
    "os.makedirs(outdir, exist_ok=True)\n",
    "os.makedirs(outdir+'af2/', exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i,row in hits.iterrows():\n",
    "    shutil.copyfile(row['folder']+'/'+row['name']+'.pdb', outdir+row['name']+'.pdb')  \n",
    "#     shutil.copyfile(row['folder']+'af2/'+row['name']+'.pdb', outdir+'af2/'+row['name']+'_af2pred.pdb')    # copy af2 model\n",
    "    shutil.copyfile(row['folder']+'../../'+row['name']+'.trb', outdir+row['name']+'.trb')    \n",
    "    shutil.copyfile(row['folder']+'../../'+row['name']+'.fas', outdir+row['name']+'.fas')    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "hits.to_csv(outdir+'metrics.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Align in Pymol\n",
    "Our script `<rfdesign repo>/scripts/pymol_align.py` can only align monomer hallucinations to the reference pdb. To align the 2-chain hallucinations from this step to reference, let's use the code below to generate commands for aligning outputs to reference structures, and paste them manually into a pymol session."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "load output/hits_sh3_r2/sh3_r1_110_r2_32.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_146_r2_5.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_11.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_146_r2_46.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_5.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_78_r2_3.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_66_r2_2.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_66_r2_17.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_35.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_32.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_120_r2_2.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_146_r2_40.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_78_r2_13.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_120_r2_36.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_66_r2_47.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_192_r2_31.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_22.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_66_r2_18.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_112_r2_19.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_146_r2_11.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_66_r2_16.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_66_r2_3.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_66_r2_4.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_22_r2_21.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_37.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_39.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_7.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_112_r2_21.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_9.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_13.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_66_r2_43.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_192_r2_49.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_110_r2_45.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_146_r2_0.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_26.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_146_r2_28.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_110_r2_15.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_87_r2_31.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_43.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_85_r2_38.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_36.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_66_r2_37.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_78_r2_18.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_112_r2_5.pdb\n",
      "load output/hits_sh3_r2/sh3_r1_69_r2_12.pdb\n"
     ]
    }
   ],
   "source": [
    "for fn in glob.glob('output/hits_sh3_r2/*pdb'):\n",
    "    print(f'load {fn}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "sys.path.append('../../hallucination/util/')\n",
    "from contigs import SampledMask # a utility for manipulating protein residue numbers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pair_fit /sh3_r1_110_r2_15//A/29-36/N, /SH3_2w0z//B/7-14/N, /sh3_r1_110_r2_15//B/96-153/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_110_r2_15_motif, /sh3_r1_110_r2_15//B/96-153\n",
      "pair_fit /sh3_r1_110_r2_32//A/29-36/N, /SH3_2w0z//B/7-14/N, /sh3_r1_110_r2_32//B/96-153/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_110_r2_32_motif, /sh3_r1_110_r2_32//B/96-153\n",
      "pair_fit /sh3_r1_110_r2_45//A/29-36/N, /SH3_2w0z//B/7-14/N, /sh3_r1_110_r2_45//B/96-153/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_110_r2_45_motif, /sh3_r1_110_r2_45//B/96-153\n",
      "pair_fit /sh3_r1_112_r2_19//A/45-52/N, /SH3_2w0z//B/7-14/N, /sh3_r1_112_r2_19//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_112_r2_19_motif, /sh3_r1_112_r2_19//B/71-128\n",
      "pair_fit /sh3_r1_112_r2_21//A/45-52/N, /SH3_2w0z//B/7-14/N, /sh3_r1_112_r2_21//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_112_r2_21_motif, /sh3_r1_112_r2_21//B/71-128\n",
      "pair_fit /sh3_r1_112_r2_5//A/45-52/N, /SH3_2w0z//B/7-14/N, /sh3_r1_112_r2_5//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_112_r2_5_motif, /sh3_r1_112_r2_5//B/71-128\n",
      "pair_fit /sh3_r1_120_r2_2//A/26-33/N, /SH3_2w0z//B/7-14/N, /sh3_r1_120_r2_2//B/97-154/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_120_r2_2_motif, /sh3_r1_120_r2_2//B/97-154\n",
      "pair_fit /sh3_r1_120_r2_36//A/26-33/N, /SH3_2w0z//B/7-14/N, /sh3_r1_120_r2_36//B/97-154/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_120_r2_36_motif, /sh3_r1_120_r2_36//B/97-154\n",
      "pair_fit /sh3_r1_146_r2_0//A/34-41/N, /SH3_2w0z//B/7-14/N, /sh3_r1_146_r2_0//B/65-122/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_146_r2_0_motif, /sh3_r1_146_r2_0//B/65-122\n",
      "pair_fit /sh3_r1_146_r2_11//A/34-41/N, /SH3_2w0z//B/7-14/N, /sh3_r1_146_r2_11//B/65-122/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_146_r2_11_motif, /sh3_r1_146_r2_11//B/65-122\n",
      "pair_fit /sh3_r1_146_r2_28//A/34-41/N, /SH3_2w0z//B/7-14/N, /sh3_r1_146_r2_28//B/65-122/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_146_r2_28_motif, /sh3_r1_146_r2_28//B/65-122\n",
      "pair_fit /sh3_r1_146_r2_40//A/34-41/N, /SH3_2w0z//B/7-14/N, /sh3_r1_146_r2_40//B/65-122/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_146_r2_40_motif, /sh3_r1_146_r2_40//B/65-122\n",
      "pair_fit /sh3_r1_146_r2_46//A/34-41/N, /SH3_2w0z//B/7-14/N, /sh3_r1_146_r2_46//B/65-122/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_146_r2_46_motif, /sh3_r1_146_r2_46//B/65-122\n",
      "pair_fit /sh3_r1_146_r2_5//A/34-41/N, /SH3_2w0z//B/7-14/N, /sh3_r1_146_r2_5//B/65-122/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_146_r2_5_motif, /sh3_r1_146_r2_5//B/65-122\n",
      "pair_fit /sh3_r1_192_r2_31//A/16-23/N, /SH3_2w0z//B/7-14/N, /sh3_r1_192_r2_31//B/65-122/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_192_r2_31_motif, /sh3_r1_192_r2_31//B/65-122\n",
      "pair_fit /sh3_r1_192_r2_49//A/16-23/N, /SH3_2w0z//B/7-14/N, /sh3_r1_192_r2_49//B/65-122/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_192_r2_49_motif, /sh3_r1_192_r2_49//B/65-122\n",
      "pair_fit /sh3_r1_22_r2_21//A/61-68/N, /SH3_2w0z//B/7-14/N, /sh3_r1_22_r2_21//B/97-154/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_22_r2_21_motif, /sh3_r1_22_r2_21//B/97-154\n",
      "pair_fit /sh3_r1_66_r2_16//A/48-55/N, /SH3_2w0z//B/7-14/N, /sh3_r1_66_r2_16//B/79-136/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_66_r2_16_motif, /sh3_r1_66_r2_16//B/79-136\n",
      "pair_fit /sh3_r1_66_r2_17//A/48-55/N, /SH3_2w0z//B/7-14/N, /sh3_r1_66_r2_17//B/79-136/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_66_r2_17_motif, /sh3_r1_66_r2_17//B/79-136\n",
      "pair_fit /sh3_r1_66_r2_18//A/48-55/N, /SH3_2w0z//B/7-14/N, /sh3_r1_66_r2_18//B/79-136/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_66_r2_18_motif, /sh3_r1_66_r2_18//B/79-136\n",
      "pair_fit /sh3_r1_66_r2_2//A/48-55/N, /SH3_2w0z//B/7-14/N, /sh3_r1_66_r2_2//B/79-136/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_66_r2_2_motif, /sh3_r1_66_r2_2//B/79-136\n",
      "pair_fit /sh3_r1_66_r2_3//A/48-55/N, /SH3_2w0z//B/7-14/N, /sh3_r1_66_r2_3//B/79-136/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_66_r2_3_motif, /sh3_r1_66_r2_3//B/79-136\n",
      "pair_fit /sh3_r1_66_r2_37//A/48-55/N, /SH3_2w0z//B/7-14/N, /sh3_r1_66_r2_37//B/79-136/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_66_r2_37_motif, /sh3_r1_66_r2_37//B/79-136\n",
      "pair_fit /sh3_r1_66_r2_4//A/48-55/N, /SH3_2w0z//B/7-14/N, /sh3_r1_66_r2_4//B/79-136/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_66_r2_4_motif, /sh3_r1_66_r2_4//B/79-136\n",
      "pair_fit /sh3_r1_66_r2_43//A/48-55/N, /SH3_2w0z//B/7-14/N, /sh3_r1_66_r2_43//B/79-136/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_66_r2_43_motif, /sh3_r1_66_r2_43//B/79-136\n",
      "pair_fit /sh3_r1_66_r2_47//A/48-55/N, /SH3_2w0z//B/7-14/N, /sh3_r1_66_r2_47//B/79-136/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_66_r2_47_motif, /sh3_r1_66_r2_47//B/79-136\n",
      "pair_fit /sh3_r1_69_r2_11//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_11//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_11_motif, /sh3_r1_69_r2_11//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_12//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_12//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_12_motif, /sh3_r1_69_r2_12//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_13//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_13//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_13_motif, /sh3_r1_69_r2_13//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_22//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_22//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_22_motif, /sh3_r1_69_r2_22//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_26//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_26//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_26_motif, /sh3_r1_69_r2_26//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_32//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_32//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_32_motif, /sh3_r1_69_r2_32//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_35//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_35//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_35_motif, /sh3_r1_69_r2_35//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_36//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_36//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_36_motif, /sh3_r1_69_r2_36//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_37//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_37//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_37_motif, /sh3_r1_69_r2_37//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_39//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_39//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_39_motif, /sh3_r1_69_r2_39//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_43//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_43//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_43_motif, /sh3_r1_69_r2_43//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_5//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_5//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_5_motif, /sh3_r1_69_r2_5//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_7//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_7//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_7_motif, /sh3_r1_69_r2_7//B/71-128\n",
      "pair_fit /sh3_r1_69_r2_9//A/19-26/N, /SH3_2w0z//B/7-14/N, /sh3_r1_69_r2_9//B/71-128/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_69_r2_9_motif, /sh3_r1_69_r2_9//B/71-128\n",
      "pair_fit /sh3_r1_78_r2_13//A/45-52/N, /SH3_2w0z//B/7-14/N, /sh3_r1_78_r2_13//B/96-153/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_78_r2_13_motif, /sh3_r1_78_r2_13//B/96-153\n",
      "pair_fit /sh3_r1_78_r2_18//A/45-52/N, /SH3_2w0z//B/7-14/N, /sh3_r1_78_r2_18//B/96-153/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_78_r2_18_motif, /sh3_r1_78_r2_18//B/96-153\n",
      "pair_fit /sh3_r1_78_r2_3//A/45-52/N, /SH3_2w0z//B/7-14/N, /sh3_r1_78_r2_3//B/96-153/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_78_r2_3_motif, /sh3_r1_78_r2_3//B/96-153\n",
      "pair_fit /sh3_r1_85_r2_38//A/50-57/N, /SH3_2w0z//B/7-14/N, /sh3_r1_85_r2_38//B/76-133/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_85_r2_38_motif, /sh3_r1_85_r2_38//B/76-133\n",
      "pair_fit /sh3_r1_87_r2_31//A/47-54/N, /SH3_2w0z//B/7-14/N, /sh3_r1_87_r2_31//B/86-143/N, /SH3_2w0z//A/1-58/N\n",
      "sele sh3_r1_87_r2_31_motif, /sh3_r1_87_r2_31//B/86-143\n"
     ]
    }
   ],
   "source": [
    "for fn in sorted(glob.glob('output/hits_sh3_r2/*trb')):\n",
    "    name = os.path.basename(fn).replace('.trb','')\n",
    "    trb = np.load(fn,allow_pickle=True)\n",
    "    \n",
    "    i_binder = [i for i,idx in enumerate(trb['con_ref_pdb_idx']) if idx[0]!='R']\n",
    "    i_receptor = [i for i,idx in enumerate(trb['con_ref_pdb_idx']) if idx[0]=='R']\n",
    "\n",
    "    idx_binder_ref = [idx for i,idx in enumerate(trb['con_ref_pdb_idx']) if i in i_binder]\n",
    "    idx_binder_hal = [idx for i,idx in enumerate(trb['con_hal_pdb_idx']) if i in i_binder]\n",
    "\n",
    "    ref_str = SampledMask.contract(idx_binder_ref)\n",
    "    hal_str = SampledMask.contract(idx_binder_hal)\n",
    "    binder_str = f'/{name}//{hal_str[0]}/{hal_str[1:]}/N, /SH3_2w0z//{ref_str[0]}/{ref_str[1:]}/N'\n",
    "    print(f'sele {name}_motif, /{name}//{hal_str[0]}/{hal_str[1:]}') # create selection for the hallucinated linear motif\n",
    "\n",
    "    idx_receptor_ref = [('A',idx[1]) for i,idx in enumerate(trb['con_ref_pdb_idx']) if i in i_receptor]\n",
    "    idx_receptor_hal = [('B',idx[1]) for idx in trb['receptor_pdb_idx']]\n",
    "\n",
    "    ref_str = SampledMask.contract(idx_receptor_ref)\n",
    "    hal_str = SampledMask.contract(idx_receptor_hal)\n",
    "    receptor_str = f'/{name}//{hal_str[0]}/{hal_str[1:]}/N, /SH3_2w0z//{ref_str[0]}/{ref_str[1:]}/N'\n",
    "\n",
    "    print(f'pair_fit {binder_str}, {receptor_str}') # superimpose hallucination on reference structure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "SE3 (Python 3.8.2)",
   "language": "python",
   "name": "se3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.2"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}